Some Observations on the Problems of an Automated Case Processing and Information Management System for the Economic Stabilization Program

Introduction

As we have seen in preceding sections of this discussion, there were many problems that the Program agencies had in common during all phases of the Economic Stabilization Program. We have spent considerable time discussing case processing and have discussed the automated computer systems used by the Program. We have also referenced the Data Analysis paper to emphasize the expectation that the automated computer system would become a primary tool by which information management and some case processing would take place. It was originally believed that an automated system would be the panacea for information management by these agencies, and would significantly reduce the manpower needs of the stabilization Program. In fact the automated systems did not live up to expectations, and may have even contributed some problems of its own.

In very general terms, one can say that any information management system, be it automated or not, must accomplish only three basic functions. There must be an input of information in some fashion, the system itself must manipulate this information in some preconceived manner, and finally there must be an output of the manipulated information in some form which the user desires. It should be recognized that there are many complexities involved in the performance of such basic functions.

For instance, consider the input of the information into the system. Some of the questions that need to be answered with respect to input are: In what form will this input information be presented to the computer system? Will all input information come into the system in exactly the same fashion? As time progresses will there be changes in the type of input information? As the system grows will it be necessary to reconsider an input format? Are there any particular spatial requirements for consideration with reference to the input information? Will all information to be entered into the system come from a single location or from multiple locations? If the information is to be entered from multiple locations will those locations be in the same room, the same building, or the same city? Is it possible that input information may have to be entered from locations throughout the United States?

By considering the answer to just this small subset of questions it is apparent that the input of information into a computer system

can represent a major problem of and by itself. Further, the answers to all these questions must be translated into appropriate computer language commands.

To continue this discussion, consider the manipulation of information that the computer system is required to do. It is in this field -the manipulation of information-that computer systems can play an important role.

Not only can the computer perform typical arithmetic functions, but the computer can be programmed to make decisions based on either initial information "programmed" into the computer or on the basis of information resulting from manipulation within the computer. For instance, the computer is capable of comparing input information with any given set of established parameters to determine if the input information falls within certain ranges.

As an example of a computer performing this comparison function, the Pay Board, in the later stages of Phase II, utilized a computer system to check the information presented in a PB-3 wage and salary form. Utilizing a "PB-3 screening mask," information from a submission on a PB-3 wage and salary form was fed into the computer. The computer had been programmed to make certain computations in order to compare the computed information with previously established parameters. This operation determined whether or not the submission was within the generally applicable 5.5 percent wage and salary guideline established by the Pay Board. If the submission fell within this established guideline the computer would indicate this fact by generating a form letter to the submittors indicating that their submission had been received and was within the guidelines of the Program. In essence, the computer was actually programmed to "adjudicate" a very simple wage and salary

case.

While such an action of information manipulation is not an overly difficult program for a computer, it still represents considerable time and effort on the part of systems analysts to conceive the program and for computer programmers to write and produce a program and make it an operational system. However, if changes are required as a result of either a variation of policy or a need to alter operational assumptions, then alterations must be made in the computer program.

The input manipulation of information into and by a computer is done for the purpose of getting some form of useful output. The form of output required can take many forms: a simple listing of cases; or perhaps a listing of numbers, percentages, dollar figures, or dollar averages. In fact, almost any information format can

theoretically be presented. However, if the information desired for output has never been placed in the computer, or cannot be achieved by manipulation of input within the computer, then the desired information will never come out. It is, therefore, incumbent on the architect and users of any computer systems to fully and completely identify all input and output requirements of the system and, in essence, to lock these requirements "in concrete" before starting to build the system.

This was perhaps the biggest failure of the information management systems at the Pay Board and the Price Commission and hence the reason for the initial failure of the automated system to live up to expectations. The Economic Stabilization Program was such a dynamic program, with its information needs constantly changing, that computer operations were not capable of keeping up on a day-to-day basis with the voluminous programming changes required in the input, manipulation and output portions of the system. The inherent inability of an automated system to keep up with rapid Stabilization policy changes meant that the Program's computer was consistently limited in its ability to produce the information required by the management and policy makers of the Economic Stabilization Program.

Management Systems Personnel

Information management experts at the beginning of the Economic Stabilization Program were recruited from all sections of government and private industry and readily teamed together to become the operations or program offices of their respective Stabilization agencies.

In the hectic days at the beginning of Phase II, there was very little opportunity for these experts to interact in any larger circle than their own immediate offices as they attempted to fulfill the requirements of their respective agencies. There was for instance a very little interface between the Office of Operations Control at the Pay Board and the Office of Program Operations at the Price Commission. During that time simply getting the agency operation took eighteen hours a day, seven days a week.

It is perhaps unfortunate that the Economic Stabilization Program did not staff its operations and program organizations with a fully functioning, already established operations unit. Whenever a new group of people come together in a new endeavor, there must be a time period in which these individuals get to know and understand each other's capabilities and limitations. If one further compounds