- .^?. « ; PS-?//6 tJort. T Econqmic Impact ™ T 5F" " I I ' s nFPARTMFNT Kit U. S. DEPARTMENT OF COMMERCE Maritime Administration Digitized by the Internet Archive in 2012 with funding from LYRASIS Members and Sloan Foundation http://archive.org/details/porteconomicimpaOOarth PORT ECONOMIC IMPACT KIT Prepared by Arthur D. Little, Inc. Pacific Coast Association of Port Authorities U.S. Department of Commerce Maritime Administration Office of Commercial Development Office of Port and Intermodal Development September 1979 rj. FOREWORD Port activities serve as vital catalysts to the generation of employment, taxes, revenues and business for local communities. Virtually every major metropolitan region of the United States centers around a port, or is closely linked by rail or highway with a port. As waterborne trade in the United States approaches 2 billion tons of cargo per year, the rip- pling economic effects of vibrant port activities will continue to provide jobs, dollars and local community development. It is, thus, important to understand and appreciate these economic impacts of local port develoDment. The Port Economic Impact Kit was developed to facilitate the preparation of economic impact reports by local communities or port authorities. This step-by-step manual is designed to standardize a methodology that will enhance the credibility, clarity and comparability of port economic impact studies. Employment of this Kit should involve existing staff and reduce the preparation costs of economic impact reports. Additionally, use of the Kit should allow periodic updates of impact estimates at minimal expense. This report was prepared under the sponsorship of the Maritime Administra- tion, the Canadian Ministry of Transport and the Pacific Coast Association of Port Authorities. The consulting firm of Arthur D. Little, Inc. prepared the Kit and worked closely with members ports of PCAPA in adapting and testing the impact methodology. This comprehensive end product is applicable for the analysis of maritime-related activities at all United States and Canadian ports. Office of Port and Intermodal Development Maritime Administration U.S. Department of Commerce September 1979 TABLE OF CONTENTS Page I . OVERVIEW A. Background 1 B. Uses and Scope of the Kit 2 C. Organization of the Kit 3 II. PORT ECONOMIC IMPACT ISSUES A. Overview of Economic Impact Theory 8 B. Methodologies for Estimating Direct Impacts 9 C. Methodologies for Estimating Secondary Impacts 10 D. Relationship Between Ports and Port-Dependent Activities 14 E. Geographic Area of Analysis 15 III. DETAILED METHODOLOGY A. Initial Considerations 17 B. Design Survey 27 C. Survey Distribution 48 D. Survey Result Compilation 54 E. Development of Secondary Impact Multipliers 66 F. Calculation of Secondary Impacts 96 G. Preparation and Dissemination of Report 111 IV. GLOSSARY 116 APPENDIX A. Background H8 B. General Problems and Solutions 119 C. Specific Case Studies 122 ±i Arthur D Little Inc. LIST OF FIGURES, EXHIBITS, AND TABLES Figures 1 2 3 Economic Impact Study Elements Program Schedule Staff Responsibility Page Exhibits A:2-l A: 3-1 A:3-2 B:l-1 B:2-l B-.3-1 B:3-2 B:4-l B:4-2 B:5-l C:2-l E:3-l E:4-l E:7-l E:7-2 E:8-l E:8-2 F:4-l F:5-l Guidelines for Cost of Economic Impact Study 19 Estimated Staff Time Requirements to Complete Analysis 21 Calendar Time Schedule 23 Suggested Survey Questions 28 Possible Optional Questions 30 Sample Survey - Port Industry 33 Sample Survey - Port-Dependent Industry 34 Coding Instructions - Port Industry Survey 36 Coding Instructions - Port-Dependent Industry Survey 37 Sample Survey Cover Letter 40 Sample Survey Response Chart 51 Concentration Technique Calculation 73 Minimum Requirements Technique Calculation 76 Example of BEA County Income Data 83 Calculation of Income Multiplier 84 RIMS Industry Classifications 87 Waterborne Commerce Related Industries with RIMS Codes 95 Payroll/Sales Ratios for Major Industry Divisions 103 State Agencies and Universities Receiving BEA Local Income Estimates 106 List of Data Requirement Publications 114 Tables Port of Grays Harbor Direct Activity 124 Port of Grays Harbor Total Economic Activity 125 Ports of San Francisco and Oakland Direct Activity 128 Ports of San Francisco and Oakland Total Economic 129 Activity Port of Vancouver, British Columbia Direct Activity 132 Port of Vancouver, British Columbia Total Economic 133 Activity in Arthur D Little Inc I. OVERVIEW This "economic impact kit" was prepared for the Pacific Coast Asso- ciation of Port Authorities (PCAPA) , the U.S. Maritime Administration (MARAD) , and the Canadian Ministry of Transport. The purpose of this assignment was to develop an effective methodology which member ports could use to conduct their own economic impact assessments. Therefore, the resulting "kit" would have to be designed so that it did not require extensive technical skills and background or substantial financial and time commitments. At the same time, the methodology would have to produce reasonably accurate impact estimates when applied to port districts of different sizes with different cargo mixes. The end product of the assessment methodology would be an economic impact report which is both credible and able to be understood by the general public. A. BACKGROUND The first step in the assignment was a review of port economic impact studies and other information to develop a common set of defini- tions of port-related terms and an inventory of economic impact metho- dologies available. The glossary was prepared to insure proper under- standing of the terms used in the kit and comparability among ports. (Existing studies show that the port industry may be defined to include only as little as the actual waterborne transport of goods or as much as all activity which occurs on port district property.) Common definitions tend to increase the credibility of individual impact studies by pro- viding a body of studies for comparison. The inventory of direct and secondary impact methodologies included both those used in port economic studies and those which are commonly available but have not been used for ports to date. These techniques were characterized by, among other things, the skills needed to implement them, time and financial outlays required, data requirements, and their applicability to PCAPA member ports. This inventory was reviewed by the PCAPA Economics Committee and the techniques to be developed for the kit were selected. It was decided that direct port industry impacts would be identified by means of a survey and that secondary (indirect and induced) economic impacts would be estimated by two techniques — an economic base multiplier and an input/output model derived multiplier. The methodologies selected for use by PCAPA members are presented in a series of 40 detailed steps describing data requirements and proce- dures. (Special sections to adapt the methodology for Canadian ports are also included.) The draft "economic impact kit" was reviewed by the Economics Committee for clarity and completeness. Finally, the metho- dology described in this kit was tested on four West Coast ports — Grays Harbor, Washington; San Francisco and Oakland, California; and Vancouver, British Columbia. B. USES AND SCOPE OF THE KIT The procedures in the PCAPA economic impact kit will yield estimates of the current economic impact (employment, payroll, etc.) of the port on its community. These estimates can be derived from readily available data by existing port staff who may not have extensive skills or training in economics or mathematics. In particular, the methodology has been designed to eliminate, to the extent possible, the need for in-depth knowledge of and access to economic and statistical data sources and "expert" judgment about economic issues. While this has required some simplification of methods and procedures, we believe the methodology outlined in this kit will yield results sufficiently accurate for the majority of port district uses. The principal use of this type of port economic impact study is to enhance community relations and improve public education efforts. At a time when taxpayers are becoming increasingly chary of all publicly funded enterprises, the benefits of showing a return on public investment — not only revenues, but also jobs and income — are obvious. The same approach can be used to project the community economic benefits of port expansion or other increase in port activity, although a major redi- rection of port operations or change in cargoes handled may require a more detailed economic assessment. Finally, the economic impact metho- dology will help policy makers identify the role of the port and its related activities in the overall local or regional economy as an input to broad planning decisions. The economic impact methodology contained in this kit will most pro- bably meet the needs of small and medium-sized ports which serve rela- tively simple economies and have limited financial resources for such efforts. Larger port districts with more complex economic inter- relationships may prefer to use the method as an annual reporting tool and spend additional funds on a more detailed in-depth economic study based on additional area-specific data and employing additional area- specific assessment techniques (e.g., a local or regional input/output model) . These districts can use the kit to update some of the results of the detailed study periodically at relatively low cost by using their own staff and published data. Thus, this economic impact kit should prove useful to all PCAPA member ports. While this kit is designed to yield reasonably accurate results with the use of available resources, it is important to recognize the limita- tions of the methodology. Perhaps most important, because a survey is required to measure direct activity, the estimate of total port industry impacts will be only as good as the survey. In other words, the credi- bility of the overall assessment will be enhanced if the time and effort necessary to increase the reliability of the survey are committed (e.g., by conducting a pretest of the questionnaire). Similarly, every effort must be made to compile a complete list of port industry firms from which the sample will be drawn. The development of reliable survey data is the single most important element of the economic impact study. "... ... Because the kit is designed to be applicable to many different ports, it may not answer all the questions of interest to a particular port district. The kit's results are expressed in terms of income and output with additional steps available to estimate employment and some tax revenues. The secondary economic effects of port-dependent indus- tries are not estimated because it is difficult to assess the true depen- dence of these firms on the specific port. Most tax impacts (sales, property, etc.) are not carried beyond the direct level because of double counting and other problems. This conservative bias (i.e., possibly underestimating the actual total impact of the port) usually tends to increase the credibility of an economic impact assessment. If a port needs information not provided by the basic methodology, it may be possi- ble to add optional questions to the survey or develop additional ratio multipliers (see Step F:6 in Chapter III). One specific ratio frequently used in port impact studies is the cargo multiplier, which is an estimate of the employment, income, revenue, tax paid, and similar economic measures associated with each ton of cargo shipped through the port. While the simplicity of this approach is attractive, closer examination of the relationships involved suggests that cargo multipliers may be quite misleading. For example, some ele- ments of port industry employment may be directly related to tonnage (e.g., loading and unloading operations), but other elements are more closely tied to the number of ships moving through the port (e.g., pilotage and tugboats) . Other elements of total employment tend to remain fixed no matter how much cargo moves through the port (e.g., port management and administration, freight forwarders, etc.). Other measures would be affected by the value of cargo (almost impossible to determine for containers), cargo origins and destinations (e.g., transportation savings) , and other factors for which data are not readily available. Because we believe that the methodology described in this kit will yield more accurate and thus more useful impact estimates, a procedure for cal- culating cargo multipliers has not been developed. C. ORGANIZATION OF THE KIT The PCAPA economic impact kit is made up of three principal sections in addition to this overview. Chapter II presents a general discussion of economic interrelationships and economic assessment methodologies which explains some of the theory on which the kit is based. It may also serve as a technical appendix to a port's economic impact report. It is not necessary for a user to be familiar with this material before begin- ning the steps described in Chapter III. Chapter III is the detailed methodology "workbook" which describes the procedures, data sources, cal- culations, and other information necessary to conduct the port economic impact study. Chapter IV is a short glossary of economic and port industry terms used in the kit. The methodology is organized into seven major elements, each of which contains several specific steps (see Figure 1) . Each element is designed to accomplish one of the major tasks of the study which are: FIGURE 1 ECONOMIC IMPACT STUDY ELEMENTS A. Initial Considerations 1 . Determine Public Purpose of Analysis 2. Obtain Funding to Conduct Analysis 3. Allocate Staff Time to Analysis 4. Determine Scope of Study 5. Select Geographical Area for Analysis 6. Prepare Schedule for Project B. Survey Design 1. Adopt Required Questions for Surveys 2. Develop Optional Questions for Surveys 3. Adjust Questionnaires for More Than One Port in Region 4. Prepare Coding Instructions for Questionnaires 5. Prepare Cover Letter for Questionnaires 6. Identify Firms in the Port Industry 7. Select Firms for Pretest of Survey Instrument 8. Select Interview Format 9. Conduct Survey Pretest and Revise Questionnaire C. Survey Distribution 1. Advertise Study Prior to Survey Distribution 2. Prepare List for Respondent Checkoff 3. Conduct Survey D. Survey Result 1. Code Returned Surveys 2. Check Consistency of Survey Responses 3. Estimate Value of Missing Responses 4. Collect Port Employment and Income Data 5. Tabulate Survey Responses by Industry and Impact Category 6. Aggregate Survey Responses and Estimate Total Direct Impact E. Development of Secondary Impact Multipliers 1. Select Methods for Secondary Impact Analysis 2. Collect Income and Employment Data for Economic Base Multipliers 3. Develop Employment Multipliers Using "Concentration" Approach 4. Develop Employment Multipliers Using "Minimum Requirements" Technique 5. Develop Employment Multipliers Using "Experience" Technique 6. Develop "Concensus" Employment Multipliers 7. Develop Economic Base Income Multipliers 8. Select Industries for Analysis and Order RIMS Multipliers F. Calculation of Secondary Impacts 1. Determine Share of Port Industry Income That is Basic 2. Estimate Total Income Supported by Port Activities Using Economic Base Multipliers 3. Estimate Total Sales Supported by Port Activities Using RIMS Multipliers 4. Compare RIMS and Economic Base Techniques 5. Update RIMS Estimates G. Preparation and Dissemination of Report 1. Prepare Report 2. Disseminate Report Initial Considerations Survey Design Survey Distribution Survey Result Compilation Development of Secondary Impact Multipliers Calculation of Secondary Impacts Preparation and Dissemination of Report Each step is described in terms of its data requirements (including spe- cific source citations) , recommended action (the specific calculations or other work required) , and output. Many steps also contain a discussion describing possible alternatives or problems, additional background information, constraints on the use of results, or other relevant issues. Each step represents a discrete work item so that the user can complete the program one step at a time. Figure 2 shows the estimated project schedule. It is estimated that a maximum of about eight months will be needed to complete the economic impact study program. The first two months are allocated to determining the scope of the study and other initial considerations; the actual data collection and calculation work is expected to require about six months. Actual implementation time may vary depending on staff availability, the size of the port district, etc. (A small district may be able to complete personal or telephone surveys in less time than a larger district would need for mail surveys, for example.) Three levels of port staff are required to implement the economic impact study — management, a project director, and technical staff. The port district's management — possibly the executive director or a com- mittee of the governing board — should make the initial decisions about the study's scope, funding, etc., and have review responsibility for all materials distributed to the public in the name of the port (question- naires and cover letters, press releases, and the final report). The project director will have principal responsibility for the conduct of the study including supervision of technical staff and writing the impact report. Ideally, this individual would have some knowledge of economics and be well acquainted with the operation of the port. At a minimum, the project director should have read and understood all of the steps of this kit before work is begun. Technical staff would be responsible for data collection, calculations and tabulations, and, if necessary, conducting telephone and personal interviews. For ports with limited staff avail- ability, a single person may serve as project director and also perform the technical staff tasks. A port might also use economics or planning students from a local college as interns to perform the technical staff functions. Figure 3 shows the level of staff responsibility associated with each step. Before beginning this economic impact study, it is recommended that all of the steps be read. The characteristics and requirements of each individual port district will determine the way in which the study should be approached, including level of detail, approximate cost, and which steps may be considered optional. FIGURE 2 PROGRAM SCHEDULE Step No. Description Week DateComp.eted I I | | | | I I I I I I I I I I I I I I I I I I I I I I I I I I I I I 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32 34 A:1 Determine Public Purpose of Analysis A. 2 Obtain Funding to Conduct Analysis A:3 Allocate Staff Time to Analysis A:4 Determine Scope of Study A:5 Select Geographical Area for Analysis A:6 Prepare Schedule for Project B: 1 Adopt Required Questions for Surveys B:2 Develop Optional Questions for Surveys B:3 Adjust Questionnaires for More than One Port in Region B:4 Prepare Coding Instructions for Questionnaires B:5 Prepare Cover Letter for Questionnaires B:6 Identify Firms in the Port Industry B:7 Select Firms for Pretest of Survey Instrument B:8 Select Interview Format 8:9 Conduct Survey Pretest and Revise Questionnaire C:l Advertise Study Prior to Survey Distribution C:2 Prepare List for Respondent Checkoff C:3 Conduct Survey D:1 Code Returned Surveys D:2 Check Consistency of Survey Responses D:3 Estimate Value of Missing Responses D:4 Collect Port Employment and Income Data 0:5 Tabulate Survey Responses by Industry and Impact Category 0:6 Aggregate Survey Responses and Estimate Total Direct Impact E:1 Select Methods for Secondary Impact Analysis E:2 Collect Income and Employment Data for Economic Base Multipliers E:3 Develop Employment Multipliers Using "Concentration" Approach ■E:4 Develop Employment Multipliers Using "Minimum Requirements" Technique E:5 Develop Employment Multipliers Using "Experience" Technique E:6 Develop "Consensus" Employment Multipliers E:7 Develop Economic Base Income Multipliers E:8 Select Industries for Analysis and Order RIMS Multipliers F:1 Determine Share of Port Industry Income that Is Basic F:2 Estimate Total Income Supported by Port Activities Using Economic Base Multipliers F:3 Estimate Total Sales Supported by Port Activities Using RIMS Multipliers F:4 Compare RIMS and Economic Base Techniques F;5 Update RIMS Estimates F.6 Calculate Other Impact Measures G: 1 Prepare Report G.2 Disseminate Report 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32 34 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 ii ii i 1 1 1 1 1 1 Estimated Date FIGURE 3 STAFF RESPONSIBILITY Step No. Description A:1 Determine Public Purpose of Analysis A:2 Obtain Funding to Conduct Analysis A:3 Allocate Staff Time to Analysis A:4 Determine Scope of Study A:5 Select Geographical Area for Analysis A:6 Prepare Schedule for Project B:1 Adopt Required Questions for Surveys B:2 Develop Optional Questions for Surveys 8:3 Adjust Questionnaires for More Than One Port in Region B:4 Prepare Coding Instructions for Questionnaires B:5 Prepare Cover Letter for Questionnaires B:6 Identify Firms in the Port Industry B:7 Select Firms for Pretest of Survey Instrument B:8 Select Interview Format B:9 Conduct Survey Pretest and Revise Questionnaire C:1 Advertise Study Prior to Survey Distribution C:2 Prepare List for Respondent Checkoff C:3 Conduct Survey D:1 Code Returned Surveys D:2 Check Consistency of Survey Responses D:3 Estimate Value of Missing Responses D:4 Collect Port Employment and Income Data D:5 Tabulate Survey Responses by Industry and Impact Category 0:6 Aggregate Survey Responses and Estimate Total Direct Impact E:1 Select Methods for Secondary Impact Analysis E:2 Collect Income and Employment Data for Economic Base Multipliers E:3 Develop Employment Multipliers Using "Concentration" Approach E:4 Develop Employment Multipliers Using "Minimum Requirements" Technique E:5 Develop Employment Multipliers Using "Experience" Technique E:6 Develop "Consensus" Employment Multipliers E:7 Develop Economic Base Income Multipliers E:8 Select Industries for Analysis and Order RIMS Multipliers F:1 Determine Share of Port Industry Income That is Basic F:2 Estimate Total Income Supported by Port Activities Using Economic Base Multipliers F:3 Estimate Total Sales Supported by Port Activities Using RIMS Multipliers F:4 Compare RIMS and Economic Base Techniques F:5 Update RIMS Estimates G:1 Prepare Report G2 Disseminate Report Project Technical Management Direction Staff P R P R P P R P R P P R P P P R P R P P P R P P R P R P R P P R P P P R P ■ performance R ■ review II. PORT ECONOMIC IMPACT ISSUES A. OVERVIEW OF ECONOMIC IMPACT THEORY The purpose of this kit is to estimate the economic impact of ports on the surrounding community — that is, what the port contributes to the community in terms of income, jobs, taxes, and other economic measures. A complete accounting of these impacts must include not only the direct impacts of the port industry (and port activities) , but also the indirect and induced effects created by the port industry. As discussed below, various methods have been developed for using observed economic rela- tionships to determine the secondary activity associated with a given type and amount of direct activity. Economic impacts, and economic activity in general, are determined by demand for goods and services. The direct impact of the port industry as defined in this kit is thus the output, employment, payrolls, and other economic activity created by the demand for the movement of water- borne commerce. As demand rises — as producers and users ship more cargo through the port — direct economic activity also increases. The demand for port industry services is translated into sales of several direct sector industries which provide stevedoring, pilotage, bunker fuel, marine insurance, and other goods and services. To provide these services, the direct sector enterprises must obtain necessary inputs from several sources. For many, the largest purchase is the time of their employees; these purchases are reflected in both pay- rolls and employment. In addition, they must purchase commodities and services from other firms. Port industry purchases include such things as tugboats, cranes, office equipment, cleaning supplies, telephone service, and many, many others. These purchases are the indirect impacts of the port industry; if demand for port services stopped, demand for these services would also cease. Indirect demand does not stop at this initial round. As an example, the supplier (manufacturer) of the cranes must purchase steel, machine tools, power, office supplies, and a variety of other equipment and services in addition to making payroll payments. Similarly, the firms which supply the crane manufacturer must also obtain inputs. The result is a continuous interchange of goods and services among the sectors of the economy in order to meet the demand in the port industry. Along with these interindustry sales and payrolls, each round of spending also generates profits and taxes. The port industry's direct and indirect activities contribute to sales, income, property, fuel, and other federal, state, and local tax revenues. In addition to the direct and indirect impacts, the port industry also has induced impacts created by the expenditures of direct and indirect payrolls. Payrolls and proprietors' income are spent for con- sumer goods and services including food, clothing, transportation, housing, and entertainment. People are employed to provide these items and more indirect sector inputs must be purchased, resulting in still more rounds of profits, payrolls, and taxes. Tax revenues purchase government services which require supplies resulting in yet another round of impacts. In theory, these rounds of impacts never end, but they do approach a maximum level. The ratio between the total volume of sales (or income or employment) generated and sales directly related to the pert is the multiplier for the port industry. The size of the multiplier depends on the structure, size, and diversity of the port district's (or region's) economy . It is not likely that the port district's economy can provide all of the required goods and services. For example, heavy cargo handling equipment is typically produced by a few large manufacturers which serve national markets. Similarly, many consumer products are manufactured for national rather than local markets at one central plant or a few regional plants. These and many other items must be imported from other regions. These imports represent regional leakages from the multiplier impacts associated with the demand for port services. For example, a port's work orders, bills of lading, and similar forms could probably be produced by a local printer, but cranes, forklifts, or other heavy equipment would probably be purchased from a manufacturer outside the port district. The printer in turn, might buy his paper and ink from suppliers outside the port region, so that each round of impacts would be accompanied by leak- ages. These leakages vary in magnitude depending on the ability of the local economy to meet the demand for goods and services. The more self- sufficient an economy, the smaller the leakage and consequently the greater the local multiplier effect. Traditionally larger, more diverse economies have higher local multipliers than smaller ones simply because they can provide a greater percentage of locally demanded goods and services. The port-related economy is not easily recognized because it is intertwined with the remainder of the regional economy. Few firms affected at the indirect and induced levels are totally dependent on the port. Their sales would decline if the port closed, but would not drop to zero. B. METHODOLOGIES FOR ESTIMATING DIRECT IMPACTS Because the port industry as defined in this kit does not conform to Standard Industrial Classification (SIC) categories, but includes parts of many SICs, there is no secondary source from which direct impacts can be determined. (As an example, an assessment of the local impacts of the primary metals industry could use all activity listed in SIC 33 in County Business Patterns or a similar source as the direct sector and proceed from there; no similar indirect approach is possible for the port industry.) In addition, as discussed further below, port-dependent activities must be addressed not only by identifying users of the port (shippers and receivers) but also by estimating the extent to which these users are truly dependent on the availability of the port. For these reasons, and because any mistakes made in the assessment of direct activity will be multiplied in the estimate of total economic impacts, a direct survey of port industry and port-dependent firms is necessary to measure direct impacts. Survey techniques also have the advantages of being able to be tailored to provide specific types of information which may not be available from other sources. For instance, in addition to the basic information needed for the economic impact study, questions could be added concerning estimated employee residential distribution (inside or outside the county), business expansion plans, or the need or desire for additional types of port services. As discussed elsewhere, shorter questionnaires are more likely to be completed, but a few optional questions may be very useful to the individual port district. C. METHODOLOGIES FOR ESTIMATING SECONDARY IMPACTS The basic assumption of economic impact analysis is that observed relationships in a region's economy can be used to determine the effects of changes in the level of economic activity on the regional economy as a whole. Simply, if activity in one industry increases by a given amount, the region's total economy can be expected to change in some predictable manner. These relationships are given in the multipliers described above, although they may be expressed as simple multipliers or a complex, sectorally disaggregated matrix of multipliers. There are three principal types of methods used as classification systems for the relationships in a region's economy and as techniques for estimating the indirect and induced effects of changes in direct activ- ity. These are: • Economic base multipliers • Interregional trade multipliers, and • Input/output models. Each of these methods can be classified according to its overall theoretical validity, the accuracy of its estimates, the skills required to develop coefficients and apply the methodology, and the cost, time, and amount of data required to implement the methodology. As might be expected, there is generally a tradeoff between the desire for a simple, low-cost approach and a more complex, complete, and accurate description of a region's economic system and interactions. Moreover, these consi- derations affect not only the choice among methods, but also the way 10 in which each method is applied. Each individual technique can be imple- mented with different levels of disaggregation, and the greater the disaggregation, the greater the usefulness, accuracy, cost, and data requirements. Therefore, the methodologies developed for this port economic impact kit and described in detail in Chapter III represent a reasonable compromise based on the expected level of resources available to the majority of users. It is important to remember that each of these systems represents a "snapshot" of a region's economy and interrelationships at a single point in time (the period for which the data were collected) . The direct application of multipliers and models to predict the overall effects of changes in port activity implicitly assumes that the relationships con- tained in the "snapshot" will not change over time because of structural changes in the region's economy (e.g., introduction of new industries), technological changes in products and processes, economies of scale, price effects, interfactor substitutions, or other forces, any of which could, of course, substantially alter economic interrelationships. The analyst who applies the techniques described in this kit must be aware of any changes which have occurred since the data were collected and any which might occur during the period for which forecasts are being made. For example, aggregate relationships between the port industry and the region's economy are likely to be affected by increasing use of container and ro/ro shipments, changes in the mix of shippers, or large increases or decreases in the amount of cargo handled. Since these changes cannot be explicitly incorporated into the estimating procedures described in this kit, it is important to understand the limitations of each of the impact methodologies. 