A Forseeable Crisis? Lessons from Subprime: The Sheep on the Hill
Author: Dr Jacques Pézier is a Visiting Professor at the ICMA Centre - the business school for financial markets - University of Reading
First Published: December 2007
If nothing else, the
subprime storm that is still rolling as we write illustrates the
importance of market sentiment. Market sentiment does not necessarily
mean irrational, emotional behaviour – although it sometimes does. What
is irrational in economics and finance is to ignore the personal views
and risk attitudes of the economic agents that drive the markets.
The subprime storm built up over years. Taking
advantage of low interest rates, engineered to lessen the aftermaths of
the previous TMC (Technology, Media & Communications) crisis,
lending institutions started to extend their facilities to people with
inadequate ability to repay in less favourable circumstances. These
loans were repackaged in attractively named investment vehicles and
sold to investors keen to secure a substantial interest margin in
excess of pallid base rates.
Borrowers started to spend beyond their means until
inflation crept in and central banks felt obliged to raise base rates.
Defaults and foreclosures ensued and investors in ‘high yield’,
‘enhanced leverage’ and such funds are now paying with their capital.
Money is not lost, only redistributed; yet economic dislocations may be
substantial.
A major consequence is that financial institutions
suddenly become wary of the quality of some assets they hold, and, by
extension, even more wary of the lesser-known quality of assets held by
their competitors. In these circumstances, extension of credit between
financial institutions is restricted and becomes more expensive.
We now have a liquidity crisis. Central banks ‘pump’
money in the system, but uncertainty and fear are narrowing the pipes
and interbank lending rates have surged above the base rates. The TED
spread – the difference between the three-month Eurodollar interbank
rate and the US Treasury bill rate, a barometer of credit risk spread –
has never been so high in more than twenty years. Institutions that
rely heavily on interbank loans are threatened. Panic could erupt.
Who or What is Responsible for the Liquidity Crisis?
How is it that eminent banking regulators,
supervisors and central bankers have not seen this crisis looming and
are finding it so difficult to control? How is it that sophisticated
minimum regulatory capital requirements designed to cover potential
losses at the 99.9% confidence level over a year have not signaled the
dangers? Armies of well-trained risk managers, equipped at great
expense with powerful computers fed by huge historical databases,
dutifully and continually carry out these calculations. How is it that
quantitative analysts, or ‘quants’, with PhDs in mathematics from the
best universities have not priced the risks correctly?
The blame game is in full swing. Some individuals
from chief executives of banks to central bank governors and Treasury
ministers have been accused personally of recklessness, negligence, or
simply bad judgment. Some have resigned. That is the honourable thing
to do in business and politics; but does it solve any problems? Have
they learnt anything more than the importance of covering their own
backsides? Have we purged the system?
Or could it be that what is more fundamentally at
fault are the very quantitative methods that contribute to the success
of modern finance? Could modern finance contain the seeds of its own
destruction? It is facile to either dismiss this opinion instantly
because modern finance is based on rational theories, or to embrace it
wholeheartedly because people do not always appear to act rationally.
These absolutes dispense from further thinking. Reality may be more
complex.
The pillars of modern finance – portfolio theory,
market equilibrium models, the pricing of derivative products – were
erected more than a generation ago. The edifice grew under the twin
influences of financial deregulation and globalization. It has been
equipped with powerful computers and the best information and
communication systems. Since the mid-eighties, traditional bankers have
had to make room for a new breed of financiers with diplomas in
mathematics, physics and engineering rather than Latin or medieval
history. Surprisingly, many are French, perhaps a conspiracy?
The Objective but irrational View of Risk Management
Central to the activities of financiers are the
concepts of risks and returns. Risks are continually assessed, priced
and transferred. Modern economies benefit enormously from this ‘risk
market’. Banking regulators and supervisors focus mainly on risks. They
set minimum capital requirements for individual financial firms
commensurate with the risks these firms take. But regulators must be
seen to act fairly and objectively. Therefore the assessment of risks
has to be codified, supported by historical data, limited to areas were
quantification can be objectively validated (market, credit and
operational risks), and aggregated into standardized ‘worst case’
measures according to simplistic rules. No French conspiracy here,
rather good old-fashioned British empiricism with a tad of conservatism.
