AR AEALYTICAL TECHNEQUE FOR "FREE OFTEWZATION 0F PRECEI SENSE MK 0F AMEMOAN BROKERAGE FIRMS ' r - Bissertation for me fieg’ree of Ph. D. - MECHEGAN STATE UREVERSETY , , BREW jGEORGE mm ‘ 1975 Erh-“WM- r4 v I W er‘h'fihuk .,.t . . I' 7 ' hint. £55331 33:11”, University ENE: ‘- This is to certify that the thesis entitled A Analytical Technique For The Optimization of Price/Service Mix of American Brokerage Firms presented by BR I AN GEORGE LONG has been accepted towards fulfillment of the requirements for Ph . D . Marketing (is-gee 111 //§7 Major professor 0 Date 2/20 75'- ‘ ‘ IINDING ”S = HIM ~' BLINK umnfutawg. "mam“. Ilfllil' a 0 £\:17 ABSTRACT \ Q AN ANALYTICAL TECHNIQUE FOR THE OPTIMIZATION OF PRICE/SERVICE MIX IN AMERICAN BROKERAGE FIRMS BY Brian George Long Purpose There are two purposes for this study. First, an analysis is made of the current service offerings of American brokerage firms in order to determine the value that brokerage customers attach to these services. Second, the dissertation deve10ps an analytical model for the determination of an optimum price/service mix. This technique may be applied to a variety of price/service mix determination problems in other service industries. Frame of Reference The study begins by describing the profitability problems of the American brokerage industry. To add to those problems, a recent action by the Securities and Exchange Commission has dictated that brokerage fees, which are currently fixed in most categories on an industry-wide uniform scale, must now become competitive. Brokerage firms must soon begin making separate charges for special services is , Brian George Long beyond order execution--services which were previously available at no extra charge. Through the use of the "four P's" managerial mar- keting model, an analysis is made of the current marketing strategy of the industry. Factors discussed include the industry's use of advertising, the functions of the broker as a salesman, the location of brokerage offices, the dilemma of pricing services in light of the SEC action, and the nature of services provided by most brokerage firms. The elements of price and service are selected for further analysis. Model Construction The dissertation develops an analytical model for the Optimization of the price/service mix. The model is designed to consider the costs, revenues, and profits associated with all possible combinations of services, and to consider the interactive effects of various service offerings. Simple accounting procedures may fail to take into account the customer's perception of the entire service package. It may be necessary to include seemingly unprofit— able services in certain service offerings because of the customer who will not purchase the service package at all unless a particular set of services is included. Therefore the total revenue generated by offering some unprofitable services may far exceed their cost. Brian George Long The output yields the most profitable configuration of services for each of the following three sets of assump- tions: (1) that all services may be purchased independently, (2) that all services must be purchased together in a pack- age, and (3) that all services not offered in the package may be added on. A fourth output yields the maximum poten— tial revenue obtainable if all services were offered to all customers. Collection of Data For the purpose of developing a data base to examine customer attitudes toward brokerage services and to provide an input for the model, a mail survey was conducted on six of the most significant (and costly) services which are currently being offered by most brokerage firms. The services studied include: (1) security custody, (2) office contact, (3) specialty orders, (4) telephone contact, (5) odd lots, and (6) research information. A seventh service, portfolio management, was added in order to assess its potential. The survey was mailed on November 1, 1974 to a random sample consisting of 500 subscribers to Barron's. Respondents were asked to indicate which of the seven ser- vices they use, which services they consider to be absolute "musts," and how much they would be willing to pay for each service. A total of 195 usable responses was obtained. Brian George Long Results Based on the results of the survey, a serious question must be raised as to whether it is profitable for all firms in the brokerage industry to operate local branch offices, or to make their brokers available for telephone contact. It is therefore recommended that those firms interested in improving their efficiency and effectiveness should survey the attitudes of their customers, either through the use of the model presented in this dissertation or by some other means. The survey is necessary to deter- mine if certain branch offices should be closed, and to ascertain the degree of telephone contact necessary. Limitation This dissertation is a preliminary rather than a conclusive study. A single attempt at the construction of a model seldom results in a perfect replication of the real world, although it forms a basis for the development of a new, more refined model. No study as modest in scope as this one can hope to fully reshape the brokerage industry, the theory of cost/revenue/profitability analysis, or the theory of price/service mix determination. AN ANALYTICAL TECHNIQUE FOR THE OPTIMIZATION OF PRICE/SERVICE MIX OF AMERICAN BROKERAGE FIRMS BY Brian George Long A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Marketing and Transportation Administration 1975 ii C) Copyright by BRIAN GEORGE LONG 1975 To my “the/L, the. [axe 06cm 0. Long iii ACKNOWLEDGMENTS Although it will never be possible to thank all of the people who aided in the successful completion of this research work, I would like to mention several in particular. First, I wish to thank my dissertation chairman, Dr. Leo G. Erickson, Professor of Marketing, whose research experience and broad understanding of many elements of this dissertation are reflected throughout. His many positive suggestions and constant notes of encouragement were inspirational. Second, I acknowledge with gratitude the assistance of Dr. Alden C. Olson, Professor of Finance, who served as a member of the dissertation committee. Dr. Olson's many helpful suggestions greatly improved the quality of this work. His extensive knowledge of the brokerage industry was called upon to sharpen many points. Third, I express my sincere thanks to Dr. Frank R. Bacon, Professor of Marketing, who served as a member of the committee and was helpful in developing the research design and statistical phases of this dissertation. Fourth, I am very grateful to James C. Lockwood, Director of Consulting Services for E. F. Hutton and Company, iv for the initial idea for this dissertation. The impact of Mr. Lockwood on the author extends beyond the chapters which follow. For computer programming aid, I am thankful to Mr. Curtis L. Hoitash, and for the typing, to Mrs. Grace S. Rutherford, whose superior professional ability speaks for itself in the pages which follow. A special thanks goes to Mrs. Betty Peacock who proofread the manuscript. I would also like to sincerely thank the people who offered me both encouragement and employment during my doctoral program, especially Dr. Dalton E. McFarland, Associate Dean James F. Rainey, and Dr. Donald A. Taylor. Finally, I would like to thank my fellow doctoral candidates and many other friends who offered encouragement and companionship during my years at Michigan State. TABLE OF CONTENTS LIST OF TABLES O O O O O I O O O O O C O O O 0 LIST OF FIGURES O O O O O O O O O O O O O O O 0 CHAPTER I. II. III. INTRODUCTION . . . . . . . . . . . . . Purpose . . . . . . . . Problem Background . . . Hypotheses . . . . . . . General Framework . . . . Scope . . . . . . . . . . . . . . . . Methodology . . . . . . . . Alternative Research Design Limitations . . . . . . . . Terms and Definitions . . . Uses of the Study . . . . . Presentation Overview . . . MARKETING IN THE BROKERAGE INDUSTRY . . The Operating Environment . . . . . . The Securities and Exchange Commission . . . . . . . . . . Competition . . . . . . . . . . . The Location of Brokerage Facilities Promotion . . . . . . . . . . . . . . Advertising . . . . . . . . . . . The Sales Function . . . . . . Brokerage Fee Determination . . . . Brokerage Service . . . . . . . . Services Offered by Most Firms Unique Services . . . . . . . . Summary . . . . . . . . . . . . . . THEORY OF PRICE/SERVICE MIX DETERMINATION Classical Theory of Pricing . . . . . Service Marketing . . . . . . . . . . vi Page ix xi 24 26 29 32 37 38 44 49 50 52 54 56 56 62 CHAPTER Page Services and the Wheel of Retailing . . . 64 Unique Aspects of Service Marketing . . . 67 Franchising and Company Ownership . . . . 69 Level of Service . . . . . . . . . . . . . . 71 Cost/Revenue Analysis . . . . . . . . . . . . 75 Summary . . . . . . . . . . . . . . . . . . . 80 IV. RESEARCH DESIGN . . . . . . . . . . . . . . . . 82 The Concept of Primary and Secondary Service . . . . . . . . . . . . . . . . . 82 The Service Cost/Revenue Analysis Model . . . 85 Outputs . . . . . . . . . . . . . . . . . 86 Inputs . . . . . . . . . . . . . . . . . 87 Processor . . . . . . . . . . . 89 Description of Test Questionnaire . . . . . . 95 Sample Selection . . . . . . . . . . . . . . 97 Survey Procedure . . . . . . . . . . . . . . 99 Survey of Non-Respondents . . . . . . . . . . 101 Tabulation . . . . . . . . . . . . . . . . . 104 Computer Programming . . . . . . . . . . . . 104 V. FINDINGS O O O O O O O O O O O O O O O O O O O 106 Survey Results . . . . . . . . . . . . . . . 106 Suggestions for Improved Broker Performance . 116 Cross Company Comparisons . . . . . . . . . . 119 Consideration of Hypothesis 1 . . . . . . . . 125 Arguments Favoring Rejection of Hypothesis 1 . . . . . . . . . . . . . 128 Arguments Favoring Acceptance of Hypothesis 1 . . . . . . . . . . . . . 129 Conclusion . . . . . . . . . . . . . . . 129 Consideration of Hypothesis 2 . . . . . . . . 130 Cost Assumptions . . . . . . 130 Examples of SCRAM Model Applications . . 134 Conclusions . . . . . . . . . . . . . . . 140 VI. SUMMARY, CONCLUSIONS, AND RECOMMENDATIONS . . . 141 smary O O O O O O O I O O O O O I C I O O O 141 Recommendations for the Brokerage Industry . 143 Other Possible Model Tests . . . . . . . . . 146 Possible Extensions of SCRAM Model . . . . . 149 Limitations 0 O C O O C C C O O C O O O O O O 151 Final Conclusions . . . . . . . . . . . . . . 152 vii APPENDIX A. SURVEY AND COVER LETTER . B. SUPPLEMENTAL SURVEY RESULTS C. SCRAM OUTPUT D. COMPUTER PROGRAMS E. NEW DATA BIBLIOGRAPHY viii Page 154 160 167 176 181 188 LIST OF TABLES TABLE Page 2.1 Minimum Commission Rates Effective November 19, 1974 . . . . . . . . . . . . . . 44 4.1 Survey Response . . . . . . . . . . . . . . . 102 4.2 Survey of Non-Respondents . . . . . . . . . . 103 5.1 Services Used . . . . . . . . . . . . . . . . 107 5.2 Insisted Services . . . . . . . . . . . . . . 109 5.3 Average Preference Ranking . . . . . . . . . 109 5.4 Ranking Comparison of Preferences and Usage . 110 5.5 Dollar Value Response . . . . . . . . . . . . 112 5.6 Transaction Response . . . . . . . . . . . . 112 5.7 Merrill Lynch, Pierce, Fenner, & Smith . . . 120 5.8 Dean Witter and Co. . . . . . . . . . . . . . 121 5.9 Hornblower & Weeks-Hemphill Noyes . . . . . . 122 5.10 E. F. Hutton & Co. . . . . . . . . . . . . . 123 5.11 Cost Assumptions . . . . . . . . . . . . . . 131 5.12 Second Cost Assumptions . . . . . . . . . . . 137 5.13 Accounting Analysis . . . . . . . . . . . . . 138 3.1 Preference Ranking Response . . . . . . . . . 161 8.2 Dollar Response--Security Custody . . . . . . 162 B.3 Dollar Response--Specialty Orders . . . . . . 162 ix TABLE B.4 8.5 8.6 3.7 B.8 B.9 8.10 Dollar Response--Odd Lots . . . . . . Dollar Response-~Portfolio Management Dollar Response--Research . . . . . . Dollar Response--Office Contact . . . Dollar Response--Market Contact . . . Transaction Response . . . . . . . . Transaction Dollar Value Response . . Page 163 163 164 164 165 165 166 FIGURE 1.1 3.1 3.2 3.3 3.4 4.1 4.2 C.l C.2 LIST OF FIGURES The McCarthy 4 P's Model . . . . . . . . . The Marshallian Supply-Demand Cross . . . . Relationship Between Prices of Complements Summation of Demand Curve . . . . . . . . . Logistic System Costs . . . . . . . . . . . Standard Model . . . . . . . . . . . . . . The SCRAM Model . . . . . . . . . . . . . . Potential Revenue--First Cost Assumption . Independent Service Output--First Cost Assmnption O O O O I I O O I O O C O O O 0 Package Output--First Cost Assumption . . . Package Plus Output--First Cost Assumption Potential Revenue-~Second Cost Assumption . Independent Service Output-—Second Cost Assmnption O O O O O C O O O O O O O O O 0 Package Output--Second Cost Assumption . . Package Plus Output--Second Cost Assumption xi Page 58 6O 61 74 86 90 168 169 170 171 172 173 174 175 u CHAPTER I INTRODUCTION Purpose The dynamic nature of business results in an almost perpetual creation of new problems requiring solutions. Because of a series of changes, the past year has witnessed what might appear to be the beginning of the end of the brokerage industry. Despite the fact that volume on the New York Stock Exchange has quadrupled over the past ten years, brokerage firms, caught by a series of changes, have been among the noteworthy money losers of an other- wise profitable economy. Many knowledgeable people have suggested that the traditional firm will soon cease to exist.1 Although the demise of this industry is not centered in any one operational segment, a portion of the problem can be attributed to the marketing function of the firms. One of the areas of recent change within the marketing structure relates to several decisions by the Securities 1John F. Lyons, "Wall Street Discovers Competition," Financial World, 3 April 1974, pp. 28-29. and Exchange Commission calling for the discontinuation of fixed brokerage fees, which will in turn require that firms begin making individual charges for certain services. This dissertation will therefore center upon the problem of the determination of a new and more profitable price/service mix. To aid in the solution of the foregoing problem, this thesis will utilize an analytical model. Models are constructed to reduce an abstract problem to a scale small enough to be manipulated for prediction purposes. "Indeed, they are vital to an intellectual attack on any problem."2 It follows that every model must have a purpose if it is to be of value in solving business problems. The purpose of the model in this dissertation is to analyze and deter- mine an optimal price/service configuration based on data input. Although other models are available for such an analysis, none has yet utilized the techniques of comparing the revenue generated with the cost of the individual ser- vice segments on the basis of the total service package.3 The successful implementation of the model by this disser- tation will therefore advance the state of the art in 2Elwood S. Buffa, Models for Production and Opera— tions Management (New York: JOhnFWiley & SOns, Inc., 1963), p. 9. ’See for instance the model proposed by Stanley F. Stasch, Systems Analysis for Marketing Planning and Control (Glenview, 111.: Scott, Foresman and Company, 1972), pp. 492-493. methodology for optimum price/service determination, as well as provide useful information concerning a potentially optimal mix for the brokerage industry. Finally, this dissertation is a preliminary rather than a conclusive study. A single attempt at the construc- tion of a model is unlikely to result in a perfect replica- tion of the real world, although it forms the basis for the development of a new, more refined model.“ By the same token, the results of the test of the model on the brokerage problem will yield only a suggested view of the adjustments which should be made to brokerage service mixes. Problem Background One of the most salient representatives of the American free enterprise system is the American brokerage industry. Nearly every city in the country with a popula- tion in excess of 20,000 boasts at least one or several local offices of a national brokerage firm. These offices are often characterized by high-cost stock market tickers, electronic quotation machinery, stock brokers in gray, pin-striped suits, and thick carpeting on the floor. A serious question must be raised as to whether the brokerage function as it now exists is even necessary. “Buffa, Models for Production and Operations Management, p. 10. Noting the sophistication of modern communications equipment it must at least be conceded that the technological feasi- bility exists for a trading market without the traditional Wall Street trading floor. It has been proposed that all brokerage houses could be tied into a centralized clearing computer system which would match buy and sell orders from everywhere in the country. Part of the current profit slump stems from a change in the rate structure for institutional transactions which took effect in 1971.. Today's institutional trading is far less profitable than in past years. Since 1971, transac- tions in excess of $300,000 have been subject to negotiated rates, rather than the traditional fixed percentages. With part of their revenue gone, the problems of brokerage man- agement began to appear. "During the fat years, brokers made unbelievable amounts of money, with partners of firms each taking home millions of dollars. Few firms invested money back into making their businesses more efficient.5 An additional problem comes from a new class of cut- rate firms in the so-called "third market." These firms, who are neither members of nor bound by the rules of the organized exchanges, are offering discounts of up to 60 percent of normal NYSE brokerage rates in exchange for . 