1. Economic Base Multipliers Economic base theory assumes that all economic activity in a region can be divided into two sectors, the export or basic sector which pro- duces goods and services shipped out of the region and the local or non- basic sector which produces goods and services consumed within the region. The size of the non-basic sector is assumed to be a known func- tion of the size of the basic sector. Export markets are considered the prime mover of the local economy. If employment serving this market rises or falls, employment serving the local market is presumed to move in the same direction. The theory further assumes that a region's eco- nomy changes slowly and that, as a result, the ratio of basic to non- basic employment, income, or other indicator is relatively stable over long periods. The principal advantage of the economic base technique is its con- ceptual and operational simplicity. Economic base multipliers can be developed and applied by persons who do not have extensive backgrounds in economics or mathematics. Furthermore, these multipliers can be derived from a relatively small amount of readily available data, using standard procedures to determine "basic" employment in each sector. 11 In many cases it is the very simplicity of this technique which has led to its abuse — most frequently by assuming that employment in cer- tain sectors (typically agriculture, mining, and manufacturing) is basic and the remainder non-basic and using these totals to develop an employ- ment multiplier. In fact, each sector of the economy serves both export and local markets, and these shares must be determined and summed to develop a multiplier. (It should also be added that income is preferable to employment as a measure of economic base relationships, and that ratio analysis can be used to derive employment from income.) In addition, economic base multipliers developed for regions in countries that have a large share of foreign exports will tend to be too high. This can be remedied by a relatively simple procedure, however, as described in the Chapter III discussion on use of the economic base technique for Canadian ports (see Step E:3, page 70). Economic base multipliers are generally best limited to use as des- criptors of present conditions rather than as a basis for projections. The introduction of a large new basic economic activity into the region could have significantly different non-basic effects than the existing aggregate basic sector, so that the previous multiplier would no longer be appropriate. Also, the economic base approach is most appropriate for small regions. As a region's economy becomes more diverse and complex, it contains a larger number of interindustry relationships tied to local demand which may not be affected by export demand. As discussed in greater detail in Chapter III, the special nature of the port industry leads to certain difficulties in the use of the eco- nomic base approach. Most industries whose impacts are analyzed by this methodology are considered wholly basic activities. However, the port industry includes some income and employment which could be classified as basic (export-related) and some which could be termed non-basic (import- related) . Therefore, it is necessary to estimate the basic component of port activity before the economic base multiplier can be used. None- theless, because of the attractiveness of this technique for small ports with limited resources, this technique is suggested in the kit. 2. Interregional Trade Multipliers Interregional trade multipliers are developed from Keynesian income determination theory, which is based on an identity relationship among total income and the various components that constitute it — i.e., con- sumption, investment, government spending, imports, and exports. The technique has structural similarities to input/output methodology, and the results produced are similar to those of economic base techniques. Unlike economic base methods, interregional trade multipliers make speci- fic assumptions about the utilization of income for various purposes and from these estimates determine the effect of a change in exports. Like economic base multipliers, however, interregional trade multipliers do not provide any sectoral disaggregation, and so provide little insight as to the interrelationships of different sectors. Because of the formulation of the underlying model, the inter- regional trade multiplier approach is capable of providing more accurate forecasts of economic activity than the economic base method, but requires far more imformation to do so. As a result, it costs far more 12 to implement than the economic base approach and requires a greater sophistication on the part of the analyst. Properly used, the two methods will provide similar estimates of point-in-time impacts, the type which are to be prepared with this kit. Therefore, a detailed method for using interregional trade multipliers has not been developed. 3. Input/Output Models The basic premise of the input/output model framework is that each industry sells its output to other industries and final consumers (personal consumption, business investment, government spending, and exports) and, in turn, purchases goods and services from other industries and primary factors of production (such as capital and labor) . There- fore, the performance of each industry can be determined by changes in both final demand and the specific interindustry relationships. Simi- larly, the economic impacts of each industry can be expressed in terms of increased demand for industrial production and the primary factors of production. The I/O approach is the most appealing from a theoretical point of view since it allows the sectoral disaggregation of impacts and is thus most sensitive to the multi-industry nature of port activities. The I/O framework enables the analyst to avoid the problems associated with sepa- rating port activities into export and local since, as done in the MARAD study of the U.S. port industry, the port industry can be treated as a single industry in the I/O matrix. Input/output models are generally considered the most theoretically accurate of the several tools available for describing economic interrelationships in a regional or national economy . As might be expected, just as the disadvantages of the economic base multiplier are associated with its simplicity, the disadvantages of the I/O approach are due to its complexity. The amount of data needed to construct an I/O table and the associated time, cost, and technical skill requirements are enormous. Available state or regional I/O tables cannot be used for the area of interest to a port district (usually a single county) because of the likely structural differences in the two areas' economic relationships. Even when an I/O table for a suitable geographic area does exist, it must be used carefully because the data needs of I/O tables generally cause them to be prepared relatively infrequently, and existing tables are often several years out of date. The smaller the region, the more likely this time-related error is to cause significant problems (e.g., a county's interindustry flows could be substantially affected by the opening or closing of one or two large plants). Finally, with the exception of the recent MARAD study, I/O models do not typically treat the "port industry" as defined in this kit as a single component of the interindustry matrix. One possible method for using the I/O approach to estimate the eco- nomic impacts of small ports involves the use of the RIMS (Regional Industrial Multiplier System) data developed from the national I/O model by the Bureau of Economic Analysis. The RIMS procedure currently uses the 1967 national I/O table in conjunction with industrial output infor- mation for counties throughout the nation to derive industry-specific multipliers for regions ranging in size from counties to multi-state 13 areas. Multipliers based on the 1972 national I/O table are expected to be available by mid-1979. Since the technology in the port industry and economic interrelationships in many counties and regions have changed rapidly since the 1967 and 1972 I/O tables were developed, it may not be appropriate to use the RIMS approach by itself. Nevertheless, it may still serve as a useful check on an economic base multiplier as recommended in this kit. D. RELATIONSHIP BETWEEN PORTS AND PORT-DEPENDENT ACTIVITIES Since the purpose of this kit is to assess the full impact of port activities, including port-dependent activities, it is important to exa- mine the nature of the relationship between ports and port-dependent activities. Our review of port economic impact assessments has shown that researchers frequently allocate the total impacts of port-dependent or related activities to the port without any attempt to assess the importance of port availability to these industries. In point of fact, many industries which are generally considered port-dependent from a location decision point of view (coal, grain, iron and steel, lumber and pulp) typically occur at substantial distances from actual port facili- ties (where resources are available) and are port-dependent in that cost- effective access to a suitable port is required. This is most clearly the case when mini-bridge cargos are consi- dered. Here the particular port is probably only selected for conve- nience and its absence would have little, if any, effect on the opera- tions of shippers and receivers whose primary concerns are the most cost- efficient methods of transportation. In these cases, alternative ports would serve as well. While there are undoubtedly some port-dependent activities tied to and unable to survive without a specific port, it is necessary to attempt to assess the importance of a port to its users so that the amount of dependent activity can be appropriately allocated. Even a firm which customarily ships all of its output through the local port might consider only 10% of its sales actually dependent on the particular port; the remaining 90% of its business could be handled through other ports or via alternative modes of transportation. In addition to gauging how dependent the port-dependent industries are, it is also important to look critically at activities which take place on port property. Even though some studies have treated all activ- ities on port property as part of the port industry, it is clear that some of them may not be port-related at all except by accident of loca- tion. As an example, several of the old finger piers in San Francisco are used for parking or storage of recreational vehicles, simple economic uses of otherwise vacant space rather than port-related activities. The economic development role of Washington public port districts leads to the location on port property of industries which may be completely, par- tially, or not at all dependent on port availability. Therefore, we believe it is critical that ports attempt to assess accurately the true dependence of "port-dependent" activities on the local port and reflect these findings in their analyses. 14 E. GEOGRAPHIC AREA OF ANALYSIS In the past, studies of port economic impacts frequently included the concept of the port "hinterland" — a natural market area for cargoes moving through the port. The prevalance of the hinterland concept has declined in recent years, largely as a result of mini-bridge and land- bridge practices in which ports are used as transshipment points rather than primary origins and destinations. Any "hinterland" defined to in- clude the majority of port users would probably be so large that impacts measured for this study area would be meaningless to port district resi- dents and decision makers. Because of the expected uses of this kit, we believe it is appro- priate that impacts be developed for relatively small geographic areas. While a small geographic unit will not encompass all of the economic activity generated by its port, this is the area of analysis most rele- vant to port district residents and decision-makers. Because of the need to prepare an analysis directed toward local concerns and because of the level of geographic disaggregation of available statistical data, we recommend that individual counties or small, economically related groups of counties (e.g., Standard Metropolitan Statistical Areas) be used as study areas for the purposes of this kit. 15 III. DETAILED METHODOLOGY This section is the detailed guide for undertaking a port economic study. Each of the required and optional steps is described in terms of its purpose, data requirements, significant issues, action to be taken, and expected use of results. In those steps which require different data sources or procedures for Canadian ports, appropriate alternatives are described. All other steps should be completed in the same manner for Canadian and U.S. ports. 16 INITIAL CONSIDERATIONS Step A;l PURPOSE: Identify the public purpose of the analysis. All subsequent steps should be completed in a manner to achieve the purpose. DATA REQUIREMENTS: None. RECOMMENDED ACTION: There are three basic uses for documents developed using this kit: in- formation dissemination, policy guidance for port planning and develop- ment, and assessment of the impacts of changes in port activities. The user should identify which of these is appropriate for the application, as the care and level of effort should reflect the purpose. The greatest accuracy would be necessary for those projects that will influence port policy. USE OF RESULTS: Decisions regarding use of the analysis should guide subsequent steps in terms of level of effort and thoroughness of analysis. 17 INITIAL CONSIDERATIONS Step A:2 PURPOSE: Acquire sufficient funding to complete the analysis. • DATA REQUIREMENTS: Approximate number of firms on port property, the number of local firms who ship or receive goods through the port, purpose of the analysis (see Step A:l), wages of persons assigned to the analysis, and the cost of required input data. Knowledge of potential funding sources. RECOMMENDED ACTION: Funding source should be identified. Options normally include: • Internal funding, and • External grant. Any requirements to obtain funding should be identified and met. The cost of analysis fluctuates depending on the size of the port, the desired level of accuracy and comprehensiveness, wage rates of staff, etc. Guidelines for cost estimation are shown in Exhibit A: 2-1. After cost and funding source are identified, necessary funds should be committed to complete the analysis. If appropriate, a separate internal budget should be allocated to the project. USE OF RESULTS: Acquired funds are used to finance completion of the analysis. 18 ITEM Exhibit A:2-l GUIDELINES FOR COST OF ECONOMIC IMPACT STUDY COST Materials 1. Collect general data $ 100 (assumes most data obtained from sources not requiring purchase) 2. Survey costs (survey printing, mailing, etc.) $ 10 per firm 3. Data for estimating indirect impacts Income multipliers $ 150 RIMS multipliers $ 1,100 Staff $1,600 4. Report Production (100 copies) $ 1,000- $ 3,000 $4,500 $4,500 $3,000 $2,500 The numbers reflect expected costs for an average West Coast port that will contact 200-300 respondents. It is assumed that all analytical work is conducted in-house, that the work is done within the timeframe sug- gested in Step A: 3, and that no significant problems are encountered in implementing the methodology. Staff time is estimated at a rate of $2,000 per person month. The total cost will depend on the methodology selected to estimate indirect and induced impacts and the degree of accuracy. To give an example of total costs the following estimates were prepared assuming 200 survey respondents, a $2,000 report production cost, and use of dif- ferent impact methodologies. Income Multipliers RIMS Multipliers Materials $4,250 $5,200 Staff $13,100 $11,600 Total $17,350 $16,800 19 INITIAL CONSIDERATIONS Step A: 3 ■ PURPOSE: Allocate sufficient staff and calendar time to complete the analysis. DATA REQUIREMENTS: Approximate number of firms on port property, the number of local firms who ship or receive goods through the port, purpose of the analysis (see Step A:l), familiarity of persons assigned to the analysis with techni- ques and data to be used, and other commitments of this staff. RECOMMENDED ACTION: Have staff who will complete the project review the entire kit and assess their ability to implement it. Estimates of staff time required to complete the analysis for ports of varying sizes are shown in Exhibit A:3-l. Adjust these estimates based on staff abilities. Calendar time completion estimates are shown in Exhibit A: 3-2. These should be adjusted based on level of staff commitment and ability. Based on estimated staff and calendar time requirements make necessary adjustments in scheduling chart, Exhibit A: 3-2. USE OF RESULTS: The scheduling chart enables the analyst to track completion of tasks and ensure the start and completion of tasks in the required sequence. 20 EXHIBIT A: 3-1 ESTIMATED STAFF TIME REQUIREMENTS TO COMPLETE ANALYSIS The basic variables which affect the amount of staff time required are the number of respondents to the survey, the desired accuracy of the results, and the methodology used to estimate the indirect and induced impacts. Staff time requirements are separated into three staff levels: Senior port management, project director, and technical staff. The table below shows estimated time requirements for ports of different sizes using dif- ferent impact methodologies. Port staff time requirement estimates (person-days) Number of Survey Respondents Project Phase A. Initial Considerations 50 51-100 101-200 201-350 351-500 Senior Management 3 3 4 4 4 Project Director 4 4 6 6 7 Technical Staff B. Survey design Senior Management 1 1 1 1 1 Project Director 10 12 15 18 20 Technical Staff 3 3 5 7 10 C. Survey Distribution Senior Management Project Director Technical Staff D. Survey Result Compilation Senior Management Project Director 3 Technical Staff 7 6 9 12 15 11 18 23 30 21 E. Development of Secondary Impact Multipliers* Senior Management Project Director 10 10 10 10 10 Technical Staff 4 4 6 6 8 F. Calculate Secondary Impacts* Senior Management Project Director 11 11 13 15 17 Technical Staff 4 5 9 10 15 G. Prepare and Disseminate Report Senior Management Project Director Technical Staff 2 2 2 2 2 6 8 8 10 12 1 1 1 2 2 Total Time Senior Management 7 7 8 8 8 Project Director 46 53 64 75 86 Technical Staff 21 27 42 53 72 NOTE: Assumes all optional steps are performed. * Assumes that both income multiplier and RIMS multiplier techniques are used. 22 EXHIBIT A:3 2 CALENDAR TIME SCHEDULE Step No. Description A:1 Determine Public Purpose of Analysis A; 2 Obtain Funding to Conduct Analysis A:3 Allocate Staff Time to Analysis A:4 Determine Scope of Study A:5 Select Geographical Area for Analysis A:6 Prepare Schedule for Project D,„ Completed | | | I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32 34 8:1 Adopt Required Questions for Surveys 8:2 Develop Optional Questions for Surveys 6:3 Adjust Questionnaires for More than One Port in Region B:4 Prepare Coding Instructions for Questionnaires B:5 Prepare Cover Letter for Questionnaires B:6 Identify Firms in the Port Industry B:7 Select Firms for Pretest of Survey Instrument B:8 Select Interview Format B:9 Conduct Survey Pretest and Revise Questionnaire C:1 Advertise Study Prior to Survey Distribution C:2 Prepare List for Respondent Checkoff C:3 Conduct Survey D:1 Code Returned Surveys 0:2 Check Consistency of Survey Responses D:3 Estimate Value of Missing Responses D:4 Collect Port Employment and Income Data 0:5 Tabulate Survey Responses by Industry and Impact Category D:6 Aggregate Survey Responses and Estimate Total Direct Impact E:1 Select Methods for Secondary Impact Analysis E:2 Collect Income and Employment Data for Economic Base Multipliers E:3 Develop Employment Multipliers Using "Concentration" Approach E:4 Develop Employment Multipliers Using "Minimum Requirements" Technique E:5 Develop Employment Multipliers Using "Experience" Technique E:6 Develop "Consensus" Employment Multipliers E:7 Develop Economic Base Income Multipliers E:8 Select Industries for Analysis and Order RIMS Multipliers F:1 Determine Share of Port Industry Income that Is Basic F:2 Estimate Total Income Supported by Port Activities Using Economic Base Multipliers F:3 Estimate Total Sales Supported by Port Activities Using RIMS Multipliers F:4 Compare RIMS and Economic Base Techniques F:5 Update RIMS Estimates F:6 Calculate Other Impact Measures G:1 Prepare Report G:2 Disseminate Report Estimated Date 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32 34 | | | | | | | | | | | | I I I I I I I I I I I I I I I I I I I I I I I 2 5 INITIAL CONSIDERATIONS Step A; 4 PURPOSE : Determine whether port industry activities, port-dependent industry activi- ties, or a combination of both will be addressed in the analysis. DATA REQUIREMENTS: Results of Steps A:l-A:3. DISCUSSION: The majority of port industry firms are classified in SIC 44, Water Transportation, and SIC 47, Transportation Services. This includes tug and towing services, stevedores, marine terminal operators, pilots, vessel operators, freight forwarders, custom house brokers, shipping agents, export/ import firms, and ports. Also included are trucking firms that interface with waterborne carriers. Port-dependent industry firms are those that utilize water transportation. RECOMMENDED ACTION: For applications where the purpose of the analysis incorporates a des- cription of the activities to be included, the stated purpose should be fol- lowed. Otherwise, available time and budget should serve as guidelines. Port industry firms should always be included and treated separately from port-dependent firms. Not including port-dependent firms can often reduce budget and staff time requirements by as much as 60%. USE OF RESULTS: If it is decided to include port-dependent industries and firms then all sets of instructions for those activities must be completed. Otherwise, they can be deleted. 24 INITIAL CONSIDERATIONS Step A: 5 PURPOSE: Select the geographical area within which port industry and port- dependent industry impacts will be measured. DATA REQUIREMENTS: Knowledge of location of port industry and port-dependent industry firms and of political boundaries. RECOMMENDED ACTION: Sometimes the desired geographical boundaries are defined in Step A:l; these should be followed as far as practical. Because most of the secondary data inputs are defined along political boundaries, the use of such regional definitions is convenient and saves time and effort if adopted. For U.S. ports, a county or group of coun- ties comprises the recommended region. For Canadian ports a census metropolitan area would define the region. Some ports will wish to consider only their own area of jurisdiction or actual land ownership. Such a selection can be made and appropriate firms for contact can be identified. However, it is likely that secon- dary impacts will be incorrectly estimated if a port's boundaries are not the same as those mentioned above. USE OF RESULTS: In the final report, measured impacts should be described with reference to the impact area. Also, surveys should be sent only to those firms located in the selected impact area, regardless of whether there might still be port industry or port-dependent industry firms outside its boundaries. 25 INITIAL CONSIDERATIONS Step A: 6 PURPOSE: Prepare a detailed schedule for all tasks that will serve as a guide during completion of the analysis. DATA REQUIREMENTS: Results of Step A: 3 and suggested schedule chart shown in Exhibit A: 3-2. RECOMMENDED ACTION: A flow chart such as the one shown in Exhibit A: 3-2 can be useful for ensuring the proper sequencing of tasks. The tasks in Exhibit A: 3-2 are structured so that the entire analysis can be completed within 7-8 months. It is unlikely that the analysis can be completed in less time if all steps are followed in the recommended sequence. However, a longer time might be required. In such event, all tasks should be rescheduled. The greatest additional time is normally required for those tasks relating to distribution, receipt, and coding of surveys. USE OF RESULTS: The final scheduling chart provides an outline and checklist for the analyst. It should be used throughout the study to ensure on- time com- pletion and proper sequencing of tasks. 26 SURVEY DESIGN Step B;l PURPOSE: Review and adopt suggested questions for port industry survey. DATA REQUIREMENTS: Suggested questions shown in Exhibit B:l-1. RECOMMENDED ACTION: At least the first five questions shown in Exhibit B:l-1 should be included in a survey of port industry firms. They should be reviewed, reworded if the pre-test shows that to be necessary, and coupled with optional questions selected in Step B:2. Sample survey forms are included as Exhibits B:3-l and B:3-2. While questions 6-8 are not absolutely required they should all be included in order to facilitate the checking of responses for consistency and to provide a wider variety of impact measures. USE OF RESULTS: The suggested questions form the base of the survey instrument for firms in the port industry. The survey instrument is the primary means for obtaining information about these businesses. PORT DEPENDENT INDUSTRY: The suggested questions for the port dependent industry survey are the same as those for the port industry with the addition of those listed in the bottom of Exhibit B:l-1. The recommended action for the use of these questions is the same as that for the port industry. 27 EXHIBIT B:l-1 SUGGESTED SURVEY QUESTIONS PORT INDUSTRY 1. Description of waterborne commerce related activities (used in defining the firm by industry) . 2. On what date does your fiscal year end? (signals differences in time periods for which data are assembled) . 3. Annual sales or revenues (used in conjunction with input/output multipliers and for determination of sales impacts) . 4. Average annual number of employees (used in determination of employ- ment impacts and in expansion of survey results to the total uni- verse of waterborne commerce related firms) . 5. Annual payroll (used in determination of income impacts and in development of income multipliers) . 6. Annual purchases of goods and services from local region (used to provide a direct measure of local impact) . 7. Types of goods and services purchased locally (assists in deter- mination of basic/non-basic distinction of firms and industries) . 8. Taxes paid to local governments (used to develop tax impacts) . PORT-DEPENDENT INDUSTRY In addition to the port industry questions listed above, the following quetions are required: 1. Percentage of total product movements through local ports (assists in the allocation of impacts to the region and ports within the region) . 2. Percentage of total supplies moving through local ports (assists in the allocation of impacts to the region and ports within the region) 3. What alternative transport facilities could be used (assists in the determination of the dependence of the region's industries upon the port) . 4. What percentage of the firms' s sales would be lost without the port's facilities (determines dependence upon port facilities in quantitative terms). 23 SURVEY DESIGN Step B;2 PURPOSE: Define and incorporate optional questions into the survey form to be sent to firms in the port industry. DATA REQUIREMENTS: List of optional questions shown in Exhibit B:2-l. RECOMMENDED ACTION: Optional questions are used both to develop more detail about the issues raised in the required questions and to develop information about topics of interest to the port but not directly required for an economic impact assessment. Several possible optional questions are shown in Exhibit B:2-l. The analyst should incorporate these questions only if a specific use can be made of the answers. Some of the optional questions would not be mean- ingful for every firm — e.g., the primary commodities shipped or received would have little meaning for a derrick barge firm. The analyst can also develop additional questions which are of particular relevance to the port under analysis. These should be placed so as not to interrupt the flow of the required questions. In general, we recommend that the shortest possible questionnaire form be used as it is more likely to draw a response. USE OF RESULTS: The optional questions combined with the required questions form the survey for firms in the port industry. The survey instrument is the primary resource for obtaining information about these firms. PORT-DEPENDENT INDUSTRY: Optional questions that could be used in the port-dependent industry survey questionnaires are also provided in Exhibit B:2-l. It should be remembered that longer and more involved questionnaires will lower the response rate. This is true of both the port industry questionnaires and port-dependent industry questionnaires, but the greater number of re- quired questions for port-dependent industry questionnaires increases the problem. 