Chris Matten’s useful little book on Capital
Management in Banking has a telling diagram. It shows credit ratings
and capital relative to regulatory capital requirements for a sample of
78 banks. There is no visible pattern, no correlation between the two.
Ask a banking analyst whether excess capital over minimum regulatory
requirement is important for credit rating, he will say yes,
undoubtedly, but so are many other factors, among which the ability to
generate regular profits over the long term.
The objective view of supervisors is shortsighted.
It leaves unexplored large areas of business risks, reputational risks
and systemic risks. Systemic risk is the danger that a large number of
financial firms might be exposed simultaneously to the same risk
factors and that failures could be contagious; a liquidity crisis is a
type of systemic risk. These risks are not overlooked because they are
less important, but simply because they are more difficult to quantify
objectively and manage. For example, regulators are aware that the
pro-cyclical effects of capital requirements could increase systemic
risks, but they do not know how to counter this effect, except by
relaxing the rules in exceptional circumstances; they also have to
rethink the pros and cons of public transparency versus covert
intervention.
Quants, traders and financial engineers can hardly
be blamed unless they engage in illegal activities. It is in the nature
of competitive capitalistic economies that their incentives are to make
profits for themselves by making profits for their firms. When they act
as intermediaries, they must design products that look attractive to
both borrowers and investors. The onus for understanding the products
rests with the clients; the reputation of the firm is at risk if
craftiness leads to blatant deception. When they trade, they use, as
they should, all the intelligence, analytical tools and computational
resources they can marshal to design and implement profitable trading
strategies. Mistakes do happen by accident, but the stability of the
financial markets is not a concern for traders. And, by and large,
financial market crises have not become more frequent or deeper with
the introduction of derivatives and more efficient trading mechanisms.
The Rational and Subjective View of Risk Manaagement
So who is to blame and how could one improve risk
management in financial firms as well as ensure better market
stability? First one should recognize that, in an uncertain world (and
the financial markets are by construction quite random), good results
do not equate with good decisions; they are mostly due to chance.
Therefore, rewarding or penalizing people mostly on the basis of their
results rather than on the quality of their decisions is a management
failure. It does not improve the likelihood of better results in
future. On the contrary it creates perverse effects such as herd
instincts or excessive risk taking, when the situation is already poor.
But how does one recognize good decisions? That is the challenge. It comes down to three things:
- Creativity: the quality of the alternatives
- Preferences: the clarity and adequacy of the objective
- Knowledge: the quality of the information and assumptions (models)
Creativity is fostered by
unfettered communications among people with a variety of backgrounds.
Preferences, in a financial context, are essentially about trade-offs
between risks and returns. Yet trade-offs are often poorly articulated;
many financial firms still prefer to set limits to risks rather than to
price them. Finally, knowledge consists of views about uncertain
quantities (e.g., future prices) as well as models for relating such
quantities to the outcomes of a decision. Now, there is little wrong
with financial models; they are the best approximations we have for
specific problems; if they need be improved, they will. But there is
something wrong about taking into account only objective empirical data
and neglecting the more subjective but no less important information.
Risks are about the future, not the past. Behaviourists would have us
throw away formal models, but we would be left with nothing to organize
our thoughts. Rationalists prefer to use logical models with subjective
information as may be relevant. At least, this is what we try to teach
our students so that they do not end up as the fabled British
accountant:
A young British accountant visits his uncle, a sheep
farmer in New Zealand. “So, I hear you are an accountant now, my boy;
very useful” says the uncle “Perhaps you could tell me how many sheep I
have?” “12,123” comes the reply after a minute. “How did you get that?”
exclaims the uncle. “Simple”, says the accountant, “I just counted 123
sheep in the pen near the farm, and I guess you have about 12,000 on
the hills.”
For more information on:
ICMA Centre - University of Reading