5"The Crowd Gets Lonelier and Unhappier A11 The Time," The Economist, 25 August 1973, p. 82. for the absence of any service other than processing the customer's order and mailing him the certificate (or check).6 A significant piece of anti-trust legislation has also shadowed the potential demise of brokerage firms. If passed, a bill introduced by Senator Harrison A. Williams (D-N.J.) would break up the traditional fixed rate structure of the brokerage industry. The proposed legislation would allow financial institutions to buy seats on the Exchange thereby eliminating at least a portion of the profitable institutional business.7 This bill, known as H.R. 5050, would therefore dissipate the ability of the brokerage industry to monopolize many of the intermediate sized block trades. An even more ominous "dooms day" prediction has been made in response to a recent SEC ruling: In the canyon of gloom that is Wall Street now- days, one of the gloomiest scenarios envisioned by the financial community's seers begins on May 1, 1975. On that date, by mandate of the Securities and Exchange Commission, all rates charged by brokerage houses will be freely competitive, thus brining to an end the historical system of fixed, uniform fees. (The competitive rate structure is already in effect for trades over $300,000 and under $2,000.) And according to the gloom-and- doom scenario, rate competition among brokerage 5Margaret D. Pacey, "Cut-Rate Commissions," Barron's, 6 August 1973, p. 11. 7"The Crowd Gets Lonelier and Unhappier All The Time," p. 83. houses will thenceforth be ruinous to the point that a number of competitors will be swatted down like so many f1ies.° However, marketing theory suggests that brokerage firms will continue to exist as long as their customers perceive the service rendered as being worth the price paid. In the long run, it will be the customer who decides what services they are willing to purchase. To some brokerage customers, the service utilized consists only of a method of placing an order to buy or sell a listed security. To others, service consists of a full array of activities ranging from margin accounts to information services. To the former group, the fact that a brokerage firm discon- tinues or institutes a special charge for a special service would be of little significance. To the latter group, such a move may prompt a discontinuation of patronage altogether. It is indeed possible that some form of low service firm will dominate the brokerage industry of the future. But an equal possibility exists that the present firms will continue to prosper by differentiating themselves through a unique variety of service offerings. The current problem is one of deciding how to predict and adjust to the future. E 'Richard Rustin and Kenneth Bacon, "Regional Brokers Say End of Fixed Rates Won't Bring Collapse," Wall Street Journal, 29 May 1974, p. 1. It would be incorrect to suggest that brokerage firms are unaware of these problems. Nearly every major firm is in the process of collecting data in an effort to readjust their price/service mix in order to meet the impending changes. "The keynote is to develop as many profitable and diversified services as possible," according to Robert C. Hill, president of one of the smaller brokerage firms.’ One need only glance at the advertising themes of companies like E. F. Hutton, Bache, and Merrill Lynch to see that many firms are in a readjustment process. Although these changes appear to be well planned and well researched, nearly all of these studies are of an "in-house" or private nature. They are neither accessible to the public nor broad enough in scope to solve the general problem of service determination for all firms. Indeed, the academic world can only guess as to how these firms are arriving at their decisions. But the current literature fails to provide a general technique for the determination of a specific price/service mix for any industry. Hypotheses It is now in order that the hypotheses of this dissertation be precisely set forth: ’Ibid., p. 23. 1. American brokerage firms are currently offering a service configuration which is not of optimum value to brokerage customers in the sense that most customers are either (1) engaging in transactions without using certain services or (2) are unwilling to pay for some of the services they are using. 2. An analytical cost/revenue model developed for the determination of optimum price/service mix is superior to accounting methods which consider profit on a service by service basis and do not take into account the interaction of the demands for various services. An evaluation of these hypotheses will be made in Chapter V. General Framework As with most managerial marketing problems, a traditional model is usually available for the purpose of providing direction to the study. Since the current marketing problems of the brokerage industry can be easily separated and analyzed by the categories of PLACE, PRICE, PRODUCT, and PROMOTION, this dissertation will use the so- called "four P's" approach to managerial marketing as a theoretical base.1° As can be noted in Figure 1.1, the -___ l°E. Jerome McCarthy, Basic Marketing: A Managerial Approach, 4th ed. (Homewood, 111.: Richard D. Irwin, 1971), p. 44. '“ifilI' I four P's (controllable factors) are surrounded by five uncontrollable factors. The main concentration of the analysis will be on the controllable factors, although a portion of a subsequent chapter will be devoted to the uncontrollable factors. UNCONTROLLABLE cultural and social environment CONTROLLABLE political FACTORS resources and and legal objectives environment PRODUCT PLACE of firm PRICE PROMOTION MARKETING MIX economic existing business environment situation FACTORS Figure 1.1 The McCarthy 4 P‘s Model.11 llIbid., inside cover. 10 To the brokerage firm, the category of PRODUCT takes the form of the brokerage service itself, although even brokers sometimes refer to their services as products. In accordance with the hypothesized problem, product deter— mination as it relates to the brokerage firm must involve the decision of what configuration of services to offer and in what quantity they should be offered. The problem of PRICE relates to the process of establishing and adjusting commission rates for each service or group of services. Until recently, price was a "given" factor in this industry because of the uniform minimum rates charged by all firms. Because of the recent SEC rulings, the firms are now faced with a problem of price/service disaggregation. PROMOTION in the brokerage may be divided into advertising and personal selling. Many of the larger firms have recently undertaken extensive advertising campaigns, as exmplified by the Merrill Lynch "bull" commercials. It may be predicted that all firms may soon have to undertake some form of advertising in order to survive. Brokerage firms have always relied on personal selling as their pri- mary means of promotion, but it is possible that the future may call for less reliance on the traditional broker- salesman. 11 Since brokerage firms are not plagued with extensive physical distribution problems, PLACE in this industry is (primarily concerned with the location of brokerage house branch offices and customer service facilities relative to the proximity of present and future customers. Most of the physical transfer of securities takes place through the 0.8. mail. Since broker-customer contact can be easily main- tained via telephone between any two points in the country, the primary purpose of the branch office is to provide the customer with face-to-face local contact with the firm. Customers insist that their ability to communicate with the firm is timely and convenient regardless of whether contact is maintained by telephone, mail, or face-to-face. Once again, the four P's model serves only as a starting point for the analysis in this dissertation. A detailed analysis of each of the previous elements will take place in Chapter II. Scope It is a widely accepted fact that a major portion of the current problems in the brokerage industry stem from factors beyond the control of the firms. For instance, investors have become apathetic. Because of the uncertain- ties of the domestic and international economic frontiers, the stock market has recently undergone a series of severe 12 declines. Individual investors, puzzled by the complexion of the market forces, have tended to react negatively to further investment in the stock market. Many investors have sustained extensive losses in their portfolios which inhibit their desire and ability to trade. Even those investors who have not lost money are disenchanted with the poor performance of traditionally "safe" investments such as blue chip utilities. Neither the gloom of public disfavor nor the problem of inflation will be dealt with at any great length in this thesis. Instead the major thrust of analysis will center upon several elements which are within the firm's grasp of control. The proposed model will utilize inputs which can be controlled by the firm. Because of limitations of time and money on the part of the researcher, only a preliminary analysis will be made of the PROMOTION and PLACE elements of the marketing mix in Chapter II. The bulk of the effort, including the proposed model, will concentrate on the PRICE/PRODUCT (service) mix beginning in Chapter III. It is the author's premise that it is here that a major portion of the current problem lies, although it must be recognized that no major marketing problem can be limited to any one or two elements of the nmrketing mix. 13 Both the model and the model application will require the input of certain fixed and variable cost components. Procedures for cost accounting are highly complex and require the analysis of costs and procedures within individual firms. Such an analysis is beyond the scope of this dissertation. It will nonetheless be neces- sary to obtain cost estimates in order that the model may be implemented. Methodology This thesis purports to set forth a general model for the determination of service mix and test it on the brokerage industry. This model will require the customer inputs of (1) services ranked in order of preference, (2) determination of revenue generated by each service, and (3) determination of the threshold number of services which a customer(s) will insist upon. Keeping in mind the pre- liminary nature of this study, it would appear that the survey method of research may be useful in exacting hypotheses in that it is a fairly elementary, straight- forward way of simply asking customers what they prefer. Although survey responses are subjective and may fail to reflect the final actions of the respondents, they are nonetheless relatively inexpensive and may be used to form the justification for advancing to a more sophisticated 14 research design. Therefore, it would appear to be most appropriate to utilize such a method for this dissertation. In addition to prescribing the required inputs, the proposed model will describe in detail the arithmetic manipulations and processes necessary to produce the desired outputs. Particular emphasis will be placed on formulating the inputs, processes, and outputs in such a manner that the model may be universally applied to any situation where it is necessary to determine a service mix. Finally, the proposed model will be subjected to a test in hopes of solving the previously described problem in the brokerage industry. It is hoped that the results of the test will prove the usefulness of the model by shedding light on the problem of brokerage service determination. Alternative Research Designs Another possible preliminary research method which could be utilized is of an ex post facto design. If it were possible to obtain the records of various service charges and service packages offered by a cross-section of firms, extensive amounts of data shifting would potentially yield useful information concerning the impact of various service charges. But this method is somewhat like shifting a hay stack in search of a needle, largely because of the failure of most firms to keep sophisticated accounting records. 15 Other research techniques such as controlled experiments are too sophisticated for this study. Designs of this nature usually involve selecting two comparable markets and varying the price/service mix in one market for the purpose of comparison with the control market. The major disadvantage of this testing system is cost. Administrative expenses are extensive for developing, controlling, and analyzing most any form of a "test market." It is therefore almost essential that some form of prelim- inary study (such as a survey) be conducted prior to the commitment of large amounts of money. Furthermore, the new rates are not yet effective and cannot legally be experimentally implemented for a test market. In recent years, simulation has advanced to the state of solving many complex problems. Once the simulation model is built (usually in computer form), it is possible to perform a variety of experiments with the modular system and observe the results. The main disadvantage of the use of simulation for the problem which has been proposed is the problem of constructing the computer model itself. It took the General Electric Company five years to devise and refine a simulation model suitable for sales forecasting. For the proposed problem, it would be necessary to know the service elasticities in advance which are to be tested. Once again, it appears that some form of an initial survey is necessary 16 for the definition of the variables to be manipulated. In view of these reasons, it would be easier and faster to wait until the new rate structure becomes legal and experiment with the real world than go to the extreme expense of building a simulation model. Limitations Since this dissertation will utilize a model with inputs from a survey, such a study is inherently subject to several limitations. One major problem of such a methodol- ogy is the gap which might occur between perception and action. Simply because a survey respondent states that he would like to purchase a given service for $10 per year does not necessarily dictate that he will do so. But in fact, it is nearly impossible to predict the future in any other way. A second problem involved in a survey of this nature is that of the variability of the interaction of the components. To hold all other factors constant while observing the impact of the variability of one component would be to thwart reality. This dissertation must there- fore test several service components at once, recognizing that the whole of the service package is greater than the sum of the parts (synergistic). 17 A third limitation to this methodology is the nature by which surveys oversimplify reality. Exact perceptions are often lost in verbal descriptions and questions. It is often necessary to dichotomize and categorize responses. For instance, it may be necessary to assume that either a service will or will not be offered. In other instances, it may be necessary to assume a finite series of responses such as $10, $20, or $30, rather than the entire range of numbers. In short, it is difficult to convert perceptions to finite numbers. Fourth, it should be recognized that mail surveys are of a lesser degree of accuracy than personal interviews. It is because of cost limitations that the mail survey must be utilized in this dissertation. Finally, since this study involves the construction and use of a model, it should be recognized that models by their nature are abstractions of reality. The model in this study will be of a conceptual nature in that it will present a method for determining optimal price/service potential. It is not reality itself, but an attempt to measure and predict reality. Although the model will be utilized in this dissertation, no single application will ultimately confirm or disaffirm the model's value. In the long run, only the repeated successful usage of the model will attest to its usefulness. 18 Terms and Definitions Certain key concepts will be reflected throughout this study which require definition at this time: Securities: The securities referred to in this dissertation are listed and unlisted common and preferred stocks, bonds, and warrents, all of which are handled by American brokerage firms. Brokerage customer: For the purpose of this dissertation, the brokerage customers will be defined to be those individuals who buy and sell listed or unlisted securities in their own name or family name through the use of a brokerage firm. The obvious exclusion from this definition are the institutional brokerage customers who buy and sell securities for corporations, trusts, mutual funds, and other entities. Some brokerage firms also engage in the execution of orders for the commodity and futures markets, but this dissertation will be limited to customers of securities. American brokerage firms: American brokerage firms are those firms possessing memberships in the New York Stock Exchange, American Stock Exchange, or various regional exchanges for the purpose of executing buy and sell security orders for their customers. 