29 EXHIBIT B:2-l POSSIBLE OPTIONAL QUESTIONS Please provide an estimate of your employment in terms of the following: Executive/Professional % Management % Operations % Please provide an estimate of the percentage of your employees living in: City % County % Region % 3. Three most common cargo origins: Location Percent Three most common cargo destinations: Location Percent 4. Port facility (ies) utilized most frequently (i.e., pier, terminal, shed, crane, etc.): Primary advantage of this particular port for movement of your goods (rank 3 elements): Service Efficiency Convenience Cost Landside Transportation ^___ Market Proximity Other (please comment) 30 Arthur D Little Inc. EXHIBIT B:2-l (Continued) 6. Have you any plans for a significant modification of port usage in terms of cargo volume, frequency of facility use, type of landside transportation, etc.? 7. In considering existing port operations, what area of service should be expanded or improved? Waterside Landside (e.g., storage, cargo handling, transportation services, etc . ) The primary commodities you ship or receive through ports in the region are: a. Commodities shipped b. Commodities received 9. If port facilities were not available, what per- centage of your sales/revenues and jobs in the region would be lost? a. Sales/Revenues % b. Jobs % 31 Arthur D Little Inc. SURVEY DESIGN Step B:3 PURPOSE: Adjust the survey questions for cases where there is more than one poi. in a region. DATA REQUIREMENTS: Identification of all public and private ports in the region selected for the analysis in Step A:5. RECOMMENDED ACTION: If there is more than one port in a region and firms in the port industry work with more than one port in the region, it is necessary to have them allocate their impacts among ports. If information about firms' activities for all ports is desired, the questionnaire should have separate reporting spaces for their activities with each port or at least an additional question that asks for the percentage of total business associated with each port. If information for only one port is desired, that should be so noted at the beginning of the survey in bold type. A recommended survey form in shown is Exhibit B:3-l. An alternative format that would be useful in obtaining information from firms whose management is reticent about providing information would include multiple choice responses. For example, employment categories of 1-5, 6-15, 25-40, etc. could be provided on the survey form, the appropriate category to be circled by the respondent. Suitable multiple choice categories could be constructed for almost every question. USE OF RESULTS: The modified survey is used in all subsequent steps where appropriate. PORT-DEPENDENT INDUSTRY: Step B:3 is unchanged for the port-dependent industry, although a sepa- rate survey form is included as Exhibit B:3-2. 32 EXHIBIT B-.3-1 PORT INDUSTRY QUESTIONNAIRE ALL INFORMATION WILL BE HELD IN STRICT CONFIDENCE Company Name City Respondent Name Telephone Title ) 1. Briefly describe your waterborne commerce-related operations in the region: PLEASE ANSWER QUESTIONS 2 THROUGH 8 USING INFORMATION FROM YOUR LATEST FISCAL YEAR. PLEASE PROVIDE INFORMATION ONLY FOR THOSE ACTIVITIES WHICH OCCURRED IN THE REGION. 2. On what date does your fiscal year end? 3. Annual sales or revenues from operations: £_ Average annual number of full-time equivalent employees: detailed information, answer the following: a. Average annual number of permanent full-time employees: b. Average annual number of permanent part-time employees: c. Average annual number of contract employees: 5. Annual payroll for permanent and contract employees (including benefits): *_ 6. Annual purchases of goods and services from regional businesses: a. Goods: $. b. Services: :£ c. If you lease, annual rental payment: S_ 7. Type of goods and services purchased from regional businesses: 8. Annual taxes and fees paid to local governments: a. Property tax: $. b. User fees for public services (water, trash disposal, port fees, etc. c. Sales tax: $. d. Other local government taxes and/or fees: §. ): 1 9. What percentage of your operations are provided for or conducted in: a. Port of San Francisco _J % Port of Oakland 33 OR, if you have more Other Please DO NOT Write in this Space 6. a. b. i.a. b. c. d. 9.3. b. c. EXHIBIT B:3-2 PORT-DEPENDENT INDUSTRY QUESTIONNAIRE ALL INFORMATION WILL BE HELD IN STRICT CONFIDENCE Company Name City Respondent Name Telephone Title 1. Briefly describe the business in which your firm is involved: PLEASE ANSWER QUESTIONS 2 THROUGH 13 USING INFORMATION FROM YOUR LATEST FISCAL YEAR. PLEASE PROVIDE INFORMATION ONLY FOR THOSE ACTIVITIES WHICH OCCURRED IN THE REGION. 2 3. 4. On what date does your fiscal year end? Annual sales or revenues from operations: §L Average annual number of full-time equivalent employees: detailed information, answer Vie following: a. Average annual number of permanent full-time employees: . b. Average annual number of permanent part-time employees: c. Average annual number of contract employees: OR, if you have more 5. Annual payroll for permanent and contract employees (including benefits): *_ 6. Annual purchases of goods and services from regional businesses: a. Goods: 5 b. Services: §. c. If you lease, annual rental payment: 7. Types of goods and services purchased from regional businesses: 8. Annual taxes and fees paid to local governments: a. Property tax: $. b. User fees for public services (water, trash disposal, port fees, etc.): sL c. Sales tax: * d. Other local government taxes and/or fees: § i The primary commodities you ship or receive through ports in the region are: a. Commodities shipped b. Commodities received 10. Approximately what percentage of your total product movements are shipped through the following ports in the region: a. % b. % c. °A 11. Approximately what percentage of your supplies move through the following ports in the region: a. (2. b. i° c. % 12. If you were denied the use of these port facilities, which, if any, alternative ports (other than those listed above) or means of transportation would you use to ship your products and/or receive your supplies: 13. If these port facilities --were not available, what percentage of your sales/revenues and jobs in the . region would be lost: a. Sales/revenues: . •"% Jobs: % 34 Please DO NOT Write in this Spac 6. a. b. c. 7. 8.3. b. c. d. 9. a. lO.a. b. 11. a. b. c. 12. 13. a. b. SURVEY DESIGN Step B:4 PURPOSE: Prepare instructions for coding survey responses. DATA REQUIREMENTS: Completed survey form, Standard Industrial Classification Manual , and Exhibit B:4-l. RECOMMENDED ACTION: The responses to the survey must be summarized for use in the analysis. The coding instructions are a specific format for reporting survey res- ponses. In general, coding instructions are straightforward — e.g., employment is coded as reported in person-years of employment, income is coded as dollars (rounded to the nearest thousand) , etc. If an SIC code was not provided by the respondent, the most difficult part of the coding is the allocation to an activity. Here the Standard Industrial Classification Manual and Exhibit B:4-l are used. The analyst should use the SIC manual to determine if the activity is part of the port industry and, if so, select the appropriate classi- fication and if not, disregard the return. Coding instructions for the other required questions are also listed in Exhibit B:4-l. The analyst should develop coding categories for any optional questions used. USE OF RESULTS: The coding methods are used when the questionnaires are returned and the results tabulated (see Step D:l). PORT-DEPENDENT INDUSTRY: Coding instructions for port-dependent industry surveys are presented in Exhibit B:4-2. Those questions that would have similar responses in port industry and port-dependent industry surveys can be coded using the in- structions found in Exhibit B:4-l. 35 EXHIBIT B:4-l CODING INSTRUCTIONS - PORT INDUSTRY SURVEY Responses should be coded in the box along the right hand edge of each survey form. The following numbering system corresponds to the numbering on the attached survey form (Exhibit B:3-l). 1. Interviewee industry - use four-digit Standard Industrial Classifi- cation Manual codes from the 1972 edition. 2. Fiscal year ending date - code month, day, and year in six places, e.g., 123178 being the code for December 31, 1978. Insert zeros if necessary — e.g., 060178 being the code for June 1, 1978. 3. Sales or revenues - code amount in thousands of dollars or dollars, but be consistent. Coding in thousands of dollars reduces the key- punching requirements if the data are to be manipulated by computer. 4. Employment - code the number of employees as given. 5. Payroll - code amount as thousands of dollars or dollars, consistent with question 3. 6. Goods and services purchased; lease payments - code amounts on appro- priate lines as thousands of dollars or dollars, consistent with question 3. 7. Types of goods and services purchased - use Standard Industrial Classification Manual to determine industry that provides the listed goods and services; codes should be as detailed as is prac- tical from the information given (i.e., only one- or two-digit codes may be possible in some cases) . Space is provided for four codes on the sample survey form. 8. Taxes and fees - code amount as thousands of dollars or dollars, consistent with question 3. 36 Arthur D Little Inc. EXHIBIT B:4-2 CODING INSTRUCTIONS - PORT-DEPENDENT INDUSTRY SURVEY Responses should be coded in the box along the right hand edge of each survey form. The following numbering system corresponds to the numbering on the attached survey form (Exhibit B:3-2). Questions 1-8 should be coded in the same manner as questions 1-8 of the port industry survey; see Exhibit B:4-l for instructions. 9. Commodities shipped and/or received - A pre-existing coding system, such as the Corps of Engineers' Waterborne Commerce commodity classification o_r the Bureau of Census Schedule A and B codes, can be used or a less detailed port-specific coding system can be established. 10. Distribution of product movements among ports - percentages should be coded as given — e.g., 15% coded as 15. 11. Distribution of supply movements among ports - percentages should be coded as given — e.g., 15% coded as 15. 12. Alternate ports or transport modes - the simplest coding scheme would be: 1 — rail 2 — truck 3 — air 4 — barge 5 — other ports This could be expanded by having separate codes for various other ports, various rail lines, various truck lines, etc. 13. Percentage of revenue lost without port - percentages should be coded as given — e.g., 15% coded as 15. 37 Arthur D Little !nc SURVEY DESIGN Step B:5 PURPOSE: Develop a cover letter to port industry firms that will explain the survey and generate a high response level. DATA REQUIREMENTS: None. RECOMMENDED ACTION: The survey should be accompanied by a one-page cover letter that: • Explains the purpose of the study, • Promises confidentiality of information, • Stresses the importance of a prompt response and sets a cut-off date, • Expresses thanks for cooperationand promises a copy of the study results, • Provides a contact person at the port who can ex- plain the meaning of the questions in greater detail and answer other inquiries about the study, and • Is signed by the chief administrative officer of the port or the chairperson of the port commission. • Provides the potential respondent with a return address, especially if the completed questionnaires are not to be sent to the port. To prevent the possibility of public disclosure, it may be wise to have the forms sent to a private individual or firm, such as the port's legal counsel. Exhibit B:5-l shows an example of a letter having these characteristics, 38 USE OF RESULTS: The letter should be attached to and sent with all mail questionnaires. If either personal or telephone surveys are used, a brief opening state- ment conveying the same information should be incorporated. PORT-DEPENDENT INDUSTRY: Exhibit B:5-l is suitable for the port-dependent industry firms as well as the port industry firms. The guidelines suggested under "RECOMMENDED ACTION" should be adhered to for both the port industry and port- dependent industry cover letters. 39 EXHIBIT B:5-l SAMPLE SURVEY COVER LETTER Dear Mr. Jones: The Port of Daly City is engaged in a study to determine the economic benefits of waterborne commerce activity on the San Francisco Bay region. The results of this study will benefit the Port of Daly City and consequently benefit firms and individuals doing business with the Port. The collection of basic economic data from local firms is the most critical component of this study. The questionnaire attached to this letter contains the questions necessary to obtain data that will allow employment, income, and local tax impact measures to be derived. All of the major port-related firms in the region are being sent this questionnaire. All of the information derived from the returned questionnaires will be held in the strictest confidence by the Port. The study's findings will be presented in aggregated forms — no information for an indivi- dual firm will be released or used in any way other than in the deter- mination of the highly aggregated impact measures (e.g., total income, total employment, and total tax impacts). If you have any questions concerning the questionnaire or how you should design your responses to the questionnaire, please contact Mr. Delbert Doe at the Port of Daly City. His telephone number is (415) 987-6543. Please return the completed survey by August 12. Thank you very much for your cooperation. Completion of this question- naire will assist Port efforts in providing improved services for its clients and in increasing economic benefit to the surrounding region. A copy of the results of this study will be sent to you in approxi- mately six months when the study is completed. Sincerely, Bill "Bilge Rat" Brown Director, Port of Daly City BBRB/ck Attachment 40 Arthur D Little Inc. SURVEY DESIGN Step B:6 PURPOSE: Identify firms and potential respondents within the port industry. DATA REQUIREMENTS: Lists of potential respondents from the following sources: knowledge of port employees (e.g., the traffic manager's client list), West Coast Marine and Transportation Directory, Dun & Bradstreet Directory, tele- phone directories, business license records, local Chamber of Commerce. RECOMMENDED ACTION: The purpose is to identify all respondents within the geographic area to be covered by the study. Therefore, it is necessary to review several sources to insure that all possible respondents are listed. Standard Industrial Classification (SIC) categories for the port industry are given in the glossary. Separate lists of firms in each category (and even for firms in subcategories — e.g., tug operators, barge operators, etc.) should be prepared from the above sources. Any information avail- able about the firms — e.g., annual sales, employment, etc. — should also be included on these lists. The firms should be ranked within each category according to number of employees, from largest to smallest. If employee counts cannot be ob- tained from published sources or estimated from experience, phone calls should be made. Lists should include names, addresses, telephone numbers, and a contact person within each company. Because of the information to be requested the contact person should be involved in either the financial or opera- tional affairs of the company. USE OF RESULTS: The lists will be used both as master lists from which survey respondents will be drawn and as the basis from which the impacts represented by sur- vey responses will be translated into total direct impacts. 41 PORT-DEPENDENT INDUSTRY: A complete list of port-dependent industry firms will generally be more difficult to obtain than a similar list for the port industry firms. This is especially true if containers are moved through the port since there may be less certainty concerning the identity of the shipper. The same sources of information should be used to obtain the port-dependent industry list. 42 SURVEY DESIGN Step B:7 PURPOSE: Select respondents from the list prepared in Step B:6 for the pre- test of the survey instrument and for implementation of the survey. DATA REQUIREMENTS: The list of potential respondents in each category developed in Step B:6. RECOMMENDED ACTION: For the pre- test some respondents in all categories should be contacted. We suggest distribution of approximately 15-20 pre- test questionnaires. USE OF RESULTS: The list produced in this step will be used for the pre- test. PORT-DEPENDENT INDUSTRY: The recorjnended action should be followed for the port-dependent industry survey and the port industry survey. 43 SURVEY DESIGN Step B:8 PURPOSE: Determine whether mail, telephone, or personal interviews, or a com- bination of these methods will be used for both the pre-test and the actual distribution of survey instruments. DATA REQUIREMENTS: None. RECOMMENDED ACTION: There is a trade-off among the alternatives in terms of cost versus per- centages of response and accuracy of responses. Mail interviews are the least expensive but also have the lowest response rate and tend to have the least accurate responses. Personal interviews are the most expensive but have substantially higher response rates and generally provide more accurate information. Telephone interviews strike a middle ground. If telephone or personal interviews are to be used the interviewer must have some knowledge of the purpose of the survey and of the appropriate way for recording information. Otherwise, the port's reputation might be damaged and inaccurate information collected. Because the justification for a pre-test is to determine whether the sur- vey instruments can be understood and completed by respondents, some of the pre-test surveys should be conducted in person. However, if it is decided that mail surveys will be used for the main surveying effort, some pre-test surveys should be mailed and their responses compared with those developed during personal interviews to assess the appropriateness of the survey and the respondents' ability to comprehend it. A port's selection of a survey method should be guided by its budget limitations, the number of potential respondents, and the desired degree of accuracy. Where the number of potential respondents is small, a personal survey should be conducted because of the better response rate and greater accuracy it should generate. The next best alternative is a telephone interview. If personal or telephone surveys are to be used, the survey instrument should be distributed in advance, in order to allow the respondent to collect the necessary information. An appointment for the personal or telephone interview should be made. 44 USE OF RESULTS: Both the pre-test and the actual survey will be conducted according to the decisions made in this step. PORT-DEPENDENT INDUSTRY: The recommended action should be followed for the port-dependent industry survey and the port industry survey. 45 SURVEY DESIGN Step B;9 PURPOSE: Conduct the pre-test and alter the questionnaire as required. DATA REQUIREMENTS: The survey for port industry respondents developed in Steps B:2 and B:3. Also, decision reached in previous steps regarding method of interviews, etc. DISCUSSION: The purpose of the pre-test is to insure that respondents are capable of understanding the questionnaire as prepared and that they will provide the desired information. It is imperative that a pre-test be conducted since without one, a survey that results in incorrect information or does not generate a response from the firms surveyed could be conducted. RECOMMENDED ACTION: The pre-test should be conducted during one- two week period using the interview techniques selected during Step B:8. Potential respondents should be informed through a variation in the cover letter that this is a pre-test and that the port is as interested in their comments about the questionnaire and the method of conducting the survey as it is in their actual responses. If some of the pre-test firms are to be contacted by mail, the port should telephone those who have not responded within a reasonable time and ask them to complete the survey. As the pre-test forms are completed, they should be reviewed to determine which of the responses are similar to those anticipated by the port and any differences should be resolved by recontacting the respondent. If the responses to the pre-test are correct, they should be retained for tabulation along with the responses to the complete survey. This is par- ticularly important for small ports where the number of potential res- pondents is limited and large percentages of them may be contacted during the pre-test. 46 USE OF RESULTS: If respondents are found to have any problems understanding and com- pleting the survey forms regardless of the method of interview, the sur- vey forms should be revised to correct the problems. If some of the dif- ficulty relates to the unwillingness of the pre-test firms to divulge some information, the port must determine whether the possibility that some firms will refuse to complete any part of the questionnaire because of inclusion of this information is greater than the value that receipt of the information from some firms might have for completion of the analysis. PORT-DEPENDENT INDUSTRY: The recommended action is the same for the port-dependent industry. The pre-tests can be conducted concurrently to compress the time required for the study. 47 SURVEY DISTRIBUTION Step C:l PURPOSE: Advertise the study and its value so that potential survey respondents will know about it before the surveys are distributed. DATA REQUIREMENTS: None. DISCUSSION: A problem with the use of a survey to collect information about port industry and port-dependent industry firms is that the response rates to such surveys are usually low. By promoting the study before the surveys are distributed, the port can generate interest in and knowledge of the study, thereby increasing the response rate. RECOMMENDED ACTION: If the port has a newsletter, mention of the study and survey should be made in the edition prior to the start of the survey. Use of other media, such as newspapers, etc., is warranted if the size of the survey is large or as part of a general promotion effort. A local chamber of commerce or business association might also be an appropriate forum for promoting the study. USE OF RESULTS: If successful, the publicity will generate interest about the study and lead to a higher response rate for the survey. 48 SURVEY DISTRIBUTION Step C:2 PURPOSE: Prepare a list of respondents selected from those identified in Step B:6 to be used to list contacts, responses, etc. DATA REQUIREMENTS: The list of potential respondents selected in Step B:6. DISCUSSION Since follow-up contact with firms that do not immediately respond to the survey is recommended it is necessary to prepare a list which can be used to monitor the dates of survey distribution, follow-up contacts, survey receipt, and recontact with the firm to check responses. RECOMMENDED ACTION: The number of questionnaires to be distributed for the full survey should be determined through consideration of the number of firms on the list, the expected response rate, the time and money available for the study, and the desired accuracy (confidence) of the estimates provided. In other studies the response rate of firms in the port industry to mail questionnaires has been about 15-50%, depending on the classification of the business, length of the survey, follow-up efforts, etc. The sample size of a survey does not necessarily influence the accuracy of the survey; more correctly, it affects the confidence the researcher has in the accuracy of the derived estimates. Generally, the more in- formation obtained from the survey procedure the more confident the researcher is that the sample information accurately reflects the total population. The level of confidence the researcher has concerning the accuracy of a given sample is generally expressed in percentage terms. For instance, a 95% confidence interval is a range of values — e.g., number of employees — between which the average value will fall 95% of the time. This means that if new samples were randomly drawn and average values computed, the average value would fall in that range 95% of the time. The number of respondents necessary to achieve a 95% confidence level (or any other confidence level) is dependent upon the distribution of values from which the researcher is sampling. Without having some information 49 about the characteristics of this distribution, it is impossible to determine the minimum number of respondents to achieve a given confidence level. For surveys aimed at studying a very large population, 30 respondents called serve as a bare minimum for the researcher. Assuming a 15-50% response rate 60-200 questionnaires would have to be distributed. Ob- viously, for industries with less than 30 firms, questionnaires should be sent to all firms in the industry and extra effort expended in assuring a high response rate. For the rare cases in which there are larger popula- tions judgment must be utilized in determining the number of surveys to be mailed and the amount of effort to be expended in assuring a high res- ponse rate. If the researcher feels that the firms within a given popu- lation (i.e., industry) are all similar in size and operation then a smaller sample (less than 30 firms surveyed) can be justified. If, on the other hand, there is a wide disparity in the size of firms within an industry more than 30 surveys must be collected and more care taken in their analysis. In the case of a large number of firms within an industry, those to be sent a survey should be randomly selected — e.g., assign each firm a number and then select firms through the use of a random number table. In such cases, the accuracy of the results can be improved if all those firms thought to account for a significant percentage of the business are contacted outside of the randomly generated list. The port should prepare two separate charts, one for the port industry and one for port-dependent firms, which will allow the port to keep dates of important events in the survey. An example is shown in Exhibit C:2-l. When any of the events mentioned in the form occurs, it should be marked in the chart. USE OF RESULTS: The chart should be used throughout Parts C and D to record the date of events concerning the surveys. It can also be used to determine the percentage responses to the survey. 50 EXHIBIT C:2-l SAMPLE SURVEY RESPONSE CHART FIRM/CONTACT/PHONE # SURVEY DISTRIBU- TION DATE FOLLOW-UP CONTACTS SURVEY RECEIPT DATE RECONTACTS 1. Acme Trucking Willie Smith 849-6573 8/4 8/13 8/20 8/22 9/5 2. JGV Barge Co. Hector Viez 547-1100 8/4 8/13 8/19 *** 3. Bay Tug Co. Simon Peet 445-2300 8/5 8/14 8/22 NR 4. Chang's Chandler ing James Chang 926-4210 8/6 8/15 8/22 8/29 *** 5. Union Stevedores Wilma Jones 392-1000 8/6 8/15 NR - No response *** - Recontact not necessary 51 SURVEY DISTRIBUTION Step C:3 PURPOSE: Distribution of the survey instrument and assurance of the highest possible response rate. DATA REQUIREMENTS: The list of respondents selected in Steps B:6 and C:2. DISCUSSION: This step is very similar to Step B:9 with the exception that more effort should be extended to assure a high response rate, especially among firms that constitute a large share of employment in their industries within the region. RECOMMENDED ACTION: The survey should be distributed in accordance with Step B:9. It is ex- pected that a mail survey will be used for the majority of firms, although firms that are particularly large, important, or sensitive should be con- tacted personally. An intermediate alternative would be telephone inter- views, but it should be remembered that they are more costly than mail surveys. The cover letter states that the surveys should be returned within one to two weeks. Firms that have not returned surveys by the specified cut-off date should be recontacted by telephone to ascertain the reason the survey was not returned. If the firm is having difficulty interpreting the ques- tionnaire, efforts should be made to explain what further information is required. If the firm does not want to divulge information it believes is sensitive, an effort should be made to explain that the information will be aggregated so that individual firms cannot be identified. It may be impossible to convince certain firms that this information will not be misused. In this case, an effort should be made to obtain a minimum amount of information. This would include the activities of the firm, so as to determine the SIC code, and sales/revenues employment and payroll data. This will allow employment and payroll impacts to be calculated. It is most efficient to attempt obtaining the reduced amount of information during the follow-up telephone call to firms that did not respond to the interview. 52 *- USE OF RESULTS: These surveys will provide base data for compilation of direct impacts and the calculation of secondary impacts. 53 SURVEY RESULT COMPILATION Step D:l PURPOSE: Put survey responses in a form more convenient for "by hand" and computer analysis. DATA REQUIREMENTS: Completed survey questionnaires, Standard Industrial Classification Manual , and Exhibits B:4-l and B:4-2 (for the port-dependent industry surveys) . DISCUSSION: In order to reduce errors in the tabulation of data from a large number of firms, questionnaire coding is recommended. It is necessary if the responses are to computerized. RECOMMENDED ACTION: It is strongly recommended that the coding strip along the right hand edge of the questionnaire be used to record responses. This permits rapid keypunching of the data if computer tabulation is to be used. Mis- sing data can be noted with paper clips along this edge. The coding instruction provided in Step B:4 should be followed in this step. USE OF RESULTS: The coded questionnaires will be checked and modified in the following steps. Any changes in responses or additional responses should be coded using the format of Step B:4. 54 SURVEY RESULT COMPILATION Step D;2 PURPOSE: This is an optional, but strongly recommended , step aimed at checking the consistency of responses within individual surveys and among surveys. DATA REQUIREMENTS: Coded survey questionnaires; additional financial information for speci- fic firms; general information for individual industries from government documents — e.g., income/employee ratios. DISCUSSION: Consistency checks at this stage of the analysis are important for two reasons. First, checks must be conducted to prevent misinterpretations from being carried through the remainder of the analysis; questions that have been answered incorrectly must be found and the reason for the mis- interpretation must be determined to make certain that the survey is functioning as anticipated. Second, it is critical that errors be cor- rected so that they are not carried through the remainder of the analysis. A variety of checks can be instituted; the extent of testing for con- sistency will vary according to the time and money allocated for the study, the confidence the researcher has in the accuracy of the survey responses, and the quantity and quality of corroborating information that can be obtained for the respondent firms. RECOMMENDED ACTION: The following consistency checks can be undertaken: 1. A comparison of annual sales or revenues (3) with the sum of payroll (5) , purchases (6a,b,c) , and taxes and fees (8a,b,c,d) . If the numbers are widely divergent the firm should be recontacted to determine if the disparity is due to business con- ditions or error in completion of the survey. 