Steering Portfolios on a Steady Course: Tools and Techniques for Tougher Times
Gerry Gunn; Signals, Autumn 2001
Forecasting of losses is only the tip of the iceberg...
managers can expand and contract staff and facilities, change product
and offerings, revise service standards—and generally do a better
job for the customer and the institution. Managing credit portfolios is often described—in something of an
industry-wide confession—as analogous to driving by looking in the
rear-view mirror. This provides a vivid picture to most of us.
We use historical data that comes from past events, actions and decisions
to determine what will happen going forward. In some times and places
this has worked very well and our portfolio performance (and therefore
our performance) has lived up to, or exceeded expectations. The outcome
may have been influenced by excellent analysis, or luck, or shaped by
events completely out of our control—but overall it worked... most
of the time. Along the way there were portfolio "accidents". Some
were small—others were
large enough to get noticed. The size of the problem was usually
directly related to who noticed. If very small, it stayed within
the business and was managed on the local P&L. If bigger, it became
a topic for executive management.
Larger problems drew the interest of outside auditors, the board of directors,
regulators and finally, if large enough, investors. When share price is
affected everyone is involved. The most difficult questions to answer
are usually preceded by a generic "How could this possibly have
happened to us?!"
Some portfolio problems have been high profile, and the involved firms
suffered fatal or near fatal blows as a result. The fatalities often resulted
from yet another driving analogy—the portfolio manager's ability
to exceed the speed limit and still be lost. Knowing the right direction
to travel is as critical as moving swiftly. Speed and nimbleness has great
value to increase market share, generate increasing profits, grow the
business and reward investors, but the market today also tells us that
earnings growth is valued over speed every time with little tolerance
for surprises.
The industry has moved at an increasingly fast pace over the last decade.
Time to market on new products and variations on old ones is a fraction
of what it was ten years ago. The residential real estate market provides
a good example. The 30-year fixed rate mortgage remains the staple of
the industry but dozens of rate, index, term and qualification variations
have sprung up. The need to enter new geographic and demographic markets
has put great pressure on the Credit/Risk functions to combine with Marketing
functions to solve the puzzles of not only "who will repay us?" but,
as importantly, "who will use the product and generate a fair return
for the institution?" The development and use of credit scoring
as a tool for assessing risk has migrated further to determination of
pricing and down payment criteria for the borrower and portfolio quality
for the investor. The result–more people than ever are being granted
credit and our portfolio profiles have the potential to change along with
the mix of borrowers.
Rapid growth, consolidations of financial service companies and an uncertain
economic future have combined to create very exciting times. We may find
that the portfolio now on the books does not behave as past portfolios
did. Surprises may lurk in how the book of business was originated, or
how it was managed, or even in factors external to the portfolio or the
institution. Yet commonly deployed techniques and methods assume that
how our customers will behave in the future is based on how they have
behaved in the past. If the mix of customers is different this time, or
if the external situation is different, or both, then our rearview mirror
approach may not be very accurate.
The need for more accurate and robust forecasting techniques has been
recognized for years. In more stable times, one of the attractive features
of consumer portfolios was that while they would always have losses, those
losses could be reliably anticipated and planned for by the business.
But as the environment changes, the ability to project losses may be affected.
The consistency that is attractive to management and investors goes away.
Clearly some of the "accidents" referred to earlier were caused
by things not going as expected or as promised. But forecasting of losses
is only the tip of the iceberg. Knowing in what direction they are headed,
managers can expand and contract staff and facilities, change product
and offerings, revise service standards and commitments to do a better
job for the customer and the institution. Regulators have also taken a
much greater interest in where a given book of business is going and how
that book might be affected by various events along the way. The ability
to "stress test"—long recognized as "nice to have"—becomes
essential as events and changes come ever faster.
Can we continue to use the "rearview mirror" to navigate?
No, we cannot. In the past ten years there have been three episodes of
massive refinancing by borrowers in the residential mortgage business.
Because the lifecycle of a mortgage is now closer to five years, few have
gone through the normal maturation process. While most are probably good
risks, the refinance boom has clouded the use of historical data. A similar
case may be made in the card business with the aggressive use of balance
transfer offers. The normal behavior patterns may have been disrupted
as a result. Additionally the rapid movement of people among firms across
the industry and the relatively long run of economic prosperity has potentially
reduced corporate memory and perhaps the reflexes to avoid future problems.
Today there is a need to get much better at understanding our portfolios.We
can't rely on history or growth or luck to keep us from experiencing
problems. The use of data and analysis to promote growth has been very
successful over the last decade. Some firms have made great strides using
data to manage the portfolios booked, but as an industry it is not evident
that portfolio management abilities have kept pace with our ability to
grow. Advanced analytics will continue to evolve to serve this need, improving
substantially on the rear-view mirror, and enabling forward-viewing portfolio
management practices to meet today's challenges and reality. Such
techniques are long overdue. Practitioners will benefit from enhanced
abilities to understand and manage portfolios in good times and bad. Those
abilities may prove, in the long run, to be the most critical skills of
all.
Gerry Gunn retired in mid-2000 after a 36-year career
in the financial services industry. Gerry's primary focus has been credit
risk management associated with credit card, consumer, real estate and
small business portfolios. He also has led the melding of acquired portfolios—a
key player in three acquisitions that were the largest mergers of consumer
portfolios for their time and has worked with regulators, investors,
analysts and industry leaders on issues of importance to banks and the
industry in general. He has been a member of CBA, RMA, BAI, ABA and California
Banker's Association committees related to Retail Banking and
Consumer Lending. He and his wife, Carole, reside in Tiburon, CA and spend
their leisure time traveling, sailing, volunteering and watching their
grandchildren grow.
|
Top of page