by
GEORGE E. SNIVELY
Manager - Credit & Collection
Finance Section
Computer Department
General Electric Company
Phoenix, Arizona

c. February 1963, by Credit Research Foundation Inc.

All Rights Reserved

The Credit Research Foundation is pleased to acknowledge the accomplishment of George E. Snively, the author of this report. His study presents some aspects of computer applications which will become increasingly important to the operation of credit departments as electronic data processing becomes more common in the business world.

The study was prepared in partial fulfillment of the requirements for graduation from the Graduate School of Credit and Financial Management sponsored by the Credit Research Foundation of the National Association of Credit Management at the Amos

Tuck School of Business Administration, Dartmouth College.

Mr. Snively is Manager, Credit and Collection, Finance Section, Computer Department, General Electric Company, Phoenix, Arizona. He graduated from the Graduate School of Credit and Financial Management, Class of 1962.

INTRODUCTION                                                          Page

CHAPTER I    -CLASSES OF BUSINESS PROBLEMS                     1

CHAPTER II   -EQUIPMENT AVAILABLE OR IN DEVELOPMENT        4

CHAPTER III  -RECEIVABLES MANAGEMENT                              7

CHAPTER IV  -CONCLUSIONS AND RECOMMENDATIONS             13

INTRODUCTION

Chapter I

Business problems are being re-examined in the light of the benefits to be derived from the use of computers or computer-like devices. These problems ca be categorized as follows:

1. Data Processing

2. Simulation

3. Information Storage and Retrieval 4. Information Communication

The listing is in the order in Which computers have been used to attack these problems. It is also the general direction in which computing hardware has been developed. However, this is not necessarily the order of importance of these

problems. It can be argued, in fact, that this listing is in the reverse order of importance. Equipment presently available or being developed can be used in the solution of all four classes of problems, and significant breakthroughs are constantly being made.

Data Processing

Problems involved in processing and recording masses of data are in this category. The preparation of a payroll, payment of bills, issuing of invoices, and inventory record keeping are all examples.

Because large numbers of clerical workers are required for the manual processing of data, these problems presented early opportunities for computers to justify themselves through savings in clerical costs. As a result, computers have been used extensively in this type of application. Computers are the solution of some data processing problems where they process more data than is possible by any other means and still provide timely information.

Data processing applications are also being developed in the direction of collecting data. The equipment used for these applications is generally referred to as production control equipment and falls into two classifications; automatic monitoring and data accumulation systems.

The monitor keeps track of a production machine's running time, down time, and quantity made, and brings individual counts of these variables for all production machines to a central control location. Accumulation systems convert data inserted by a machine operator close to the production tool into a format that can be processed further by electronic computers.

The main job of the monitors has been to reduce the down time. One user claimed to have paid for an automatic monitoring system in a month from savings in reduced down time. Data accumulators have been applied to many functions that are part of the overall job of production control. They emphasize attendance reporting, labor distribution, order (or job) status and location.

-1-

Simulation is the process of substituting a more easily observable event or model for an actual event that is not as easily measurable. This is done in order to understand the event or to predict the future.

For example, scale models of airplanes are built and tested in wind tunnels before large sums are expended in constructing a full-size prototype. The wind tunnel and model" simulate" the operating conditions to be encountered by the plane in actual operation.

A breakeven chart "simulates" the expected operations of a business enterprise under varying costs and volume. A credit man is "simulating" a busines s when he projects the financial statements to determine if funds will be available to pay for merchandise.

The popular practice of playing business games on a computer is an example of business problems being "simulated." Player "executives" face such management problems as sales forecasting, inventory management and profit optimization.

Simulation problems present management with more than interesting games. The challenges and opportunities created by the introduction of engineering techniques into the areas of business and economic research have been whetting the interest of management and university faculty members for a number of years. Some educators are predicting that the merging of the capabilities of the new high speed computers with the current level of simulation methodology will create the foundation of a new" scientific institutional economics". Also, as a result of the growth of operations research and management science, we can expect data organization and information processes to be revised drastically in the years ahead.

