measure up to those of the Region as a whole and how to resolve their interface problems with neighboring political jurisdictions. At any rate, regional aggregates of various types of data must be built up, in many instances, from the detailed data of lessor/political or economic units. Thus, a most fascinating aspect of developing a regional economic data bank is how to design the data content to meet the overlapping requirements of a wide variety of agencies within the St. Louis Region. Also involved is the matter of setting priorities for data collection and identifying the responsible channels for gathering or generating the data in the first place in the avoidance of costly duplication. The establishment of advisory/user committees as outlined above should be the chief instrument for attaining agreement on an overall design. D. PHASING THE WORK PROGRAM OF ESSENTIAL TASKS There are many circular relationships among the various kinds of data requirements to serve the many purposes indicated in the prior section. For example, labor force projects depend upon population projections which depend upon projections of economic and industrial developments of the Region which depend upon the availability of labor force which depend upon . . . etc. While this is a seriously over-simplified view of interrelationships among these factors, the point is that many kinds of data of a regional nature must be assembled conjointly before various planning requirements can be met. This truism gives rise to the notion that in the establishment of a data bank a threshhold of useful data must be assembled regarding circular relationships such as these before very much effort can profitably be devoted to analyzing or disseminating the data. Exceptions to the contrary-and some such assignments can be anticipated-will only serve to delay the larger benefits to be derived from the data bank. In order to highlight the evolutionary, phased development of the data bank, it is proposed that the latter be designated according to whether it is a first, second or third generation data bank (see FIGURE 2). The distinction between the generations of data banks are as follows: 1. The first generation data bank is basically an assemblage of primary data inputs irrespective of whether these have been manually prepared for storage or whether they are in machine-readable form (e.g., punched card, tapes, etc.). 2. The chief new ingredients of a second generation data bank is that the latter includes design information on input/output interindustry models, forecast models, etc., as well as related computer programs and data transductions or "massagings" as developed for economic or industrial analyses and reports. Thus the second generation data bank has now become computerized though not necessarily for all data entered in the bank. However, with regard to the non-computerized elements, even these may be distinguished from their counterparts in the first generation bank in that many have since been put into an enhanced form for purposes of analysis or decision making. 3. The third generation data bank is characterized by the comprehensiveness of data and its relative maturity. By the latter is meant that in comparison with the second generation data bank it has relatively fewer data gaps, higher quality data, more time series data and a wider variety of data outputs, especially those resulting from different computer runs of input/ output models, forecast models, etc. Indeed, the models themselves will have undergone a substantial degree of refinement and the third generation data bank will realize a higher degree of mechanization as registered by a reduction in the ratio of manual to machine-readable data stored in the bank. At this point it is desirable to go back to the problem of identifying work tasks essential to the establishment of the first generation data bank; for this is the key to the development of later generation data banks. Basically there are two types of activities involved in gathering this data. A rather obvious activity is to retrieve data of importance to the economic development of the St. Louis Region from published sources, or from the records of files of various agencies when such data is unpublished. Some of the latter data will already be on punched 3 See Appendix for a "Preliminary List of Data Services for a Regional Economic Data Bank." cards or tapes so these may be reproduced at very low cost. This collection type of activity is enumerated in Figure 2 as tasks 8, 19 and 20. The second type of activity involves the generation of data through the conduct of regional surveys, often on a sampling basis. One such same sample survey is shown in Figure 2 as an input/output interindustry survey as previously described. This has been denoted by task 21 and involves numerous preliminary efforts as shown on the left of the chart. It should be noted that a survey which is very similar to this one has already been conducted in the Philadelphia Region and another is under way at Boston. Such a survey is very costly and complex in its implementation as the data is often not organized in the records of firm in the specific form required and, moreover, will generally pose problems of eliciting cooperation because of the confidentiality with which the entrepreneur regards the data. Because of the limitations of RIDC's budget we are planning to survey only eight key industries this summer through the employment of graduate students from the universities of the Region. It is imperative, however, that all industries be studied in this manner to "close the loop", so to speak, in our analysis of the Region's potential for industrial development as well as to assess the impact of the latter on other economic growth, population growth, tax revenues, etc. Thus, additional funds will need to be committed over and above that which can be made available from RIDC's annual operating budget to complete this survey. It is planned to code this data, key punch and file it as it comes in. Hence, whenever its comprehensiveness reaches the threshold level as required for designing the input/output and forecasting models as in the second generation data bank, the lag will be minimized in arriving at the third generation data bank. Likewise, as time and resources permit, it is desirable to code and put into machine-readable form the economic and other data as assembled in tasks 8, 19, and 20 as mentioned above. E. THE SELECTION OF METHODOLOGIES For a community or region which is dissatisfied with its past developmental trends the methodologies of greatest importance are those which center around estimating potentials rather than reliance on the usual economic projection techniques that are chiefly based on historical data. This is not to say that historical data is of little significance to the analysis but rather that, in addition, much thought must be given to the capability and desire of the Region in solving its problems and in identifying and fulfilling its opportunities. Statistical