During the last 10 years, the reduction and control of loan losses and chargeoffs has become a major managerial challenge for financial institutions.
The root of the problem has been failure to adapt the credit review process to a wide variety of analytical challenges.
How many times at loan committee meetings have you or someone you know been asked a question you couldn't answer fully, because information access was inadequate?
Maybe the problem was with physical access, such as someone else having portions of the credit file. Maybe some legwork analysts usually do to fill information gaps was lacking.
And maybe the right kind of strategic information is not being gathered.
Research by my associate Hassell H. McClellan of Boston College's Wallace E. Carroll School of Management has shown that application of strategic analysis techniques to credit analysis would improve the decisions made.
In his book, "Managing One-Bank Holding Companies," Prof. McClellan combined a review of bank managementt and lending practices with an examination of the loan-loss patterns of a sample of banking institutions.
A key conclusion was that loan analysis based primarily on traditional financial and historical credit information gives short shrift to strategic factors that identify borrowers whose operations will remain credit-worthy over the life of a loan.
Prof. McClellan suggests that major institutions can lend more efficiently by integrating strategic analysis better into the credit approval process and expanding the information basis upon which loans are extended.
For several reasons, it is hard to implement strategic analysis techniques in the loan review analysis process.
* Traditional lending practices tend to be plagued by "analysis myopia," relying primarily on historical financial information instead of forward-looking data.
Historical information does not identify the projective factors of change that contribute to flaws in initial loan decisions and undetected erosion in outstanding loans.
* Many banks have tried to capitalize on learning curves in analysis by creating lending groups that specialize in particular industries.
A better idea would to specialize the analysis function, creating a strategic analysis unit on which individual lending groups could rely.
* If the lending process becomes overly routine, economies of scale in credit analysis can have diminishing marginal returns.
New loans tend to receive the most scrutiny; ongoing review is exemplified by monitoring the loan covenants, not by evaluating threats to the success of the loan by unforeseen factors.
Many Data Sources
Computer systems are available today that can provide timely, reliable information and that let the user perform a series of structured analyses in minutes.
A wealth of business information is available via data bases residing within the institution as well as from external data sources. Compustat, Lexis/Nexis, J.D. Power & Associates, Robert Morris Associates, Dun & Bradstreet, and Standard & Poor's are some of the more commonly known sources.
But many more sources of strategic and competitive information can play a critical role in decision-making.
Data retrieval can now take the form of looking at an electronic copy available via a personal computer screen. This will eventually cut down on the mountains of back-room paper files and provide more efficient access to this type of information.
Optical character recognition technology can now translate "hard copy" into computer-readable form. As a result, credit files of the future will be much slimmer.
Pulling It All Together
Software systems that banks use for loan analysis focus on financial analysis; lacking automated data retrieval, they give limited attention to industry and competitive analysis.
My firm and AI Corp. in Waltham, Mass., are jointly developing an expert system to integrate strategic analysis with automated data retrieval. The technology can use a bank's mainframe to access external data bases with a wide variety of formats.
Our Strategic Loan Analysis product might access such data about the borrower as:
* Competitive position in the industry and ability to move to better strategic groups.
* Cost advantages.
* Product innovation capabilities.
* Track record in implementing business strategies
* Significance of research and development.
* Appropriateness of current and foreseeable strategies.
The system could also find information on:
* Driving forces in the industry and their implications for the borrower's future financial and strategic position.
* Threats of new entrants and substitute products.
* Potential adverse demographic changes.
* Changes in buyer and customer power and needs.
Often such information must be compiled from sources that banks rarely used for credit analysis, including industry and investment reports, industry journals and special reports, and data bases used for venture capital and forecasting purposes.
Science Enhances an Art
With pretty straightforward language and a relatively simple set of instructions, a product like ours would enable loan officers or analysts to generate reports they have designed themselves.
In addition, an automated follow-up system can be easily integrated to watch problem credits.
Also, such a system could help address a problem that regulators cite at many banks: inconsistency in the enforcement of loan policy.
Though a new perspective on analysis will mitigate some loan quality problems, a system that automates credit review and applies policy consistently will go further, bringing science to the art of commercial lending.