Information Storage and Retrieval

The problems in storing and retrieving technical data have become so important that major efforts in time and money are being allocated by computer manufacturers, government agencies, and libraries to solutions. The credit function will be a direct beneficiary of this work.

Scientists, businessmen and engineers cannot keep up with the mass of published information. They need a fast way to lay hands on specific subjects. A practical method of assimilating material, classifying it, and compressing it into an index is essential for research. With such a facility accessible to him, the researcher can focus on literature pertinent to his general problem, then hunt up the articles that sound promising. This can spare him months or years of duplicating research that someone else may have done.

Information Communication

-2-

Business machines will have access to communication networks through lightning fast teleprinters or orbiting transmitters in the sky to exchange data over thousands of miles without human intervention.

Selective dissemination offers to free the executive from studying much nonessential information to determine that it is not useful to him. Executives who are interested in particular topics or have a need to know will automatically receive information and reports in those fields and will not receive material which is not pertinent.

-3-

Chapter II

EQUIPMENT AVAILABLE OR IN DEVELOPMENT

Exciting equipment is already available, is in development, or is being planned for applying electronic solutions to business problems. Equipment will be discussed as it relates to the classes of business problems previously defined. Since the same equipment may be used in the solution of each class of problems, only those with unique applications to the respective problems will be mentioned. Computers or central processors will not be discussed as equipment already available is limited only by the ingenuity of the user. The input-output devices will be of most interest in developing a receivables management system.

Data Processing

In order to process data, it must first be put into a form that can be "read" by the computer; holes punched in cards; magnetic printing on a check; magnetic impulses on magnetic tape; or holes punched in paper tape. Today many of these input media must be manually prepared. However, advances are being made to eliminate this process through devices that will read the printed word.

The devices read pages of printed characters. Work is also underway on devices that will read handwriting by recognizing patterns in much the same manner as the human brain does.

In this field as in many others, technological competition from other countries is speeding developments.

Listening Devices. An experimental machine, that does arithmetic on voice command, has been developed by one computer manufacturer. The device recognizes up to 16 words spoken into a microphone, including 10 digits. When the words, "plus", "minus", and "total" are spoken, it instructs an adding machine to calculate and print out correct answers to simple arithmetic problems. There is no present plan to manufacture such machines, although they may some day provide access to information systems for people who are too busy to prepare information for computers.

-4-

With one such system, a draft of any book or document can be typed on a paper tape punching typewriter which produces typewritten pages and, at the same time, punches the text in code on a paper tape. The typewritten draft then can be proof read and editing instructions typed out with a correction tape produced at the same time.

Working on the information from the two tapes, a computer performs all deletions, insertions and other editorial changes at high speed and itself turns out a master tape of a completely correct text, which is then used to produce a finished positive on film or paper, ready for reproduction by offset or letter press printing.

Simulation

Generally most computers used in data processing applications can also be programmed to solve simulation problems. However, many simulations require equipment especially designed for the problem at hand.

Not surprisingly, computer systems are being designed which simulate computer systems. A new type of computer program that reduces time and effort required to simulate information handling systems before they are built is being developed. The new method makes simulation a routine technique that can produce results in a few days instead of requiring months of special programming efforts. The first general purpose simulation program for systems has already been used experimentally to analyze and evaluate a variety of system designs.

Information Storage and Retrieval

Still another system is used to locate information at ultrahigh speed, searching through 50,000 technical articles an hour. Its use marks a big step toward mechanizing the laborious and time consuming job of searching for specific information in a growing mas s of technical literature.

Information Communication

This is among the latest problems to be attacked by computer manufacturers, but development efforts have already made many systems available.

An order and supply system, said to be the largest application yet announced, has been placed in operation by one company to link its service center with many

of its distributors. The heart of the network is data transmission equipment

-5-

A communications system that can move information between distant points more than 1, 000 times as fast as the human voice has also been demonstrated. Computers in a data processing center in New York City and a point 68 miles distant were linked by a method which permits an operator at one point to operate a computer at the other. By this means a large organization may distribute jobs among its several computer facilities without regard to physical distance.