19 Marketing efficiency: Marketing efficiency refers to the performance of a marketing function, activity, or process at minimum cost. Marketing effectiveness: Marketing effectiveness refers to the degree to which a marketing function, activity, or process is successful in accomplishing its purpose. Optimization: All references to optimization will imply profit maximization, unless otherwise stated. Service mix: Service mix will be inferred to be that combination of activities, large and small, which the brokerage firm offers and the consumer finds useful. Con— centration is placed on the amounts of various interrelated services that might be offered, which will be referred to as the service configuration. Epigg; Price will refer to the dollar charge to be paid by the customer for various services or groups of services. Price/service mix: Price/service mix will therefore refer to the manner in which prices and services are combined. Uses of the Study It is hoped that this study will have uses beyond those heretofore set forth. For instance, it is the author's hope that one future value of this dissertation 20 will be to point out the need for further academic research in the brokerage industry, particularly the need for the study of more effective market segmentation and target marketing. Since no study of this nature is currently available to the public, it is hoped that the work provided herein will become a base for future similar studies. Since the proposed model will be of a general design, it should be useful for the development of service configurations in many industries. Since all potential configurations are considered in the model, a change in any one factor can be noted as it relates to the other elements of the service mix. The change of a firm to, say, quarterly billings may have little impact on sales. A reduction in some other small service may also have a negligible impact on sales. But when both actions are performed at once, it may be found that the net reaction is greater than the sum of the independent reactions. Altogether too often, no attempt is made to measure the interrelationship of various actions. Perhaps most importantly the model will zero in on the massive numbers of ambivalent customers who are not motivated to buy any one product. In an environment where "all aspirin is alike" and no brand distinguishes itself from the proliferate mass, the customer may reach for which- ever alternative is available. Such a customer is neither 21 motivated nor dissatisfied. He is merely shopping for a threshold level or service which, when satisfied, terminates his search pattern until one or another of the prescribed services drops below his prescribed minimum service level 95 some other motivation causes him to resume his search pattern. Most prudent marketing organizations have long since abandoned any notion of being able to please "all of the people all of the time." In every area of marketing expense, there is an expenditure point beyond which the revenue stream generated by the additional expense fails to justify any attempt to create total customer satisfaction by all customers. Too often, managers decide what the proper level of service should be by trial and error. In recent years, a large quantity of literature has noted that perfect satis— faction is unnecessary and unprofitable.12 In recognition of this fact, this dissertation will attempt to provide a method for defining a profitable service mix. Finally, it should be noted that the importance of this study is magnified by the predictions for the growth of service industries. The growth of services is a logical result of several converging forces. As the marginal attraction of services increases faster than that 1zSee, for instance, "The Case for 90% Satisfaction," Business Week, 4 November 1961, pp. 108-110. 22 for more goods, entrepreneurial capital and business managements are turning more to the development of service markets. Because the expansion of these markets requires business organization, managements borrow applicable techniques that were successful in the mass production of goods. In the short run, the emerging service systems both limit the range and impersonalize the nature of services. In the long run, it is possible that a desirable proliferation of services will develop, paralleling that for commodities.13 Presentation Overview Chapter I has presented an introduction to the topic. Chapter II will be devoted to a marketing overview of the brokerage industry. A brief analysis will be made of each component of the industry's four P's marketing mix. The selected topics will include such factors as personal selling, advertising, and office location. Particular attention will be devoted to the current debate concerning brokerage pricing practices. The final portion of the chapter will be devoted to a description of the services currently being offered by the brokerage industry. Since any new contribution to the state of the art must first begin with an assessment of the status quo, Chapter III will present a review of various methodologies 13William J. Regan, "The Service Revolution," Journal of Marketing, 27 (July 1963), 57. 23 for the determination of the service mix and will be entitled "Theory of Price/Service Mix Determination." Attention will ultimately be centered upon the cost/revenue analysis model. Chapter IV, entitled "Research Design," will present a modification of the cost/revenue model for the purpose of solving the proposed price/service mix problem. The model will then utilize a survey input of a representative sample of brokerage customers concerning their attitudes toward their most preferred service configurations. In order that the effectiveness of various firms may be assessed, the data will be reclassified and analyzed according to the partic- ular brokerage houses used by the responding customers. Chapter V will report the "Findings" of the research study. This chapter will analyze the customer response to the preferred service mix configuration and will draw tentative conclusions concerning the most preferred service mixes. The analysis will present a series of customer pro- files for each of the major brokerage firms represented in the sample Chapter VI, entitled "Summary, Conclusions, and Recommendations," will appropriately wrap up the study. The implications of the impact of the study will be made. The potential for future research will be discussed. CHAPTER II MARKETING IN THE BROKERAGE INDUSTRY From the viewpoint of a marketing manager, this chapter will present an overview of the current operating posture of the American brokerage industry. The analysis will begin with a description of the environmental setting in which the industry is currently operating, including its relationship with the Securities and Exchange Commission and the nature of the competition within the industry. Each of the elements of the strategic marketing mix will then be considered, including the location of brokerage facilities, the nature of the industry's sales promotion, brokerage fee determination, and brokerage services. The Operating Environment Although the year 1975 will probably be remembered as a year of severe recession for American business, the brokerage industry will recall that the recession for many Wall Street firms began some two years earlier. Although members of the largest exchange, the New York Stock Exchange, earned over $877 million in 1972, they lost $150 million in the first five months of 1973 and continued to lose at the 24 25 rate of about $40 million per month for the remainder of the year.1 In the year of 1973 alone, Wall Street witnessed the disappearance of 106 firms, 73 of which discontinued operations altogether. The remaining 33 were absorbed into other firms.2 Although 1974 was still a red ink year, the results were not as bad as‘l973. An aggregate profit was earned in the months of March and September by the members of the New York Stock Exchange. Several companies, including Merrill Lynch, First Boston Corporation, E. F. Hutton, and Reynolds Securities all announced substantial increases in third quarter profits over the previous year.3 But overall losses by all NYSE firms were still $92 million for the first three quarters of 1974, although the same period in 1973 had shown a loss of $162 million.“ Although certain brokerage houses are currently profitable, none of the firms are anywhere near their peak profitability levels of the late 1960's, and many others are continuing to lose money. A glance at the poor performance 1"The Crowd Gets Lonelier and Unhappier All the Time," The Economist, 25 August 1973, p. 82. 2"Wall Street Woes," Barron's, 10 June 1974, p. 5. ’"E. F. Hutton Profits Increased Sharply in Quarter, 9 Months," Wall Street Journal, 17 October 1974, p. 6. l"'Big Board Had Indicated September Profit," Wall Street Journal, 26 November 1974, p. 2. 26 of the stock market over the past several months provides a clue to a portion of the problem. In any declining market, brokerage customer confidence in the entire financial system falls, resulting in a decline in business for many firms. Another portion of the problem relates to rising costs. Like many businesses which are labor and equipment intensive, the brokerage industry is subject to the problems of inflation. In the period between June 1973 and September 1974, brokerage firms averaged a 9.2 percent increase in costs.5 Two other environmental factors which have con- tributed to the problems of the industry are the nature of the competition among certain classes of brokerage firms and the policy stance of the Securities and Exchange Commission toward price regulation of the industry. These factors warrant closer consideration. The Securities and Exchange Commission Certainly one of the most important factors in the operating environment of the brokerage market is the Securities and Exchange Commission, the watchdog arm of the federal government. It is the purpose of the Commission 5"Big Board Firms Urgently Need 8% Rise in Brokerage Rates, Needham Tells SEC," Wall Street Journal, 25 September 1974, p. 2. 27 to protect the investor and company from unfair practices, and to maintain confidence in publicly held corporations. During the late 1960's, the SEC began to notice that an increasing portion of the trading on the NYSE and other exchanges was being conducted by large institutional in— vestors. In 1973 about 70 percent of all trading on the NYSE was institutional. But institutions became increas- ingly dissatisfied with the fixed commissions of the major organized exchanges, and they began to take more of their transactions to the regional exchanges, where commissions could be negotiated. They also took more of their business to discount brokers who were not members of organized exchanges and not subject to the fixed commission schedules of the large organized exchanges. The SEC felt compelled to save the Street from self-destruction. Trying to c0pe with these cycles, the Commis- sion worked frantically against itself. Since 1970 it has granted three increases in the fixed-price structure; a fourth is expected this month. But, like a father who fears he has been too generous, the Commission took back with one hand what it granted with the other. In rulings highly popu- lar with Congress, the SEC in 1971 made commission rates competitive on all portions of orders exceed- ing $500,000--which led, as the Commission intended to a reduction of those rates. The cutoff was lowered to $300,000 in 1972. It was due to come down to $100,000 last April. But that move was postponed out of fear that it would bring down the whole Street with it.6 6Walter Guzzardi, Jr., "The SEC's Crusade on Wall Street," Fortune, November 1974, p. 246. 28 By May 1, 1975 the SEC has ruled that all brokerage rates must be fully competitive, despite the apparent negative effect on the profitability of the firms which the Commission is supposed to protect. This subject will be further discussed in the section on pricing. Another action by the SEC relates to the nature of the services offered by the brokerage industry. Last September [1973], when the Securities and Exchange Commission approved higher commission rates it also requested the exchanges to revise their rules to permit, but not require, broker- age houses to provide less than the full range of services, in return for discounts of up to 10 percent by April 1, 1974. According to SEC Chairman Ray Garrett, Jr., the stock exchanges and their members are ex- pected to "assume the initiative" in developing a limited form of competitive brokerage fees to begin in April 1974. The SEC envisioned this as a means of gradually enabling the industry to gain experience in the "unbundling" of services prior to the imposition of fully negotiated rates after April 30, 1975.7 Several firms, including Merrill Lynch, have begun to experiment with unbundled services as a prelude to a future system of fully competitive rates and service offer- ings. It may be inferred that if the rate structure does become competitive on May 1, 1975, then all firms will have to develop some form of a plan for deciding upon which services to unbundle and what prices to charge for the services. 7"New SEC Rulings Encourage 'Unbundling' of Broker— age Services," American Investor, June 1974, p. 4. 29 One pending action by the SEC relates to an eventual plan to eliminate the trading floors of all organized exchanges and substitute a centralized market. The idea is that all the present markets--the stock exchanges in New York, the regional exchanges, and the third market--would be linked in an electronic network. The network would provide information about the prices of securities everywhere and would shunt orders to the market in which the price was best. All orders awaiting execution at prices as good as or better than, a newly entered order would have to be given priority. The exact nature of a plan such as this is unclear, especially with regard to the role which would be played by the existing stock exchanges. It is equally unclear who will provide the capital for such a venture, since estimates 9 But it is obvious that run at upwards of $100 million. such a plan would have a profound impact on the competitive nature of stock prices by exposing a customer's order to many different markets. Competition There is no present law stating that securities must be bought and sold through an organized stock exchange. A sale can be concluded between any mutual buyer and seller by simply obtaining a notarized endorsement on the back of the stock certificate and sending it to the transfer agent who aGuzzardi, "The SEC's Crusade on Wall Street," p. 246. 9Ibid. 30 maintains the record of the firm's stockholders. The membership in an organized exchange by a brokerage firm is therefore only one means of providing a medium for joining buyers and sellers. Of the stocks listed on the NYSE in 1971, about 15 percent of the total trading volume took place on regional exchanges and away from Wall Street, and 7 percent took place in the so-called "third market."‘° The third market.--The third market is a term used to describe a growing number of brokerage firms who do not belong to any organized exchange but conduct an over-the— counter business in stocks that are listed on other ex- changes. Because they do not belong to an exchange, they are subject only to the rules of competition in determining their rate structure. These firms exist because of the discount fees they offer their customers in exchange for the absence of the services associated with most brokerage houses. In short, they offer cash-and-carry service for a minimum price--no research, advice, sales- men, margin accounts, safe-keeping or discretionary accounts. The buyer (or seller) puts down his money (or stock), is given the best price available and is mailed the certificate (or check). That's all there is to it.1 l°Donald E. Farrar, "The Coming Reform on Wall Street," Harvard Business Review, September—October 1972, p. 110. ll"Cut-Rate Commissions," Barron's, 6 August 1973. p. 11. 31 Third market firms have traditionally dealt with the institutional market, but recent increases in the fix-fee schedules of the major exchanges have resulted in the patronage of even many small investors seeking to save commissions.12 Their manner of operation is modest. Most of these firms occupy only one office and are normally linked to the outside world only through the telephone. They do not fear the competition of the big firms even if services become unbundled because of their modest costs and modest commissions.13 The growth of the third market does, however, pose a threat to the New York Stock Exchange, and Chairman James J. Needham would like to have it abolished.1“ The fourth market.--Another source of competition to the organized exchanges and their member firms is the so-called "fourth market." This market is a step beyond the third market in that two institutions get together without the aid of a broker and make a transaction. In practice, few firms are able to locate each other without the aid of one of several firms who make a business of computer match- ing buyer's and seller's trading intentions. The fee 12"Discount Brokers: Don't Discount Them," Financial n.“ 1.4] World, 7 February 1973, p. 13. 13"New Specialty Houses Fill the Gaps," Business Week 15 July 1972, p. 71. 1""Discount Brokers: Don't Discount Them," p. 13. 32 charged for this service is small even when compared with the discount prices of the third market. Although the current volume in this specialized market is light, its proponents predict it will grow.15 The Location of Brokerage Facilities Whereas the function of a commercial broker has been associated with commerce for centuries, it was not until the beginning of the nineteenth century that the stock broker as he is known today came into existence. Somewhere in the late 1700's, a group of New York security dealers began to interact with each other on behalf of their customers, and it was decided that it would augment the brokerage function if they were to form an organized exchange. In 1817 the door was Open to a rented room at 40 Wall Street under the name of the New York Stock and Exchange Board, which later became the New York Stock Exchange.16 In these early beginnings, there were few logistics problems because of the local nature of the operation of the brokers who belonged to the exchange. But as the nation grew, so did the need to raise capital. During the drive 15"The 'Fourth Market' Is Real," Commercial and Financial Chronicle, 5 August 1974, p. 1. 16George L. Leffler, The Stock Market (New York: The Ronald Press Company, 1963), pp. 81-85. 