55 2. A comparison of the average annual wage (response 5 divided by response 4) with published wage informa- tion for the specific industry. Industry specific wage information (for production workers) is obtain- able from Employment and Earnings published monthly by the U.S. Department of Labor, Bureau of Labor Statistics (for Canada — Employment , Earnings, and Hours published by Statistics Canada) . Obviously full-time equivalent employees would have to be used in this determination. Wide divergence would point toward recontact with the respondent firm. 3. The responses of firms within an industry, espe- cially firms of similar size, should be compared to determine divergences. Any "oddities" should be recontacted to assure response accuracy. USE OF RESULTS: The recommended actions listed above should be used to check the accuracy and consistency of the responses so that errors are not carried through the analysis. Recoding of the questionnaires should be done where necessary. 56 SURVEY RESULT COMPILATION Step D;3 PURPOSE : Estimate the value of missing responses on partially completed question- naires. Incomplete responses reduce the quality and quantity of the data available for analysis. DATA REQUIREMENTS: Coded survey questionnaires. DISCUSSION: This step necessitates the use of a great deal of judgment. Only basic guidelines can be provided with respect to the estimation of missing data. Above all it should be kept in mind that the aim of this step is to retain information that would otherwise be lost if a partially com- pleted questionnaire had to be discarded because of missing responses. If the questionnaire has very little information (i.e., many uncompleted portions) , attempts at estimating the non-responses should be forgone. In addition, firms which are unique, because of either their size or pro- duct/service rendered, should not have missing questions estimated unless there is a strong feeling that the estimate is reasonably accurate. RECOMMENDED ACTION: It should first be determined that estimating missing responses is of sufficient importance to justify the effort. Often it is more advan- tageous to discard the questionnaire; if the response rate is high no problem will be created by this action. There are two approaches to estimating missing information after ex- hausting the possibility of obtaining the information directly from the firm. The first approach is to obtain the information from other sources — e.g., total sales and employment could be obtained from the firm's latest 10K form (available upon request) if the firm is required to sub- mit the form, from its annual report, or from Dun and Bradstreet's Million Dollar Directory , and property tax information could be obtained from the county assessor's office in the United States or from the Attorney General's Department in Canadian provinces. 57 The second approach entails the use of information from other completed surveys to estimate missing information on partially completed response forms. If, for instance, payroll data were not presented for firm X, it is possible that the average wage of firm Y employees could be multiplied by the number of employees listed for firm X in order to derive a total payroll for firm X. Great care must be taken when estimating missing information in this manner, however. If possible, average values derived from a number of similar firms should be used rather than information from a single firm. In addition, the firms should be in the same indus- trial category and of approximately the same size. This procedure should not be employed in the derivation of property tax information, rental payments, or services purchased from regional businesses (unless the researcher can be certain that the same share is purchased locally) . USE OF RESULTS: This step acts to reduce the number of questionnaires that must be sent out and prevents the potential loss of information from discarding par- tially completed questionnaires. After coding the changes the question- naires are ready for tabulation in Step D:5. 58 SURVEY RESULT COMPILATION Step D;4 PURPOSE: Collect employment and income information for the port conducting the economic impact study. DATA REQUIREMENTS: The port's most recent annual auditor's report supplemented by employment data. DISCUSSION: The operation of the port itself represents a significant impact in the local community; therefore, similar information to that obtained from the surveys must be collected and coded. RECOMMENDED ACTION: The short survey form should be completed with the port operation being treated as a firm. Care must be taken to exclude non-waterborne commerce related costs and employment. These would include costs, revenues, and employment attributable to aviation or airport operation, industrial land leasing, and non-waterborne commerce related economic development activi- ties. There will be difficulty in separating out non-waterborne commerce related costs and especially employment since administrative activities are often not separated into waterborne commerce versus non-waterborne commerce categories. Attempts should be made to estimate the share of time spent on non-waterborne commerce related activities for individuals whose activities span many areas. USE OF RESULTS: After this information is coded onto a survey form the information will be tabulated along with other survey information in the following step. 59 SURVEY RESULT COMPILATION Step D;5 PURPOSE: Tabulate survey results by industry and impact category as a prelude to the determination of the direct economic impacts, including employment, income, and tax impacts. DATA REQUIREMENTS: Coded surveys separated by industry (i.e., type of firm) and ordered by number of employees; original list of firms from which the sample was drawn; and the latest issue of County Business Patterns published annually by the U.S. Department of Commerce (obtainable from a local library, college library, or directly from the U.S. Government Printing Office) . DISCUSSION: This tabulation step must be conducted prior to the final tabulation of the direct impacts since it is necessary to use information from the sur- veys to estimate the non-surveyed firms' sales, employment, payroll, taxes, etc. This is the purpose of doing a sampled survey of firms rather than an exhaustive survey of all of the waterborne commerce related firms. If the survey has been correctly administered and the response rate was high then a large percentage of the total waterborne commerce related economic activity will have been accounted for in the survey responses. RECOMMENDED ACTION: Partition the original master list of firms by industry into employment size classifications (by industry) . An example of a suitable parti- tioning is used in County Business Patterns. 1-4 Employees 5-9 Employees 10-19 Employees 20-49 Employees 50-99 Employees 100-249 Employees 250-499 Employees 500-999 Employees 1000+ Employees 60 The County Business Patterns breakdown of firms by employment size class can also serve as a useful check on the total number and size of firms in the port industry or port-dependent industry. There is no Canadian pub- lication that provides similarly formatted information, so this optional checking procedure cannot be performed in Canada unless locally produced listings of firms and their employment are available. Note that the number and size of firms could have changed from the publication of the latest County Business Patterns and compilation of the original master list of firms in Step B:6. There is also a potential technical problem with County Business Patterns . If a firm does not have individuals on the payroll during the time that the County Business Patterns' survey of firms is conducted, but has a payroll during other times during the year the firm will be lumped into the "1 to 3" category. The completed surveys should be partitioned in the same manner — i.e., by industry and employment size class. This allows a comparison of the survey sample with the total population of firms by industry and employ- ment size class. Smaller regional economies will, of course, demand more aggregated industrial and employment categories. A tabulation of survey responses would then be conducted by industry and employment size class. This would allow average values of the responses — e.g., sales, employment, payroll, and taxes to be determined. These average values would be utilized as surrogate values for firms that were not surveyed, thereby "expanding" the survey information to the entire universe of firms. It is critical that the industry and employment size classes be constructed to reflect small differences among the firms. The finer the categories, the more accurate the impact study; unfortunately, increasing the number of categories also increases the number of firms that must be surveyed. In addition, it is also extremely important that the larger firms be surveyed since the accuracy of information obtained about them has a much greater effect on the final economic impacts than information concerning smaller firms. In industry categories for which no questionnaires were returned, the analyst has three alternatives: (1) to attempt to obtain information through telephone or personal interview if these have not been tried; (2) to estimate sales and payroll from employment by using state (province) or national sales/employee and payroll employee ratios; or (3) to disregard the impact from that industry category. In the special case where there is more than one port in a region the step involving expansion of survey results to the entire universe of firms must be followed by an allocation to specific ports. This can be accomplished in a variety of ways. The simplest, although least desir- able, method is to allocate a firm's employment income, etc., by the relative tonnage share of cargo moving through each port. If firms could be disaggregated by the type of cargo they are involved with (containers, etc.) a similar procedure could be done with the relative tonnage share of each generic type of cargo. In a similar manner, firm's dealing with vessels only (i.e., tug companies) could have their estimated responses 61 allocated by the number of vessels calling at each port. A completely different allocation method uses the "average" allocation derived from response to the survey question asking for the percentage of business attributable to each port. All of these allocation methodologies can and should be modified by any special knowledge the analyst has concerning individual firms. USE OF RESULTS: The survey responses and estimated "responses" of the firms which were not surveyed represent the basic information used in the estimation of total direct impacts in the next step. 62 SURVEY RESULT COMPILATION Step D;6 PURPOSE: Aggregate survey responses and calculate a variety of direct impact measures. DATA REQUIREMENTS: Coded survey forms and estimated responses for non-surveyed firms (see Step D:5) . DISCUSSION: If the preceding steps have been executed with care the calculation of the direct impacts will be easily accomplished and will be accurate within the limits accepted for this study. Obviously, if the list of firms is incomplete the impacts will be understated. The coding strip along the edge of the survey form is a data processing aid. Computer- ization of the survey information may allow for the realization of time savings in impact calculation for larger ports. This is not a necessity, however . RECOMMENDED ACTION: 1. Employment Impacts The employment impact is simply the sum of employees listed in response to question 4 plus the estimated number of employees for the firms not surveyed. The most accurate impact measure is the number of full-time equivalent empoyees, since the addition of part-time employees directly into an impact measure overcounts the number of jobs related to water- borne commerce. If the respondent uses blanks 4a, 4b, and 4c instead of blank 4 the number of full-time equivalent employees must be estimated. If the respondent cannot be recontacted to obtain the information — i.e., what is the average amount of time part-time and contract employees spend at your firm compared to full-time employees — estimates must be made. This can be accomplished by obtaining the average wage for the industry (state average wage data or national average wage data, from Employment and Earnings , or in the case of Canadian ports from Employ - ment, Earnings, and Hours ) and comparing this figure with the average wage determined for the surveyed firm. Simply divide the total payroll (f5) by the sum of 4a, 4b, and 4c; if the calculated wage is lower than the published figure the sum should be reduced to reflect the existence 63 of part-time employees. If the calculated average is higher than the published average the calculated employment should be considered full- time equivalent, unless it is substantially higher. In the latter case the survey should be examined for coding errors and consistency (see Step D:2) . 2. Income Impacts Total direct payroll impacts are the sum of responses to question 5 plus the estimated responses of non-surveyed firms. This impact can also be called an "income" impact, as long as it is noted that benefits have been included in the figure and that non-payroll types of income (e.g., pro- perty income and profits) have not been included. 3. Tax Impacts Tax impacts should be broken out by type of tax since the tax revenues accrue to a variety of agencies. In summing the tax revenue survey res- ponses and estimates of non-surveyed firms* responses it is important that the researcher be certain to whom the tax accrues. For example, it would be incorrect to state that waterborne commerce related activities generate $xxx,xxx in sales tax revenue for Alameda County, California, since the local governments receive only 16.7% of the sales tax revenue. Similarly, property tax revenues must be attributed to the various county and municipal governments. Moreover, unless responses have been received from firms representing the vast majority of economic activity in an industry, attempts to estimate total tax impacts are probably unwise. 4. Sales Impacts Total direct sales impacts — i.e., sales generated by waterborne com- merce related activity -- can be calculated from responses to question 3 of the survey, but care should be exercised to not double count sales. This is a special problem with respect to freight forwarders; their sales may include a wide range of costs that are counted as sales by other waterborne commerce related firms. Sales/revenue responses must be adjusted to remove double counting. The responses to question 7 will be of assistance in determining the presence Of double counting. For instance, if tug boat services are included as part of the answer to question 7 on a freight forwarder's questionnaire it will be necessary to reduce the firm's sales revenue to reflect this. Since the questionnaire does not provide specific purchase information, assumptions must be made as to how much revenue is double counted. It may be necessary to sub- tract the figures provided in response to questions 6a, 6b, and 6c in order to remove double counting. In addition to double counting problems with freight forwarders it should be remembered that questionnaires from firms leasing facilities or being provided services by the port will have to be scrutinized to prevent double counting of sales/revenue impacts since the port will have counted the cost of these services as part of its revenue. If direct sales/revenue impacts are to be used as input for the RIMS methodology (see Step F:3) summations must be done by industry rather than by aggregating all industries. 64 If the RIMS methodology is going to be used, it may be advisable to con- duct an additional test for double counting of sales. Two or three res- pondent firms in industries for which this may be a problem (e.g., freight forwarders) can be contacted and interviewed to determine what, if any, other port industry revenues have been included in reported totals. A sales/employee ratio can be developed from this information and used to check the responses of other firms in the same industry for consistency. In the case of regions with more than one port, allocations of impacts to each port should be made. This allocation is done prior to the summation of total direct impacts through the use of responses to the survey ques- tion asking for an allocation of activity to various ports. See Step D:5 for additional information. USE OF RESULTS: The aggregated impact measures will be used as data for publication and as data needed for the determination of secondary impacts. 65 DEVELOPMENT OF SECONDARY IMPACT MULTIPLIERS Step E:l PURPOSE: Select that method or methods which will be used to calculate secondary impacts. DATA REQUIREMENTS: None. DISCUSSION: This kit contains directions for calculating secondary economic impacts by two methods — economic base multipliers and input/output multi- pliers. The advantages and disadvantages of each approach are discussed in Chapter II. The specific techniques described below are: • An economic base income multiplier derived from the consensus of three different employment multipliers, and • An output multiplier derived from an input/output table (Bureau of Economic Analysis RIMS multipliers for United States "kit" users and regional input/output tables for Canadian kit users) . We believe the most accurate results can be obtained by following all of the steps and comparing the results of the two approaches. However, a less ambitious program may be more in keeping with an individual user's needs. If the purpose of the study is to describe direct activity with little emphasis on secondary impacts, the time and effort involved in completing all of the secondary impact steps is probably not warranted. Similarly, time, staff, or budget constraints may require a modified approach. In such cases, it is reasonable to select a single approach, and even a single technique for calculating the employment multiplier, as long as the limitations and probable biases of each approach are recog- nized and cited in the impact report. Some Canadian ports may not have suitable regional input/output studies available to them. Large ports could make use of the interprovincial input/output model published by Statistics Canada. The structural Analysis Division of Statistics Canada in Ottawa should be contacted for further information concerning the use of the interprovincial model. 66 RECOMMENDED ACTION: If the port chooses not to follow the complete methodology, the selection of the individual technique can be based on available resources. The techniques for calculating an economic base multiplier require very little money to obtain required data, but do require a fairly substantial investment of staff time. The RIMS input/output multipliers require an initial purchase price of about $1,000, but no staff time is required for their calculation. An examination of the remaining steps in this element will help the user decide which technique is more compatible with the port's purposes and resources. USE OF RESULTS: The decisions made in this step will guide the implementation of the secondary impact analysis. 67 DEVELOPMENT OF SECONDARY IMPACT MULTIPLIERS Step E;2 PURPOSE: Collect income and employment data necessary to calculate economic base multipliers. DATA REQUIREMENTS: U.S. Department of Commerce, Bureau of the Census, County Business Patterns , available from Superintendent of Documents, U.S. Government Printing Office, Washington, D.C., 20402. U.S. Department of Commerce, Bureau of Economic Analysis, Regional Econo- mic Measurement Division, Local Area Personal Income , available from National Technical Information Service, Springfield, VA, 22161. For Canada — Statistics Canada, Canadian Census, Vol. Ill, Part 4. Industries by Sex for Census Metropolitan Areas, Place of Residence and Place of Work (Bulletin 3.4-5) , and Industries by Sex for Census Agglomerations of 25,000 and over (Place of Residence) , and Census Agglomerations of 50,000 and Over (Place of Work), (Bulletin 3.4-6). RECOMMENDED ACTION: The basic published data needed to calculate employment and income multi- pliers for the port district county (s) should be obtained. The most recent issues of County Business Patterns for the state in which the port is located and the United States as a whole should be ordered from the Government Printing Office. The BEA income data for the port county should also be ordered. The NTIS publication numbers are: • Volume I — Summary Including Methodology and Clas- sification of SMSAs and BEA regions — PB291164 Volume II — Northeast Region — PB291165 Volume III — Mideast Region — PB91166 Volume IV — Great Lakes Region — PB291167 Volume V — Plains Region — PB291168 Volume VI — Southeast Region — PB291169 Volume VII — Southwest Region — PB291170 Volume VIII — Rocky Mountain Region — PB291171 Volume IX — Far West Region — PB291172 Entire Set — PB291163-SET 68 Publication of both sets of data takes about one and one-half years (i.e., 1977 data should be available in mid-1979). Since it is important to use employment and income data for the same year, the user should check with the Department of Commerce to find out the most recent year for which both are available before ordering publications. The Canadian Census Vol. Ill, Part 4 Bulletins 3.4-5 and 3.4-6 should be obtained by Canadian ports. The volumes contain employment data by industry for urban areas. The latest census information is from 1971 as of this printing, making it somewhat dated, but no other publication provides sufficient detail. USE OF RESULTS: These data will be used in the calculations of employment and income multipliers for secondary impacts. 69 DEVELOPMENT OF SECONDARY IMPACT MULTIPLIERS Step E:3 PURPOSE: Develop employment multipliers based on the "concentration" technique. DATA REQUIREMENTS: Most recent County Business Patterns for the port's state and the United States summary; Census Vol. Ill, Part 4 Bulletins 3.4-5 and 3.4-6, and Canadian Statistical Review for Canadian port users. DISCUSSION: This technique, also known as location quotients, is used to determine the basic and non-basic shares of employment in each industry by com- paring the port community or region's employment distribution to national employment distribution. The underlying assumption is that if a com- munity is highly specialized relative to the nation in the production of a particular commodity, that commodity is presumed to be an export item (e.g., automobiles from Detroit). If local residents have the same demand patterns that prevail nationally (as is assumed) , a self- sufficient community should have the same employment distribution as the country as a whole. If 2% of U.S. employment is in apparel manufac- turing, for example, then a 2% share in a given community would indicate just enough employment to meet local demand. If the local share exceeds 2%, the excess employees are assumed to serve export demand. (Also, if the local share is less than 2%, imports would be required to meet local demand, but this is not of concern in this calculation.) The most significant criticism of the concentration technique is that it understates basic employment and thus overstates the employment multi- plier because it cannot adequately treat differences in product mix. In the case described above, even though 2% of the community's employment was in the apparel manufacturing industry, it might all be accounted for by a single firm which produces swimwear , most of which would be ex- ported. Even data disaggregated to the four-digit SIC level group together several different products and often several brands of each pro- duct. This product mix factor causes the concentration technique to understate the volume of exports. Comparisons of basic employment derived through this approach and estimates derived from direct surveys of local industries indicate that in each case the concentration techni- que yields a lower estimate of export levels. As a result, the employ- ment multiplier and the estimate of secondary impacts will be over- stated. Nevertheless, this approach is an inexpensive way to estimate the employment multiplier as long as its limitations are recognized. 70 RECOMMENDED ACTION: To calculate the number of "basic" employees in each SIC industry, it is necessary to apply the national percentage of employment in the industry to total employment in the county and compare this number to actual county employment in the same SIC. If actual employment is less than or equal to the calculated number of employees, this employment is consi- dered completely non-basic. If actual employment is greater than the calculated number, the difference (actual minus calculated) is basic employment. Exhibit E:3-l illustrates the application of this techni- que. After the process has been repeated for each local SIC industry, the results are summed and all employment is assigned to either the basic or non-basic sector. The economic base multiplier itself is calculated by dividing total employment by basic employment. For the purposes of this analysis, basic and non-basic employment should be summed for each of the major industrial divisions — agriculture, forestry, and fish- eries; mining; contract construction; manufacturing; transportation, com- munications, and utilities; wholesale and retail trade; finance, insur- ance, and real estate; and services. It is important to note that national employment data are disaggregated to the four-digit SIC level, but data for counties are frequently pre- sented in more highly aggregated form because of disclosure problems or the absence of certain industries. The procedure described above must therefore be adjusted to make sure that similar employment groupings are being compared. In the example in Exhibit E:3-l, for instance, national and county employment in SIC 209 can be compared directly. However, no county data are reported for the remaining three-digit SICs in this cate- gory (201-208) . The proper comparisons would therefore be between U.S. and county employment in the two-digit SIC 20 minus employment in SIC 209 . In the example, U.S. employment for this group would be 2.14% (2.36%-0.22%) for a calculated county employment of 362. As a result, all 121 actual county employees (285-164) would be assigned to the non- basic sector. Canadian ports must make a further calculation in order to adjust basic employement to account for foreign exports. Since approximately one-fifth of Canadian GNP is accounted for by exports, the economic base multipliers will tend to be overstated if not adjusted to account for production for foreign export, which is often incorporated into the non-basic sector of the regional economy. The simpliest adjustment procedure uses an export/GNP ratio (as derived from data in the Canadian Statistical Review ) to determine the number of employees producing products for foreign export. The remaining employment is disaggregated into basic and non-basic categories as above, but before the multiplier is calculated the foreign export employment is summed with basic employment. This can also be accomplished by the following computation: multiplier = T (e x T) + (1 - e) x B) where T = total employement B = basic employement e = exports/gross national product 71 If more accuracy is desired, a regional exports to gross regional product ratio can be used. Provincial data may have to be used if regional exports and product statistics cannot be obtained. A further elaboration would be to use the same procedure at an industry level, i.e., to net out foreign export-related employment for each industry. USE OF RESULTS: The employment assignments developed in this step will be used in the derivation of the income multiplier in Step E:7. 72 i.MllitLi !•: 1-1 CONCENTRATION TECHNIQUE CALCULATION V.R.WS MAUBOR COUNTY 09 15 17 173 20 209 24 241 242 2429 243 2436 26 27 35 42 421 44 48 50 501 508 51 514 5146 517 52 521 53 54 541 55 551 553 554 562 57 571 5712 58 5812 5813 59 591 594 60 602 61 612 64 65 651 70 72 723 73 739 75 753 79 799 80 80 1 805 81 83 8)1 8b 863 864 89 sic u.s Percent Calculated Actual TOTAL 00.00". 16,902 16,902 Agriculture, Forestrv, Fisheries 0.36 61 176 Fishing, hunting, and trapping 0.03 5 116 Contract Construction 5.50 9 30 501 General contractors 1.47 248 239 Special trade contractors 2.84 480 216 Electrical work 0.51 86 69 Manufacturing 30.27 5,116 8,204 Food and kindred products 2.36 399 285 Misc. foods and kindred products 0.22 37 164 Lumber and wood products 1.01 171 5.893 Logging camps 0.13 22 2,766 Sawmills and planing mills 0.33 56 1,915 Special product sawmills 0.01 2 1,264 Mlllwork, plywood & structural members 0.27 46 1,003 Softwood veneer and plywood 0.06 10 674 Paper and allied products 0.98 166 1,262 Printing and publishing 1.73 292 111 Machinery, except electrical 3.18 537 294 Transportation and Other Public Utilities 6.33 1,070 690 Trucking and warehousing 1.73 292 338 Trucking, local 4 long distance 1.56 264 338 Water transportation 0.30 51 65 Communication 1.81 306 204 Wholesale Trade 7.11 1,202 563 Wholesale trade - durable goods 3.89 657 308 Motor vehicles & automotive equipment 0.65 110 76 Machinery, equipment & supplies 1.50 234 93 Wholesale trade - nondurable goods 2.88 487 255 Groceries and related products 0.93 157 113 Fish and seafoods 0.03 5 60 Petroleum & petroleum products 0.30 51 62 Retail Trade 20.71 3,500 3,297 Building materials 0.71 120 127 Lumber & other building materials 0.37 63 51 General merchandise stores 3.07 519 203 Food stores 3.08 521 616 Grocery stores 2.62 443 571 Automotive dealers & service stations 2.76 466 645 New and used car dealers 1.20 203 315 Auto and home supply stores 0.31 52 69 Gasoline service stations 1.06 179 244 Women's ready-to-wear stores 0.48 81 95 Furniture and home furnishing stores 0.79 134 121 Furniture and home furnishing stores 0.48 81 82 Furniture stores 0.34 57 74 Eating and drinking places 5.50 9 30 1,078 Eating places 4.64 784 764 Drinking places 0.50 85 138 Miscellaneous retail 2.63 445 332 Drug stores 0.74 125 155 Miscellaneous shopping goods stores 0.69 117 69 Finance, Insurance, and Real Estate 7.02 1,187 698 Banking 1.99 336 202 Commercial and stock savings banks 1.83 301 183 Credit agencies other than banks 0.73 123 101 Savings and loan associations 0.27 4b 71 Insurance agents 0.58 98 59 Real estate 1.31 221 278 Real estate operators and lessors 0.68 115 228 Services 21.29 3,598 2,718 Hotels and other lodging places 1.42 240 299 Personal services 1.41 2 38 177 Beauty shops 0.44 74 87 Business services 3.39 573 173 Miscellaneous business services 1.50 254 117 Auto repair, services, & garages 0.71 120 79 Automotive repair shops 0.45 76 68 Amusement & recreational services 0.90 152 145 Miscellaneous amusement services 0.51 86 95 Health services 6.53 1.104 1,117 Offices of physicians 0.90 152 101 Nursing and personal care facilities 1.18 199 32 3 Legal services 0.58 4 8 63 Social services 1.15 19i. 98 Social services, nee 0.92 105 98 Membership organizations 1.71 289 440 Labor organizations 0.26 44 73 Civic and social associations 0.43 73 322 Miscellaneous services 1.02 172 60 Basic 8. 