Factory Monitoring

The ultimate of a completely integrated control system would feature a centralized "dynamic management" staff room where total information would flash on screens for top executive decisions. Alternate courses of action, based on precise and timely information, could be pre -tested before committing manpower and materials to projects under management consideration.

Information Processing Centers

This survey of equipment would not be complete without noting that it is not necessary for a businessman with a problem to acquire equipment of his own. He may well be able to find the problem solving help he needs at an information processing center. The high cost of electronic data processing equipment for lightly loaded systems has given rise to a new industry - the renting of computer time. Computer service centers are operating in more than 200 locations around the country.

Once the computer leaders were boasting that nothing was too big for a computer. Now, the computer centers are saying nothing is too small.

-6-

Chapter III  Receivables Management 

A broad system of men, machines, and technics that might be employed in a future Receivables Management system can now be outlined. Only the barest outline of such a system will be attempted, as credit managers have yet to "think through" the requirements for performing their expanded role. Much hard work will be required to determine what it is that is wanted and needed out of such a system. Then, once the system requirements have been determined, much more work will be necessary to design the system.

The increasing importance of the concentration of current assets in receivables will necessitate intensive work on the problem. It remains to be seen whether credit managers will rise to the challenge to design and manage the system or permit some other function of the business to do so.

Shift of Industrial Assets

During the past few years, accounts receivable have become an increasingly large proportion of the assets of manufacturing concerns. At the same time relative inventories have been declining. What factors have been operating to cause this shift of industrial assets from inventories to receivables? Have credit men allowed their receivables to get out of hand?

Inventory Management

One concept of inventory management recognizes that inventory is not controlled from the receiving dock to the shipping dock but is controlled from the supplier's receiving dock to the customer's receiving dock. It focuses the attention of businessmen on the larger problem which can be solved with the aid of computers, often with spectacular results. The businessman may be able to nearly eliminate standing inventory, reduce the need for warehouses, yet provide better service to customers by better scheduling of production.

The importance of this development to the credit man is twofold. While he has not lost control of his receivables, he can expect increasing management attention to receivables as they continue to represent a larger percentage of business assets. He should also broaden his concept of his own function, and recognize

the existence of new tools which make it possible to solve the problems introduced by the broadened concept.

-7-

Credit decisions should take into account:

1. The funds already committed to a line of products; 

2. The marketing goals of the business;

4. The cost/volume relationships of the business; and

5. The capacity at which the factory is operating.

Receivables Management is a concept which embraces all these considerations, not only on the part of the credit man' s own business, but, to the extent possible, also his customers' business.

It should not be assumed that the use of a broader concept and the use of new tools in applying the concept will reduce the size of industrial receivables and credit losses. Indeed, it m1.ay well be that larger commitments of resources to receivables and larger losses will contribute to greater profits for the enterprise. If a "receivable" is traced back to its source it will be found that it originated in market research. At that stage, it was established that a product at a certain price and terms would satisfy the needs of a large enough group of customers to make it profitable to produce the product. Our receivables management system must therefore start at this point. It can aid market research efforts in establishing price, terms, classes of customers, anticipated lock-up of funds in receivables, credit losses, and sales volume. However, the portion of the system involved in credit decisions will be examined first.

Information Collection

-8-

Upon receiving this information, the machine will sort and file it by customer, by subject, and by chronology, to be available upon request. In addition to this information, the device will be linked through communications channels with sales offices and warehouses to collect information on new orders, status of shipments, and other pertinent data.

Analyzing Stored Information

The second step in a credit decision is to analyze the information that has been collected. Here is where credit managers must make a major contribution

to the system requirements. The credit manager, if he chooses, can sit back and watch the many other people interested in information storage and retrieval do the work necessary to satisfy the information needs of their departments. If he does that, however, the system will not work for him.