33 to raise money for the expansion of the railroads in the mid-1800's, it became essential that the New York brokers contact investors all over the country, and the need for correspondent brokers in other cities became necessary. With the invention of the telegraph and later the telephone, these branch offices became more closely linked to their parent firm and the New York Stock Exchange. Besides the search for new capital, the brokerage firms had other reasons for opening branch offices. By its nature, the brokerage function is a service, and there was a demand for this service in all major population centers. The growing level of American affluence resulted in a growing desire to invest. The fever of speculation in the 1920's resulted in a national interest in the stock market. Even after the depression of the 1930's, the mass financing of corporations through the sale of securities continued to characterize the American industrial financial structure. The postwar period of the 1950's and 1960's saw the price level of stocks quadruple. In any era of advancing market prices, the rumors of vast fortunes being made on the stock market fans the fires of the profit-minded American investor. Many people who had never purchased securities set out to seek a competent stock broker through the recommendations of their friends or financial associates. Since most of these customers preferred to do business on 34 a face-to-face basis, most customers patronized local branch offices. Between 1950 and 1970, the number of branch offices of the NYSE members alone grew from 1,661 to 3,636.17 In general, the new offices were built by the brokerage firms for the expressed purpose of gaining a larger share of the existing market or capturing a portion of the expanding market. Many of the firms were successful in locating these new offices for several reasons. First, one element of the marketing mix--brokerage fees--was fixed for all major firms. Second, the concept of mass national advertising had not really caught on. This left the brokers of the new offices with the more limited task of selling themselves and their firm to new customers, while perhaps explaining the convenience of the location of their new office relative to the offices of other firms which may not even have offices in that particular town. Third, the climate of the times was right. New brokers seemed to have little difficulty in attracting new customers. Between 1952 and 1970, the number of American shareholders in public corporations grew from approximately 6 million to 30 million.10 17The 1974 Fact Book (New York: New York Stock Exchange, 1974), p. 84. '°Ibid., p. 50. 35 A great deal of money was expended in this era on lavish office facilities. Most firms furnished their offices with walnut desks, wood paneling, thick carpeting, and plush "theater"—sty1e seats which faced a panoramic stock ticker. The purpose of this expensive decoration was to create an image of security and well—being to the custom- ers who visited the offices. Such a strategy has been used for years by banking institutions, who learned after the depression of the thirties that large marble buildings and plate glass windows lent to a feeling of security and well being in the minds of the customers who entrusted their money to these institutions. Since brokerage customers often invested even more money with their broker than with their bank, it may be implied that a certain amount of plush imagery was necessary, if for no other reason than to appear on an even par with competing brokerage offices. It is fairly obvious to even the casual observer of the operation of a local brokerage office that these houses do not themselves normally possess vast numbers of securi- ties in their own name waiting for customers to purchase them. Instead, the national firm bearing the name of the local house acts as an agent for the customer and attempts to obtain (or sell) the security in the marketplace at the best possible price. This usually requires that the firm be a member of the New York Stock Exchange or some other 36 organized exchange in order to have contact with the market place. The purpose of an organized stock exchange is to provide a central clearing market for the joining of buyers and sellers. Logistically, this means that a customer's buy or sell order is transmitted from his local branch office to the floor of the exchange, where a representative of the brokerage firm carries the order across the floor of the exchange to a specified post where the particular stock is traded. The customer's order is executed by the floor broker negotiating the transaction with either the person operating the post (called a specialist or "market maker") or with a floor broker of another firm who may already be at the post with a matching (buy or sell) order from his customer. The news of the floor broker's success is then transmitted back to the local office of the customer, and a record of the sale almost simultaneously appears on the stock market ticker. One of the biggest internal logistics problem of the securities industries is communication. Because of the remoteness of most branch offices from the trading floors of the New York and American Stock Exchange, extensive out— lays must be made to maintain the communications' equipment which provides the vital information needed by brokers and customers. Indeed, communication costs represent approx- imately 10 percent of the total operating expenses of the 37 average brokerage firm." This includes the cost of operating the quotation machinery, as well as telephone and teletype expense. In summary, it may be concluded that brokerage facilities are located to accommodate customers in the manner in which they wish to be served. In the past, most customers have insisted upon direct contact with a broker in a local office, and new branch offices were located to meet this market. But the recent downturn in the industry has resulted in the merger and acquisition of the branch offices of many firms, resulting in the possibility of a mismatch between the current pattern of branch office locations and the concentrations of brokerage customers. The pattern is further distorted by the desire of some customers to do business with firms with no branch offices, namely, the third market. Promotion Promotion refers to any form of communications between the buyer and seller of a product or service. "The basic, broad objectives of promotion are to inform, persuade, or remind target customers about the company's marketing mix 1’Security Commission Business (New York: New York Stock Exchange, 1971)T’ 38 "2° The discussion which follows and the company itself. will center upon two of the most important forms of pro- motion in the brokerage industry, namely, advertising and the sales function. Advertising For the majority of the lives of most of the current leading firms, little attempt has ever been made to utilize any form of a mass advertising campaign. The notable lack of advertising in this industry is the result of several factors. First, many firms have in the past felt that they were "above" such forms of promotion and that advertising might hurt their image. Second, many firms specialize in institutional customers who are more difficult to reach and persuade with advertising. Third, the overwhelming success of many firms in the 1960's made the generation of addi- tional non-institutional customers unnecessary and almost unwanted. It may be recalled that the NYSE adOpted a schedule of reduced hours in 1968 to help the member firms dig out from under their backlog of paperwork. It was not until after the golden age of the 1960's that Wall Street saw the first major advertising campaign launched by a brokerage firm. On October 6, 1971, Merrill 2°E. Jerome McCarthy, Basic Marketing, 4th ed. (Homewood, 111.: Richard D. Irwin, Inc., 1971), p. 514. 39 Lynch, Pierce, Fenner & Smith launched a multi-media campaign which carried a theme of optimism about the future of America. The phrase, "Merrill Lynch is Bullish on America" is now a familiar slogan to many Americans. The advertisements appeared on television, in magazines, and in newspapers.21 Aside from the "bullish" theme, the ads also attempted to explain the brokerage functions and services which the firm could provide. The overall campaign was judged to be so successful that it has been expanded over the last three years. The total expenditures for 1974 were about $128 million, of which $8 million was television, magazine, and newspaper ads, and $48 million of which was divided among educational literature, special exhibits, and direct mail.22 Not to be completely outdone by Merrill Lynch, two other firms, E. F. Hutton and Bache & Co., have launched mass advertising campaigns with annual budgets of $2 million and $1 million, respectively.23 One key advertisement used by Hutton features a personal testimonial from J. Paul 21"Merrill Lynch Campaign Carries a 'Bullish' Theme," Advertising Age, 11 October 1971, p. 3. 22"Big Brokers Plan No Ad Cutbacks; Wall Street Names Three Uptown Shops," Advertising Age, 2 April 1974, p. 2. 23Ibid. 40 Getty, who praises the company for the service he has received. A primary function of advertising is to create a distinct image of the advertiser's firm. Therefore other brokers are attempting to use advertising to inform brok— erage customers of their unique services, such as Dean Witter's ads concerning their biweekly research letter.2” It is difficult to tell the degree to which the current advertising campaign is creating a generic interest in stock and securities rather than a particular customer interest in the firms conducting the ad campaigns. The profitability levels of both Hutton and Merrill Lynch have improved over the past year, but Bache continues to lose money. But the word from all firms who are advertising is that the campaigns will continue. The Sales Function Few brokerage firms like to think of their brokers as salesmen in the traditional sense of the term. The industry therefore refers to them as "account executives," "registered representatives," or just "brokers." Yet their primary function is to sell their service and the services of the firm to brokerage customers. This function is 2""Thank Goodness It's Thursday," Wall Street Journal, 19 September 1974, p. 13. 41 usually performed in a local branch office where customers come to meet the broker face-to-face, or by telephone, especially when the customer is not local. After a local customer has met and is familiar with his broker, he is likely to contact him more by telephone than in person when he wishes to place an order. Because of the decline in overall volume in the industry, over 3,000 brokers lost or left their jobs last year, and the 33,000 who remained were forced to get by on much less income.25 "Men who had styled their lives around six-figure incomes have turned to cab driving, bartending, and other less remunerative occupations."26 Part of this problem stems from the system of compensating brokers. There is no incentive paid to the salesman for the quality of the job which he does. As long as the customer does not become dissatisfied and seek another firm, the broker continues to earn revenue each time a trade is made, but gply when a trade is made. The concentration of the effort of the brokers is often on the high-volume traders, regardless of whether they are winners or losers. They tend to become concerned with 25"Broker Steve Karelity Continues to Prosper Despite Bear Market," Wall Street Journal, 11 December 1974, p. l. 26Stanley H. Brown, "A Random Walk Among the Wounded," Fortune, October 1974, p. 154. 42 doing whatever is necessary to maintain the customer's short-term volume, rather than considering his long-term investment objects and best interest. E. B. Weiss, vice president in charge of creative merchandising at the advertising agency of Doyle, Dane, and Bernbach told a New York Stock Exchange conference on marketing that: "The main problem with the commission struc- ture," he said, "is that the whole focus is on sales. The concept of service is neglected. And your future will be service-oriented." "Surely total financial planning and the commission fiyptem of compenation are not compat1ble. On the other hand, brokers can be the best promotion lever of the firm they represent. A good broker is always prospecting for new customers, which often requires night work on the telephone. Especially when the broker is backed up by an advertising campaign, he can sell the potential customers the secure image and reputation of the firm. Like other salesmen, he can present unique features of the ser- vice which he and his firm can provide. Like successful practitioners in other professions, brokers can attract a permanent following of customers, who in turn refer their broker to their friends. 27"Wall Street Marketing: Can It Sell the New Line?" Commercial and Financial Chronicle, 22 June 1972, p. 13. 43 In the summer of 1973, Financial World conducted interviews with the five most successful brokers in the securities industry in an effort to determine why they were still successful despite the decline in the industry. Several common factors appear behind their success. All of these brokers have diversified the scope of financial services they perform for their clients. More than being security salesmen, they have become money managers in the true sense of the word, steering their clients into tax shelters, corporate bonds, insurance and commod- ities. Secondly, each one of these brokers has developed a sense of rapport with his clients so that, at least at this point, the clients are not second—guessing him, even though portfolio values are down. Past success seems to be having a tranquilizing effect on current apprehensions. Finally, each of the five men is deeply committed to his business and to his clients and has refused to use the current market performance as an excuse for declining values.28 Therefore a good broker may be a man who is well versed in all aspects of the investment business. He is willing to sacrifice a Short-term commission for a long-term level of customer confidence. He is truly concerned with the long run welfare of his client, and earns his commissions as a result of this concern. 2°Barry Tarshis, "How Top Brokers Prosper in Bear Markets," Financial World, 8 August 1973, p. 20. 44 Brokerage Fee Determination For the members of most of the major exchanges, brokerage fees are currently determined by a minimum fixed fee scale which corresponds to Table 2.1.2’ Revision of this scale is always subject to the approval of the SEC. Table 2.1. Minimum Commission Rates Effective November 19, 1974 The rates shown are minimum and are scheduled to be abandoned May 1, 1975 in favor of completely competitive rates. Actual rates currently charged on orders up to $300,000 may be higher than the minimums. In addition, reduced rates are available on certain types of orders of $2,000 or less. SELLING AT $1.00 AND ABOVE Commissions on stocks selling at $1 per share and above are to be computed on the basis of the amount of money involved in an order. The schedule of rates should be applied to each sin- gle order, as defined. The Constitution sets forth the minimum commission as follows: On 100 Share Orders 5nd Odd Lot Ogders Moggy Involyed In Thg Ordg; Minimum Commission $ loo-but under $ 800 . . . . . . . . . . . 2.0% plus $ 6.40 $ BOO-but under $2,500 . . . . . . . . . . . 1.3% plus $12.00 $2,500-and above . . . . . . . . . . . . . . . 0.9% plus $22.00 Odd Lot-$2 Less Multiple Round Lot Orders Money Involved In The Order Minimum Commission $ lOO-but under 5 2,500 . . . . . . . . . . 1.3% plus $ 12.00 $ 2,500-but under $ 20,000 . . . . . . . . . . 0.9% plus $ 22.00 $20,000-but under $ 30,000 . . . . . . . . . . 0.6% plus $ 82.00 $30,000-to and including $300,000 . . . . . . 0.4% plus $142.00 Plus (for Each Round Lot) First to tenth round lot . . . . . . . . . . . $6 per round lot Eleventh round lot and above . . . . . . . . . .$4 per round lot 2’Security Owner's Stock Guide, December 1974, pp. 253-254. 45 Beginning in 1971, the SEC ruled that fixed fees could no longer be applied to transactions below $2,000 or above $300,000. This resulted in the establishment of competitive rates among firms for transactions under $2,000. Customers with orders in excess of $300,000 most Of which are institutions, are now in a position to bargain for the best rates among various firms. Fees in the third market run between 20 percent and 80 percent below the minimum fees that NYSE member firms must charge.30 To the institutional investor who makes daily trades, such savings are very significant. But even the small investor can save commissions by doing business with a third market firm, as evidenced by the following comparison of three firms with the NYSE fixed commission scale: Which house gives the biggest discount depends in part on the size of the trade. For example, on a $1,500 trade involving a sale or purchase of 100 shares at $15, the NYSE commission is $31.50, StockCross $28, Marquette de Bary $25.83 and Odd Lots $25.20. (Source Equities is not included because its rates apply to a whole year's trading, not individual trades.) From there on up to a $20,000 trade, where Odd Lots starts to negotiate its commission, StockCross usually has the edge because of its flat $28 fee per 1,000 shares: For 1,000 shares at $15, the commissions are NYSE, $217; Odd Lots, $108.50; de Bary. $86.80 3°James P. Roscow, "Cutting Down Your Commission Charges," Financial World, 26 September 1973, p. 4. 46 and StockCross, $28. For 5,000 shares at $20, the figures are NYSE, $862; de Bary, $384.80 and StockCross, $140, while Odd Lots' fee would be negotiated.31 On October 17, 1974, the New York Stock Exchange announced that its board of directors had voted to refuse to comply with the SEC ruling calling for fully negotiated rates by next May 1. The Big Board reaffirmed the principle of having fully negotiated rates as a long-term goal, but only if the so-called "third market" were eliminated.32 James J. Needham, Chairman of the Big Board, fears that the removal of fixed fees too quickly would "remove the incentive brokers currently have to remain members of exchanges."33 Needham also feels that the abolition of fixed rates should be accomplished by establishment of several SEC rules that would: 0 Require brokers that aren't stock exchange members to expose all orders to an exchange before executing them Off the exchange in the over-the-counter market. 0 Retain a Big Board rule that requires mem- ber firms to take all orders in listed stocks to exchange markets before deciding whether to execute them Off the exchange. 0 Retain fixed minimum floor brokerage rates-- fees that members charge one another--even though rates charged on public trades become competitive. 31"The Discount Brokers," Forbes, 15 June 1972, p. 28. 32"Big Board Refuses to Heed SEC Order on Negotiated Rates," Wall Street Journal, 17 October 1974, p. 2. 