4 fit) 4 111 127 162 2,744 597 1,262 283 664 1,096 74 14 55 11 19 128 112 17 65 14 17 115 53 30 25 113 59 13 Non-Baa ic 8,502* 56 5 46 239 147 69 359 121 37 47 22 54 2 46 10 166 111 29i 83 26i 51 20i 34 249 139 7b 93 80 53 5 51 80 57 51 203 45 443 17 203 52 179 81 39 8 57 61 764 85 108 125 69 53 19 183 30 4b 59 50 lis 67 240 90 74 5b 1 17 11 h8 50 86 693 101 199 63 98 40 44 73 bO Includes non-classlflahle establishment! 73 AnhurD Little Inc DEVELOPMENT OF SECONDARY IMPACT MULTIPLIERS Step E;4 PURPOSE: Develop employment multipliers based on the minimum requirements tech- nique. DATA REQUIREMENTS: Most recent County Business Patterns for the port's state, the United States summary, and other states, as determined below; Canadian Census Vol. Ill, Part 4 Bulletins 3.4-5 and 3.4-6. for Canadian port users. DISCUSSION: This approach attempts to overcome the principal problem of the con- centration technique, which systematically overestimates the multiplier by underestimating basic employment. The underlying assumption of the minimum requirements approach is that the amount of employment in a given industry necessary to serve local demand in a community is equivalent to the smallest amount of employment in that industry in a group of counties of similar size. Any employment in excess of this minimum is assumed to serve export demand. The problem with this technique, particularly when relatively small com- munities are being analyzed, is that the sample of communities can be constructed so that there is very little or even no non-basic employment in each industry, so that basic employment is overstated and the multi- plier understated. For this reason it is often argued that some cutoff other than the lowest employment share be used — e.g., the second lowest, the share corresponding to the 5% level, etc. The difficulty with this approach is that by arbitrarily establishing a higher cutoff, the basic share of employment will be lower. Unless some judgment is exercised about what truly represents the minimum requirement for each industry, this technique can be misleading. RECOMMENDED ACTION: The first step in determining the minimum share of employment required to serve local demand in each industry is to identify the group of counties (or SMSAs) to which the port community will be compared. The object is to select counties of similar size (measured by employment) with dif- ferent economic structures. The port analyst should thus select at least some counties in non-coastal states (or some coastal counties in the case of river ports) . 74 The source for selecting these counties is the U.S. summary volume of County Business Patterns . Table 3A presents a list of all United States counties and their total employment. Ten counties (no more than two in any state) with employment similar to the county being analyzed should be selected and the appropriate volumes of County Business Patterns or- dered. (This publication is available for examination in Department of Commerce field offices and public libraries in many larger cities. If feasible, it is probably quicker and less expensive to photocopy the few pages of data needed for each of the ten counties than to order County Business Patterns from the Government Printing Office.) In some cases, data availability may dictate a smaller number of comparisons or even suggest that this technique is inappropriate. For example, there are fewer than ten SMSAs with employment close to or larger than one million. Exhibit E:4-l is an example of the minimum requirements calculation. For each industry, the lowest percentage of employment (in italics) repre- sents the non-basic share of employment. When this minimum share is found in the port county, there is assumed to be no basic employment in the industry (SIC 17) . When port county employment exceeds this minimum share, the difference (1.41% - 0.82% = 0.59% for SIC 15) represents basic employment. After this calculation has been completed for each industry, the results are summed to determine total basic employment for each major industry division. The employment multiplier is calculated by dividing total employment by basic employment. As discussed in the previous step, county data may have to be adjusted so that similar employment categories are compared. A similar approach should be taken by Canadian users of the kit. Canadian Census employment data for similar "urban areas" should be used in lieu of "county" employment data. The adjustment for foreign exports (see Step E:3) will also have to be made by Canadian users. USE OF RESULTS: The employment assignments developed in this step will be used in the derivation of the income multiplier in Step E:7. 75 EXHIBIT E 41 MINIMUM REQUIREMENTS TECHNIQUE CALCULATION Tottl Aqnc- . Forestry. Fisheries Contract Coostn 209 24 2ii 24? 2429 2j3 ^436 on and Other Public Uttlim Wholesale Trade 50 501 508 514 5146 517 52 521 55 3 554 562 57 5812 581 1 591 594 60 60 2 72 723 73 739 75 75 3 79 rag BO 8.11 805 81 83 831 8fi 86) B64 e. and Real twain Percent Dutnbut Grays Pinal Imperial Napa Maui Bannock Bonne.. Harbor AZ CA CA HW ID ID 100 0% 17,156 16.275 17.320 17.461 15321 18.775 1 05 051 5.34 072 036 019 0.71 0.69 " - - " - " 2.96 2.41 5.14 5.64 736 6.38 9.16 1.41 0.82 1.16 1 82 239 1.76 2.09 1.28 1.37 2 55 3.23 3.37 3.79 335 0.41 - 0.45 0.41 0.46 1.13 0.72 48.54 14.59 8.87 24.47 1266 22.51 10.72 1 69 0.99 3.34 535 933 3.05 537 097 - - - 0.30 - _ 34.87 - - — - _ 0.46 1636 - - - - - _ 11.33 — - - - _ _ 7 48 - — - - _ - 5.93 - - - - - 28 359 _ _ _ _ _ _ 7.47 _ - _ _ _ _ 0.66 0.43 0.82 1.13 0.69 0.85 1.12 1.74 0.32 36 - - - 1.11 4.08 298 9.79 803 8.74 7.91 500 2.00 0.74 4 24 _ 153 3.57 194 2.00 — 2.75 - 1 23 - 130 0.38 - - - _ _ _ 1.21 0.S6 - - - - - 3.33 2.63 15.71 2.49 3 30 797 13.61 1 82 1.64 - 1.42 _ 4.03 6 49 0.45 - 033 - 0.35 032 0.67 0.55 1.16 2.00 0.88 - 192 3 28 151 1.09 1030 1.07 2 0? 3.94 - 0.67 - 686 - 1 41 1 82 435 35 - — — _ _ - 0.37 0.41 053 ~ - 053 1.16 19.51 te.28 34.49 26 67 30.10 27.00 2235 75 0.75 1 49 1 72 0-8O 139 77 71? — 0235 0.76 - 82 31 1.20 031 627 2.41 3.38 2.77 3 22 3.64 352 6.05 4 06 4 51 3 17 3 11 3.38 3.49 5.67 3.27 4.27 3 05 2 56 3.82 339 5 73 4.61 2.50 4.35 3.83 1.86 1.16 1.95 235 0.79 1.61 139 0.41 0.60 1.11 06 0.34 057 051 1.44 1.32 243 1.13 1.26 190 1.08 56 0.33 1.07 0.32 043 36 0.69 0.72 - - - - — 39 0.49 - 0.73 0.71 0.57 0.61 061 0.44 - 60 0.54 - 046 43 6.38 4.83 739 7 62 1130 8.50 6 28 4.52 3.70 6.72 6.25 835 751 5 78 0.82 0.72 0.70 0.80 0.52 0.49 043 136 1.82 3.78 3.61 397 386 322 032 0.77 157 1.24 74 1.72 1 36 0.41 036 030 033 2.22 1 13 t 31 4.13 3.11 4.59 4.92 6.59 9 64 4.27 1 20 0.98 237 134 1.10 2.23 1.61 1 08 98 2 37 1.82 1 09 223 1.61 60 — 62 083 C92 0.91 091 42 - - 0.38 - - - 0.35 — 057 035 - 064 38 1 64 1 07 0.42 1 33 3.99 155 0.62 1 35 - ~ 0.95 234 076 - 1608 8.83 16.57 26 40 29 84 17 26 33 63 1 77 2.03 1,83 1.59 1592 1.57 2 40 1.05 0.72 1 49 138 1 74 2 20 117 051 0.38 0.49 105 62 54 0.40 1 02 64 1 08 2 64 1.44 1 06 - 69 0.50 061 081 60 38 — 0.4! _ 1.07 0.79 2.02 061 1 06 40 _ 087 063 63 054 IB 086 - 1 09 1 09 1 29 84 052 056 _ 057 - 85 39 - 661 1.61 4 16 12.20 2 34 5.42 - 060 041 1 89 1 82 1 05 1 20 93 1 91 — - - - - - 037 - 67 47 33 56 48 53 089 085 58 1 57 - 69 058 - - 52 1 57 - - 260 106 1 16 1 J6 13 1.90 99 046 - - 0.40 ~ 051 - 1.91 41 051 - - 63 52 JS - 83 0.55 144 1 30 111 Cravi Harbor County 2.36 060 134 202 0.53 35.71 236 130 28 78 8.77 11.11 3.71 3.71 1.03 288 0.72 1 18 25.08 1 71 89 2 38 3 33 3.18 4.16 133 60 85 66 6.11 4 66 1 28 2 76 1 00 42 4.72 239 69 0.89 064 1451 76 1 22 63 1 26 85 45 4.19 152 1 78 46 1.11 054 JS 1 46 20.799 039 221 1.78 0.55 4456 1 20 36 44 10.57 18.08 1631 1.22 131 431 130 3 40 2.27 036 086 1.13 21.31 80 33 1 95 7 06 2.75 4 66 2 00 0.75 1.71 038 0.67 44 32 7.30 554 1.02 204 0.6S 0.78 3.12 1 62 1 62 0.44 039 44 1 07 1 14 057 2 20 1 03 63 056 82 52 6 78 1 10 33 068 92 025 14583 0.7 7 300 7.34 036 038 130 031 0.37 432 089 234 5.73 4 98 4 59 238 1.34 1 09 3 47 2.09 1.06 2.74 2.74 1.19 1.06 42 I 45 074 1.37 078 076 59 256 62 62 1 78 42 72 Thuriton WA 4.75 36 237 320 055 4.63 039 106 1.06 1.00 2 40 0.34 102 1 36 051 366 358 347 4 69 2.07 43 1.81 057 1 34 0.70 050 9.49 6.98 62 4.21 1.12 1 95 1 86 1.47 035 0.76 1.60 083 2573 2 82 1.49 083 2 20 078 88 77 37 9 74 1.49 1 89 071 1.30 I 12 292 59 089 1 31 Total 16302 176 116 239 216 8204 Non -basic 5392 19 46 139 147 285 _ 121 164 164 - 5393 133 76 2.766 2.766 - 1315 651 - 1.264 1264 - 1003 329 - 674 •74 - 1.262 1562 - 111 30 81 294 240 54 690 - 83 338 213 125 338 - _ 65 65 - 204 109 95 563 - - 308 19 120 76 19 57 93 30 63 255 - 80 113 - 53 60 60 — 62 " 62 3597 - 80 127 _ 76 51 - 61 203 49 154 616 - 45 571 138 433 645 - 17 315 181 134 69 12 5? 244 100 144 95 41 54 121 - 39 82 - 8 74 20 54 1078 58 118 764 139 625 138 65 73 332 - 106 155 45 110 69 8 61 698 - 58 202 19 - 183 17 166 101 27 3 71 71 - 59 — 59 278 50 - 228 157 71 2.718 - 67 299 171 128 177 33 57 87 23 64 17J 12 44 117 S3 64 ?9 - 11 68 - 68 145 25 25 95 32 63 1.117 490 203 101 32 69 323 323 - 63 7 56 98 - - 98 10 88 440 23 17 78 36 42 322 258 64 60 - 60 76 DEVELOPMENT OF SECONDARY IMPACT MULTIPLIERS Step E:5 PURPOSE: Develop employment multipliers based on the "experience" technique. DATA REQUIREMENTS: Most recent County Business Patterns for the port's state; Canadian Census Vol. Ill, Part 4 Bulletins 3.4-5 and 3.4-6 for Canadian port users. There are no other specific data requirements for this step. However, local and regional planning agencies, economic development districts, state or community college eco- nomics departments, and similar sources should be contacted to locate any economic base studies which may have been done for the study area and to identify persons with extensive knowledge of the area's economic structure. DISCUSSION: As discussed in the previous two steps, the available indirect techniques for determining basic and non-basic employment may yield misleading results. This technique, admittedly much "softer" than the others, relies on the user's knowledge of the local economy. The assumption is that familiarity with local businesses and conditions is a better predic- tor of basic employment than any approach based on averages and aggre- gates. It should be mentioned here that the ideal economic base study would be a direct survey in which each local firm reported its local and export sales. If a port community has a recent study of this type avail- able, no other techniques would be needed to derive economic base multipliers. The advantage of this approach is that it focuses attention on the speci- fic characteristics of the local economy which may not be apparent in either of the other approaches. Both the concentration and the minimum requirements calculations begin with the percent distribution of local employment reported in County Business Patterns (or the Canadian Census Bulletins) . The problem with using this distribution is that it may be distorted by a single special circumstance. For example, large amounts of construction employment involved in building a power plant would drive down the share of employment in all other industries, causing an over- estimate of non-basic employment by either of the other techniques. 77 RECOMMENDED ACTION: The user begins with the detailed information on port county employment presented in County Business Patterns (or the Canadian Census Bulletins) and estimates the amount of employment in each industry that is basic. Sources which may be used for these estimates include: • Existing economic base studies, particularly those based on surveys, • Persons knowledgeable about the local economy such as planners or economists, who may either provide estimates or check the reasonableness of the user's estimates, and • Optional questions added to the port and port- dependent industry surveys. Estimates of basic employment are summed and divided into total employ- ment to calculate multipliers for each major industry division. USE OF RESULTS: The employment assignments developed in this step will be used in the derivation of the income multiplier in Step E:7. 78 DEVELOPMENT OF SECONDARY IMPACT MULTIPLIERS Step E:6 PURPOSE: Derive a single "consensus" employment multiplier. DATA REQUIREMENTS: Results of Steps E:3, E:4, and E:5. DISCUSSION: Each of the three approaches described in the previous steps is likely to result in a different level of basic employment and thus a different employment multiplier. The concentration approach generally yields a higher multiplier (lower number of basic employees) than the minimum requirements approach. While it cannot be determined on an a priori basis whether the "experience" technique will yield a higher or lower multiplier, it is likely to be lower because the other two methods tend to bias the total multiplier upward. In order to present a conservative estimate of port-related impacts, we recommend that the lowest multiplier be selected. In general the total employment multiplier (that is, the combined multiplier for all indus- tries) should be between 2 and 3, and probably closer to 2 for most coun- ties. For relatively small, undeveloped areas, the multiplier may be as low as 1.5. RECOMMENDED ACTION: From the employment distributions calculated in Steps E:3, E:4, and E:5, select the largest of the three numbers of basic employees in each SIC category reported in County Business Patterns (or Canadian Census Bulletins 3.4-5 and 3.4-6) for the port county. These will form the basis of the con- sensus multiplier. Basic employment in each industry is summed and divi- ded into total employment to calculate the total multiplier and the multiplier for each major industry division. USE OF RESULTS: These consensus employment assignments will be used to calculate the income multiplier in Step E:7. 79 DEVELOPMENT OF SECONDARY IMPACT MULTIPLIERS Step E:7 PURPOSE: Develop economic base income multiplier. DATA REQUIREMENTS: Bureau of Economic Analysis personal income data for port county or SMSA (see Step E:2) and results of Step E:6; Employment , Earnings and Hours for Canadian ports. DISCUSSION: Income is generally a better measure of downstream economic impacts than employment. Moreover, the employment multiplier excludes the effects of non-labor income (interest, dividends, and transfer payments) which is primarily basic and supports non-basic employment. (Transfer payments alone currently account for close to 15% of total personal income nation- wide.) Using an income multiplier to capture this additional component of the basic sector is therefore a more theoretically justifiable approach. The reason that the income multiplier is not computed ini- tially is that the available data are too highly aggregated to use the indirect techniques for determining distributions between the basic and non-basic sectors. Large Canadian ports will be able to utilize this technique, but only with wage income being included. Small ports, or more correctly ports from less developed areas, will not be able to use this step unless disaggregated income and employment by industry can be obtained for the area. RECOMMENDED ACTION: The first element of this step is to determine the percentage of basic and non-basic employment in each major industry division for which income data are reported. Major industry divisions and corresponding two-digit SICs are: 80 Agriculture, forestry, and fisheries 01-09 Mining 10-14 Contract construction 15-17 Manufacturing 19-39 Transportation, communication, and public utilities 40-49 Wholesale and retail trade 50-59 Finance, insurance, and real estate 60-67 Services 70-89 The consensus allocation of employment to the basic and non-basic sectors (Step E:6) should be aggregated to correspond to these industry divisions and the percent in each sector determined. Exhibit E:7-l is an example of the BEA income data available for each county. The income multiplier is calculated by determining the basic and non-basic components of per- sonal income by place of residence. First, labor and proprietors' income by place of work is divided accord- ing to the employment data. Income for each major industry division is split into basic and non-basic components according to the percentage of employment in each sector. Since farm and government employment are not covered in the County Business Patterns data, income for these industries is treated separately. Farm activity (and income) tends to be basic; it is reasonable to assume that at least 70% of total farm income is related to export demand. A brief discussion with the County Extension Service would provide an idea of product destinations which can be used as a basis for the allocation. Federal Government civilian and military income is generally considered completely basic, although a small portion of federal income related to such things as the postal service could be considered non-basic. Most state government income is also considered basic. Local government employment is usually considered non-basic, but it could reasonably be argued that those jobs supported by federal reve- nue sharing, CETA, or similar programs are basic. Therefore, state and local government income can be split equally between the basic and non- basic sectors. The sum of basic and non-basic income for all industries is labor and proprietors' income by place of work. The remaining elements of the personal income by place of residence cal- culation are divided according to the assumptons below. Personal con- tributions for social insurance are divided in the same proportion as labor and proprietors' income. The residence adjustment, which accounts for people who live and work in different counties, is treated the same way. Dividends, interest, and rent are derived from both local and non- local sources, and should be considered 50% basic and 50% non-basic. Transfer payments represent monies paid to individuals by federal and state government (social security, welfare, etc.) and thus should be con- sidered basic. The sum of these four components plus labor and propri- etors' income is total personal income by place of residence. The income multiplier is calculated by dividing total income by basic income. An example of the entire process is shown in Exhibit E:7-2. 81 For Canadian ports the weekly earnings data from Employment Earnings and Hours should be multiplied by the appropriate number of employees in each industry category. The resulting total wage and salary income should then be expanded to acount for firms not covered by the larger firm survey that forms the basis of Employment , Earnings , and Hours . The expansion is accomplished by dividing the total income for each sector by the survey coverage percentages provided in the "Concepts and Methods" section of Employment , Earnings , and Hours . The results of this expansion should then be allocated between the basic and non-basic sectors according to the employment assignments (Step E:6). The income multiplier is then estimated by dividing total earnings (basic plus non-basic) by basic earnings. The resultant multiplier will be somewhat higher than if these other income sources were included since they would add to the basic income share. The foreign export adjustment detailed in Stept E:3 will have to be utilized if a consistent income multiplier is to obtained. USE OF RESULTS: The income multiplier is used to estimate the total impact of direct port activities. 82 EXHIBIT E:7-l EXAMPLE OF BEA COUNTY INCOME DATA GRAYS HAROCR •ASHINGTCN PERSONAL IKCFE et PAJCR SCURCES 1971-76 ITKLSANC! Cf COLLARS) 1971 1/ 1972 1/ 1973 H 197* 1/ 1975 2/ 1976 2/ TOTAL LAecR ANJ PROPRIETORS INCCME 87 PLACE CF wCRK 3/ 8Y TYPE MACC ANO SALARY CISUURSEMENTS OTHER LADOR INCCE PROPRIETORS IhCCC */ Farm NCN.FARM t/ BY INCUSTHY FARM NCN.FARH PRIVATE AG.SEMV..FCR..FISH..AHC CTFER 5/ MINING CONSTRUCTION MANUFACTURING NCN-CU&ABlE OCCDS CLRAULE GCCCS TRANSPORTATION AflC PUBLIC LTILITIES WHOLESALE TRACE RETAIL TRACE FRANCE, INSURANCE, AND REAL ESTATE SERVICES CCVEH.-.HEM ANC CCvERNMENT ENTERPRISES FECEHAL, CIVILIAN FECERal, MILITARY STATE ANC LOCAL OERIVATICN CF PERSONAL INCOME BY PLACE CF RESICENCE TOTAL LABOR anO PROPRIETORS INCOME BY PLACE CF «CRK LEJS! PERSCNAL CONTRIBUTIONS FOR SOCIAL INSURANCE eY PLACE CF kCRK MET LABCR ANO PROPRIETORS INCOME 8Y PLACE CF *CRK PLCS: HESICENCE ADJUSTMENT HET LAbCR ANO PROPRIETORS INCOME BY PLACE CF RESICENCE PLLS: CIVICENOS, INTEREST, ANC RENT 7/ PLLS: TRANSFER PAYMENTS PER'CNAL INCOME 67 PLACE CF RESICENCE PER CAPITA PERSONAL 1NCC*E (CCLLAP.S1 TCTal POPULATION ITHCLSANCSI 1*5.51* 1*2. »99 I82.03C 19*. 9)* 199.21! 2*t .*ai ♦ ,288 7,9*3 9,313 10.3C6 12,122 15,(2! 21,931 25,362 31,929 33.7C9 33,3t( 36,162 3.S03 *.963 9,3*3 9.575 7,049 7,117 U.C3J 2C.399 22,586 2*. 13* 26,319 29, C*! *,37T 5,*32 9.911 10.257 7,886 8,151 169.361 ISC. 37* 213,361 229.192 236.839 292.115 1*1,727 WC154 18C.178 193.937 196.133 2*5, cia 2.313 2,770 3.75C *,591 4.337 *,57l (LI (L) ID 191 199 9C 12.919 11.C7C 12.765 15.395 13.99E 18.9CC 66.C21 ?9,*92 86.C83 91.676 9C.3E9 121,321 17,523 2C.35C 2C.591 23,*C3 21.6 36 3C.C39 *«.*98 58, 9*2 67.*92 66.273 68.751 91,282 9.328 11.601 12,871 13.252 13.*6« 17.27J $.130 5.*35 6.318 8.C72 8.152 9.153 21.232 23.161 25,855 28,109 29,*29 33,*33 4,229 *.i06 4,770 *.«93 3.02C 5.572 2C,!2« 22.C72 25,721 27,758 31.1*3 3*. ICC 27,63* 3C.218 33,183 35.255 4C.7C6 *7.C91 3«*66 3,562 *,C31 *.*78 5, '59 6.C46 l.tes 2,101 2.57C 2.276 2,32c 2.693 22.203 2*, 555 26,582 28. SCI 32,921 38,358 173,738 19). (06 223,272 239. **9 2**. 72! 30C.J66 1.171 9.*73 12.32C 14,7*0 15.16* 18,131 16), 567 -1.32) 116.333 -1.60* 21C.932 .1.912 22*.7C9 -1.9*9 229.561 -1.323 282,13! -2.63* 16*, 2*2 30,98 7 33.737 18*. 729 J*.C57 36,2)5 209. C*C 38.*97 *0.90* 222.760 44,422 49.99* 228. 23C *t.*6« 61.13* 279, 5C1 51 .fOC 62,*6I 228.966 3.833 JS.* 2)),C*t 4,272 59.7 286,**! *,63) 49.7 317.176 9*2 19 6C.8 335. 8*C 5.»9! 61.1 391.769 4,-27 41.3 SEE FCCThCTEJ At E*0 CF T*eiES, 83 CM w w M P-i /-^ 1-1 CO H H hJ co O S O 4-1 o o S3 H Cfl O co 0) 2 3 O O M J3 < O -H CJ "- ' •H m 00 vO St O CM CM CM 00 m en NO rH H O o nO CM 00 O H vO cn 00 rH N cd •H c QJ ,Q CJ QJ 4-1 4-1 CO e o XI M CO c J-i QJ CO aj 3 4J CJ E u -a •H e CO c c u o H C c c C H c o rH g a) rH rH rH •H •H •« £ > rH T3 CO CL 4-1 CO ■> o CO c c •H — CJ J-l o CJ CO u o •H CO o 4-1 to 3 rH O OJ CO Ph 0) c c c UJ •H •H •H •H O J-l B QJ CO QJ •H •H o cfl c 4-1 4J CJ s rJ o 4-1 J-l 4-1 rH g M rJ CJ H cfl •H UH CO OJ c CO o a 3 QJ 4J H 1 1 T3 3 4-1 QJ 4J CJ o ^ 4LI 4-1 rH ^ J-l •H CO C CO •r - ) c 6 o c >-i u CJ o CO UJ o 4-1 OJ rH rH CO a X) rH >. H rH Ph o CO 60 CO CD CJ a. 3 u CO CO O 0) OJ aj 4-1 4-1 OJ CO C UH J-l C c c o C CO C >-i JS XI TO CO 3 CJ -a J-i CO O CO CO •H o j3 •H U CO 0J J-l QJ QJ 4-) .a c c OJ tH S s o 3 Pb H CO o Pm U-, CO •H 0) OJ UH j-i ai J-l TJ TJ Cfl o a 4-1 •H •H C x> ca C CO > a CO rH o QJ •H j-i rJ Ph CJ ai Q H 84 DEVELOP SECONDARY IMPACT MULTIPLIERS Step E:8 PURPOSE: Select industries whose multipliers will be derived by RIMS and instruc- tions for ordering RIMS data. DATA REQUIREMENTS: RIMS industry classification chart (Exhibit E:8-l); original list of waterborne commerce related firms including the relevant SIC codes for each firm. Canadian ports have nothing completely comparable to RIMS so their alternatives are to obtain local or regional input/output studies (e.g./ for Vancouver, British Columbia — • An Interindustry Study of the Metropolitan Vancouver Economy by H. Craig Davis, School of Commerce and Regional Planning, University of British Columbia, 1972) or, in the case of large ports in the principal metropolitan area of a province, to use the interprovincial input/output model of Statistics Canada . DISCUSSION: If the RIMS methodology is adopted it is important that the multipliers be ordered in advance since a six-week turnaround time should be expected. The cost was approximately $1,000 as of April 1979. Specific local or regional Canadian input/output studies can be located, if they exist, by contacting local community planning boards, the local office of the Ministry of Economic Development, and/or local college or university economics departments. The interprovincial input/output model and a dis- cussion of its use can be obtained from the structural Analysis Division of Statistics Canada in Ottawa. RECOMMENDED ACTION: The waterborne commerce related industries located in the region under study should be associated with RIMS industry classifications (Exhibit E:8-l) to determine the industry-specific multipliers that are needed. The $1,000 purchase price allows for the derivation of 50 multipliers at no additional charge. A sample listing of waterborne commerce related industries with attached RIMS code is included as Exhibit E:8-2. Further information concerning the RIMS methodology can be obtained by contacting Joseph Cartwright (202-523-0594). To write for RIMS informa- tion or to order a series of multipliers for a particular region send to: 85 Joseph Cartwright BE - 61 Bureau of Economic Analysis Department of Commerce Washington, D.C. 20230 Requirements for a special set of multipliers to be formulated include: 1. A check or money order for $1,000. 2. A list of the counties to be included in the region; the counties must be contiguous. 3. A list of the industries for which multipliers are to be determined. USE OF RESULTS: The RIMS multipliers will be used in Steps F:3, F:4, and F:5 to determine secondary and total impacts of waterborne commerce activity. The impact measures will include gross output, income, and employment. 86 CO w (Q u W r^ r*. r—t . — 1 M n • • CO LO 4-1 4-1 o VT a Cl I-H I-H A ^ ft p t_i « M f* r~- vD ^O r~. r^r^r^r~-r~~r-~r-r^ r~^ i~~ r-- o CO »n Cu •— 1 CUr H Cu Cu P. P.PP.P.P.PC.CU P p p uo ^ LO r * * MMf%«k*t*A« » * * at co iJ 3 r— 1 CM iH m vd in V4D m un LO Lommmmmmm m m m •— 1 •> •> r-l •— 1 rH —i r—l l-H •— 1 • 1 l— 1 ■ li li IrH ^H rH i-H CO CO * CTi .— 1 «00 • • rH C cu cu OS u CO O •> r— 1 i— 1 * •U Cu 4-1 X 4-1 4-1 4-1 4J4-I4-14-14J.U4J4J 4-J 4-1 4-1 t— 4 t— 1 I-H 1— 1 i— i a 1 a Cu a 1 PP-PP.CuP.P-Cu p P P 1 1 1 1 1 1 1 00 1 i i >~. i i i 1 1 1 1 1 1 1 1 1 c i rH 1 i 1 I I 1 1 1 co I ft 1 co in t-H 1 Sn 1 ■H 1 i -0 1 1 1 1 1 t C 1 1 60 1 cO >■> CO 1 rH 1 B 1 I 1 1 1 1 1 1 O 1 1 c 1 00 u U 1 •H 1 CO 1 I co I 1 1 1 1 1 1-1 1 1 I co •H 1 H CU 1 B I Uh 1 i C 1 111114-111 1 60 cu c 1 rH Cfl C 1 CO 1 1 CU i O 1 II 1 1 CO 1 1 C ,— | •H 1 co 3 •H 1 4-1 1 r4 1 1 CO i •H 1 I 1 co 1 i 4-i cn 1 «H 4J E I 1-4 cr E 1 1 3 1 1 -H i 4-1 1 1 1 60 1 1 CO 4-1 CO cn TJ •H 1 3 CU 1 O 1 C 1 r4 i CO 1 1 1 C 1 1 C 60 60 co i— i 4J CO i 4-> -a U 1 tH 1 M-l 1 CU 1 1 i U 1 1 1 i-l 1 CU CO C C 60 -H cu 1 co OJ 1 60 1 1 1 •X3 1 rC i cu I 1 *3 CO 1 CJ U i-l •H C 3 *o JH 1 C 60 N 1 C 1 O 1 r4 | 60 i 4-1 1 CO 1 rH 00 1 iH 3 13 -a •H Xi c o 1 c •H 1 •H. & 1 CO 1 -H i rH 1 rH I -H C 1 > CO rH rH T> 1 u3 iH ^H 1 to e 4-1 1 60 1 £. I CO 1 CU 1 3 -H 1 U 4-J -H •rl rH £ H c •H 1 H i 1 ■U XI "O 1 0) CO 3 3 •H iH Vj « 1 6 -H 4-1 1 r-l CO rH 1 r-l 1 H i rH 1 O CO rH CO CU rO ^ 3 CO cu 4-1 M i 3 e l-l 1 CO U-l CO 1 CO 1 CO i CO 1 E a) t-a iH cn >h XI "4H ,n CU 01 1 CU CU 1 •H C •H 1 •H 1 -H i 1-1 1 •H CO 3 0J <■£ CO C c E 3 E P rH >> U-) I 4J O 4-1 1 4-i 4-1 4-1 1 u3 1h iH XI cn <-3 3 T-i O p 60 O CO c c C 1 C CO c in c O w 3 CO O •H CO C C co o 3 Vj rH io 1 CU CU 1 01 4J 01 4-1 cu cn C04-I4JO1OO1C0T-I 4H 4-1 3 o •H 4-) CJ -o •> T3 ■O CO c -a c t-A -H cn CJX 600) 60 crj •H 1h )H O C cu ^H •H 60 •H 60 •H 0) -H CU •H O 0) E 3 -H Ol CO Sh -H CJ P CU Vj 4-1 •H P ofl CO 60 CO C CO C CO E CO B cn ■H JJlHTJMHiHlHOrH 3 CO X 4J h Pu 6 CJ C CU -H CU -H CU 4-1 CU 4-1 0) 4-1 OOC^hcOcOJHCU TJ O 4-1 CO CO CU CU cu cu •H -H l-l CO r4 CO r4 (H V-l rJ H -H rC r a-HOr*60C0r4 O X o c o 3 <4-l CJ Hfl C B c e 3 3 CO CO T3 T3 C X CO 3 O CU -H o & o S o 3 Cu ^ Cu 3 -3 3S3SSS33 2 3 3 c o 0) O M 4J X B ■H CU X. 0) rC cu cO cu RJ 0) CO OIOIOICUOIOIOIO) a 0J 01 4J co M z CJ CJ C/3 CJ 4-1 3 U 4-1 2 52 rS rS r3 ZSZZZZrSZ z Z rS o •H *4-l CM O O O o CO vr m vo t*< oocr>rHCMco •H O o o o o CO o o o o O OOrHrHrHrHi— li— 1 • — i rH i-H co VO r-~ oo cn o c m in m m m minminmminLn LT, in in CO O o o o i-H o ■ — i f—i r—i i— 1 . — i i — IrHi — li — IrH. — li — li — I • — 1 —h . — i - 1 u . cj o o - o o * * l-i O CO CJ 0) i— I . • r- oo •—> r— 1 — i o , — i 1—14-14-1 i — i r-l 4J 4-1 ■ — i CO o O o O O o o O Cu Cu o O Cu Cu o o c m -a CO O CJ 4_) JJ • 4-1 4J (X 4-J 4-1 4-1 " vjD CO CTN rH CO CM CM o _- P. CM CL CN Cu CM a < N Cu H — * rM Cu CN CU rH rH CM P, CM CO o r» r~- ^ H ^3 CU CO o o *■ o < O O r-i O O O r-4 O J O O CJ M 4-1 3 03 CO H C cu cu pi O CM . CO • » CO C^ CTn CM o . CO • • CM O . CO • . CM . -. a. . — i P r-- O r-~ p O o o o o o 1 o O Cu o O Cu 1 1 c O O 1 1 1 1 ! 1 1 1 1 >^ 1 ^ 1 co 1 U 1 1 CJ 1 CO 1 Cu 1 1 01 1 1 o 1 TJ 1 O 1 1 rC 1 60 1 CO 1 4-1 1 01 1 U 1 r^ 1 co I C 1 cu •H I co 1 CU 1 a rW | CO -H 1 iH 1 I cu 1 co I CU 1 4-1 UH 1 C 1 cu 1— 1 CU I > 1 -H 1 1 CO CJ cn I U 1 CJ 1 3 u3 I ■H 1 E 4-1 4-1 x CO •H UH 1 rH 1 CJ 1 1 1 1 CO 1 CO 1 U 1 60 1 co i •H 1 E 3 1 C 1 O 1 TJ 1 O ^ 1 U U 1 P 4-1 1 CO 1 cu I i- I c CO c CO co 4-1 CJ 1 CO 1 -H e 1 1 1 ^ en 4J U CO CU 1 cn i CO 1 >, CU 1 Sh U 1 O 1 >-, 60 3 1 CO 1 D CO Pu 3 1 0) O 1 o c u cu T3 u3 1 C 1 c 60 O O 1 X Uh 1 rH -H >, o CO 1 •H 1 3 U r= 1 cn I i— 1 C 3 c (-1 4-1 U P 60 CO 60 rH 1 1 CO 1 U 1 01 CU cn o C 1 CU 1 iH « 1 <4H rH | (0 *H O E 1h CO CU CO 1 60 1 u 60 CU 1 CO 1 CU u e u e en T3 1 4-1 co e CU t4 u 60 CO 3 CO 1h 0> OJ In >~, o CO C 4-1 01 ^5 rH tn 4-J >^ 4-1 CU Uh O Id cm CM >-> CO CJ cu o X CO •> o |h rH CJ 4J M 3 c U-l CJ cn CO CU 4J 3 4J 3 -H l3 In CO ^ >s 4J •> -a o a 4-1 4-1 rd cn -a cn cj > 0) 3 CU >h ^H 4-1 o 4-1 T3 CO •H 0) cu o CU H h C P u iH 3 co u 4-1 o x> 3 60 rH in 1h U U 0) C P c H 3 CO O CU Cu O o o l-i 0) i-l O P. o oo cn u o 4J Q P- S CJ P* H fo > o tu u- < rH CJ t— 1 3 o CM CO CM CO . (a M H J-J B to 3 w -a c H to g BU o CO u qj H 13 CO o u (1) CO 4-J 3 (t) 'Si rH C . M 4-1 CO 3 -a c 1-4 o (0 U CD i— i -a C/j o a a W to H 3 < to hJ C W CD OS CJ 0) rH 4-' •H T3 C 03 U QJ ,0. E 3 c >^ 1-1 4-1 CD 3 •V G a, a VO 4-J Cu Cu in 4-J vD i— i • 4J ■u O- CO CO 4-J Cu 4-1 Cu Cu 4-1 4J a cu Cu vo r^ O CM CM CM CM ON ON CO a. en CM o X -i •4) rH u3 I ^3 -H ><3 •H CJ CO CD I I co co CO S -3 00 .