How does a credit man use the available information to make a credit decision? Books and articles have been written on analyzing financial statements. Ratios are calculated, compared with prior years, and projected into the future in rather precise ways. After this is all done, a "credit limit" of a definite dollar amount is established and the account is classified as prime, marginal, or submarginal. The writer has never seen demonstrated a rational procedure to arrive at the amount of a credit extension from all the analyses. Invariably the process of jumping from the analysis to the credit is called an exercise in judgment.

In many cases judgment needs to be exercised as a result of the insufficient development of analytical methods for bridging the gap between statement analysis and the amount of credit. It is hoped that attempts will be made to develop such methods for then the computer can analyze the data it has stored on an account, calculate all the ratios, compare the ratios with industry standards, calculate source and application of funds, forecast the financial status of the account at the time payment is anticipated, and calculate the amount of credit.

It is possible for the computer to learn how to set limits by analyzing the information present in a number of "judgments" made by a credit manager and then making the same judgments under similar circumstances. The problem with this procedure, of course, is to find someone whose judgment is sound enough in all situations. Only through analytical means can conclusions be reached which can be consistently revised in the light of better techniques.

At this point an approach to the problem might be suggested. Presumably, the system has stored all pertinent information relating to an account, including several prior financial statements. Also stored is information relating to estimated future requirements, and several trade clearances. The computer is now put to work forecasting the financial status of the account at the time in the future when payment is expected. This can be done in several ways:

-9-

A forecast of the financial condition of the account could be made by assimilating various pieces of in-file information such as newspaper clippings forecasting future sales volume, SEC registrations of additional equity, or agency reports of additional borrowings. The computer program would assemble these bits of information in their proper place in a forecasted financial statement.

Projection of Prior-period Trends

Projection Based on Industry Ratios

This method could be used to simulate future statements of an account by applying standard ratios of its industry to the company's figures, then adjusting them for the account's historical deviation from those figures.

Determining Credit lines and Classifying Accounts

Now that a projection is available on an account, the computer is requested to predict how the company will pay its bills by relating the projected financial ratios to present or past conditions. Did trade clearances show the same promptness or slowness when conditions similar to these existed in the past? Can payment be expected to be prompt or slow 30, or 60, or 90 days?

Once the process has gone this far, the problem can be solved by application of policy. If the seller's margins are small and net working capital limited the policy might be to classify any account as a marginal account where payments are expected to be 30 days slow.

The account has now been classified, but the amount of credit has yet to be established. This can be approached by referring to the "intelligence" information stored from market research studies and salesmen's reports. Information regarding the per cent or quantity of this account's volume supplied by the seller will be retrieved. The system will again fall back on policy regarding the desired share of this account's business, based on the expected promptness of payment. The policy might be to supply requirements if payment is anticipated within terms, seventy-five per cent of requirements if 30 days slow and twenty-five per cent of requirements if 60 days slow.

Anticipated losses

Anticipated losses can be forecast by a computer analysis of losses from previous decisions as to amounts and promptness of payment in relation to the projected statements. The system has now established both the classification of the account and the amount of credit involved. Judgment in the form of "policy" has been interjected in the system at this point but the system will also provide information to aid in policy formulation.

-10

The receivables management system begins with market research where credit analyses, as outlined above, are performed on selected accounts in the market survey. These analyses will determine:

1. If credit policy will permit the volume forecast with these customers; 

2. The extent of the lock-up in receivables;

3. Anticipated losses.

At this point the total business planning effort comes into play and the receivables forecast is balanced against the price, volume, cost relationships of the product, engineering and facilities risks, working capital restraints, and cash flow. The receivables management system has created an important input to this planning process and will produce an important output. The final plan will provide the credit policy to be followed which permits the accomplishment of the other objectives (volume, income, investment) of the plan.

Collecting Information in Advance of Orders

The next step in the receivables management program is to begin gathering information sufficiently in advance of receipt of an order. Prospects might be statistically selected by market re search studies and the information storage device instructed to begin storing information about these accounts. Salesmen's call reports may then be stored and analyzed to determine the number of accounts which, called upon for the first time, ultimately become customers and how soon after the first call.