33"SEC Says Order to End Fixed Broker Fees Isn't Negotiable With New York Exchange," Wall Street Journal, 11 December 1974, p. 16. 47 0 Place exchange and nonexchange members under equal rules.3“ Noting the action of the Big Board, Ray Garrett, Jr., chairman of the SEC, declared that "we are not in the posture Of negotiating an arrangement."35 This infers that the entire issue of negotiated rates may be headed for a showdown in the courts. The proposed SEC action has little support anywhere in the industry. In a survey conducted by the Securities Industry Association, 309 of the 354 members who responded to a survey concerning their Opinion toward competitive rates responded that they favored the retention of some form of the traditional fixed fee scale. Of these same respondents, 125 favored the return of a full fixed rate system similar to that which existed prior to 1971 when the SEC made rates outside of the $2,000 to $300,000 range competitive.36 Despite the negative reactions, the SEC appears intent upon its course of action. One criticism lodged against the SEC is the possibility that they may have misjudged the nature of the resulting competition. The 3"Ihid. 35Ibid. 36"SEC's Plan for Competitive Broker Rates Is Opposed by 309 of 354 Firms in Poll," Wall Street Journal, 27 September 1974, p. 2. 48 financial institutions, which comprise 70 percent of the trading volume on the NYSE alone, are Often much larger and more powerful than the brokerage firms. They can already use their power to force commissions to the sub- sistence level for trades over $300,000. The brokers are already blaming this loss in revenue for part of their loss in profits. When all rates become competitive, they will be able to squeeze the brokers even further. Other critics have asserted that the competitive situation will become one in which only the largest firms will survive. But a recent study concludes that "the introduction Of competitive rates would not appreciably increase the concentration of economic power in the brokerage industry. The exception is the national retail firms where there is probably a better than even chance that there would be some modest but not alarming increase in concentration."37 The same study also concludes: The main argument in favor Of competitive rates is that they can be expected to appreciably raise the market's efficiency. The market for outstanding stocks will clearly be improved, and there does not appear to be any basis for the allegation that new equity investment would be adversely affected. Virtually all classes of 37Irwin Friend and Marshall B. Blume, "Competitive Commissions on the New York Stock Exchange," Journal of Finance, 28 (September 1973), 818-819. 49 investors are likely to benefit. These gains appear possible without any major damage to the structure of the brokerage industry.38 Where the controversy will end is uncertain at this time, and it will probably be necessary to wait until May 1, 1975 to see whether the SEC action will take effect on schedule. But it is apparent that at some point in the future the industry will be forced to adopt competitive rates. Brokerage Services Prior to the advent of price competition and advertising, the nature and quality of the services given to the brokerage customers provided one of the few means by which firms would compete for differential advantages. Salesmen frequently referred to services such as research as being superior in their particular firm. Almost no brokerage customer uses all of the services available to him from his broker. In fact, some customers may be unaware that certain services are available at no extra charge under the fixed fee structure. Under the proposed new system of competitive rates, services will become unbundled and special charges will be instituted for at least some of these services. 3°lbid., p. 819. 50 This section will look at two classes of services, the first of which is composed of services which are offered by most firms, and the second of which is composed of more unique services. Services Offered By Most Firms By definition, the service of order execution is offered by every brokerage firm in the industry. This service connects a customer who wishes to buy or sell a given security with another person somewhere in the country who wishes to sell or buy. The nature of this service is such that it must always be offered or the firm can no longer be defined as a brokerage firm. Specialty orders.-—Most full-service brokerage firms also allow the customer to place a variety of dif- ferent types of orders. The most common order is one which directs the broker to buy or sell at the best market price. If the customer desires, he may place a slightly different type Of order which specifies the maximum or minimum price at which the order is to be executed. If the customer feels the price of a given stock is about to fall, he may place an order to sell "Short," which allows him to borrow stock to sell now. Security custody.--Another service offered by most firms is security custody. This service allows the customer to maintain the securities he has purchased in Special 51 account maintained by the brokerage firm so that the necessity of keeping stock certificates in a safe deposit box is avoided. When securities are held in "street name," a great deal Of paper shuffling is avoided for the customer, brokerage firm, and stock transfer agent when stocks are rapidly traded by the customer. One disadvantage to the customers who use this service is the delay in the receipt of information such as annual and quarterly reports from the companies because of the necessity Of brokerage firm to receive and readdress all correspondence to the stock— holders.39 Research.--Any organized attempt by a brokerage firm to create a base of information concerning recommenda- tions of whether to buy, sell, or hold certain securities may be classified as research. This information may be provided to the firm's brokers for dissemination or be published in newsletter form for the direct use of the customers. A weakness of this service is the low quality level Of many of the recommendations. A 1973 Harris poll revealed that about one in every five investors felt "over—promised" by their broker in the purchase of their last security."0 3’“The Silence Imposed by Street Names," Business Week, 8 December 1973, p. 40. . I'°Lee Berton, "Locked Out From the Street's Best," financial World, 30 May 1973, p. 5. 52 Office contact.-—Most brokerage firms provide the services of a local office for the purpose of maintaining face-to-face contact with their customers. These offices usually provide a comfortable place for the customer to sit and talk with his broker, as well as a certain amount of reading materials. Most Offices have a panoramic display ticker which keeps pace with the current trading action on the floor of the New York Stock Exchange (some firms also maintain an American Exchange ticker). The local Office location may also simplify the delivery of securities and payment for customers who prefer not to use the mail. Market contact.--One of the primary functions of the broker is to give his customers the best knowledge which he possesses about the market. When a customer calls, the broker should be able to provide him with information about market conditions and trends, as well as potential trading actions which the customer may wish to consider. The broker should also provide up-to-the-minute stock price quotations by use of special electronic quotation machinery. Unique Services These services are offered by a minority of firms in the industry. Although some are part of normal brokerage duties, others are fairly specialized and are independently priced under special rate scales. 53 Option trading.--"Options enable investors to speculate on changes in stock prices without actually purchasing the stock.“1 An option is therefore a right to buy or sell a stock at a specified price at some future point in time. The largest Options market is the Chicago Board Options Exchange which began operations in April 1973. Many firms are now connected to the CBOE. Portfolio management.--In order to link the functions of the broker with those of the money managers, several firms including E. F. Hutton and Company now offer the service of taking over the management of an entire portfolio. The firm is responsible for making all buy and sell decisions for which a flat rate of k of 1 percent of the assets of the portfolio is charged each year. If the use of the service catches on, it may replace a portion of the traditional brokerage business altogether. Commodity futures.--Many of the larger brokerage firms (including Merrill Lynch) have trained a portion of the sales staff to handle the trading Of commodity futures, which is a form of buying a contract for the purchase of a commodity at some future point in time with the hopes that market price may be higher so that a profit can be made by selling the contract. The dollar volume of trading in the l”"SEC Clears AMEX To Begin Pilot Plan for Options Trades," Wall Street Journal. 20 December 1974, p. 3. 54 commodity exchanges now exceeds the level Of trading in the New York Stock Exchange."2 Many future brokerage customers may wish to participate in commodity speculation. Insurance. Since March Of 1972, members of the New York Stock Exchange have been permitted to sell life, accident, and health insurance. It has been proposed that they be allowed to sell all lines of insurance, although only 83 of the NYSE's 521 members are currently engaging in this segment of the investment business."3 The selling of insurance by brokerage firms is a step toward a one-stop financial supermarket. Summary This chapter has presented an Operational overview Of the brokerage industry from the vieWpoint of a marketing manager. A description was made by each element of industry's marketing mix. Up until several years ago, this industry operated with little advertising and pricing was conducted under a fixed fee scale. Promotion was carried out largely by “26era1d Gald, "Commodities' Trillion Dollar Future," Commercial and Financial Chronicle, 22 July 1974, p. l. “3"N.Y. Exchange Members Seek to Sell All Lines of Insurance," The National Underwriter (Life), 16 March 1974, p. 4. 55 salesmen, and companies differentiated themselves with their service offerings and their branch office locations. The past three years has seen the introduction of large advertising campaigns by several major firms. In addition, the SEC has begun moving toward the unbundling of services and the gradual elimination of the fixed fee scale under the assumption that competitive rates are necessary for the survival of Wall Street. Firms will soon have to begin making some hard decisions concerning which services to unbundle, which services to discontinue, and how much to charge for the unbundled services. The chapters which follow will therefore attempt to develop a methodology for the determination of an optimum price-service mix and to apply this method to the brokerage industry. CHAPTER III THEORY OF PRICE/SERVICE MIX DETERMINATION "Theories without facts may be barren, but facts without theories are meaningless."1 It is therefore essen- tial that a theoretical base be established for any research study. This chapter will explore the prominent literature in service pricing, service marketing, service levels, and cost/revenue analysis. Classical Theory of Pricing From its inception, economic theory has been con- cerned with the efficient allocating Of scarce resources. Traditional economic theory is based on the Operation of the price system as a means of allocating land, labor, and capital among alternative uses. Price theory grew out of the need to explain, predict, and describe the actions of the growing number of industrial firms spawned by the industrial revolution. It was an admission by the econ- omists of the day that pure competition was no longer the dominant economic force. 1Kenneth E. Boulding, Economic Analysis (New York: Harper and Row, 1966). P. 5. 56 57 One Of the first writers to successfully contribute to price theory was the French economist Leon Walras (1834- 1910). Although his approach was strictly mathematical, he successfully developed a complex set of simultaneous supply and demand equations which depicted the general equilibrium Of an economy at any given point in time.2 The contemporary science of econometrics owes its foundation to Walras. But since econometric analysis is designed for the general equi- librium of an entire economy and requires the definition of all economic elements, such a theoretical approach is beyond the scope of this dissertation. Alfred Marshall (1842-1924) felt that a more prag- matic method Of economic analysis was necessary. His famil- iar intersecting supply and demand curves form the basis of the contemporary pedagogical approach to economics, and his textbook is still often quoted as the bench mark of economic analysis3 (see Figure 3-1). Since the evaluation of commerce it had been recognized that an increase in the price of a good or service would generally result in a sales decline, but the question of the rate Of the decrease was seldom formally considered. Marshall coined the term "elasticity" to 2Walter Nicholson, Microeconomic Theory (Hinsdale, ;L11..: The Dryden Press, Inc., 1972), p. 10} 3Alfred Marshall, Principles of Economics, 8th ed. {New York: Macmillan and Co., 1938). 58 Price Equilibrium Price ---------- Equilibrium Quantity Quantity Figure 3.1 The Marshallian Supply—Demand Cross. describe the responsiveness of the changes in quantities purchased with respect to changes in price. "Marshall defined elasticity as the ratio of the proportioned change in quantity (demanded or supplied) to the proportional change in the price."“ Demand or supply is said to be elastic if the proportional change in the quantity demanded exceeds the proportional price increase, and conversely inelastic if the change in the quantity demanded is less than proportional. Referring to elasticity in an absolute sense, some economists simply define elasticity as the ratio Of change in quantity to the change in price. It should be noted that the shape and degree of ealasticity Of any demand curve is determined by the numerous ’— I’Boulding, Economic Analysis, p. 182. 59 individuals who compose the market in question. Their individual actions often take the form of a binomial deci- sion: whether to buy or not to buy. In some instances, elasticity is determined by the quantity purchased by individuals. But in the market for consumer goods and services, the impact of a price increase is one of causing the customer to contemplate a shift to a substitute product or to a similar product produced by a competing company. Marshall's theory can therefore provide useful insights for the pricing of goods and services, providing that the supply and demand curve be defined. The theory can also be extended to the determination Of the price of several goods or services offered jointly. For example, if two goods are jointly Offered, that is, the sale of one good is complemented by the sale of others, then a change in the supply or demand for one will alter the demand for the second as indicated by Figure 3.2.5 If several goods or services are put together into a single package and must be purchased altogether, then all of the supply and demand curves blend together and become like a supply and demand curve for a single product. The exact nature and elasticities of the new supply and demand 5Donald 8. Watson, Price Theory and Its Uses (Boston: Houghton Mifflin COmpany, 1968), p. 271. 60 QA QB Good A Good B Figure 3.2 Relationship Between Prices Of Complements. curves will be some form of an average between the supply and demand curves for the individual products as indicated in Figure 3.3.6 Although contemporary economics has fully refined the theoretical constructs of supply, demand, and elasticity, practically nothing has been done to set forth a methodology for determining the curves and the elasticities themselves. McCarthy discusses a methodology for estimating a demand curve, but even the casual Observer will note that a demand curve developed by this method is so subjective as to limit its application.7 6Boulding, Economic Analysis, pp. 236-238. 7E. Jerome McCarthy, Basic Marketing, 2nd ed. (Home— wood, 111.: Richard D. Irwin, Inc., 1964), pp. 799-804. 61 A+B+C QA+B+C Figure 3.3 Summation of Demand Curve. Except for providing a theoretical base, the application Of economics for the solution of the price/ service mix determination problem proposed by this disser- tation is of little value. In this sense, economic theory is not so much wrong; it simply does not gO far enough. It invariably begins with the demand curve as a "given," and it is excessively concerned with price as the sole determinant of buyer behavior. Many Of the elements of economic theory could be projected to other elements of the marketing mix, but the current thrust of microeconomics centers only on price. 62 It is a further premise of this dissertation that price/service mix cannot be separated, ceteris paribus, into individual components for analysis. It would seem useless to define the demand elasticity Of one service component without exploring its impact on the total package of com- ponents. Herein also lies a weakness in the contemporary economist's supply-demand analysis for joint products: any truly useful model must take into account and measure the synergistic impact of any change on the entire service package. Especially in the case of a large number of goods or services being Offered as a group, certain customers may be attracted or repelled by the nature and price of any single component. The methodology of this dissertation shall attempt to overcome this problem. Service Marketipg It is unfortunate that very little of the marketing literature is concerned with the marketing of services. The bulk of the writing by most marketing scholars has been aimed at the marketing of products. Converse et a1. states: Most books on general marketing begin by saying that the work of marketing includes both goods and services. The good intention to con- sider both goods and services may be preserved 63 during the first few chapters but soon the awkward and roundabout phrase falls and the discussion—- and unfortunately the thought--is concentrated on commodities.a Most writings in the functional areas of advertising, personal selling, pricing, and physical distribution are concerned with consumer goods, despite the fact that 30 to 40 percent of the consumer's dollar is spent on various forms of services in the domestic economy.’ Converse et a1. is one of the few principles textbooks to devote an entire chapter to service marketing,‘° although Beckman and David- son also devote a section to the subject.11 Entire books written on the marketing of services tend to encompass a specific class of services.12 There appears to be no widely circulated books on the general subject of service marketing. °Pau1 D. Converse, Harvey W. Huegy, and Robert V. Mitchell, Elements of Marketing, 7th ed. (Englewood Cliffs, N.J.: Prentice-Hall, Inc., 1958), p. 495. ’John M. Rathmell, "What Is Meant by Services?" Journal of Marketing, 30 (October 1966), 32. l°Converse et a1., Elements of Marketing, 8th ed., pp. 397-413. 11Theodore N. Beckman and William R. Davidson, Marketing (New York: The Ronald Press Co., 1967). pp. 97-98. 12See, for example, W. J. E. Crissy and Robert J. Boewadt, Marketing of Hospitality Services (East Lansing: American Hotel and Motel Association, 1971); and A. Wilson, The Marketin of Professional Services (London: McGraw-Hill Bock Co.,il 2). I. ‘14 .au ..- (Q. a‘h \ 4 AU rVfi .Fu\.\u\ 3d L . su a3 Av OH H... Dun .9 P» nit 64 There are, however, several significant journal articles which deal with the basic question of what con- stitutes a service as related to the service industries. This portion of the literature warrants further discussion. In order to organize the presentation of this literature such that it relates to the problems associated with the brokerage industry services, it appears worthwhile to consider (1) the ramifications of the "wheel of retailing" as it relates to service,13 (2) the unique aspects of mar— keting services, and (3) the use of franchising in service industries. Services and the Wheel of Retailing Services have traditionally evolved around the sale Of products. For instance, many customers expect the sale to include the services of credit, delivery, special orders, and repair. It is Often assumed that these services augment the sale of products in both the consumer and industrial Inarkets, but the question of whether or not they increase sales relative to their cost is seldom measured. It is Often only when some service is eliminated that the impact of that service on sales is noted. 13The term "wheel of retailing" first appeared in M. P. McNair, "Significant Trends and DevelOpmentS in the Postwar Period," in A. B. Smith (editor), Competitive 21stribupion in a Free, High—Level Economy and Its Impli- Cdtions for the Univgrsity (Pittsburgh: University of Plttsburgh Press, 1958), pp. 1-25 at pp. 17—18. 65 Large numbers of services are often fairly expensive and cannot be easily absorbed into the general cost struc- ture. This requires that either the price structure be adjusted to reflect the cost, or that a service charge be made to cover the additional cost. When a service charge is instituted for a service, certain consequences may result. Many times the service charge may eradicate the service altogether if an insuffi— cient number of customers are willing to pay for it. Such a pricing strategy may even anger the customer into discontinuing his patronage altogether. Another problem of service charges is determining the price at which they are to be offered. One strategy would be to add a small charge with the intent of discour~ aging wasteful and excessive use Of the service. A second strategy might call for pricing to cover the marginal costs. It follows that a third strategy may be to cover marginal costs and spread a portion Of the fixed costs. A final service price strategy might call for covering both fixed and variable costs plus make a profit. Altogether too often, it is infeasible to make individual charges for minor services associated with the sale of goods and major services. It is equally infeasible (and sometimes illegal) to offer these minor services to some customers and not to others. Consequently, the service 66 package attached to a product is often treated as purely a cost of doing business, and the service offering simply duplicates the package offered by the competition. There is Often an attempt to outdo competitiOn by offering one more service in the hopes of attracting some unknown portion of business away from the rest of the market. An escalating service package must eventually lead to higher costs for the firm and, hence, higher prices. At this point, an innovator usually enters the market with the promise of offering sub- stantially lower prices in exchange for little or no service. The most notable example of this type of change in recent years is the emergence of the discount department store under the names of Woolco, Kmart, Zayer, and Korvette. Today these stores are characterized by adding new services a little at a time. The wheel of retailing hypothesis states that these institutions will eventually add new services (and more expense) to the point that a new class Of discounters will once again sweep the market.‘“ The theoretical "wheel" is then unmistakably complete, but neither the high service nor the low service firm is necessarily sure exactly 32123 services should or should not have been eliminated or added. ll'Stanley C. Hollander, "The Wheel Of Retailing," in Marketing in Progress, ed. by H. C. Barksdale (New York: 4 Holt, Rinehart and Winston, Inc., 1966), pp. 441-449. 67 Unique Aspects of Service Marketing The current body of marketing theory possesses a degree of universality which may be applied to many com- mercial activities. In this context, there are many marketing principles which are equally applicable to the marketing of products and services alike. For instance, similarities can be found in product and service selling techniques, promotion, and advertising.15 It is in the differences of service from product marketing that many problems lie. There are at least seven unique aspects to the marketing of services: 1. Lack of taggibility. Whereas the results of some services (such as house painting) are readily available for the world to see, the invisible nature of other services makes them difficult to promote, advertise, or demonstrate. Unlike consumer goods, they are Often sold by description only.16 Cognitive dissonance. Because of the intangible nature of services, the customer must often rely on his memory to be the judge of the quality of the 15Robert C. Judd, "Similarities or Differences in Product and Service Retailing," Journal of Retailing, 43 (Winter 1968), 1-9. 16Ibid., pp. 4-5. 68 service. Although satisfied at the time of receiving the service, the passage of time may cause the customer to recall elements of the service that he did not like.17 3. gocalized Operation. Most services do not lend themselves to Shipment or storage like their tangible counterparts. A service must generally be performed at or near the point of consumption.18 4. Qualipy control. Because most services are per- formed in total or in part by people, the quality of the service may be no better than the training, temperament, effort, and concentration of the person performing the service at any given point in time. Because of the subjective nature of service quality, it may be difficult to establish any form of a quality control standard.9 5. Propensity toward upper incomes. Whereas food, automobile, and housing tend to be consumed by all income groups, many services tend to be consumed primarily by middle and upper income groups.20 17Richard M. Bessom, "Unique Aspects of Marketing Services," Arizona Business, Fall 1973, p. 11. "rbid. "John M. Rathmell, "What Is Meant by Service?" p. 35. 2°W’illiam J. Regan, "The Service Revolution," Journal of Marketing, 27 (July 1963), 60. 69 This is especially true for travel, education, legal, and stock brokerage services. 6. Postponement potential. Unlike certain consumer essentials such as food and clothing, many services are of a non-essential nature and can be postponed for long periods. This is especially true of services related to leisure, such as movies and travel. 7. Lack of marketing. Perhaps the largest difference between the service and consumer goods industries is the failure of the service industries tO devote sufficient attention to marketing. "It is ironic that service businesses, which are necessarily in the most direct contact with the consumers, seem to be the last kind of firm to adopt a consumer- oriented marketing concept.21 Franchising and Company Ownership Perhaps the most successful attempt in recent years to combine the advantages Of mass marketing with the neces— sity of local Operation of various service industries has taken the form of the franchise.22 The use of the franchise 21Bessom, "Unique Aspects of Marketing Services," p. 14. 22Shelby D. Hunt, "The Socioeconomic Consequences of the Franchise System of Distribution," Journal of Marketing, 36 (July 1972), 32-38. 70 allows a nationally recognized institutional name to be placed on what might otherwise be an ordinary local service business. Several important advantages accrue to service organizations who are linked by franchises. First, the name can be easily recognized by travelers and transients. Second, uniform service Offerings can be established so that the customer knows in advance what to expect. Third, cen- tralized facilities for personnel training, purchasing, and warehousing can Often be established. Finally, the fran- chise organization can undertake a broad-based advertising campaign which may be impractical for independent local units. Another method Of developing a series of service outlets is through centralized ownership of the units by a parent company. The casual Observer may easily fail to notice the difference between the operating practices of a company-owned and a franchise service Operation, Since both types of institutions may appear under national names and utilize national advertising. The classification distinc- tion is usually made on the basis of ownership and control of the local operation. The brokerage industry is clearly of the centrally owned type Of Operation, but many of the advantages asso— ciated with franchising also apply. Many brokerage firms 71 make use of national and regional advertising and provide uniform services between branches. Especially in the case Of the large firms, their names bear national recognition among brokerage customers. Level of Service A significant portion of the writings concerning services has been in the area of physical distribution and business logistics. Since many of these same concepts can be applied to services in general, it is worthwhile to explore the state of the art in the physical distribution concept Of "level of service." Physical distribution, in a very basic sense, is concerned with moving finished products to the right places, at the right time, and in the right quantities. It is therefore a type of service which is concerned with those marketing activities which take place after the product is produced. In the past fifteen years, a great deal of atten- tion has been devoted to performing these tasks with a greater degree of efficiency and effectiveness. The primary purpose Of physical distribution from the standpoint of the customer is delivery of the product. When questioned more closely, the customer is usually able tc> define some specific characteristics which he expects the IthISical distribution service to possess. Hutchinson and Strolle identify seven elements of customer service to be: 72 1. Order processing time: elapsed time from receipt of customer's order until it is ready for assembly. 2. Order assembly time: time required to prepare the order for shipment. 3. Delivery time: time in transit to the customer. 4. Inventory reliability: stockouts, back orders, percentage of demand filled, omission rate, per- centage of orders shipped complete, and so on. 5. Order-size constraint: minimum order size and minimum frequency allowed. 6. Consolidation allowed: ability to consolidate items from several locations into a single shipment. 7. Consistency: range of variation in each of the preceding elements.23 In this light, customer service may be defined as ". . . a complex collection of demand-related factors under the control of the firm, but whose importance in determining supplier patronage is ultimately evaluated by the customer receiving the service.2“ 23William M. Hutchinson and John F. Stolle, "How to Manage Customer Service," Harvard Business Review, November- December 1968, p. 88. 2"Ronald H. Ballou, Business Logistics Management (Englewood Cliffs, N.J.: Prentice-Hall, Inc., 1973), p. 96. 73 All elements of customer service come at a cost prOportional to their quantity and quality. By inspecting the foregoing seven item list, it may be concluded that some threshold number and quantities of these services must be provided if any reasonable amount of customer demand is to be obtained. As may be observed by Figure 3.4, a certain trade- Off exists between the amount of customer service Offered and logistic system costs. As the level of service is increased by improving the quantity and quality Of existing services and adding new services, the additional number of customers who are not lost to competition allows the service cost to be spread over more units. But as the service level is extended beyond a certain point, the number of additional customers attracted or retained by the increased service is not sufficient to offset the increased costs, and the total cost curve begins to rise. According to this example, the firm would do best to remain at the customer service level of Sx (see Figure 3.4).25 Several points are worthy of note from this analysis. First, an extremely high level of service is seldom the point of lowest cost. The number of customers attracted by 25A similar analysis is presented by Donald J. Bowersox, Edward W. Smykay, and Bernard J. LaLonde, Ph Sical Dlstribution Management (New York: Macmillan Company, I968), pp. 209-212. 74 5 1 Total Costs Costs Seerce Costs per Unit | of Sales 1 Lost Sales Costs 5)! Level of Service Figure 3.4 Logistic System Costs.26 offering perfect service are simply not enough to justify the cost in most instances. Second, this analysis is concerned with minimization of logistics costs. The question of total costs and total revenues for the whole firm was not considered. New cus- tomers allow the spreading of logistics costs over a wider base, but they also represent new revenue for the entire firm which may exceed the cost Of attracting them with increased services. Third, this analysis is theoretical and does not present a methodology for actually determining an optimum service level. One analytical technique which has the potential for operationalizing this concept is cost/revenue analysis, which will now be discussed. 26Adapted from J. L. Heskett, Robert M. Ivie, and Nicholas A. Glaskowsky, Jr., Business Logistics (New York: The Ronald Press Company, 1964), p. 30. 75 Cost/Revenue Analysis It has been estimated that 50 percent of the cost of all products sold in the United States are marketing costs.27 The impact of the energy crisis and inflation may have driven this percentage even higher. Since the refine- ment of the American mass production system has trimmed manufacturing costs to near-optimum levels, it seems appar- ent that future improvements in the efficiency of many firms will have to come from the more effective allocation of marketing costs. Marketing cost/revenue analysis forms a new and growing body of theory and knowledge which aims at accomplishing this goal. Although marketing cost/revenue analysis grew out of the need to define and analyze physical distribution costs, it has subsequently been applied to all marketing costs.28 It may be broadly defined as any method, system, technique, or model designed for the purpose of measuring efficiency and effectiveness by comparing any marketing cost or cost grouping with the revenue stream it generates. 27E. Jerome McCarthy, Basic Marketipg, 4th ed. (Homewood, 111.: Richard D. Irwin, Inc., 1971), p. 21. 28See, for instance, Charles H. Sevin, "Some Aspects of Distribution Cost Analysis," Journal of Marketing, 11 (July 1947) 92-98. 76 It is a major premise of cost/revenue analysis that every expenditure which the firm makes should, in either the short or long run, generate a revenue stream which exceeds its cost. In this context, all cost expenditures which do not generate a sufficient revenue should be elim- inated. Furthermore, the impact of every decision, action, or proposed change must be considered in light of its final impact on the cost, revenue, and profitability of the firm. The difficulty of implementing this concept is the problem of attempting to define and measure the revenue stream which relates to a particular cost. A major difference between cost/revenue analysis and normal accounting procedures is the purpose for which it is intended. Accounting is concerned largely with maintaining a complete historical record of those company events thatfin any way have a financial flavor. Thus, the system feeds to management a story of merchandise sales, materials purchases, interest accruals, equipment depreciation, salaries paid, and all other activities involving financial considerations. In some contrast, marketing cost analysis is a managerial tool designed more for use in the planning and control of future Operations in a firm.29 This is not to suggest that certain aspects of cost accounting analysis as it is usually practiced are not used for planning and control purposes. Indeed, the body of *1 2’William J. Stanton and Richard H. Buskirk, Mana e— ngt.of the Sales Force (Homewood, 111.: Richard D. Irw1n, Inc-, 1969), p. 594. 