-I •H O U U £ CO 14-4 M-l I4_l U-4 o o o 1-4 c •h ai co B u u ■H -H CO CO a a a a 1-1 4-J •H -H CO iH Cu i-l CD a M CO lt-4 to CD C CO CJ i-l H C T3 CO C QJ U C 4-> c c CO i-l cfl iH -H -M CO > co CO g g g in cfl 00 CO ■U J2 C C ro CO CO CO 00 G •H U 3 4J O Cti M-l 3 C CO g to h < on >-*. -- ■& 4J 4J -U 4-1 4-J 4J 4-1 4J 4J 4J 4-1 4-1 4-J 4-J 4J 4J 4-J 4-) 4-J 4-1 4-1 4J 4-1 a Cu Cu Cu a Cu a a, Cu Cu Cu Cu Cu Cu Cu Cu Cu Cu Cu Cu Cu Cu Cu n ,.. „ ^ „ „ „ •, » „ „ 00 00 ,. r-~ « r^ •> r^ »> . « „ r, „ „ „ „ >-D vD vD ^ vD \D v£5 \D vO LO u-l CO CO UO i— i LP, i— i UO rH m uO LTi uO \jD ^D VO mD vjD 4-1 4J 4-1 4-J 4-1 4-1 4-J 4-1 4-J 4-J 4J 4-J 4J J-J Cu JH Cu 4J Cu J_i JJ 4J 4-1 4-1 4-J 4-J 4-1 4J ft Cu Cu Cu Cu Cu Cu a Cu Cu Cu Cu a a a CU a a Cu a Cu Cu Cu a a 1 co 1 1 1 1 i i QJ 1 1 CD 1 1 i CJ ^3 1 >N •H 1 1 •H 4-J 4-1 1 rH > CD CO •H e •H 1 co Cfl !-i G -o CJ rH 3 ^H CO 1 CD •H CJ o CO ■H •H QJ •H CJ 1 CO 4-1 •H JJ CO -C o i-i 4-J rH O •H 1 oo c 4J G U Cu i-l J-J 3 O Cfl 4J CO c QJ •H CD CJ S £ CJ rH CJ IH U-l ■H CO aj •H CO e rH TJ X u JH rH ■H QJ CO 4J r-4 CO QJ •H TO QJ a •H •H 4-J CO CO QJ CO rH CO CJ -C •H CJ •H 4-J rH •H c o CJ CO O tH U-l 4-J V-i QJ 00 Cu Cu U •H 4J •H •H 4-J o rH CO CD TO CO 4-1 CO •H ^-1 3 •H •H to CJ Uh U y-i tH U-l U-l U-l U-4 U-j U-l i-i >4H •H QJ r-t •H rO rH 4-J QJ > O O O o o O O O co .-1 c •H CJ •H CO •H CO 00 •N CO a> >, •H ■H u to CO rH CJ l-i J-J -a c iH !h 1-4 U U CJ 1-1 1-1 l-J S-i rH 4-J 'J ^ CO CJ U-l CO CO to o •H •H •H •H •H •H •H •H •H •H CO •H •H OJ •H CO QJ U-l •H •H M-l rH rH rH •4) TJ Cfl CO CO CO CO CO 4-J CO CO CJ CO CO 4-1 rH M-l Cu 4-J 4-J 4-J rH Cu •H rH P- Cu OO a Cu Cu •H Cu Cu •H Cu Cu •I-l •H >•. •H •H c CJ QJ X CJ c •H 0) QJ c CJ QJ QJ rH CJJ 0.1 4-1 CJ QJ o3 4-J >, CU rH ^H CO 0) o 3 CU cO o 3 !H 1-J •H u i-J i-l •H Vj 1-J •H 1-1 1-4 3 4-1 Cu •H c T3 •H U-l •H J3 X) CJ rH CJ cn •H E Cu CJ CO •H > to CO 4-1 C u3 ^3 rH "3 1-3 ^3 CO -3 ^ 1-1 ^3 K3 c -a CJ r-l 3 3 CO M to CO V-i CO CO >, cO O •H rH U-l CJ o cfl •H ■H QJ CO U-i 4-1 >. cu CJ 00 00 )-i > CO c QJ QJ 3 QJ CO QJ to QJ QJ OJ CO CJ to QJ -C o Vu J-J -H CO i-l to rd >-i CJ CJ CJ J2 U •H CJ CJ O -c a CJ Uj CJ CJ CJ IS) Cu S-i 4-J 3 O Vv U r-l 3 -3 <4) 4-J CD •H s C C c 4-J C •H C Cu C c C •H C QJ cu ^-1 CJ i-i QJ CJ CO X B 6 •H CO 4-J CD CO CO E CO G <0 4-J CO CO CO CO >* CO 4J CO C ^H ■H OJ CO 4-1 4J 3 CJ CO !-i i-i .H rH rH c •H C c c 1-4 C CJ C •H C S-i C C 4-J C •rl C ■H CJ CD .— 1 co QJ CO CJ O •H CO CO •H •H •H o .H CD QJ CO QJ -a CJ rH CJ co CU CJ •H QJ rH QJ rH 4_l U CJ co a. 3 LO rH jr M-l U-t O o e CJ •H u 4J 4-1 i^j H •H 4J •H J-J CJ 4-J 4-J rH J-J •rH 4J CJ CJ OJ C c c C to c CJ c rH c C •H C CJ c Cu 3 3 5 3 3 3 3 3 3 3 3 3 3 3 5 CO ^: •H •H o •H QJ •H <0 •H CD •H -H 4-1 •H CO ■H •H CJ 0) CJ CD QJ 01 CJ CJ CD CD CJ QJ CJ CJ CD M-l 4-J CO CO c CO in CO U-l CO 4-J CO CJ 3 CO U-l CO Cu z z z; !S X Z ■z Z ii ii 2 2 ^3 Pi Z o g g g g 2 g 2 g S _ C-J. ro »- o . — 1 CM CO -o \D ^D >£> \D vO vjD vO r- r^ r-~ r» r^« I--* )>- r^ r-« 88 U O CO u QJ 1— 1 X GO O CJ x qj CO 4-1 D 03 co rH C X C cd M m S 3 C P-. w 4J CO 3 X C H M co CJ CU 1— 1 x CO o u X CU CO 4-1 3 eg CO H C 0) cu Pi CJ X c co S-i 0) e 3 c u u CO 3 •3 C m CN CN x CM — 1 \o CN CM Csl CN CN .— I .-H t-H rH^OH-HNtnCT>CTNONCT\C^CTNCTiCTNCT\CTvCTiaNC^ CMCNCMCMCMCNCMCMCMCMCMCMCNCMCMCMCMCM ON m CN u •H X> co 00 HH C •H ^ V-l CO -3 rH CU V4 00 c •H o E CO ^3 ■H e CJ 00 -H C r4 ■H ,£) E co E 4-1 CU ■uj a co cu 00 CO c r4 -H co 3 00 CU •H X C_) cj o o 3 CJ X CO CO J3 O O w H P3 > o jr I I I CO M a o CO ex cu o co X H CU E >» U CU •H -H J-4 CO XJ o CO jl >4-l CO 3 - o c r4 CU U E cfl O Z 3 I CO CO I rH iH E CJ M • cfl CU CU • 3 c u O . 00 o C • -H QJ X. • CO C -H c U >H -H CO M-i CU cO E CU 3 cj rH u3 C CO QJ H 4J X J3 5^ 3 1-1 O CU •H 4-J CO -H O C •H E iH 00 rH C -H •H E 4-1 4J 4-1 C •H tH -H W C C C co t^ ^ ^ ?" •H 13 4-) CO CO CO X TO O O o o 00 00 I 00 c •H •H U-l CU 4-1 P^ CO U CO cu 3 4-) co cu rH i-H O -H x: 4J a x 3 cu 4-1 X CO cu oo co C CO H CU rj cj ■o o CO W I •a cu N •H u cu ,c 43 3 u o cj u - ,0 CO CO cj cm •H r4 c« Cfl T3 y-i s-i o 00 c •H e o CJ i c3 P-i PM 00 C CU •H 00 H CO 3 XI O M CJ O CO CJ cu CO • cu C •H w « CU CO a. oo co c CO I I I I I CO oo co C QJ •H -H E H E cu •H TD oo >-3 ■H c 3 •H <4-4 CO QJ CO O 3 U EC H O o CO 3 00 TJ cfl O a. cu i— i co •H CO w > X c QJ cfl o u QJ E U QJ CO ex qj a c CO -H j= CJ cfl E co <-a <^j cu > 00 -H -H C 4JH •H O <4-i 4-1 E 4-1 CO O -rH cu 4-i x: H 3 U Pi < CO CJ 3 O U a. cu rH •H ■U CU oc c •h OC 0£: o CO CX CO E >- cfl o CJ 4J u 00 cfl cn co a> CMCNCNCMCNCNCMCMCNCNCOC^cncncOrOrncO CNCMCMCMCMCMCMCMCMCNCMCMCMCMCMCMCMCN M OC o .J vo^cvim CO CO •r-t 4-1 CJ -H QJ 3 CX U CO >4-l T3 t3 QJ QJ C C C C CO CO CJ CJ I I CO oo c •H co co ,n QJ co J-i -H X) "4-4 co x) xi 4-1 CO QJ CJ r-4 00 3 cfl cfl X CO ^ O CJ s-i <^j ca a ex i i I i CO rH 4J 3 CJ o 3 Uh X O ^) H co cx co CU rH cO E _Q rH CO -H *-> E CU 00 c CU > X o o co C QJ QJ CJ N 3 O CO r4 X CO "4H QJ 4J ~ U CO CO o U CU X rH JC >» j^i co £ a m QJ -H U Q d, (b cfl O ^3 00 CO »-4 X CU QJ x: qj 4-1 UH o X <^J QJ HI X CU cfl a U QJ i CO 00 U i cx c ex CU 'rH u 3 x o )-J D- X CJ ^3 X CU X c CJ QJ CO QJ CU u QJ CJ Di P3 CJ 3 x o u cx cfl O U O CJ u 3 X O l-l a. >, (-4 QJ ^3 C O QJ •H 4-1 4-1 Cfl CJ rH CU o Mh CJ C O O CO E r4 3 O 00 D CT 00 -r4 C rH CO ^ 4J X •H 3 U CO CO •H U ^ CX J3 CO c •H r4 >% CX X X c Cfl rQ CU CJ 4J X Uh 0) o CO c« r4 X O QJ - 3 C >. cr c X -H Cfl C rH CJ CO rJ X <•« J3 CU rH X •> rH CU 3 « u u u u CO -H QJ 4-1 C CO •H -H 3 Q cu c CO cx CO <& co rH CO rH 4-1 -H cj E cfl H rH 4-1 -H X o QJ X 00 QJ C QJ •H CO M C o o > 4-1 Cfl 4-1 rH O tu cj • CO CJ rH • -H CU o C «3 - CO CO 4-1 OC c cu D c •H r4 rfl E ^3 C X CO co CU 4-1 cfl X cu E >> 00 -H O QJ C CO > < O QJ O CJ QJ CJ -H CU <4H > >-i 4J M 3 *3 c u o en u cu M T) CO O CJ 0) CO •u 3 3 CO ■H C 0) QJ Pi u -a c 3 J-j E 3 C >i J-j 4-1 to 3 •3 C HHHfjHro-HH[No\Hn<)-Hrgroc^oo NNCNNNNCSNNNrSCNlCNNCNlOJNNNfMCSNCNrNNfslNNNCNCvlCSICNlNnrO WD C rl x: co co 00 C 00 -H C 4-1 •H C x -h CO u -h c CO t-l U CO 3 3 X ex -h P. o CO -H 3 ^< o o co oo s QJ QJ " 3 l-i a an M ^ ■h a ^ rH X 1) CO Sum a to aj h c a. «h (0 3 X 00 C T3 H rH > O 0) CO 14-1 CJ l-i -H 00 CO C 3 o a) c co <-3 O O -H C CO I I CO M 0) c •H OCX C •H 14-1 X CO CO QJ •rl rH rH OJ X CO 3 O 3- O rH J-i "3 CO CJ CO M 00 o C O •H X 4-> Jsi QJ 3 QJ CO S-J rH 00 00 c c QJ w QJ a pa h i i i i 00 C co •H C D, -H >% CO 4-1 QJ o u QJ Q) 4-1 CO co H CO u3 >H 00 M C 00 QJ •H C 4J > "H CO to a e J-J >, oo 4.1 co c o o 0) U -rl O 4-1 4J 4-1 CJ CO O QJ CO X rH rH a, w pl, CO T3 o o CO 00 4-1 c c QJ O 00 -H J-l 4-1 c OJ QJ CO 00 tfl QJ x 3 4-1 r-t C H >. QJ CO C_> OJ > ■rl J-l CJ CO CO OJ OJ o3 £ CJ CO CO Oj iH U-l CO rH U O O 3 to p., co >-i co QJ X <•<-> 4J o CO 4-1 cj 3 •a co CO O rH C ^ CO o a, cj H iH 4JT) E CO QJ QJ J-J -H X CO rH CJ QJ CO T) u a, <& CO o o 3 -H 4J CO r-\ QJ 4J <-3 -H H C 4-1 •rl 'H 6 In O CO 3 QJ HPLOh C O CO 00 CJ C -H •rl e X Q) •H X! E O CO CO U U Q) 3 N 4J •H rH rH 3 •H 3 4-1 -rl M (H QJ 00 PH < co CO QJ C o M CO 00 X c •H C 4-1 O C X CO c o •H 4J CO u CO a -i WClUUQ co co .X 00 o c O —I r-\ 4-1 4-1 CO o <-3 o CO Lfl QJ J-i CO 3 4J 4J ^-i X QJ •H CM e 4-1 00 rH C co •rl J= > Cu CO CO p-l < u CO QJ OJ XI •H 3 H Pi HHHMH(OHHNO\HrO>fHNnstH CN' CO •JHHHNCTlrlNrOlflCTli-IHNHH rtNncot>J-crJvDiniOiOinifliAiniOiniOi/liArN\OvD>DiOiCiDvDvO\OvD\OvO CS|CN1CSICSIC^CN1CNCS1C\ICNCNCNCNJCS|C\1CNJCSCNCNCNCNCN<>JCNICNCNC>JCN ON oo CM 00 CN CJ CO U QJ C OO 0) QJ C • 00 -H C >-! » o - co o co rH rH rH rH UH rH •rl -H g ta E OO C CO e o U5 •H C CO CO 4-1 C CJ Q) 3 P. g TJ •n o -3 T3 CO "3 t-i O rH O •H 3 Vj H S-i CO O rl 3 3 U QJ rH CO co Q, tH to s to 2 •3 O •3 O o 3 o ^> QJ rH 4-1 Cu co a u U J3 QJ CO QJ 14-1 3 QJ QJ S-, > 0- CD X o J3 QJ c oo c «• ■H CO > 4J U CJ QJ 3 CO *T3 QJ O J-l J-4 a a. "3 TJ o o o o 3 S QJ J-J 3 CO 4-1 J-l -H QJ C 3 >-i •H 3 CO M-4 -a r4 CO O XI J-i QJ T3 a. o 3 o P- 3 0) J-l 3 4-1 •H c J-l QJ 3 Sh U-i 3 4-1 -O -H rH 3 O J-i J= 3 OJ cm co 3 X) O rH X! O x: •3 QJ QJ CO J-4 3 QJ O 4J X CO rH rH O 3 X 4-1 a. qj 33 S o QJ CO • 00 3 3 •rl J-i QJ CL J-J CO 3 -a 4-> i^3 W QJ "3 CO rH CO O QJ X J-i QJ 4-1 CO 4-1 3 3 O 2 X QJ J-l J-i 3 3 4J 4J -H •H 3 3 J-i J-i 3 3 cm CM OJ Q) CJ CJ vH •rl IM CM CM CM O O CO CO QJ QJ QJ J-i M J-l 3 CO 3 3 4-1 QJ 4-1 4-1 X T3 •H X -rl 3 3 vH «H X- U <4H CO 3 CM c3 00 CO 3 3 •rl O TJ -H t«3 co 3 co O "3 •H 3 J-i 3 I O. I 3 a i i i oo I 3 I •rl I •3 I 3 X -3 3 O 4-1 O 3 3 £ •rl 4J X> 4-1 J-l Jj 3 3 3 3. CO a. -h rH 4-1 TJ 3 QJ X O 4-1 C 3 O QJ 3 p- S s > a i 3 CJ CO X rH QJ r-\ to 00 3 •H N 3 rH oo oo C CO 00 CO X CO TJ o o oo -3 Ou t-i rH rH CO rH 3 r-< <-i •H rH -H CJ H g m e QJ J-i §a qj r-i O 3 U Pi J-i 03 O -O M QJ 3- CL 3 3 P- Ph 3 CO O 3 cj a. o J-l rH QJ 3 a > 3 3 Pl, PJ 3 3 o a. x •o i^J QJ •3 J-l rH Q) O a E 3 M E CO 4J 4-1 CJ -3 » OJ QJ 3. 4-) T3 J-i 4J C 3 3 3 3 J-i -H •> 3. U CO 4J OJ "3 CO rH CO -H > rH 00 rH QJ QJ 3 3 -rl 3 3 -M J-i 3 O 3 «30ftcflUrtH I I I I I 3 I 1-1 I 3 I 3 I 00 I J-i I O I I <# I I o I ■rl I 3 I 3 I 00 I J-< I O I C I •n i <-\ co CO i-t •H 3 J-i 3 4J -H co E 3 QJ ■3 X 3 U H^Dc^-^N(OHCj^OHN^}lOC^-JlOlO^(^-JH c\iNNnnno>^CMrtrtrtMCMCNco<3'OMNro>jjvTvj'>j^jn M (-1 tc 4J t>< C/J W 3 X) C OO rH Pi J-i o en u cd M XI CO O u T3 CD CO 4J 3 co cn rH £ CD ID Pi CJ XI c CO J-l J-i O CD M 4J C CD CJ -H X> co E £ J-4 3 HH CJ C CD -H Jh .H >, rH u >, CO ■w nJ -u en i-l CD 3 o e T) c c e o o M 3 S J-4 o en U (D l-l X cn o CJ X cd in J-i 3 co en -H C CD CD Pi CJ X) c CO u CD e 3 c >n 4J en 3 x c On CO CM r-» cnNCimvONH(sn>jWHHN on oi h h r- i.-irHcococococo-d-iiou'i m cn cmcococococococococococococo co co cococococncocnconcirococon co SHMc^i-iNONHHCiincri-HN irivO^OcDONONON<— I CN CN CN CM CO CO cncicincnnn>3jJJ l-l 01 ex 3, CD O CX o 3 'H a CJ CD CD CD W C/0 o u en 3 e o 3 ^ c J-i •H CD O 6 Mh £ 3 £ rl H O N CO C >n >n -^ ^ J-i !-i U U 4-1 00 CD C e -h 3 en co 3 ^ O X U U <-3 CD c 00 c CO (0 CO E g E P j-i PL, P-l Ph P-l 00 C i 3 cfl 3 H CO X U X 4-3 <^> 00 E 00 -H E rH J-i en 3 e o 3 J-i C J-i •H CD E 4h 3 C rH O <-* z oo £ •H 3 CO J-i X 0) J-i I eo CO C •H 4-1 cn • CO CJ CJ • CD r-- on ON ON CM CM CO CO CM CO LO NO P^ i— I CN CO 00 CJ £ en 3 ri CCO 00 c o J-I -H J-I O 00 (X M J-I O rH 3 O M J-i en a) cn TO 1-4 •JJ < CO Z M CD 4-1 cd cn cd 3 B o J-I > J-I J-I CD CO c «w e O £ -H J-i O J-i 2 P-i u CD en • CD C *o CO « i— l en x> tH O 3 O CO 4J cn CD i-3 00 X CD J-i ^ CD rH X 4-1 C 3 ca i cn X G CO a: er co U CD ^J J-l ca en • 3 co o c • >, -H CD J-i 4J • CO 4J C 4-i -H •H <4-l « c CD CO OO j-i en e CO -H 3 rH X> •3 co E J-I CO CJ c ^ , e a •H 3 a* CD CJ •H 00 J-i E 4J •H CJ 4-1 CD CO i—l CD 01 t— ICMON>— ICMONi— I i— ICOlOONi— ICMCO .—l VO'XJvOONONON'— ICMCMCMCMCOCOCO -3" COCOCOCOCOCO ONONON OOOairlHHHrtHHHHHNNNNNNNCMNNNCMNNMNCMNNNCMN COCOCOCMCOCOCOCOCOCOCOCOCOCOCOCOCOCOCOCOCOCOCOCOCOCOCOCOCOCOCOCO"' vC CO CO CO CO 1 1 1 1 1 I I i Cfl 4-1 CJ 3 TJ 1 1 1 1 1 CJ O • 1 • cn J-l CJ 1 OJ cn •H OJ 1 " TD 0) oo rH 1 c ■— i cn >H X 1 Cfl O 4-1 C 00 •H CO •H C rH cn 01 1 4-1 J-i CO •H c cn 4-1 OJ - 0) cn OJ a • 4-J 4-1 1 CJ Cl r-l JZ •H en OJ ■— 1 Cfl rH c 4-J ex • u u CO 1 3 CD cn 4-1 E en >N •H 4-1 3 0) 3 a CJ • 3 OJ 1 TJ cn J-i ■H ^H 0) J-i CO CO 4-J CJ 4-1 4^ cn • Jjd 01 X -J I o u C OJ 4-1 3 T3 • rH X 3 X) 01 CJ • c K 4-1 1 J- •H ■4 ■H XI Jh 4-1 P- O u CJ J-l -a •H o rH • ■H c 01 J-l C 1 a. 4-1 (4-1 OJ •H o - o o MH TJ c co c -J 4-1 ex c U en oo M ^ X E ^J oo 0) <-^ o )- C C4-J CJ X - 01 — 1 o j-i CO c ^3 01 CJ X ■ CJ CO rH cn D- 00 o •H - Cfl u 0) cn en ~J u 0) i—l ■H X o «£ en J-l c cn •H J-l <4H OJ c I4H OJ t-i cn -3 4-J a E 4-J 4-1 co X ai X; a. C 01 u 4-J u oo OJ M H 3 •H r^ >H rJ 4-1 4-J CJ E cn o c o . — E ^3 X •H c CO 10 CO cn CO X n OJ D 4-1 -3 r4 CD X CO CO CJ CJ .* 3 O 4-1 4-J 3 3 3 3 X 3 en 14H ■H OJ 4-1 J^ OJ XI 4-1 CO c CO CJ o rH E C 3 0) 3 U -a O CJ cn X X Cfl O 3 M 3 OJ C ^H 4-1 a OJ > T3 CO TJ >H J-l a rH 4-1 CJ JJ •H C H -a o o 13 C c c Jj rH J-i o >H c a o a o £ OJ O CO -a j- rH CJ rH X OJ Oj o rH r. XI XJ i« Ih u — 00 C -a CD CO ^H CJ CJ a. ih CO 1—1 o 4-1 f>> 4-1 CO CO r^ Q. u X u X cx CD O a ex C T3 01 CO E 1-1 (0 X •H 00 JS OC C X CO 3 u 03 4-J C a. X J-i o -3 - 3 OJ 4-1 c a •rH J-i 0) H 1-1 o 14-1 u 0] cn J-l •H OJ OJ •H D. c CD Cfl cn E CO ^H 01 J4 M CO en J-i c i en CO 1-1 CJ - ^3 CJ Qj 3 3 3 CO CO >. 4H 4-1 E o > c ec r— i r^ •H CI i—l ^H CD 4-1 OJ - CD t iO- c u 4-1 •H r- 4H O O 01 rH u OJ OJ 1 E 4-1 —• 4-J 4J CO CO CO •H OJ o _C Cfl 3 en o> X c 3 E o X cn c -« E U OJ 01 OJ Oj 1-. J-l >. 3 cn X Cfl o* J-i Vj ^H J-i CJ Jx u P i-i CD en w e JD OJ V, 4-J Cfl 01 o CO ^ 3 1-4 U OJ CJ 4-1 CJ CJ X c u cn CO 1) X OJ O CJ X U) 4-1 ca T3 O O D CO c iO E u CO CO E •H U CO (-i 4-J 4-J c J-l 4J c c CO i = ex 4-J Id X X B c OJ CO •H CD OJ C o XI O cd : 3 O CD 0) i—i CD s- CD 1— I 4H •H •H ■H o o CD •' -1 >, 3 X !fl C3 — •*-4 a: u. s P- -J I—l Cj, CO (S ►J o J 2 C- X O CJ 03 CJ CJ Cfl > > (X Pu c_ CJ CJ Pi - J CJ CJ < < C E -?: H(>0>r4r' >-> -ir-i-irtrtiHiHrtNNCNlCNCNCNCNNNCSClCNIcNNNNNNINICNCN 1 CM c^rocococococococococococ^cococococococococococococococococococococ^coc^ 91 CO C O •H 4-1 CO CJ •H t— t 14-1 1 •H oo Cfl • ■ 01 w CO rH H CJ H CQ ►. H M s 4J 01 w 3 T) C H C/1 s M O U) O CD M 73 CO O CJ TJ 0) 05 4J 3 ca cn rH C CD QJ PS CJ 0) rH 4-1 ■H TJ c CO U CD e p c !*> M 4J 05 3 T3 C )-> O cn C_> 0) I— I T3 C/1 O CJ TJ qj en 4-> 3 CO 0) H C OJ QJ OS o -a C CO >-i > t-i 4-1 cn 3 •a c o\ h n sj in \o rs LO vO vD \D vO vO VD in ia in i/i m ui iri en en en n en m en i ONHNvoo>H(Ninioc> \Or-»r~»r^r-.0000C»00c» ininininmminmmin enencnenencnenenenen cjiHCMennNcn^ OHrtHMCSNM in^DvDvDvCiflvOvD enenenenenmencn C\ ~H cm en -tf m vjO cm en en en en en en MO ^D VO O vD \D vO en en en en en en en OJ c ■H cj co en e u o >> en u en 4J OJ en u 3 a t3 g c o •H CJ en oo c •H M CO qj X) en l-< c CD CO ■a •h cn u cl rH J) g H a 3 C/j P4 c I 00 OJ c 4J c3 ej w )h en 3 3 OJ OJ CL OJ i— I • e a co C O >% ej c_; H co I I cn i aj \ C I •H X 4-1 CJ C CO OJ e e a 00 -H C 3 •h cr en oj •H TJ ^ C r4 CO T3 X C O 3 l-i CO OJ -H g iH O CO •H -H 4-> CJ CO in g OJ O g J- 1 £ 3 O < CJ t I I cn a. E 3 S- 00 OJ C 3 -H t-i en CJ 4J • a OJ QJ • o C X « aj c -a o •H <4> 4-1 CO 00 l-i c OJ T-t 00 r-l •H 3 U en 14-1 CO OJ OJ PS IS X OJ CJ C CO -H 6 X CJ >. co ii e 4-1 en cn OJ rH CJ r-l •H OJ > cj u en OJ T-I CO S I I I I I cn i 4J I C I OJ I £ I 3 I U I 4-> I cn i C I 00 I C I •H I I-i I 3 I cn co cn aj i-i e oj • -H O CJ U Ml • 4-1 0) OJ CJ C • OJ CO C H 3 CO -C TJ i-i a c CO CO *H a s-i a oo r-i co co I e i o- en I •H H I 3 O CT 4J QJ CO U 00 01 I I I I I C 00 I •H -H I A! -4 I O <4-l I O OJ I CJ r-l I TJ TJ 0) C en oj c g eo a. u-i •H 3 k3 cr oj en OJ r^ l-i CJ OJ rH C ' O O QJ CJ JS ,C N . OJ OJ QJ oj cn en qj 3 3 >-« O O M-i cn tj QJ 3 CO QJ r-H CJ B 3 en 3 QJ cj C CO -H > -c cj T3 CO W I tn cr>i— (CMr--oooooooooocT. mmmmmmLnminin enencnenenenenenmen h cn en r- icseni— lr- iCTi enenenenenencnen ^-i i-H cm en vo r-~ o> 00 CTn cja ct^ o*\ CM CM en -Hr- icTNCN^HCNen-j-mvor-.-HCMoO'— i cs en •H •H • 01 ?-> CL rH u P C aj •rH 1 U 3 CO ^ CJ QJ cfl ■H •rl •H ■H • >> e o 1 QJ Cl 01 X C g 3 cfl 4-J 4-1 E c 1-4 >-« •H X ^ ^. I cn -o 4-1 QJ 01 •H cr l-i CJ 4-1 V-i QJ 1-4 rJ -w u u QJ . C CD 4-1 o o 1 TJ 4J 3 01 C c >^ Cfl c Si CJ M-l >, I-I •H C ^ 3 3 1 QJ CO ^ ID 4-1 •H QJ l-i 4-1 00 o 4-1 u OJ >nX •H ^3 V4 •H O cn O CJ J2 QJ 01 C e rH rH CJ c I-i CJ X o rH rH 00 rH •H H 1 IH 3 l-i C C r^ •H ua cfl CO c ■H QJ Cfl o fl ^ CO CO C rH I-I 3 Cl TJ 3 O •H l-i oo >> ^3 4-1 4-1 •H X C E cfl J= 4-1 j-i •H CO X !-i 01 M-J O 4-1 •H X CD c QJ 01 QJ QJ X CJ >> •H E cn a) J^ , C •H > » At E E o Cfl )-J X 01 to 4-1 l-i e e cn en C 0) CJ Cfl E QJ CJ OJ 01 01 CO O oo •H r. cn 3 rH 4-1 3 ^ e l-i X •H OJ 00 QJ 00 XI 4-1 cfl T3 01 cn CO 4-1 4-1 r* TJ E •r-t -a o 4-1 u3 01 01 U-l CO i-3 •H QJ Cfl CO 3 OJ " o * M OJ QJ o l-i QJ u •H c CJ 4-1 rH c rH 3 Cl CJ Cfl c 6 U co •H 4-1 s*. 01 4J c C 3 CL rH o •n rH •H 4-1 •H QJ I—i •H r-l QJ -o rH CO rH cfl i—i •H 01 4-1 g rJ 4-J c 14-1 CO QJ 4-1 01 •H ■H rH ■H 3 L4 4-1 CO l-i QJ rC CJ CO AJ CO CJ O Cfl QJ CD rH cfl t-i cn c CO QJ E 01 •H > > 01 3 X -C CO -o 4-1 -a OJ c 4-1 X aj CJ 01 4-1 CO 4-1 cn l-i 4-1 a- QJ rH 4-1 JO CO o 0) iJ Lj C c rH QJ C •H -a CJ CJ 4-1 o X o CL •H QJ CO X l-i •H OJ rH QJ •H C- CJ cfl 4-1 O QJ cfl rH CJ 4-J C cfl O •H •H rH o O c Cfl a CD O QJ o Cfl 1_ S ft. c/i < S X CU IS s: 2 c/1 C/1 O £ IJL O CO H [JU CJ E o w u cc r-l s X. s Ex, H -? Cu a, CMCn-Cj-vOO-H— IC5>-H •»^jvONMio enenenenenenenenen i n en \c r» -< ON Cj^ CT> CJ^ CT^ CT\ O ^3- - c co ^ cu x £ 9 c j-i en 3 -a c oo en • • cn w ca rH H c_> oo in as -cj- sr in HHHNHN(<1C-nc^C^CT>Cj>C7NO'i ^D^oooooooooooooaNCTsCTN cx i a i i i i ca i i c ^ cj J-i X -H CJ CU u 3 CO CD S !-3 I I I I en I r-l I cfl I •H I U I cu I 4J I ca i g l I -a i i en I oo i C I •H I -a c ^ •H >-l <4-l O 3 en >> (-1 u 0) cfl tH TD CU 3 CX CU CO -> hJ Sh aj co u cx co 3 ^ en I ■u 3 CO QJ -h g cx 3 rJ c3 4J CO (3 i— I c3 H > 0) !-i CO 3 U cfl cu ej en en H CU rH en g ,-h 3 co o Cfl S O Q 4-J ex CU CJ X QJ X 0) > en - en 3 o> QJ ,-H V-i cj -a >^ iH CJ •H -H X X c_) en en ■n i-i o -H o u oo 3 QJ u a, QJ cfl rH CJ x -h 4-> c co co x c3 CJ QJ oo g 3 •H <-3 4-1 u cn O 3 CU QJ Cfl CX, I I I I I I I I I I I en i T3 I O I O I 00 I -a cu c ca en CO cj <-3 I en t U QJ >, 3 s-i o en -h -h > •H QJ c CU 00 cx 3 •H TD ,* cfl >h QJ CO ►J s In QJ CM cu 3 CX QJ rH CO t-) cfl CX -H QJ CJ 3 g -H C 3 <4H I cn u cu c 0) 4-J in cfl Uh en O M U u < <-3 en cu « x en en 3 3 •h ^ a. x « ca en en 3 qj en O rH g 4-i -a o 4-1 QJ O 3 QJ l-i S2PQ OO C •H rJ OJ > o U In o o 1) o CO u 3 en cu X ID CJ )-l 4J co co >* cn co cu >H 3 a co cn cj TD <-3 4J 4-1 oo en 3 rH •H O cn cn cn TD -H O 4J O >-i 00 QJ CJ 4J • rH CU o • X 3 v3 •> cn 01 to ^ cu o cj jz 3 3 cn O o 3 ^4 U a. a. ^ CO en qj cn 3 3 4-1 I QJ 3 cn -h oo x; 3 o •H CO w e 4-1 •H ca U-l cn CU iH cx o •H O Cu 4-1 cfl c3 en qj en t3 •> •n 3 co cn CO CO CO cn X "H cfl iH cu cn cj ih cn cu 3 s- oo x e o QJ X 3 a cfl CO rH g o cn en en cu cu en > -h rl H fl "D a 4J 3 g )-i CO CO O U hJ S Cfl 3J QJ 3 cn •> CO U cu 4-i en }-i •H CJ S cfl cu u o a. qj o •H (X 4-1 P-i Cfl HHHMHCNCOsfHHNCO UIVONrsHHHHf)^''?"* COCOCOCOC^CTiCTiC^C^CriOC^ dfonnnnnfincnnn ^HCMCiiriHMcC^CTi^O>C^ai(^C^C^(TiONCT>CJ>0>CJNC^O>^ cnnncn<»ifnconcocncocncoconfOfnn>T oo cn o o cn r~~ CQ W >> U J-i cn a 3 M Cfl M u o cn c_> cu M ID Cfl O CJ qj en 4-1 3 cfl en H 3 QJ QJ OS U c CO cu X g 3 3 u 4J cn a 3 O— iCMHCNHNsfcnHCNn^C^ co-j^inincovorviNc^ejvc^oNacvi vDiDvO\OvOcO(CvD\OvOvO(OvOvDr° cincocnncncnnmcirinrinro mtnrsHMntjiHcsHcsHrHONHHfMHrtNfn HHHCMcscNNnfO'r-~r-^r---r~.r^r^.r-r^r^r^r^r^oooooooooooooo i QJ I • I 3 I I I I I I c cu e i (X I I I I I I I I I I I en i CO 3 u cr CO 0) cx CX 3 CO O I • I I cj l I • I I CU I I • I I 3 I 3 I cn cfl I cu •H U CX CX 4-1 CX g X cfl CO 00 I 3 •h cn > x) i x -a CJ 00 •H 3 u o CJ CJ QJ QJ u u > X H CX CX CO I cfl cj | U -H | 00 3 I CU 3 I ■H g qj g en 4J o u cj o en I I i« o X QJ 4J 4-J en u x 3 0) oo O i— I -H I U iJ «3 at 00 o o •H C -a o co x oc cx, > QJ H a 3 3 XI O oS 3 c OJ 3 o cu cx -H g ^ o cu CJ 4-> 4-1 CJ Cfl •H Xl c CU & en >^ cu >-i x T3 3 3 • CU cu g • cx c CX O CJ QJ -H -H H -a g OJ cfl CU H oi tfl en ca cu •h en l-i 3 QJ 4-1 4-1 CO 4-1 U cfl to X CX a >. CO u CO ^ g CO •H tJ 1-1 I (X, x h en 3 " CU cr 4-i -h 0) 3 l-i QJ TJ H g 3 CO CX 3 CJ -H O H 3 U-l u tr U QJ rH CJ CU QJ iH 0) H Cfl 4-1 cu cj co i-j ca o c ■H CJ C 00 OJ O C rH l-l UJ CjJ rH ca 4-1 CJ CJ 3 3 VJ rJ H H I u I cfl I cx I i ca cn cn u cu QJ <-t —I CJ ■H -H Cfl X U QJ > r4 cfl cx en 4J cu u 3 co •H CX 01 00 3 ^3 3 OJ en « ca S-4 4J cu c en i— i cu CU rH g 3 cu cx ■H CX -H 00 O 3 3 In CT u 00 oo 3 3 •H -H >H ^J •H -H CO Cfl CX cx QJ QJ I I I I I I en I I 4-1 I I en i I cj I en i u I cfl CJ cn 4J 4J In QJ cfl QJ CX Ih U CO I a I I «3 I I OJ 4-1 OJ CX OJ 'H -H 4J 4-1 4-1 rH i— I c3 ca 4-1 ca en 00 00 3 3 Uh tJH CJH <4H Cfl Cfl IH |H CJ CJ U U 2 < < < < 3 3 X X CX 4J •H tfl X o Cfl PQ cn ca QJ > •H T3 4J Cfl O C g >- O >H U -h O CO rJ OS en cu x - CJ en co cu o cj CJ Sn I-i CJ QJ U rH O -H 4J Cfl o u. 4-1 CJ 3 -h QJ 'JH g -H CX 4-1 •H C 3 QJ CT -H cu u en c o ca •H 4H 00 CO 3 i 4-1 -H r4 |H O CU CX QJ cn c c -H co oo >h 3 H W cn rH en o en 01 U QJ u u to •H C 3 > O 0) 01 CJ rH qj ca 00 rJ 3 3 en •H U U Ih cfl c 3 U QJ cn cu g cfl CX 3 0) g Ih 01 4-1 co en 4-1 QJ C -H rH cn cx cu CX -H 3 rH u cn en c o cx cx 3 •h cn en rH QJ cfl co CJ y c u ■H Cfl cfl cj CJ -H 3 tfl cfl cfl g cj X O -H CJ 4-1 4-1 v 3 a 2 < O 3 TJ H HI CU rH g g CX CX CX -H -3 tfl 3 CT rH rH QJ tfl Cfl CJ U rH •H -H Cfl 00 00 4-1 Ih l-i C 3 3 QJ Cfl Cfl Q o ^. ^, 93 co c o •w 4-1 TO CJ tH r- 1 C|H 1 •H 00 co • • EO w CO i— I H c_> \-i CO >1 l-f u g 4_) CO w 3 X) c M C/1 S Vj O CO U c CO J-l 01 rO £ c >, u ■u CO 3 c vD CX O X 0) CM co vo CM r--. vC CT- co * cr* -h oo co vO vo el >j < in vo vO vO Z vO O vo cn r^ CX ^~, QJ Oi cj oi 4J m VO rH O.M ro N CO I CM ON oo CX 0) o X OJ ON CO o co co O co CM o OO o co O CO CN co vO oo ■— i in co on oo r^. r-- r~- •> >— I CM H (O M « o o r-» cn i CX 4J 0J 3 CJ CO x oj 0J XI 3 CO OJ XI CO 3 O 0) C i£ cfl I I I I I cfl C O •H CO CO OJ <4-l O U CX CO 3 00 o I CO OJ u CO OJ CJ •H > OJ CO c o •H 4J CO CO QJ 4-1 CO l-l rH OJ QJ rO O l-l CO Cfl -H CO rH o o CN CO •H C CO CO Cfl QJ •H i— I CJ 4-1 i—l -H M HI > OJ CJ l-l > CO QJ ' -H OJ I CO I I ■-3 I I !-< I •H cfl CO CX OJ 01 U U 3 4J OJ CJ t— i -i-i 4-i <-a •H (X C x cu co OJ 4J t-i c QJ ^3 T3 CO -H 01 4J CJ Cfl < X CO O -H CO 4J U o o 3 O CO CJ •H CO T3 OJ E CO 4J •H S-i a qj co jz; O 4J X O qj co co oo 5-1 H O cfl C 4J O -H •H U-i 4-1 O Cfl S-l CJ CX W 2 OO oooooooo PS o co C_) QJ i—i -a tn o o oj co 4J a CO CO ^H C QJ QJ Pi O T3 C CO U OJ £ 3 C >, t-i 4-1 CO 3 ID C CO vO CO -J" CN VD VD co co CO r-l VD vO CO CO o CO CO ex Oj CJ X QJ CO 00 St cx OJ CJ X QJ vO CO CO CO CT> 0> cr> o> 4-1 OJ TD C o 00 M C 4-1 •H CJ U QJ •H i— I 12 W CO —I o o uO r^ UO uo o cx X c CO S-i H I I I e o •H 4-1 CO CO C 4J cn o <■ -d- CO CO ON CJ> CJ> -J CO VD 1 QJ m 3 7-H ^ vo C TO in ON TO X m cn e ^ „ I s * 4-1 CO J ^^ m a> QJ 01 CO m CJ r4 OJ t- * X 3 CJ CO CO QJ 4-J •H cw UO TO M-l r r^ O CM O CM m in m Ohm vD \C vO ^J I OJ TO CJ X QJ QJ CO 4J c QJ 4-J k3 cfl rH C QJ TO t-l X 1-1 ^ 3 CO 3 TD CO CO O - U rH i—l TO •H CJ TO O Pi rJ r-H CN o o UO LO VO vD -3 I >, !-i CO U_i x >-i 00 o •H 4-1 x: o 2 I 4J O I TO -H I 4-1 4J I M TO O 4-1 oo a C CO ■H C CO TO 3 r4 M o CX CO 0) c c TO -H rJ rH 4J OJ vj a •H -H < p. OJ QJ co c co o c ■h o oo c •H 4-1 CO TO CJ TJ TO O U rO c o I CO 4-1 CO QJ CJ >H > Qj CO >^ S-i CO TO 4-1 4-1 TO lH U O iH a c CO 3 C i to e l-i O H U C rH O QJ •H 4-1 CO •H i*3 > QJ O r-i -rH OJ T3 4-1 TO Pi 3 4J TO •H CO CJ rH •H -H i-3 l-l 4-1 4J 3 l-l CJ QJ QJ CO 4-1 i — I TO TO W CJ S co o m -j- in vo r^ o o o o o o uo in uo in vo vO vO vO vO vO 94 O —i CM CO O O O o r^. oo oo oo vO vO vO vO QJ TO l-i QJ TO U CO QJ -3 ■— 1 O 0) r-l s§ TO CO OJ rH rH o O ON vO QJ TO l-i QJ Pi CN O OS vO t T-l CO M OJ r* l-l co E OJ E •H O CJ cj C OJ ^ 00 TO >, 00 4J C 4J -H •H "H rJ ^•0 3 C QJ CJ TO l-i CJ ca u w •— I CM CO O O O o o o r^ r^ r^ EXHIBIT E:8-2 WATERBORNE COMMERCE RELATED INDUSTRIES WITH RIMS CODES 0300 Forestry and fishery products 1513 New warehouses 1721 Maintenance and repair of other non-farm buildings 1519 New other non-farm buildings 1627 Other new non-building facilities 1737 Maintenance and repair of other non-building facilities 3731 Shipbuilding and repairing 3732 Boatbuilding and repairing 6501 Railroads and related services 6503 Motorfreight transportation and warehousing 6504 Water transportation 6507 Transportation services 6506 Pipeline transportation 6901 Wholesale trade 6902 Retail trade 7001 Banking 7002 Credit agencies 7004 Insurance carriers 7005 Insurance agents and brokers 7102 Real estate 7301 Miscellaneous business services 7303 Miscellaneous professional services 95 CALCULATION OF SECONDARY IMPACTS Step F:l PURPOSE: Determine the share of direct income that can be considered basic. DATA REQUIREMENTS: None. DISCUSSION: Simply, total income related to the port industry is the product of port industry basic income and the income multiplier derived in Step E:7. The problem is that some of the income from port activities can be considered basic and some non-basic. The port industry may serve both export and import (local) markets equally. In general, income earned from outbound cargo movements can be considered basic regardless of whether the cargo originates inside or outside the study area. The allocation of income earned from inbound cargoes is more dif- ficult, since these are not all destined for local consumption. Mini- bridge, land-bridge, and other cargoes whose destinations are outside the port study area produce basic impacts. Similarly, inbound cargoes which are intermediate inputs to products manufactured in the study area and then exported can also be considered to have basic impacts at the level of the port industry. However, income earned from cargoes consumed locally cannot be attributed to the basic sector. RECOMMENDED ACTION: Many smaller Pacific Coast ports are engaged primarily in the export of one or two major commodities. Any imports of cargoes for local consumption are negligible in comparison to export activity. Therefore, these users can reasonably treat all port industry income as basic. For larger ports with substantial inbound traffic, the identification of non-basic income may be quite difficult. One possibility would be a review of cargo destinations to determine local consumption. Sources for this would include Customs records, knowledge of the local economy, or optional questions included in the port industry survey. If cargo movements can be divided into basic and non-basic activities, then assumptions can be made about the income associated with each. 96 Because it is likely to be very difficult to determine the non-basic share of port industry income with any degree of confidence, it is gener- ally recommended that all port industry income be considered basic for the purposes of this analysis. In most cases, this will result in a moderate overstatement of impacts, somewhat offset by the conservative approach used in calculating the multiplier. For those few cases in which non-basic activity accounts for a significant share of port indus- try income, the most useful approach may be to incorporate optional ques- tions about this activity into the survey instruments. USE OF RESULTS: The basic component of port activity income will be used to determine total port-related impacts in Step F:2. 97 Step F;2 CALCULATION OF SECONDARY IMPACTS PURPOSE: Determine total income supported by the port. DATA REQUIREMENTS: Results of Steps E:7 and F:l. RECOMMENDED ACTION: If all port industry income is to be considered basic, total port-related income is the product of direct income and the income multiplier. Secondary income is the difference between total income and direct income, If port industry income has been divided into basic and non-basic com- ponents, total port-related income is the product of basic income and the income multiplier plus the non-basic component of direct income. Second- ary income is the difference between total income and direct income. As noted in Step D:6, direct income represents payroll plus benefits. Other labor and proprietors' income is not included. USE OF RESULTS: Direct, secondary, and total income impacts can be published in the eco- nomic impact report and used to calculate other impact measures through ratio analysis. 98 CALCULATION OF SECONDARY IMPACTS Step F;3 PURPOSE : Determine total sales supported by the port. DATA REQUIREMENTS: Results of Steps E:6 and E:8 (RIMS multipliers or input/output sales multipliers) . RECOMMENDED ACTION: The products of direct final demand sales by industry (i.e., after ad- justing for double counting by subtracting freight forwarders and transportation brokers sales) and the RIMS gross output multiplier for each industry are summed to determine total gross output (or sales) sup- ported by the port. Secondary sales are the difference between total sales and direct sales. The total gross output multiplier for the port industry can be calculated by dividing total gross output by total direct sales. Canadian port regions for which there are input/output studies may be analyzed by the same procedure. The RIMS multipliers are presented in the form of direct effect com- ponent, indirect effect component, and gross output multiplier for each industry. The direct effect component j_s not the same as the direct impact of the port industry. Rather, it represents what are known as direct requirements in input/output analysis — the first round of pur- chases made to produce a given output. The gross output multiplier is expressed as 1 plus direct effects plus indirect effects, where 1 repre- sents direct port industry sales, the direct effects component represents the first round of purchases, and the indirect effects component repre- sents all remaining rounds of interindustry and consumer spending. A less expensive, but much less useful approach to using the RIMS multi- pliers is to calculate impacts based on the aggregated multipliers for BEA economic areas. These are presented in "Industry-Specific Gross Output Multipliers for BEA Economic Areas" published by the U.S. Depart- ment of Commerce. (Available from the Superintendent of Documents, U.S. Government Printing Office, Washington, D.C. 20402, No. 052-045-000-48-7, $2.00.) Most BEA areas consist of several counties and at least one large city, so that these multipliers will overstate impacts at the single-county level. Moreover, because these multipliers were originally developed to analyze water projects, the 56 sectors included provide significant detail for agricultural production but highly aggregated data for other industries. Nevertheless, if a decision has been made not to 99 order county and industry-specific RIMS multipliers for the port, the published data can serve to check the reasonableness of the economic base multipliers. Briefly, since the size of the BEA econo- mic area is likely to overstate the secondary impacts of the port indus- try, an economic base multiplier which exceeds the output multiplier derived from published BEA data should be treated with a large amount of skepticism. USE OF RESULTS: Direct, secondary, and total sales impacts can be published in the econo- mic impact report and used to calculate other impact measures through ratio analysis. 