Statistical methods can be used to identify prospects which will need additional information before they can be considered desirable credit risks. When the first order is received (fed directly into the computer from the sales office) sufficient information should already be available for analysis by the computer with the exception of financial statements direct from the customer. The system, however, will have already analyzed the information in its file for reliability and, if the information is not consistent or sufficient, will have printed out a letter requesting financial statements.

Credit Approval

If, for instance, volume is not up to expectations and facilities which were planned on those expectations are idle, the system would act to increase volume (assuming it is a primary objective) by accepting larger risks with more marginal accounts. Conversely, if volume expectations are being exceeded, limits might be reduced on marginal accounts to provide funds for increased capacity from receivables lock-up.

-11-

Receivables Accounting

The next link in the system is the familiar data proces sing function of updating the accounts for shipments, crediting payments, preparing statements, and automatically initiating follow-up letters, and telegrams.

Storing Results

The system is not finished after successful collection is made. It is hard at work comparing the actual payment experience with what was forecast and adjusting the forecasting techniques in view of the experience. It is also storing the profitability of the orders from the account for reference in establishing credit limits on future orders.

Customer Counseling

The system must also provide assistance in customer counseling. Composite statements could be readily prepared from the individual figures of accounts in the same line of business. Their composite figures could then be made available to customers for use as yardsticks against which they could measure their own performance.

-12-

Chapter IV

CONCLUSIONS AND RECOMMENDATIONS

In the near future the role of the credit manager will either be drastically reduced or dramatically increased. It depends on his ability to broaden his viewpoint and imagination in enlarging the concept of his job. He will soon have the necessary tools to do the larger job if he is ready to use them. Certainly the credit manager who is the bookkeeper of the receivables ledger will rapidly fade from the scene as will the credit manager who says "there is no substitute for judgment, " and while saying so, discourages efforts to develop aids to judgment.

To realize his full potentialities the credit manager must begin to "think through" the broader concept of his responsibilities, and develop receivables management consciousness. He must understand the systems concepts in other areas of his business so that his system will tie into them. In many instances the credit manager will find that other functions in his business are already doing considerable work on expanded systems concepts to fulfill their responsibilities. These are things that can be done only if he makes a conscious effort to delegate day to day activities and to devote the gained time in planning a receivables management system.

The credit manager must also write. Proposals, no matter how crude, should be written and circulated among credit managers. One idea sparks another and no credit manager has the ability or requisite knowledge to design a complete receivables management system by himself. Credit managers lack a purely technical journal as a forum for proposing new analytical methods or systems. Occasionally a paper, usually an important contribution, will appear in bulletins published by associations, but that is not enough.

Computer equipment can be designed to do so many different jobs that product planners, engineers, and programmers find it extremely difficult to determine exactly which jobs a new computer should be designed to do most efficiently. Design compromises must be made which may permit one job to be done more efficiently than others. As a result, many of the more interested technical societies carry on what are essentially lobbying activities to make sure that the manufacturer considers their needs and requirements in future designs. Credit managers do not appear to be engaging in these activities, as they have yet to define their needs, and their needs may not be given consideration in future designs.

In order to take full advantage of the new tools, credit managers must "think through" their systems problems and develop an organization to:

1. Provide a clearing house for contributions by various credit managers; 

2. Determine equipment capabilities required in the system;

3. Make their equipment requirements known to manufacturers.

-13

Bibliography on Equipment Leasing

Cash Flow Projection - A Tool of Credit Management

Credit Executives' Outlook

Credit Management and the Strong Vocational Interest Test

Credit Management Handbook

Credit Orientation and Training for Salesmen

Current Trends in the Practice of Charging Interest on Past Due Accounts

Customers - Your Most Important Asset

(Summary of Workshop Discussion on Customer Counseling)

Financing Industrial Equipment Leases

Guide for the Establishment of a Credit Department

Methods of Calculating Reserves for Bad Debts

Punched Card Accounting and the Credit Department (1962 Edition)