77 knowledge in cost accounting has successfully reduced many problems of production planning and control to a refined science. However, the purpose of production cost accounting is to devise methodologies for minimizing costs, i.e., improve efficiency. There is a vague notion that a "quality" product must be produced if it is to be successfully sold, but various production and quality control steps are Often added or eliminated without attempting to measure the impact Of the action on the overall cost and revenue picture of the firm. In this sense, marketing cost/revenue analysis must go beyond the information which production cost accounting normally provides. There are several reasons why marketing costs are difficult to analyze. First, the results obtained from marketing expenditures are Often difficult to measure, especially in the cases of advertising, promotion, sales, and physical distribution. Traditional accounting methods can determine if these Operations are efficient, but the question of effectiveness is not so easily answered. Second, many marketing activities take place outside of the controlled environment of the firm or plant. It is Often difficult to trace where the dollars go and exactly how they are spent. Third, many firms do not possess information systems ndaixsh lend themselves to the collection and analysis of 78 marketing costs. In most cases, the analysis must be based on the regular accounting records, which are usually inadequate for this purpose.3° A major portion of the writing in cost/revenue analysis is devoted to the theory and methodology for making the transition from traditional accounting to a system adequate for cost/revenue analysis.31 A typical approach begins with breaking the natural accounts of the accounting system (those accounts associated with the firm's profit and loss statement) into functional accounts. Functional accounts may be characterized by relating to a specific purpose, such as advertising or physical dis- tribution. Finally, the functional accounts are broken into segments for further analysis. Although the choice Of segments depends on the nature of the analysis being conducted, the most typical breakdowns are by product, customer, territory, channel, and order size.32 With costs and revenues fully disaggregated to these levels, it is then possible to compare the relative profitability Of the segments. 3°Ibid. 31"Report of the Committee on Cost and Profitability Analysis for Marketing," The Accounting Review, Supplement to Volume 47, 1974, 577-615. 32Frank H. Mossman, Paul M. Fischer, and W. J. E. Crissy, "New Approaches to Analyzing Marketing Profitabil- ity," Journal of Marketigg, 38 (April 1974), 46. 79 The foregoing system of breaking costs and revenues into segments may be termed the traditional approach.33 Contemporary writers in the field have suggested that cost and revenue source documents should be coded as they are incurred, and that storage of these documents in a modular data base information system would allow both the accounting and marketing functions of the firm to utilize the data.3“ Other writers concern themselves with the levels and degree to which costs and revenues are aggregated or disaggregated, as well as the nature of the ratios which may be used for the analysis of the result.35 There is a minor degree of variation between the writers in the field as to breadth of analysis. Mossman et a1. infers that the analysis should be made by product, territory, 25 customer, whereas Lewis and Erickson infer that all three breakdowns could be analyzed at once. They further recommend that the differences in the configuration can be analyzed through the use of analysis of variance 33Richard J. Lewis and Leo G. Erickson, "Distribution System Costing: An Overview" (unpublished, Michigan State University, 1972), p. 3. 3"Mossman, Fischer, and Crissy, "New Approaches to Analyzing Marketing Profitability," p. 44. 35"Report of the Committee on Cost and Profitability Analysis for Marketing," pp. 593-605. 80 (ANOVA), although a less sophisticated design which skips unneeded configurations may also prove useful.36 A logical extension of the current state of the art would be to construct a model which automatically encompasses the correct variables in the proper relationship to each other such that the net result provides management with the information it needs. Such a model will be demonstrated by the next chapter. Summary The economist's theory of price provides a theoretical insight to the determination of an optimal price/service mix, but it stops short of providing a technique which can easily be applied to the brokerage industry. The same limitation is found in the treatment of the physical distribution concept of levels of service. The current literature has described the problem of pro- viding a service level, but little has been done to develop a system for determining exactly which services should be provided and at what prices they should be Offered. A portion of this chapter has been devoted to a review of state of the art in the marketing of services. This literature provides insight into the unique nature 36Lewis and Erickson, "Distribution System Costing: An Overview," pp. 20—26. 81 of service marketing, but it also fails to provide a methodology for the determination of a price/service mix. The analytical nature of cost/revenue analysis makes the application of this theory most useful to the solution of the current problem. All of the foregoing concepts provide insight for the construction of an analytical technique for the Optimization Of price/service mix and may be included in a cost/revenue model. This will be demonstrated in the following chapter. CHAPTER IV RESEARCH DESIGN This chapter will set forth in detail the design of the research model, the procedure for designing, testing, and tabulating a questionnaire which specifically aims at the service problems of the brokerage industry, and the methodology for drawing the sample of brokerage customers for testing the model. The Concept of Primary and Secondary Service Most service industries are organized around one primary service which is usually obvious to even the most casual observer. For instance, bus companies are primarily concerned with the tranSportation Of people between various points on the map, and carpet cleaners are primarily con- cerned with cleaning carpets. Colleges and universities are primarily concerned with offering courses of study, and brokers are primarily in the business of executing buy and sell security orders for customers. But a complete service Offering usually calls for the variation Of other service components. For instance, 82 83 the frequency with which a bus runs and quality of the ride are subservice factors which augment the primary service. Colleges may vary the quality of their course offerings by controlling the availability Of certain courses, quality of instruction, and the availability of Special services such as libraries. All of these subservices may be lumped into a catch-all category of peripheral or secondary services. It is the tendency of the non-marketing oriented firm to concentrate on the primary service and ignore the myriad of small services which complement the organization's primary service and create differential advantage. Indeed, it must be conceded that the abandonment or severe curtail- ment of the primary service of bus companies (to transport passengers) or universities (to offer courses of study) would most probably lead to their extinction. Accordingly, a large portion of the attention of management must be devoted to the primary service. But it is often the secondary services which provide the service enterprises with their leverage in the market place. Secondary services are not ends in themselves. Their purpose is to augment the sale of the primary service, and in some cases, produce profits of their own. Stated differently, it may be viewed that the secondary service should generate a total revenue for the firm through its direct and indirect impact on the primary service which 84 exceeds the cost of providing the secondary service. Any secondary service which fails to meet this goal should be abandoned. Astute marketing oriented firms in all service industries concentrate on the augmentation Of their primary service with a carefully planned array of secondary services. There are at least three basic methods for the inclusion or exclusion Of secondary services in the service configuration, one of which is to offer the primary service as an entity by itself and offer the secondary service on an optional basis. A charge is then made for the primary service and individual charges are made for each of the secondary services selected by the customer. An example of a service industry which Operates on a somewhat similar strategy is the insurance industry. Most insurance companies bill an additional charge for each and every additional type of coverage. A second strategy is to simply include the secondary service with the primary service as a combined package. The charge for the secondary service is built into the charge for the primary service, and the customer may be unaware that a charge for the secondary service has been included in the price structure. Some customers may even be unaware that the secondary service exists. Secondary charges applied in this manner are often hidden, as with the case of a college student who may be shocked to find that 20 percent Of his tuition is used to support a walk-in health 85 service. As noted earlier, it has been the nature of the brokerage industry up until recently to simultaneously Offer all of the secondary services and include them in the overall charge for the brokerage service. Altogether too Often, complete disaggregation from the primary service as presented by the first strategy is infeasible. Services are often excessively expensive when a sufficient number of customers is not available to spread the fixed costs. Some services simply do not lend them— selves to individual billing, as would be the case with air conditioning in a bus. In all probability the quality of higher education would suffer if college libraries began charging admission. But it may still be infeasible for competitive reasons to lump all of the primary and secondary services into one package because of the resulting high price. Many service industries must therefore rely on a third configuration strategy of Offering some services in a package with the primary service while offering other secondary services independently for a corresponding fee. The Service Cost/Revenue Analysis Model The purpose of this model is to synthesize the foregoing concepts into a general framework which can be used for the analysis of this and other service mix problems. Since the model is concerned with the approximation of the 86 most profitable service mix through the use of a balance of costs and revenues, it was decided to refer to the technique as the Service Cost Revenue Analysis Model (SCRAM) in order that it might be more easily referenced and recalled. The main thrust of the model involves the (1) ranking Of various service alternatives, (2) matching each alternative with cost, and (3) matching each cost and service ranking with its revenue stream for the purpose of determining maximum profitability. This process must be accomplished in accordance with certain refinement which warrants closer examination. INPUT PROCESSOR OUTPUT Figure 4.1 Standard Model. As depicted by Figure 4.1, most models possess three basic components: (1) inputs, (2) processor, and (3) out— puts. In order that the SCRAM model may be more easily understood, it will be described in these segments. Outputs The first portion of this chapter set forth several strategies for the offering of primary and secondary service configurations. Without some reliable method of matching 87 costs to revenues, it is difficult to guess which strategy is most profitable. Therefore three of the desired outputs would include: (1) the most profitable a-la-carte or inde- pendent service configuration, which assumes that each service is offered separately, (2) the most profitable package configuration, which assumes that all services in the package must be bought together, and (3) the most profitable combined strategy, which assumes that a package Offering can be augmented with independent services. A fourth and final desired output from the model would be that of maximum potential sales revenue. This output would measure the total dollars of sales which would be generated if all customers purchased all of their desired services. Such an output would be useful for judging the most revenue a firm could expect to receive if there were no costs and all services could be freely Offered. Inputs The inputs required by the model include the results of a cross-sectional customer survey and the determination of several decision parameters. The initial requirements Of the survey are that the primary service be clearly defined and that each of the secondary services be described and listed for the inspection of the respondent. The survey must automatically assume that the primary service is always being offered, but that any and all combinations of the 88 secondary services may be made available. Next, the respondent should be asked to indicate those services which he uses. Thirdly, an indication should be made of those services used by the customer which he considers to be absolute "musts," i.e., those services which would cause the customer to seek another firm if the service were not Offered. Fourthly, the survey must indicate the dollar amounts that the customer is willing to pay for each service used. Finally, the survey must determine the volume or quantity of primary service which the customer normally uses per unit Of time, such as one year. The required decision parameters include the input of the fixed and variable costs for the secondary services. Although it would also be possible to include semi-fixed and semi-variable costs, the analysis will be limited to fixed and variable costs for simplicity's sake. An addi- tional decision parameter must be inserted to encompass the customer's price variation tolerance. This parameter assumes that a customer will continue to purchase a given service configuration as long as it does not exceed his willingness to pay for the service by more than some given percentage, such as 10 percent. This factor recognizes the fact that no service configuration can ever be offered at the exact, to-the-penny price that a significant number of customers will prefer. It is recognized that rational 89 customers can and do pay more and less than a specific optimum, and that other elements of the marketing mix besides price influence purchases. Finally, a list of all potential service configurations must be provided. Proce 8801‘ The processor section of the model converts the inputs into the desired outputs through a series of process- ing Operations as shown in Figure 4.2. As may be noted by the aforementioned figure, the model assumes the input of (A) a configuration list which contains all possible service combinations to be tested, (B) the results Of a customer survey, (C) a price tolerance coefficient, and (D) fixed (annual) and variable costs for each service being tested. Processingsteps.--Each of the processing subsec- tions may now be described, beginning with Step 1 "Compute Average Price." In this process, the price for each service in the configuration is computed by averaging the dollar response of the customers in the survey. Step 2 "Formulate Price/Configuration List" allows each service in the various configurations to be assigned a price as computed in Step 1. Step 3 "Select a Configuration" begins the computa- tion section of the model by selecting the first item on the price/configuration list. As indicated by the feedback loop, 90 A.uwmcoo phoneme BHhOmml On. I ma. SDZdez mmoam m I, A.u«usoo ouooouv .aHthm..oa..ma mmoam 02¢ NBDmZOU m mBmOU 92¢ OMXHh. .Hoeos serum one ~.e unseen MNZOBmDU fl Bumflmm v mmmZOBmDU . BmHmZH mbbtmm «mmZOBmDU Mm QmmD mMUH>Mflm add m QuadmmAOB NUHmm Reenaueeae «mozameqoe enema zHraHz \\ ZOHB¢MDGHKZOU fl Homflmm ESE: U moow>uon 1 U poms mooa>uouwuv { Quezon 3 he)? on an :aaaas A...» in.» H meow» lonesome mo .0: seize museum .mmv aqu zonaampoHezoo \monmm maeqozmoe [1L mUHmm m5¢mm>¢ mZOH84MDUHm200 mflmHmmOm Add .4 91 this step is repeated until all configurations are considered. Step 4 "Select a Customer" references the survey input and directs the first respondent to be compared against the configuration being tested. This step is repeated until all respondents have been tabulated. Step 5 "Remove Insist Customers" involves comparing the configuration being tested against the profile of ser- vices which the respondent insists upon. If a service insisted upon by the customer is not present in the con- figuration being tested, then that particular customer is considered to be lost and another customer is selected for analysis from Step 4. If all services were offered, then Obviously no customers would be lost in this step. Step 6 "Are Services Used by Customer?" compares the configuration being tested against the customer's response to the services he uses. Those services which are both used by the customer and Offered by the configuration, if any, are then subjected to the next step. Step 7 "Within Price Tolerance?" compares the customer's price for the service as indicated by the survey against the price at which it is being offered in the con- figuration being tested. If the service is being offered at a price less than the customer's response, it may easily be assumed that the customer would be willing to purchase 92 at a lower price. Even if the service were offered at a price higher than his indicated willingness to pay, the customer may still be willing to pay a slight premium rather than lose the use of the service altogether. The price tolerance coefficient is therefore a subjective estimate of price flexibility on the part of customers. This flexibility is necessary to reflect the fact that a customer's notion of the relative importance of price may be changed by other elements of the market place such as personal selling and advertising which may influence behavior to vary. The user Of the model may, however, elect to set this coefficient at zero. If the customer's selection of services are within the price tolerance, the offered price of the services are multiplied by the number Of transactions that the customer makes per year (as per the survey) and the revenue is added to the total for the configuration being tested. The process then selects another customer by returning to Step 4 and the previous steps are repeated until all customers have been tallied. Step 8 "Compute and Store TR-—TC==Profit" occurs when a configuration has been completely interfaced with all respondents. At this point, the total cost of Offering the service can be subtracted from the total revenue, and total profit (or loss) can be recorded, along with a list Of the items in the particular configuration. 