100 CALCULATION OF SECONDARY IMPACTS Step F:4 PURPOSE: Develop an income/sales ratio to compare the results of the economic base and RIMS approaches. DATA REQUIREMENTS: Most recent County Business Patterns for the port's state. DISCUSSION: Obviously, the purpose of calculating port industry impacts by two dif- ferent methods is to compare the results. Since each method is based on different assumptions and data and has different flaws, similar results would indicate a reasonable level of accuracy. However, since the econo- mic base technique used yields total impacts in terms of income and the input/output (e.g., RIMS) technique yields total impacts in terms of gross output (sales) , a common unit of measure is needed before the com- parison can be made. The means for making this comparison described below is a payroll/sales ratio which can be used to convert economic base income impacts to sales equivalents and input/output gross output impacts to income equivalents. (Note that because the income impacts are based on payroll plus fringe benefits, published data are used to estimate the wage and benefit com- ponents of total employee compensation.) The methodology described below is a simplified one based on the limited published data, money, and time available to most users of this kit. Information to calculate payroll/sales ratios for individual counties is published in the censuses of business (construction industries, mineral industries, manufacturers, retail trade, wholesale trade, selected ser- vices, and transportation) conducted every five years. The collection of data from the seven different censuses for two or more years is a tedious and expensive task. The alternative methodology should be adequate for the needs of most ports. The approach to undertaking the county-specific analysis, if a user wishes to do so, should be apparent from the recom- mended action described below. RECOMMENDED ACTION: Exhibit F:4-l shows payroll/sales ratios for broad industry groups for the three Pacific Coast states. Estimated ratios for industries not covered in the business censuses were developed from national data on 101 employee compensation and business receipts by industry. Ratios are based on 1972 data, the latest year for which business census information is available. The Vancouver, British Columbia ratios were taken from An Interindustry Study of the Metropolitan Vancouver Economy completed in 1972. Historically there has been very little change in these industry ratios over time. Since total port industry income and output impacts are not disaggregated by industry, it is necessary to compute a single payroll/sales ratio. This is done by determining an average payroll/sales ratio weighted by port county employment in each industry group. (This average ratio is the sum of employment in each industry division times its payroll/sales ratio divided by total employment.) The result is the relationship between payroll and sales for the port county for the year for which the income and output data were developed. (An advantage of this approach is that it eliminates the need to adjust for inflation, assuming that wages and prices inflate at the same rate.) In order to compare the results of the economic base and RIMS techniques, the relationship between payroll, as used in the payroll/sales ratio, and income, derived from the economic base multiplier, must be determined. Payroll (income) is defined in the surveys to include fringe benefits; payroll data used to calculate payroll/sales ratios do not. National data (published annually in the "Handbook of Labor Statistics, BLS) indicate that benefits currently account for about 20% of total employer compensation. Therefore, after adjusting total income impacts to exclude the fringe benefit component, the results of the two approaches can be compared by using the payroll/sales ratio. Because many of the steps in this analysis are rather imprecise, a 10-15% difference between the results of the two approaches is an acceptable indication of accuracy. In this case, either the average or the lower of the two totals can be considered the final estimate of port-related impacts. If the results are significantly different and remain so after checking for possible arithmetic errors, the conservative approach dictates selecting the lower estimate of port industry impacts. USE OF RESULTS: The final estimates of income and output impacts can be published in the economic impact report and used to calculate other impact measures through ratio analysis. 102 EXHIBIT F:4-l PAYROLL/ SALES RATIOS FOR MAJOR INDUSTRY DIVISIONS Agriculture, Forestry, and Fisheries 3 Mining 3 Construction Manufacturing SIC 20 Food and Kindred Products 21 Tobacco Manufacturers 22 Textile Mill Products 23 Apparel and Other Textile Products 24 Lumber and Wood Products 25 Furniture and Fixtures 26 Paper and Allied Products 27 Printing and Publishing 28 Chemicals and Allied Products 29 Petroleum and Coal Products 30 Rubber and Plastic Products N.E.C. 31 Leather and Leather Products 32 Stone, Clay, and Glass Products 33 Primary Metal Industries 34 Fabricated Metal Products 35 Machinery, except Electrical 36 Electrical Equipment and Supplies 37 Transportation Equipment 38 Instruments and Related Products 39 Misc. Manufacturing Industries Transportation, Communications, Utilities 3 . Wholesale Trade Retail Trade Finance, Insurance, and Real Estate 3 Services Vancouver British California Oregon Washington Columbia" 3 .194 .194 .194 .132 .171 .171 .171 .257 .257 .257 .524 .100 .122 .191 .200 .173 .229 — .338 .236 .286 .191 .182 .184 .262 .280 .248 .340 .159 .139 .158 .290 .294 .365 .325 .134 .032 — _ .194 .025 .052 .206 — .203 — — — .286 .222 .228 .234 .204 — .120 .344 c .237 .229 .235 .288 .279 .268 .305 .307 .226 .329 .254 — .236 .344 c .332 .302 .375 .234 — .355 .319 .319 .319 .056 .050 .061 .466 d .135 . 129 . 132 .135 .135 .135 .277 .351 .315 .342 e U.S. average ratio for California, Oregon, and Washington. b Determined from An Interindustry Study of the Metropolitan Vancouver Economy , 1972. Manufacturing, N.E.C. "Trade and transport ratio is .466, communications ratio is .409; utilities ratio is . 162. e Health and Welfare ratio is .565; education ratio is .547; business services ratio is .435; other services ratio is .456. Arthur D Little Inc 103 CALCULATION OF SECONDARY IMPACTS Step F:5 PURPOSE: Update input/output (RIMS) multipliers based on income multipliers (optional) . DATA REQUIREMENTS: Bureau of Economic Analysis county personnel income data for the RIMS base year, County Businesss Patterns for the RIMS base year for the port's state, the United States summary, and other states, as needed, or for Canadian ports, Employment, Earnings and Hours and local or regional input/output study for the port region being studied. DISCUSSION: A significant criticism of the RIMS multipliers is that they are based on 1967 data. (An updated version of RIMS, based on 1972 data, is expected to be available by the end of 1979.) Multipliers based on dated informa- tion may ignore substantial changes in the structure or size of a region's economy. In other words, the RIMS multipliers will tend to understate the impact of the port industry in a region whose economy has expanded and overstate the impacts in a region where economic activity has declined. Conceptually, at least, the input-output Type II multiplier, which consi- ders all of the rounds of spending and production created by a given direct activity, is the equivalent of the economic base multiplier. (In practice, there are data and definitional problems which preclude the calculation of identical results.) Therefore, it can be argued that changes in the I/O (RIMS) multiplier over time should parallel changes in the economic base income multiplier. By determining the change in the income multiplier from the I/O base year to the present (the most recent year for which data are available) , the change in the I/O multipliers can be estimated. RECOMMENDED ACTION: The procedure for calculating the income multiplier is described in Steps E:l through E:7. The steps should be repeated to determine the income multiplier for the RIMS base year (currently 1967). If only one of the techniques for calculating the employment multiplier (Steps E:2, E:3, and E:4) was used, it is necessary to use only one technique to determine the 104 base year multiplier. Because BEA county income data for years prior to 1970 are not published, the necessary 1967 information must be requested from the Regional Economic Measurement Division or obtained from one of the state agencies or universities to which local personal income esti- mates are sent (see Exhibit F:5-l). When RIMS multipliers based on the 1972 I/O table become available, income data can be obtained from BEA publications. The base year and current income multipliers should be compared and the percent change computed. This percent change is then used to adjust each of the RIMS multipliers. For example, if the income multiplier was 2.1 in the base year and is currently 2.2, the RIMS multiplier for each in- dustry should be increased by 5%. Total gross output impacts can then be calculated with these updated RIMS multipliers (see Step F:3). The procedure is the same for Canadian ports that wish to update local or regional input/output studies. It is necessary that income information for the same region as that used for the I/O study be obtainable. It must also be in sufficient detail to construct an economic base multiplier . USE OF RESULTS: The gross output impacts calculated with updated RIMS multipliers can be compared to total income impacts (see Step F:4) to determine the final estimate of total impacts. Use of the updated RIMS multipliers may help to narrow the difference between the results of the economic base and I/O approaches. 105 EXHIBIT F:5-l STATE AGENCIES AND UNIVERSITIES RECEIVING BEA LOCAL INCOME ESTIMATES Alaska Institute of Social, Economic and Government Research University of Alaska College, Alaska 99701 Department of Economic Development Alaska Division of Economic Enterprise Pouch EE Juneau, Alaska 99801 Office of Policy Development and Planning Office of the Governor Juneau, Alaska 99801 California Center for Business and Economic Research California State University - Chico Chico, California 95926 Department of Finance and Industry School of Business California State University - Fresno Fresno, California 93710 Institute of Business and Economic Research University of California - Berkeley 556 Barros Hall Berkeley, California 94720 Department of Finance State of California Sacramento, California 95814 UCLA Business Forecasting Project Graduate School of Management University of California - Los Angeles Los Angeles, California 90024 Hawaii Department of Planning and Economic Development Post Office Box 2359 Honolulu, Hawaii 96804 106 EXHIBIT F:5-l (continued) Oregon Bureau of Business and Economic Research University of Oregon 140 Commonwealth Hall Eugene, Oregon 97403 State of Oregon Department of Economic Development 317 S.W. Alder Portland, Oregon 97201 Budget Division Oregon Executive Department 240 Cottage Street, S.E. Salem, Oregon 97310 Research and Statistics Employment Division Oregon Department of Human Resources 402 Labor and Industries Building Salem, Oregon 97310 Washington Economic and Planning Division Department of Commerce and Economic Development General Administration Building Olympia, Washington 98501 Graduate School of Business Administration and School of Business Administration Office of the Dean University of Washington Seattle, Washington 98195 Research and Information Division Department of Revenue Olympia, Washington 98504 107 I CALCULATION OF SECONDARY; IMPACTS Step F:6 PURPOSE Collect data to develop impact measure ratios and calculate port-related impacts expressed in units of interest to the user. DATA REQUIREMENTS: State employment and payroll data for year described in the survey; U.S. Department of the Treasury, Internal Revenue Service, Statistics of Income , Individual Income Tax Returns and Business Income Tax Return s for most recent year; state and local tax data as needed. Canadian port users should obtain Revenue Canada's Taxation Statistics (No. RV44) , Cor - poration Taxation Statistics (ISSN 0576-0119) , and Corporation Financial Statistics (ISSN 0575-8262) from Statistics Canada. DISCUSSION: The procedures described in the preceding steps yield estimates of port industry impacts expressed in terms of income (payroll) and output (sales). However, these measures are often not the most useful to the port. Typically, a report directed to the general public will express impacts in terms of employment, which is the most widely understood mea- sures. Port-related tax payments are also frequently reported to illus- trate the benefits resulting from public investments in port facilities and services. Ratio analysis is used to redefine impacts in terms of various units of measure. Simply, secondary data sources are used to determine the rela- tionship between the available measure (the one in which impacts are cur- rently expressed) and the desired measure, and this relationship is used to convert total impacts. For example, once the employment/payroll ratio is determined, total port-related employment can be readily calculated from total payroll impacts. There are two principal problems associated with ratio analysis — the variation in ratios associated with the level of data aggregation and the differences between average and marginal relationships. An example of the first problem is the use of average wages per employer to calculate total employment. A county might have an overall average wage of $12,000 per employee, and this could be used to convert total payroll impacts to total employment impacts. However, construction workers might average $20,000 per year and retail sales workers $8,000. If the payroll impacts are created primarily in one of these sectors, total employment impacts would be quite different from the average. Using the overall average implicitly assumes that impact employment distribution among industries is the same as total employment distribution. 108 The difference between average and marginal relationships is of greatest importance when ratios are used to make forecasts. As noted in Step F:4 f payroll/sales ratios tend not be change over time at the national level, but this is not necessarily so for individual businesses or even coun- ties. For example, a retailer might be able to handle a 10-20% sales increase without additional employees and only then need to hire addi- tional sales clerks. A much larger sales increase could take place before additional bookkeepers, buyers, or managers were needed. Alter- natively, a relatively small increase in sales could give a business the "critical mass" it needs to hire new types of employees — e.g., computer programmers, advertising specialists, etc., — to replace outside ser- vices. These examples clearly indicate that ratios may change with changing conditions and should be used with care. RECOMMENDED ACTION: Impacts should be calculated in terms of employment and, if desired, income taxes generated. A user who wishes to express impacts in some other measure can create additional ratios following procedures analogous to those described below. However, it is important to recognize the limitations of this method and understand that each successive transfor- mation of impact estimates is likely to decrease their, accuracy. Employment is calculated by determining average annual wages per employee for the study area and dividing this into total impact payroll to derive full-time equivalent employment. The best sources for payroll and employment data are the state agencies which collect them under various federal programs. In the three Pacific Coast states, plus Alaska and Hawaii, they are: Employment Security Division Department of Labor Post Office Box 3-7000 Juneau, Alaska 99802 California Employment Development Department Post Office Box 1679 Sacramento, California 95808 Department of Labor and Industrial Relations Post Office box 3680 Honolulu, Hawaii 96811 Oregon Department of Human Resources Employment Division 875 Union Street, N.E., Room 402 Salem, Oregon 97310 Washington Employment Security Department 1007 South Washington Street Olympia, California 98501 In Canada all employment and payroll data are collected by Statistics Canada. 109 Annual data from the appropriate state agency should be used rather than County Business Patterns data because the latter represent employment for only a single week and may be distorted by seasonal variation in some industries. If employment and payroll information is not available for the year covered by the survey (that is, the year for which direct pay- roll impacts are reported) , the Consumer Price Index is a readily avail- able statistical source that can be used to adjust average payroll per employer for inflation. Statistics Canada's Employment, Earnings and Hour s is published monthly. An annual average can be estimated from the monthly statistics. Ratio analysis can also be used to determine federal, state, and local income taxes paid. Estimates of other state and local taxes (sales, pro- perty, etc.) should be limited to the direct impacts derived from the survey because of the difficulty of avoiding double counting secondary impacts. Personal income taxes can be calculated by determining the effective tax rate on wages and salaries. U.S. Internal Revenue Service, Statistics of Income , Individual Income Tax Returns , published annually, shows total income tax paid as a percent of adjusted gross income, and this can be used to estimate individual income taxes paid to the Federal government. Similar information on state income taxes can be obtained from the appro- priate state agency. Where state or local governments levy flat rate income taxes, the prevailing rate can simply be applied to payroll impacts. Revenue Canada's publication entitled Taxation Statistics pro- vides similar information for Canadian users of this "kit". Business income taxes must be calculated in two steps. First taxable income must be calculated, and then appropriate tax rates can be applied. U.S. Internal Revenue Service, Statistics of Income, Business Income Tax Returns , published annually, shows receipts and net income for major industry divisions for proprietorships, partnerships, and corpora- tions. These data can be used to calculate net income as a percent of receipts for all businesses and for each industry division. Because estimates of sales are not disaggregated, the percent for all businesses must be used. (There are no readily available local data to derive a weighted income as a percent of sales figure.) Once net business income has been estimated, prevailing federal, state, and local tax rates can be used to estimate business income tax payments. The Canadian publication Corporation Financial Statistics can be used in conjunction with Corporation Taxation Statistics to first estimate a sales-to-net-income ratio that would then be used with the data in Corpo - ration Taxation Statistics to determine tax payments. USE OF RESULTS: These ratio analysis techniques can be used to convert port-related impact estimates to the form most useful for public reports. 110 PREPARATION AND DISSEMINATION OF REPORT Step G;l PURPOSE: Prepare a report describing the results of the economic study conducted using this kit. DATA REQUIREMENTS: Results of all preceding steps. DISCUSSION: The type of report prepared depends primarily on the purpose of the study. If the port's main purpose is a public information effort high- lighting the benefits of the port to the local economy, its report should be a short document summarizing the major findings of the study and aimed at a wide readership. An economic impact study done to support policy or investment decisions about which there is likely to be debate will require a much more detailed and comprehensive report. A report of this type would describe research methods, data sources, calculation proce- dures, and, in effect, each of the steps contained in this kit. The detailed report should also contain a short summary of findings similar to that described above for distribution to the general public. If the port decides to publish only a summary report, it is still necessary to have on file the type of information required for the detailed report (e.g., copies of raw data and calculations) to answer any questions about how the findings were obtained. RECOMMENDED ACTION: The port economic report should contain, at a minimum: • A statement of the purpose of the study. • Definitions of any terms used in the report, includ- ing a specific definition of the study area selected. • A description of the methodologies used to estimate direct and secondary effects. • Port industry direct and secondary impacts and port- dependent industry direct impacts expressed in terms of employment, income, output, taxes paid, or other relevant measures. • A comparison of port industry impacts to total county or regional economic indicators. Ill The amount of material included will depend on the type of report. For example, a summary report might describe the methodology used with "A survey of firms in the port industry was used to estimate direct impacts. Secondary impacts were estimated by an economic base multi- plier." A detailed report would probably include copies of the question- naires used, the number of questionnaires mailed and the response rate, an indication of the coverage of the survey (e.g., firms employing 75% of all workers in the warehousing industry responded) , calculations of secondary multipliers similar to Exhibits E:3-l and E:4-l, and other information so that the reader can follow each step and duplicate them, if desired. However, survey data should not be presented in sufficient detail for individual firms 1 responses to be identified. When several firms in an industry respond, aggregated survey results for this industry can be presented. For industries in which there are only one or two firms or responses, data included in report tables should be grouped into an "other" category. The format of the report and the method of data presentation will be selected by the project director. It is essential, however, that the published results do not imply more precision than is warranted by the methodology. Estimates of impacts will be reasonably accurate; they will not be precise. It is important for the user to acknowledge this uncertainty. Rather than reporting "the port industry supports 642 jobs in our community," the report should state, "it is estimated that the port industry supports approximately 650 jobs" or "the port industry supports some 600-700 jobs." Since the reporting of even rounded estimates may be interpreted as precise numbers, we recommend the use of graphics, particularly in widely distributed reports. A pie chart could be used to show port industry direct and secondary employment in relation to total study area employment. A bar chart could be used to show the change in port industry impacts from one study year to the next. Presen- tations of this type are often more effective than lists of numbers in conveying information to the public. 112 PREPARATION AND DISSEMINATION OF REPORT Step G:2 PURPOSE: Disseminate the port economic impact report to interested individuals and groups. DATA REQUIREMENTS: Reports prepared in Step 6:1. RECOMMENDED ACTION: Copies of the economic impact report should be sent to local government officials, newspapers, public libraries, firms which participated in the survey, and any other interested parties. (A newspaper article might suggest that members of the public request copies from the port district.) If only a summary report is prepared, it should be sent to the indivi- duals and groups mentioned above along with the name of a person at the port (e.g., the study director) who can be contacted for additional in- formation. If a detailed report is prepared, a complete copy should be sent to public libraries, newspapers, and government agencies. Partici- pating firms and individuals should receive the summary section only. Those who request more information could then be sent the complete report. 113 LIST OF DATA REQUIREMENT PUBLICATIONS UNITED STATES Bureau of Economic Analysis, U.S. Department of Commerce, Regional Economic Measurement Division, County and Metropolitan Area Personal Income Data, available from National Technical Information Service, Springfield, VA 22161 Bureau of Economic Analysis, U.S. Department of Commerce, Regional Economic Measurement Division, County and Metropolitan Area Personnel Income Data for the RIMS Base Year, available from National Technical Information Service, Springfield, VA 22161 County Business Patterns , available from Superintendent of Documents, U.S. Government Printing Office, Washington, D.C., 20402 Dun and Bradstreet Directory, available through Dun and Bradstreet, Inc. 99 Church Street, Post Office Box 803, New York, NY 10008 Employment and Earnings , published by Bureau of Labor Statistics, U.S. Department of Labor, Washington, D.C. "Industry-Specific Gross Output Multipliers for BEA Economic Areas", published by the U.S. Department of Commerce. Available from the Super- intendent of Documents, U.S. Government Printing Office, Washington, D.C. 20402, //052-045-000-48-7, $2.00. Standard Industrial Classification Manual , published by the Office of Statistical Standards, Bureau of the Budget, Washington, D.C. Coast Marine & Transportation Directory (1978) published by Pacific Shipper, Inc., 1050 Sansome Street, San Francisco, CA 94111, $8.00. CANADIAN Corporation Taxation Statistics , ISSN 0576-0119, $2.80. Corporation Financial Statistics , ISSN 0575-8262, $7.50. Taxrtion Statistics , Revenue Canada RV44-78 Standard Industrial Classification Manual, Revised 1970 , Statistics Canada, Publications Distribution, Ottawa, Canada K1A 0T6, Cat . //12-501E , $6.00. Employment, Earnings and Hours , Statistics Canada, Publications Distribution, Ottawa, Canada K1A 0T6 , Cat.//72-002 Canadian Census Vol. Ill, Part 4, Industries by Sex for Census Metropolitan Areas, Place of Residence, and Place of Work (Bulletin 3.4-5), May 1975, $1.05. Canadian Census, Vol. Ill, Part 4, Industries by Sex, for Census Agglomerations of 25,000 and Over (Place of Residence) and Census, Agglomerations of 50,000 and Over (Place of Work ) (Bulletin 3.4-5), May 1975, $1.40. Canadian Statistical Review, Cat. // 11-206. Arthur D Little Inc 114 Local or Regional Input/Output Studies - local universities, colleges and offices of the Ministry of Economic Development should be canvassed to determine the availability of input/output studies. For example, a study for the Vancouver area was conducted by H. Craig Davis of the University of British Columbia. It is entitled An Interindustry Study of the Metropolitan Vancouver Economy and is availble from Mr. Davis at the School of Community and Regional Planning at U.B.C. 115 IV. GLOSSARY BREAK-BULK CARGO - A catch-all category for cargoes not shipped as dry bulk, liquid bulk, in containers, or in unitized form. Most commonly it includes cargoes packed in cartons or bags that are loaded or unloaded by hand. BULK CARGO - Unsegregated cargo, either in liquid or solid (dry) form, that is handled and loaded (or can be handled and loaded) in a continuous operation — i.e., there are no discrete units to be loaded. The cate- gory includes: ores, petroleum products, metal scrap, wood chips, ferti- lizers, and other chemicals. CONTAINER CARGO - Cargoes shipped inside of standardized metal boxes (most often 8'x 8'x20' or 8'x8'x40'). DIRECT IMPACT - The first round economic impact — i.e., in the case of the port industry employment, income, sales, taxes, etc. associated with the operation of port industry firms (this involves the netting out of non-waterborne commerce related employment, income, etc., for some firms, e.g. banks) . Direct impact is used synonomously with PRIMARY IMPACT. ECONOMIC IMPACT - Changes in the level of economic activity attributable to a particular event or activity. It is generally measured in terms of employment, income, tax revenues, or sales. For example, the employment at a port's container freight station is, to a certain extent, tied to the availability of container handling equipment and facilities at the port. However, the economy of a region is comprised of firms that exhib- it mutual dependence, therefore, it is inaccurate to state that the exis- tence of one firm is entirely dependent upon another. EMPLOYMENT IMPACT - The economic impact of a project or economic activity measured in terms of employment — i.e., the number of jobs or full-time equivalent positions attributable to the project or activity. GENERAL CARGO - Segregated cargo handled in containers or as break-bulk. This generic category also includes logs, lumber, pulp, motor vehicles, steel, and other metal alloy products. INCOME IMPACT - The economic impact of a project or economic activity measured in terms of income. Income can be defined in various ways — e.g., as gross pay (e.g., Statistics Canada - earnings) or as more com- prehensive measures including proprietors income, government transfer payments, rental income, and/or dividend income (e.g., BEA - labor and proprietors' income). INDIRECT IMPACT - Economic impact beyond the first round impact. For the purpose of this study this would include the employment and income gene- rated in industries or firms supplying goods to the port industry firms; a component of SECONDARY IMPACT. INDUCED IMPACT - Economic impact — e.g., employment and income — gene- rated by the spending of employees in firms providing direct or indirect economic impacts; a component of SECONDARY IMPACT. 116 NEO-BULK CARGO - General cargo that moves in unitized form often on ships designed specifically to carry one type of cargo. The category includes autos, lumber, and steel products. PORT-DEPENDENT INDUSTRY - The group of firms or activities that ship cargoes through a port(s). This includes any firms or industries that receive materials through a port or ship products through a port. If a firm runs its own terminal and employment of the terminal can be separated from total firm employment, the terminal activity can be considered part of the port industry rather than the port dependent industry. PORT INDUSTRY - The group of firms or activities that are directly needed for the movement of waterborne cargo. It includes firms falling into the following Standard Industrial Classification (SIC) categories: SIC 44, Water Transportation (Canada SIC 504 and 505); SIC 42, Motor Freight Transportation and Warehousing (Canada SIC 507 and 527); SIC 47, Transportation Services (Canada SIC 504 and 505); SIC 6512, Operators of Nonresidential Buildings (i.e., lessors of piers, docks, and associated buildings and facilities) . The 1972 Standard Industrial Classification Manual, published by The Executive Office of the President, Office of Management and Budget, can be used to obtain more detail about these categories. (Note: The Canadian SIC is similar to the U.S. SIC, but the numbering is not consistent. See Standard Industrial Classification Manual , Revised 1970 from Statistics Canada to obtain details.) PORT REGION - The geographical area to which port cargoes are sent and from which they are received; the service area(s) of the ports. This area varies for different commodities — e.g., container cargo may be collected from or shipped over much greater distances than some bulk cargoes. For the purpose of this "kit", the port region is defined as a collection of counties (for U.S. ports) and census metropolitan areas (for Canadian ports) . This definition facilitates the compilation of employment and income data for the port region. PRIMARY IMPACT - See DIRECT IMPACT RIMS - Regional Industrial Multiplier System - It is a technique for producing regional input/output tables from the national input/output table. The U.S. Department of Commerce, Bureau of Economic Analysis provides input/output sales multipliers for individual county economies or regional economies using RIMS. SECONDARY IMPACT - The sum of indirect impacts and induced impacts. SIC CODE - Standard Industrial Classification code; a classification system published by the Executive Office of the President, Office of Management and Budget, that relates specific economic activities to generalized industrial categories. The Canada S.I.C. is published by Statistics Canada. TAX IMPACT - The economic impact of a project or economic relationship measured in terms of the specific tax revenues generated. The tax impact can be measured in terms of local tax revenues, state tax revenues, or federal tax revenues. 