correlation methods can be used to some advantage as well as economic modelling, linear programing, input/output analysis and an assortment of econometric and operations research techniques. The latter, in particular, could be expected to make use of the data stored in the bank for operating and planning decisions of various agencies whose focus is on problems and decisions of more limited scope than regional economic development per se even though these activities could have a significant bearing on the latter. Since input/output analysis can be an invaluable tool for identifying developmental opportunities and estimating growth potentials, it is desirable to expand a bit on the character and implementation of this approach which has been only briefly described in an earlier section of this paper. Basically, the heart of the technique is an accounting one in which the economic interrelations are shown in tabular form called an "input/output matrix". One such table is a multi-dimensional revenue-expenditure accounting framework. It expresses in a row and column arrangement the basic relationships existing among industries as suppliers and consumers of the products of one another. As a consequence, an input/output analysis reveals the types of trade among industries within the regional economic system and between that system and other systems (e.g., the public sector, household consumer sector, etc.). The significant advantage of this method is that the matrix developed in an input/ output study establishes the degrees of interdependence betwen economic sectors and the most likely effect upon other sectors of any fluctuations in the activity of the selected sector. At present, there is no single source or study of any type which is current and which covers the entire St. Louis Region. 4 These next several pages were drawn from material prepared by Dr. Leroy Grossman Associate Professor of Economics at St. Louis Univ. A local matrix can provide a useful point of departure for making estimates of future growth patterns, both from the firm's and from the public's point of view. In such a procedure economic sectors are divided into two broad categories. The first group, termed "processing sectors", buy goods and services from the rest of the economy and process them for resale to the "demand sectors" which may be other processing sectors or to the final demand sectors such as households or governments. Purchases are regarded as inputs while sale are considered outputs. The following table is an abbreviated hypothetical matrix showing two procesing sectors and two final demand sectors. When read horizontally, it shows how the output of sectors named on the left were distributed to the sectors named at the top of each column. Alternatively, the table may be read vertically. We then have a list of the purchases made by each sector from the sector named on the left. By definition, gross inputs and gross outputs are equal. TABLE I.—Direct purchases per dollar of output (hypothetical matrix) Thus, a $1 million change in the output of the manufacturing sector would have a $100,000 first round impact upon the agricultural sector ($1,000,000×.10). By the use of high speed computers second and third round effects upon the various sectors could also be determined. This technique forces the investigation to consider the relevant variables in a systematic manner. Subjectivity would enter into the problem in several ways: the degree of refinement in which the data is fed into the martix and estimating the changes of industrial input/output coefficients resulting from technological change. Furthermore, the matrix should not be regarded as an end in itself but should be utilized along with other data for intelligent decisionmaking. More meaningful solutions to many very practical types of problems could be found if the proposed data bank and input/output study were a reality in the St. Louis area. The uses to which such information could be put are discussed in terms of two very broad problem solving categories; i.e., those that concern individual companies and those that concern the region and/or its political subdivisions. For example, input/output matrices have the advantage of showing the transaction linkages between private and public agencies within the region— as well as outside either in terms of sales or employment of certain skill categories. Thus it provides a systematic, quantitative reference scheme for discussing the soundness of economic conditions that already prevail and of policies and plans for changing these whether in the public or private domains. More specifically, regional input/output data could, under appropriate circumstances, lead to the following benefits and uses: 6 1. Making structural comparisons in the economy of the St. Louis Region with other regions or the U.S. economy as a whole. 2. Identifying structural gaps and weaknesses in the Region's economy. 3. Assessing the degree of economic self-sufficiency of the Region. 4. Evaluating the dependency of a given industry on other industries so as to better promote the establishment of industrial complexes for the St. Louis Region in maximizing the sort of external economies that industries within a complex provide to each other. 5 In practice, we would subdivide the sectors shown in this table into a large number of subsectors, particularly in the case of manufacturing. 6 Many of the ideas presented here were suggested by Professor G. Donald Hanrahan of St. Louis University. 5. To attain a more valid ranking of the importance of industries to the Region in setting priorities for developmental effort. 6. Provided that time series input/output data are available for the Region and not encumbered by: a) entry of new industries or redefinition of industries; b) changes in relative prices of inputs; or c) data unreliabilities of serious proportions, then one could draw useful inferences from the input/output coefficients in comparision with national or other regional data of like kind as to whether industry in the Region is equal to the average, or perhaps, the most advanced state of the technologies of the given industries. 7. Identifying bottleneck problems as when industrial growth is now pressing or will press hard against certain specialized resources of the Region. 8. Assessing the local impact of shifts in the nation's economy or that of other regions which supply essential materials or goods to the St. Louis Region or make significant purchases from us. 9. Providing much useful insight for preparing forecasts of the Region's economy under special assumptions with regard to key developments or progress in solving Regional problems. 10. Giving the best clues as to which sectors of the economy merit the highest priorities for assembling data in the Regional Economic Data Bank. Before leaving the subject of methodologies it is imperative to consider the methodologies or techniques of measurement to be employed. Measurements or census data of the Region already abound but these do not necessarily measure certain of the vital factors which are the central issues to economic development: 1. The attitudes and opinions of community or business leaders toward social and economic conditions prevailing in the Region, or the community's willingness to do the necessary things to achieve the fuller potential for economic development. 