Punched Card Systems in Credit Department & Accounting Operations

In Progress For The Future

Computer Applications for Credit and Collections

A Study on Bad Debt Analysis

Credit Department Expenses

Credit Department Training Programs

A Study on the Value of Credit Management to Business

-14-

CREDIT RESEARCH FOUNDATION BOARD OF TRUSTEES

Ralph E. Brown, Vice President, Marsh & McLennan, Inc., St. Louis, Missouri

Kenneth E. Campbell, Vice President, Bank of America NT & SA, ,san Francisco, California James H. Donovan, Assistant Treasurer, Jones & Laughlin Steel Corporation, Pittsburgh, Pennsyl vania

Loretta Fischer, Secretary-Treasurer, George Ziegler Company, Milwaukee, Wisconsin Fred J. Flom, Detroit Edison Company, Detroit, Michigan

Tull N. Gearreald, Vice President and Treasurer, F. S. Royster Guano Company, Norfolk, Virginia

L. B. Houghton, Treasurer, Union Oil Company of California, Los Angeles, California Harold J. Kneuker, Assistant Treasurer, American Machine & Foundry Company, New York, New York

Lloyd Sinnickson, General Credit Manager, American Cyanamid Company, Wayne, New Jersey Gilbert W. Sites, Credit Manager, The Times-Mirror Company, Los Angeles, California

Irwin Stumborg, Assistant Treasurer, Baldwin Piano Company, Cincinnati, Ohio

Paul J. Viall, Treasurer, Chattanooga Medicine Company, Chattanooga, Tennessee

J. Allen Walker, General Credit Manager, Standard Oil Company of California, San Francisco, California

James F. Welsh, General Manager, McCormick & Company, Inc., Baltimore, Maryland

L. B. Wilson, General Credit Manager, Union Carbide Corporation, New York, New York

OFFICERS

President Irwin Stumborg

Executive Vice President Alan S. Jeffrey

Vice Presidents

J. Allen Walker, Education Lloyd Sinnickson, Research

Leroy B. Houghton, Promotion James H. Donovan, Finance

Treasurer William J. McDonald

Assistant Treasurer R. M. Woodruff

Secretary Robert L. Roper

STAFF

Managing Director Peter C. Peasley, Ph. D.

Associate Director George N. Christie

As sistant Director s of Re search Lionel L. Wallenrod

Elmer T. Sivertsen

COMMITTEE ON RESEARCH

Chairman: Lloyd Sinnickson, General Credit Manager, American Cyanamid Company, Wayne, New Jersey

Vice Chairman: R. L. Carlton, As sistant Treasurer and General Credit Manager, Continental Can Company, New York, New York

George T. Brian, Jr., Vice President, Noxzema Chemical Company, Baltimore, Maryland John J. Costello, General Credit Manager, Allied Chemical Corporation, New York, New York Herbert J. Dickson, Executive Vice President, Citizens & Southern National Bank, Atlanta, Georgia

Fred Kiel, Assistant Treasurer, Owens-Corning Fiberglas, Toledo, Ohio

Harold J. Kneuker, Assistant Treasurer, American Machine & Foundry Company, New York, New York

Walter J. Naber, Jr., General Credit Manager, Monsanto Chemical Company, St. Louis, Missouri

R. A. Nunlist, Treasurer, National Supply Division, Armco Steel Corporation, Pittsburgh, Pennsylvania

Richard M. Oddie, Assistant Vice President & Director, Small Business Advisory Service, Bank of America NT & SA, San Francisco, California

Jasper C. Osborne, Vice President, Trust Company of Georgia, Atlanta, Georgia

Blanche M. Scanlon, Nash Finch Company, Minneapolis, Minnesota

W. W. Smith, Manager - Credit & Collection Service, General Electric Company, New York, New York

Robert W. Sprenger, Assistant Controller and General Credit Manager, Johnson & Johnson, New Brunswick, New Jersey

L. B. Wilson, General Credit Manager, Union Carbide Corporation, New York, New York