117 APPENDIX CASE STUDIES A. BACKGROUND The purpose of this kit was the development of an economic assess- ment methodology which could be used by ports with limited staff and resources. In preparing the kit, an attempt was made to anticipate special circumstances or problems which might arise and suggest solutions or alternative approaches. However, no document of this type can deal with the full range of issues that individual port districts may en- counter in conducting economic assessments. Four case studies were undertaken so that the procedures in this kit could be illustrated in the context of specific port impact studies. As noted, it is likely that an individual port will have unique conditions or problems not treated in the text of this kit. Therefore, this summary of the case studies focuses on the difficulties encountered in each case and the ways in which they were handled. In some instances modifications were made to specific steps in the kit to clarify instructions and avoid problems experienced by the test ports. Other problems were resolved by case- specific decisions which may provide some guidance for reasonable judg- ments by other users. Overall, we believe the case studies have been very important because they identified areas of the methodology which needed additional explanation or clarification. 1. Selection of the Test Ports The goal of this kit was to provide a methodology which could be used by U.S. and Canadian ports of different sizes with different cargo mixes. Therefore, the test ports were selected to represent a cross- section of potential users. The three areas and four ports tested were: • Grays Harbor, Washington — The Port of Grays Harbor is the smallest of the four test ports. It is pri- marily a center for the export of logs, lumber, and pulp. It can be considered representative of many Northwest ports which service the area's forest products industry. • San Francisco-Oakland, California — The San Francisco Bay Area contains two major ports, both of which handle a wide variety of commodities. The ports serve a metropolitan area of several million inhabitants and a highly developed economy. The Port of Oakland currently accounts for about 70% of the tonnage shipped through the two ports due to its function as a containerized cargo "load center." These test ports were selected to illustrate the analysis of a region with more than one port. 118 • Vancouver, British Columbia — The Port of Vancouver is the largest port in British Columbia and, in terms of tonnage, the largest port on the West Coast of North America. The major commodities shipped through Vancouver are coal, grains, oil and other bulk items. The Port of Vancouver was selected to illustrate the application of the methodology to Canadian ports. 2. Role of Port Staff As discussed in Chapter 1, the development of reliable survey data is the single most important element of the economic impact study. More- over, conducting a successful survey is typically the most difficult ele- ment of the assessment methodology. Therefore, the survey work for the case studies was conducted by the staff of the test ports in accordance with the procedures being developed for the kit. This served both to test the prescribed methodology and to identify difficulties which might arise in different situations. Secondary impact calculations were per- formed by Arthur D. Little, Inc., because of both the limited time avail- able for this work and the in-house availability of most of the required secondary data. B. GENERAL PROBLEMS AND SOLUTIONS 1. Survey The test ports' difficulties in conducting the survey fell into two categories — determining which firms should be included in the survey and obtaining an adequate number of responses to the questionnaire. a. Firms Included in the Survey Each port surveyed some firms which could not be included in the "port industry." To some extent this is a result of the lack of a com- monly accepted precise definition of the industry. For the purposes of this economic assessment, the port industry has been defined as the group of firms or activities that are directly needed for the movement of water borne commerce. The firms which produce or consume the commodities shipped through the port make up the "port-dependent industry." A port also does business with and may even lease property to numerous other firms which fit neither of these definitions. The survey work done by the test ports suggests that the majority of port industry activities fall into two major industrial classifications: • Water Transportation. This group includes vessel operators, tug and towing services, stevedores and other cargo handling, marine terminal operation, pilots, etc. (U.S. SIC 44, Canadian SIC 504 and 505) 119 • Transportation Services. This group includes freight forwarders and firms involved in the arrangement of transportation, such as custom house brokers, shipping agents, and export/import firms. (U.S. SIC 47, Canadian SIC 504 and 505) The other SICs involved in the transportation of waterborne commerce include 40 and 42 — railroad transportation and motor freight trans- portation and warehousing (Canadian SIC 506 and 507) . Additional firms typically included in the port industry are in SIC 373, Ship and Boat Building and Repairing (Canadian SIC 327), and SIC 559, which includes retailers of marine supplies (ship chandlers) . Firms identified and surveyed by the test ports but not part of the port industry include manufacturing, financial, and service activities. The forest products firms in Grays Harbor County can be primarily classi- fied as part of the port-dependent industry. San Francisco Bay region manufacturers of paint and valves used in shipbuilding do not belong to either the port industry or the port-dependent industry. These firms are not related to the area's ports except insofar as some of their economic activity will be represented in the indirect multiplier of the ship- building firms. Similarly, the effects of firms which supply business and financial services to port industry clients — insurance brokers, banks, attorneys, etc. — will be reflected in the port industry multi- plier and should thus not be considered direct activity. A more rigorous definition of the port industry (i.e., by SIC code) will help to direct the survey to appropriate firms. Some of the res- pondents to the test ports' questionnaires appeared to misinterpret the question "describe your waterborne commerce-related operations." One firm described its activities as exporting machinery but closer examina- tion of its responses to other questions indicated that the firm was actually engaged in manufacturing machinery some of which was exported. Therefore, the survey form should ask for the respondent's SIC code, if known, in addition to a description of the firm's activities. This will permit allocation of the respondents to the proper categories and serve as an additional check on the groupings used in the master list. b. Response Rate The response rates for the test ports' surveys (percent of total questionnaires returned) averaged about 20%. However, the quality of the responses was far from uniform. Some firms returned survey forms with some or even all of the requested data missing. The response rate was also significantly higher or lower than the overall average for some industries. For example, all but one of the firms in one industry might answer the port's questionnaire, but there might be no responses from firms in another industry. All of the test ports had a significant number of firms that were unwilling to disclose any financial data (sales, payrolls, etc.). 120 The most cost-efficient and effective way to increase the response rate appears to be telephone interviewing. Depending on the needs of the port conducting the study, a short format telephone interview can be used to collect essential data — employment, revenues (sales) , and payroll. This information is the minimum required to calculate direct and secondary economic impacts. A possible option for ports that meet resistance from firms unwilling to disclose information is to use a "multiple choice" response format. Based on some preliminary knowledge about the firms that are potential respondents, ranges for key data items can be developed — e.g., employment: 0-9, 10-19, 20-39, 40-59, etc. Firms which will not provide specific data may be willing to indicate the appropriate range. The midpoint of the selected range can then be considered the response for purposes of coding and tabulating. 2. Determination of Direct Impacts The calculation of direct impacts in the case studies essentially followed the procedures outlined in element D of the methodology, but the quantity and quality of the survey data required some modification of the suggested steps. Step D:5 recommends partitioning the surveyed firms by employment size classf ication for purposes of tabulating responses. However, the number of responses received by the test ports was too small for such a tabulation to be useful. Case study survey data were thus tabulated by industry, but all employment size classifications were grouped together. Some of the test ports received no responses from firms in a particular industry. If this information is omitted for a segment of the port industry, both direct and secondary impacts will be understated. Therefore, sales and payroll were estimated for these industries when only employment data were available. First, payroll per employee for the appropriate year was derived from the sources cited in Step F:6 and used to calculate total payroll. Then, the payroll/sales ratios cited in Step F:4 were used to determine total sales. While this is admittedly somewhat circular and less accurate than a survey-based approach, it is a means to complete the economic assessment when other methods are not available. The tabulation of direct sales (revenue) information was adjusted to prevent double counting by omitting the reported sales of freight forwarders, transportation brokers, and related businesses. These firms typically report sales figures which include the cost of transportation they have arranged; however, the shipping firms, stevedores, etc. report the same sales. Therefore, the true sales figures for transportation service firms should be equivalent to the value added (payrolls and profits) by these businesses and exclude amounts paid to other port industry firms. Because there is no available information from which to determine the value added portion of transportation service revenues, and because this number is small relative to total revenues, a decision was made to eliminate these firms from the calculation of sales impacts. However, payroll and employment for these firms were considered since these impacts are not duplicated in other responses. 121 3. Calculation of Secondary Impacts Secondary impacts for the test ports were calculated in accordance with the steps outlined in the kit. There were no significant problems encountered which had not been anticipated in the development of the methodology, but time constraints limited the analysis to estimation of sales, income, and employment impacts. C. SPECIFIC CASE STUDIES 1. Grays Harbor, Washington The Port of Grays Harbor identified some 72 port industry and port- dependent firms to be included in the survey. Responses were received from 19 firms, including: • One vessel repair firm (SIC 37) • Two trucking firms (SIC 42) • Four water transportation firms, including the port itself (SIC 44) • Seven transportation services firms (SIC 47) • Five forest products firms (port-dependent) Five of the transportation services firms were Seattle-based shipping/ trading companies which reported no employment and payroll in Grays Harbor. Therefore, 14 of the responses received were usable, for an overall response rate of about 20%. The Port of Grays Harbor provided additional information on direct employment to supplement the survey data. Specifically, the port esti- mated the number of employees of a major forest products firm directly engaged in waterborne commerce. While these firms would be considered port-dependent, it is reasonable to separate out those employees engaged in waterborne commerce and their prorated shares of sales and payrolls and assign them to the port industry. Table 1 shows the estimates of direct port and port-dependent acti- vity. SIC 44 includes the forest product firm activity described above and the operation of the port itself. Data for firms which did not res- pond were estimated in accordance with the procedures outlined in element D of the methodology. Almost all of the firms identified in SIC 47, transportation services, are located in Seattle. Because only two of the seven firms in this group which responded to the survey reported employ- ment and payroll in Grays Harbor, no activity was assigned to the firms which did not respond. This may undercount impacts in this industry to some extent, but when available data are insufficient to make reasonable estimates, a conservative approach is preferred. Employment and payroll estimates thus represent only the amounts reported by two survey respon- dents. Sales were not estimated because of the decision to omit this figure to avoid double counting. 122 Direct port-dependent activity represents the responses of five forest products firms (less activity allocated to the port industry) and estimates for two additional firms identified by the port. Because the Grays Harbor survey instrument did not include questions about alternate ports or the amount of the firm's business that would be lost without the port, it was not possible to ascertain the extent to which these firms are truly port-dependent. Table 2 shows the calculation of total economic activity associated with the Port of Grays Harbor. Using the economic base income multi- plier, whose development is illustrated in Steps E:3 through E:7, total payroll supported by the port is estimated to be about $20 million. Port-dependent activities contribute almost $75 million. The RIMS approach yields a total impact estimate of almost $85 mil- lion in sales. Using a weighted payroll/sales ratio developed from Grays Harbor County employment data and the industry-specific ratios in Step F:4, the sales equivalent of the income impact calculated using the eco- nomic base multiplier is $78 million, only 7% less than the RIMS-derived value. Employment impacts were estimated from the lower value (the economic base technique) . Total wages are divided by the 1978 average wage per employee for Grays Harbor County to yield a total employment impact of 1280. Since direct employment for the port industry is estimated at about 685, the implied employment multiplier is 1.87, which is within the range one would expect for an economy the size of Grays Harbor County. The 1300 port industry direct and secondary employees account for about 6% of Grays Harbor County's total employment of 22,000. When the 3200 port-dependent workers are considered as well, waterborne commerce- related activities account for some 20% of the county's total employment. 2. San Francisco and Oakland, California The Port of San Francisco and the Port of Oakland jointly conducted a survey to determine their impact on the San Francisco Bay region. Questionnaires were sent to port industry firms and to firms that use the ports in the movement of their goods or materials. We have calculated only the port industry impact due to difficulties in the determination of the port-dependent industry universe. Forty-one usable responses were received to the 140 questionnaires sent to firms defined as being in the port industry. This corresponds to a 29% response rate. The number of responses in specific industrial categories, along with tabulations of employment, payroll, and sales associated with each category, is provided in Table 3. Responses were received from: 123 TABLE 1 PORT OF GRAYS HARBOR DIRECT ACTIVITY Respondents Universe Employment Payroll Sales ($000s) ($000s) Port Industry SIC 37 1 1 275 $ 6,100 $ 15,700 SIC 42 2 2 27 546 1,216 SIC 44 4 11 372 6,760 26,996 SIC 47 2 N.A. 10 212 N.A. Total 9 — 684 $13,618 $ 43,912 Por t-dependent Industry 5 7 3,195 $73,775 $334,640 124 TABLE 2 PORT OF GRAYS HARBOR TOTAL ECONOMIC ACTIVITY (dollars in thousands) 1. Economic Base Multiplier Direct Income X Income Multiplier = Total Income $13,618 1.45 $19 f 746 2. RIMS Multiplier Direct Sales by Industry X RIMS Multipliers SIC 37 SIC 42 SIC 44 Total Sales $15,700 X 2.021 1,216 X 2.228 26,996 X 1.831 $31,730 2,709 49,430 $83,869 3. Comparison Total Income X 0.8 (wage percent of payroll and benefits) -f- Payroll/Sales Ratio for County = Total Sales $19,746 $15,797 0.203 $77,818 Total Employment Total Wages -7- Average Wage per Employee = Total Employment $15,797 $12,339 1,280 125 • Five boatbuilding (SIC 3731) and shipbuilding (SIC 3732) firms • Eight trucking firms (SIC 42) • Sixteen water transportation firms (SIC 44) , inclu- ding the ports themselves • Seven firms providing transportation services (SIC 47) • Five chandlers (SIC 55) A number of difficulties were encountered during the survey stage of the analysis for these ports. The primary problem was the low response rate. Call-back reminders did little to improve the response rate, but it was found that telephone requests for a limited number of question responses increased the information available for analysis dramatically and at a relatively small cost. The Port of Oakland was able to assign a summer university intern to the task of following up after non- respondents. He was instructed to contact a member (or members) of non- responding firms that could provide answers to the following five questions: 1. What are the firm's waterborne commerce related activities? 2. What are last year's total sales or revenues related waterborne commerce? 3. What are the total number of full-time equivalent employees engaged in waterborne commerce related activities? 4. What is the total payroll of these employees? 5. What percentage of your business is related to the Port of Oakland and to the Port of San Francisco? These abbreviated telephone interviews were very successful. Their only drawback is the limited quantity of information obtained which in turn limits the variety of impact measures that can be calculated as well as the "checks" that can be performed on the survey data. A second difficulty encountered during the survey stage was the Port of Oakland's reluctance to have the completed surveys mailed to the Port, since doing so would have placed the surveys in the public domain. This problem was handled by having the surveys sent to Arthur D. Little, Inc. An alternate approach would be to have them returned to a private firm associated with the Port — e.g., its counsel. It is not known whether having the completed questionnaires sent to a private firm affected the response rate. 126 The estimates of total direct sales, employment, and payroll presented in Table 3 were derived with the procedures outlined in element D of the methodology. The totals are for the Port of San Francisco and the Port of Oakland combined. When the San Francisco-Oakland survey work was begun, the intention was to partition the direct activity between the two ports. However, the data provided in the responses proved inadequate to make this allocation. There was significant variation in the allocation between ports reported by survey respondents (e.g., some .firms reported 100% of their business in one port, some reported 100% split between the two, and some reported only 10-20% for both) . Because responses were received from only about one- third of the port industry firms (and much fewer in some SICs) , it was decided that any attempt to allocate activity between the ports would be highly suspect. However, a direct impact split could be estimated by assuming it was equivalent to the cargo split. Table 4 provides a summary of the secondary impact calculations per- formed for the Port of San Francisco and the Port of Oakland. First total income is determined using the economic base approach described in element E of the methodology. In part 3 of Table 4 the $458 million in total income is converted to an estimate of the total sales associated with it. The $1,580 million sales estimate allows a direct comparison with the RIMS derived total sales impact of $1,250 million. The dis- crepancy between the figures is due primarily to the use of a regional average payroll/sales ratio. The large retail and FIRE (finance, insur- ance, and real estate) sectors in the region bias the ratio downward from what it should be in this application. The use of a payroll/sales ratio that is a weighted average of a waterborne commerce related payroll/sales ratio and an average regional payroll/sales ratio would be preferred. In keeping with our conservative bias total employment was cal- culated by applying the payroll/sales ratio to the RIMS derived total sales, rather than the economic base derived total sales. Using the economic base derived total sales yields an estimated total employment impact of 25,443, some 26% higher than the RIMS derived total of 20,133. These employment impacts represent 1.7% and 1.3% of total regional employment. The implied employment multipliers are 2.32 (RIMS sales impact) and 2.93 (economic base sale impact). No attempt has been made to distribute these impacts geographically within the Bay Area. The direct impact "split" between the Port of Oakland and the Port of San Francisco could be used to crudely approximate the impact of each port on the San Francisco Bay region, defined for the purpose of this study to consist of Alameda, Contra Costa, Marin, San Francisco, San Mateo, and Santa Clara counties. Generally it is not recommended that aggregate cargo tonnage or value be used to define the "split" since impacts vary by cargo type and destination; these factors differe greatly among ports. 3. Vancouver, British Columbia The Port of Vancouver, British Columbia, was included in the study in an attempt to broaden the kit's use to Canadian ports. This entailed only minor modifications of the direct impact determination portions of the methodology (sections B, C, and D) , but substantially altered the applicability of the secondary impact determination sections. Statis- tical data are collected differently in Canada than in the United States. The following discrepancies are the most important: 127 TABLE 3 PORTS OF SAN FRANCISCO AND OAKLAND DIRECT ACTIVITY Respondents Universe Employment Payroll Sales ($000s) ($000s) Port Industry SIC 3731 SIC 3732 SIC 42 SIC 44 SIC 47 SIC 55 4 1 8 16 7 5 7 1 21 37 61 13 2,094 $ 53,115 $ 99,145 90 2,529 6,400 1,291 27,515 55,640 3,588 79,880 206,151 1,394 27,482 — 238 4,284 30,256 Total 41 140 8,695 $194,805 $397,592 128 TABLE 4 PORTS OF SAN FRANCISCO AND OAKLAND TOTAL ECONOMIC ACTIVITY (dollars in thousands) 1. Economic Base Multiplier Direct Income $194,805 X Income Multiplier 2.35 = Total Income $457,792 2. RIMS Multiplier Direct Sales by Industry X RIMS Multipliers SIC 3731 $ 99,145 X 3.615 = $ 358,409 SIC 3732 6,400 X 3.204 = 20,506 SIC 42 55,640 X 3.504 = 194,963 SIC 44 206,151 X 2.736 = 574,337 SIC 55 30,256 X 3.352 = 101,418 Total Sales $1 ,249,633 3. Comparison Total Income $ 457,792 X 0.8 (wage percent of payroll and benefits) $ 366,234 -4r Payroll/Sales Ratio for Region 0.232 = Total Sales $1,578,593 4. Total Employment Total Wages* $ 289,915 -r- Average Wage per Employee $ 14.400 = Total Employment 20,133 ♦Determined by applying payroll/sales ratio to RIMS derived total sales. 129 • There is no service that provides RIMS type multi- pliers for small areas. However, it should be noted that Canada does have more current national input/ output studies than does the United States, and that it has a sophisticated model that regionalizes (pro- vincializes) its national I/O tables. • Annual Canadian employment and payroll data are pre- sented in much more aggregated form for small areas than in the United States. This increases the con- fidence the analyst has in the accuracy of the data, but does limit the use of the economic base metho- dology to census years. We have constructed an economic base multiplier for Vancouver using 1971 Canadian Census data and have used a 1972 regional input/output study of the region to estimate secondary impacts of waterborne commerce related activity in the Vancouver area. For a more current analysis, the only alternatives for Canadian ports would be to conduct a regional input/ output study or to regionalize an input/output study for a larger area of which they are a part — i.e., the national I/O tables or provincial I/O tables. It is believed that most ports would not want to commit the resources necessary for such endeavors, especially given the relatively limited changes that would occur in the impacts determined by the proce- dures we have outlined. The survey of port industry firms was very successful except for a few industry categories for which there were no responses. These included ships chandlers, trucking firms, marine pilots, and ship brokers. The overall response rate of 25% was especially good con- sidering that a local trade publication printed a notice that suggested firms not respond to the questionnaire sent by the Port since the data they would provide were sensitive. Based on this experience it is sug- gested that port personnel contact local trade journals, associations, and newspapers to solicit the assistance, or at least the blessing, of this influential group. The Port of Vancouver received questionnaires from 53 port industry firms and 16 port-dependent industry firms. Responses were received from the following types of firms: Customs brokers, 3 responses Shipping agents, 8 responses Freight forwarders, 4 responses Stevedoring companies, 4 responses Marine terminal operators, 5 responses Railroads, 3 responses Marine towing and tug lines, 6 responses Boat builders, 5 responses 130 • Grain elevator companies, 3 responses • Marinas, 6 responses • Marine service firms, 3 responses Table 5 shows the aggregated employment, payroll, and sales associated with the port industry firms. The survey results were expanded for the more detailed groups listed above; then the two aggregations, "Trade and Transportation" and "Shipbuilding", were formed so as to match the input/output categories in An Interindustry Study of the Metropolitan Vancouver Economy . Port-dependent industry impacts were not calculated because of time and budget constraints. For accuracy in the determination of port- dependent industry impacts it is very important that employment, sales, and payroll shares attributable to the existence of the port be iden- tified. This demands careful checking by the analyst and recontacts to fill in missing responses. Because we could not do this, no port- dependent industry impacts were estimated. Table 6 shows the calculation of total impacts from direct activity using an economic base employment multiplier and industry specific input/output multipliers from the aforementioned regional input/output study done for Vancouver. A recalculation of 1978 employment shares for the major industrial groupings in Vancouver showed remarkable consistency with the 1971 shares; therefore, the interindustry relationships have remained relatively constant during the intervening seven years. For this reason, and because the intervening time period is so brief, it is believed that the 1972 input/output study is still very useful. It is necessary to adjust some of the multipliers to account for inflation, however . The economic base multiplier is based on 1971 Census data, since more recent data did not provide the necessary detail. The census survey is much more comprehensive than the monthly survey designed to provide the base information for Employment, Earnings, and Hours ; hence, less aggregation of results is required. Total employment is estimated to be approximately 38,000 with this technique. The economic base derived employment estimate should be compared with total employment estimated with the I/O multipliers. Employment was estimated by first calculating indirect wages from indirect sales. Then employment was estimated through division of wages by average wage. The resulting indirect employment was added to direct employment derived from the survey to obtain total employment. Total employment was estimated at approximately 30,000 using this input/output approach. Input/output derived employment is 79% of economic base employment. Total wages were estimated using the economic base employment estimates and input/output derived sales estimates. The July 1978 average wage was multiplied by indirect employment (calculated with economic base multiplier) and added to the direct wage (compiled from the survey) to determine total wages of $710,404,000. Input/output derived indirect sales of $710,404,000 estimates were converted to wages through multiplication by a payroll/sales ratio, yielding $589,637,000. 131 TABLE 5 PORT OF VANCOUVER, BRITISH COLUMBIA DIRECT ACTIVITY Respondents Universe Employment Payroll Sales ($000s) ($000s) Port Industry Trade and Transportation 49 205 15 , 451 $354,205 $747,721 Shipbuilding _5 8 1,251 31,773 74,988 Total 53 212 16,702 $385,978 $822,709 132 TABLE 6 PORT OF VANCOUVER, BRITISH COLUMBIA TOTAL ECONOMIC ACTIVITY (dollars in thousands) 1. Economic Base Multiplier Employment Direct Employment 16,702 X Employment Multiplier 2.28 = Total Employment 38,081 Wages Indirect Employment 21,379 X Average Wage 15.175 Indirect Wages 324,426 + Direct Wages 385,978 = Total Wages 710,404 2. I/O Multiplier Sales Direct Sales by Industry X Sales Multipliers Trade and Transportation Shipbuilding Total Sales Wages Indirect Sales x Wage/Sales Ratio = Indirect Wages + Direct Wages = Total Wages Employment Indirect Wages Average Wage = Indirect Employment + Direct Employment = Total Employment 30,035 $747,721 74,988 X X 1.69 = 1.50 = $1 $1 $ $ $ $ ,263,648 112,482 ,376,130 553,421 .368 203,659 385,978 589,637 203,659 15.275 13,333 16,702 133