2. The extent of underutilization of the human and physical resources of the Region including industrial plant capacities, social overheads such as schools, hospitals, highways, streets, etc. it. 3. The productivity of the Region's resources or the factors contributing to 4. Suitable measures of the efficiency of government or the degree of Regional cooperation. 5. Entrepreneurial qualities of management such as alertness to potentials and the drive and follow-through necessary to achieve them. 6. Horizontal and vertical mobility of the labor force intra-regional and inter-regional. 7. The quality of the Region's productive outputs relative to their prices. 8. The efficacy of labor-management relations as judged from the standpoint of the community's best interests. 9. Standards of excellence with regard to the Region's educational programs, cultural resources, social institutions, etc. 10. Social, as opposed to economic, values which are "measured" in the market place. 11. Many of the direct and indirect costs and benefits to the Region of various developmental programs so as to better choose between them. 12. The extent to which non-respondence to sample surveys biases the conclusions or inferences-among other sources of unreliability of data obtained in this manner. 13. The a priori "goodness" of forecasts of Regional developments or estimates of potential. 14. The presence of spurious correlation between factors that are important to Regional development. 15. Etc. Scholars have spent many frustrating hours over the above problems of measurement and advances have often been painfully slow. It is obvious that many of these factors will never become measurable in the strict scientific sense of the word. In this event, these factors must be estimated by imperfect techniques or as a matter of judgment guided by experience and statistical observations and analysis. At any rate, the methodology of measurement of factors which are meaningful to regional development deserves to be put high on the list of the proposed university advisory/user committee of the St. Louis Regional Economic Data Bank. V.-ADVANCED DATA PROCESSING SYSTEMS HARDWARE REQUIREMENTS As mentioned above,on conceptual grounds it is technically feasible to store a broad range of data in a regional economic data bank and retrieve it by remote control in the latest push-buttoin style. Economic feasibility, however, is quite another matter and this will have to be worked out with patience and care as the data bank grows in content and conceptual sophistication. With reference to the proposed first generation data bank (see Figure 2) it is not planned that this be computerized. Indeed, only certain aspects of the second generation data bank need be computerized on economic grounds (e.g., input/output and forecasting models). However, with regard to a third generation data bank a marked increase in capability to retrieve data from the bank by computer or punched card type of data processing hardware is envisioned. By the same token, certain methodologies of data analysis or synthesis of forecasts, plans, etc., are sufficiently complex as to make the use of advanced data processing systems almost mandatory. Indeed, RIDC has already made use of punched card equipment in a local industry survey and it is contemplated that even a preliminary analysis of input/output data of a survey of eight industry groups being planned for this summer will make it profitable to use punched card equipment versus manual methods. Later on, when input/output data is available for all industries of the Region the matrix will be sufficiently large that the use of one of the larger computers of the Region will be required to manipulate the matrix. Hence, to the limit of our budgetary resources, RIDC proposes to begin organizing all of the data now being manually stored in such a manner as to facilitate its ready transformation to punched cards or tapes in meeting the requirements of the second and third generation data banks. It should be stated at the outset that RIDC has no plans to purchase or lease computer or other data processing equipment which would exceed our financial resources. Rather, we propose to accumulate much data on punched cards or tapes which can be reproduced economically for use by other institutions of the Region who are engaged in developmental activities. For our own needs to manipulate or analyze the data, etc., we would hope to receive gratis computer or other data processing services as we have in the past-from supporting institutions. As the data bank matures in its development RIDC should be able to afford, however, to leave such simple equipment as a key punch machine and a data link device. The latter is capable of transmitting (or receiving) punched card or computer tape information by telephone to the advanced data processing centers of interested users. This practice will surely grow in popularity in future years as a result of technological developments recently made in the computer industry. Hence, the benefits of the RIDC Regional Economic Data Bank would be more readily accessible at lower cost to a wider group of potential users. Thus, it may be said that RIDC's requirements for advanced data processing systems hardware will be modest in relation to the total system's requirements which take into account the user's needs. Inasmuch as all Regional universities already have much of the needed equipment and the same can be said for many governmental agencies and business firms of the Region, then additional expenditures to more fully utilize the growing volume of data to be stored in the RIDC data bank may be relatively small. VI.-FINANCIAL REQUIREMENTS The rationalization of financial requirements for the fulfillment of the proposed scheme for a regional economic data bank closely parallels that given above for advanced data processing system hardware requirements. There is already a large number of skilled analysts and professionals on the planning and operating staffs of local governments, business firms, universities, etc., whose contributions can presumably be tapped to aid in developing the methodologies for making use of the data bank and to advise on the evolutionary conceptualization of the bank, etc. In fact, one should expect that the provision of data on a highly accessible and timely basis and in an enriched and reliable form for decision making will spare many organizations considerable expense. In some instances, however, various user groups may need to upgrade certain staff positions or much of the value of the data may be lost due to inadequate training in relating the data to the decision-making process. 77-577-67 |