93 Step 9 "Store Maximum TR-TC==Profit" compares the profitability of each configuration and stores the total cost, total revenue, net profit, and a list of items in the configuration which is most profitable. The model now continues to select new configurations from Step 3 until all configurations have been tested. When the process is complete, Step 10 "Output" sets forth the maximum profit configuration which has been recorded by Step 9. Computation of outputs.--The question must now be raised as to how each of the desired outputs might be Obtained. The answer lies in the assumptions made about certain processing operations of the model itself. For instance, the first desired output which assumes all ser- vices tO be offered independently may be Obtained by assum- ing in Step 7 Of the model that 223p service for which the customer indicates usage must be compared against the price the customer is willing to pay and in turn with the price tolerance coefficient. Therefore some services in any given configuration being tested may be priced far beyond what a customer is willing to pay, whereas others may be within his price range and may be added to the revenue for that partic- ular configuration. The rest of the model is processed as previously described, but it must be recalled that the out- put under the modified assumptions will be the maximum profit configuration for independently offered services. 94 The second desired output requires Obtaining an optimum package configuration. This result may be obtained by assuming that all prices for the configuration must be added into a package price in Step 7. The customer's will- ingness to pay for the combined total of the package being tested must be compared with the price tolerance coefficient and, if the package is priced within the customer's price range, the revenue can be added. Once again, the rest of the model can be processed as previously described. The third output may be obtained by utilizing the best package configuration (from Output 2) and adding each additional configuration not included in the package under the set of assumptions used to Obtain Output 1. This process allows the best of both outputs to be combined. The fourth output of maximum potential revenue may be Obtained by simply assuming costs to be zero and input— ting only the configuration which includes all services. This process will allow the output Of the greatest sales volume that could be Obtained under the assumptions that all services are being Offered and that prices are based on a mean average of the survey results. Although only four outputs from the model were requested in this instance, it can be seen that minor changes in the underlying assumptions Of the model or its inputs can result in many different outputs. For instance, 95 it might be desirable under certain circumstances to eliminate the "insist" function or some other parameter altogether. It might also be useful to input and test Specific prices in order to ascertain the impact of a price change. Other potential uses will be described in Chapter VI. Description of Test Questionnaire In the American brokerage industry, the primary service performed by the firms is the execution of buy and sell orders for listed and unlisted securities. This ser— vice is usually performed through the brokerage firm's membership in various organized exchanges, such as the New York and American Stock Exchanges, The National Asso- ciation of Security Dealers (for over-the-counter business), and various smaller, regional exchanges. The secondary brokerage services are those facil- itating functions which make it easier or more profitable for the customer to utilize the primary service. Although the secondary services Offered by any one firm are limited in number, the services Offered by all firms are seemingly endless. In order to test the model, it was decided to select six of the most significant (and costly) secondary services which are currently Offered by nearly all brokerage firms. Specifically, the services selected for testing 96 include (1) security custody, (2) Office contact, (3) specialty orders, (4) telephone contact, (5) Odd lots, and (6) research information. An additional seventh service, portfolio management, which is not Offered by most firms was added in order to assess its potential. In accordance with the survey parameters of the SCRAM model, the first page of the questionnaire (see Appendix A) begins with a description of the primary service and a descriptive list of secondary services. A series of response blanks are then provided to allow the respondent to indicate the services he uses, rank the services used, and indicate which services he insists upon having. The second page of the questionnaire provides blanks for the indication Of willingness to pay for unit services under the assumption of a series of $4,000 transactions. A similar series of blanks is provided for annual services. The respondents are then asked to indicate the number of transactions which they actually made last year and their average dollar value. The final page of the questionnaire calls for a response indicating the brokerage house most often used by the customer. This question is not intended to fit the model. Its primary purpose is to facilitate the comparison of the variation in survey results between the customers of the various firms. 97 A final open-ended question is provided to allow the respondents to state their suggestions concerning additional services which they would like to receive. By its nature, the question provides an Opportunity to critique the present service offering. Sample Selection It is unfortunate that there exists no universal list of American brokerage customers, although there are many ways of Obtaining lists which contain segmental por- tions of the brokerage customer population. Examples of bias lists such as these include the subscription lists of brokerage and financial trade publications such as The Wall Street Journal, Barron's, Fortune, and Financial World. Each brokerage firm also possesses what they feel to be a list of their current customers. But since the firm really has no more than a record of past transactions, even these lists are less than accurate. A customer who has made no transactions in the past year (1) may still consider him- self to be a loyal customer of the firm, (2) may currently be doing business with another firm, or (3) may plan to do business with several firms. . . . We never know when a customer stops being a customer of Merrill Lynch and becomes a customer Of some other firm. . . . If a person wants to buy $10,000 worth of stocks, and he orders three dif- ferent issues, of round lots, then accepts 98 deliveries Of those issues, and we ship the securities to him, then he has invested. He has no more money to invest now, although he has the three stock certificates. He is off our books and no longer a customer of Merrill Lynch. It is only when he gets his next $5,000 or $10,000 to invest that he will come back to us. We don't know whether we lost that customer or not. That is my difficulty with these net figures.l It should also be noted that a universal sample Of brokerage customers selected from brokerage firms would require that the firms themselves be sampled. It further assumes that the firms will be willing to cooperate. With all of these uncertainties about the quality of a sample obtained from brokerage houses, it was decided to purchase a portion of the subscription list to Barron's. Although Barron's is a widely circulated and well respected weekly newspaper, it is Obvious that any sample selected therefrom is subject tO a bias. But as already indicated, there exists no feasible means of Obtaining a completely unbiased sample. It was then necessary to select an appropriate sample size. Under the assumptions that an error factor of e==.10 is to be maintained on the confidence intervals 1Testimony by Donald T. Regan, Chairman of the Board, Merrill Lynch & Co., Inc., before the Financial Market Subcommittee of the Senate Finance Committee, July 24, 1973, as reported in Some Concrete PrOposals for the Individ- ual, the Institution, and the Markepplace (New York: Merrill Lynch & Co., 1973), p. 8. 99 Of the binomial questions, the worst possible case exists where p and q both equal .5, and Z is equal to 1.96, then from the formula it may be approximated that 96 valid responses would be necessary to assure a 95 percent level of confidence. It should also be noted by inspection of the ques- tionnaire that not all of the questions yield a response from every customer. Since not all brokerage customers make use Of every service, it was estimated that only 40 percent of the respondents would respond to some of the questions. Thus approximately 250 responses would be required to main- tain the same level of significance. Since a 50 percent return on the questionnaire was assumed, it was decided to use a total sample size of 500. Survey Procedure When considering survey procedures, it was a major purpose Of the author to Obtain an accurate response from the questionnaire. This was accomplished through the expenditure of additional effort to make the questionnaire more professional looking and appealing to the potential respondent. 100 Appendix A contains a copy Of the cover letter which accompanied each questionnaire. Although the body of the letter was printed on a high quality bond paper, each name and address in the sample was individually typed. Each letter was hand signed. Rather than use address labels, it was also decided to individually address each envelope. A return envelope, addressed to the author, was included with each survey and cover letter. To add another personal touch, both the internal and external envelopes were hand stamped . Although this questionnaire matches all of the parameters of the SCRAM model, it was still necessary to perform a careful pretest. Because of the necessity for absolute clarity, this action was accomplished in two steps: After clarifying a preliminary draft as much as possible, a non-random sample of 30 brokerage customers was selected as trial recipients of the survey. Seventeen Of the 30 persons responded. Minor corrections were made, and a second draft was constructed for the purpose of making another test. The second test was sent to 50 people selected at random from the sample list received from Barron's, which yielded a response of 23 usable questionnaires. In order that a follow-up letter might also be pretested, a follow- up questionnaire was sent to each non-respondent. This 101 action yielded an additional 7 respondents. The results of these tests pointed out the need for several more minor corrections, as well as serving as a final check on the actual collectability of the data. It was fortunate that the list Of customers Obtained from Barron's came in the form of a scientifically random sample as generated by a random number selector in the computer of the circulation department. The questionnaire was prepared and sent to the 500 persons in the sample on November 1, 1974. Although the receipt of 130 usable re— sponses was good for a mail survey, a follow-up letter was sent to each non-respondent about 15 days after the non- receipt of the first questionnaire. Appendix A contains a copy Of the follow—up letter. For convenience an addi- tional questionnaire and stamped return envelope was again included in the mailing. This procedure yielded an addi- tional 65 usable responses, which brought the total response to the survey to 195. Table 4.1 presents a summary of all valid and invalid responses. Survey of Non-Respondents In order to test for the presence of bias outside Of the foregoing responses, it was deemed essential that a portion of the 221 non—respondents be sampled in order that their similarities or differences to the main 102 Table 4.1 Survey Response Valid Responses . . . . . . . . . . . . . . . 195 Invalid Responses Invalid addresses . . . . . . . . 26 Addressee deceased . . . . . . . 10 Respondent disqualified self . . 30 Incorrect or incomplete . . . . . 18 Non-respondents . . . . . . . . . 221 Total invalid responses . . . . . . . . . 305 Total salnple O O O O O O I O O O O O O O O O 500 population might be ascertained. The procedure involved an attempt to place a telephone call to the non-respondents, whereby they were simply asked if they recalled receiving the survey and if they would be willing to answer ques- tion 1 (services used) over the phone. In order that the scientific accuracy of the result might be preserved, these persons were selected with a random digit table interfaced against the list Of non-respondents. Table 4.2 details the complete results of an attempt to contact 20 (or 10%) of those persons from whom no response was received and neither the primary nor follow-up questionnaire was returned by the post office because of an invalid address. It is noteworthy that only one of the 20 people in the table turned out to be an active trader. He stated that other pressures on his time had precluded his response. A second individual simply did not wish to comment on the questionnaire and did not care to discuss any aspect of the study. A third person was reported to be out of town 103 Table 4.2 Survey of Non-Respondents Respondent trades; forgot questionnaire . . . . Respondent had no comment . . . . . . . . . Respondent not available . . . . . . . . . . Respondent not active in market . . . . . . . . NO listing by telephone company . . . . . . . . Unlisted telephone . . . . . . . . . . . . . . Poor address . . . . . . . . . . . . . . . . . No answer . . . . . . . . . . . . . . . . . . . Total Sample . . . . . . . . . . . . . . . k) oLer»:a.ernI~sard by his family. Five other persons stated that they had not returned the questionnaire because they were not currently trading in the stock market. The remaining twelve Of the non-respondents were not contacted, but the failure to make contact with them perhaps provides a clue to the reasons for their failure to reply. For instance, four of the persons had no telephone listing at all, and a fifth person had an unlisted phone number. Three of the addresses were judged to be poor in that two were the addresses of companies rather than people. Another address possessed the name of a town that did not exist, according to the telephone company. Finally, four of the non-respondents were not home, despite repeated attempts to reach them. Based on the foregoing 10 percent sample of the non- respondents, it may therefore be inferred that 25 percent of the other non-respondents failed to answer because of a lack Of current interest in the stock market. Perhaps another 40 percent failed to respond because Of the questionnaire being 104 sent to a dead or inaccurate address. The remaining 45 percent or so of the non-reSpondents may have the potential to possess a slight bias to the study, but the cost of obtaining responses from this group did not seem to be justified in view of this slight probability of error. Tabulation Because of the extensive number of responses contained in this study, it was decided to use the computer for statistical tabulation. Specifically, each of the first seven items of the questionnaire was processed by the CODE- BOOK Option of the Statistical Package for Social Science, a standard program used by many researchers in the social sciences. In order to facilitate further computations in Chapter V, the computer was instructed to tally Observations, generate histograms, means, frequencies, standard deviations, standard errors, and ranges for each set of responses. Computer Programming By its nature, the SCRAM model does not require a complex tabulation system. But the extensive number of responses contained in the test of the model again deemed the use of the computer a necessity. Appendix D sets forth two Fortran IV programs which operationalize the SCRAM model when both programs are sequentially run. 105 The first program, entitled PREP, is especially designed to prepare the data Obtained from the survey for processing. The program contains a compression subroutine which automatically generates a list Of all possible con- figurations. For seven services, there are 127 possible configurations. The PREP program also computes an average price for all services (Step 1 in SCRAM) and formulates a price/configuration list (Step 2). Finally, the survey data are stored for further processing. The second program, entitled CRAN, picks up where PREP leaves off. This program requires the input of a price tolerance coefficient and the fixed and variable costs of each of the seven services. It then performs all of the remaining functions of the SCRAM model as detailed in Figure 4.2. The program is automatically designed to produce all four of the previously described outputs. Appendix C presents a portion of the actual output from several runs Of the program. Rather than list only the most profitable combinations, the CRAN program presents a ranking Of the ten most profitable configurations for each of the four outputs. Since models which yield outputs require inputs, attention will now be directed to the results of the survey which will provide the input for testing the computerized version of the model. CHAPTER V FINDINGS In this chapter, the results of the survey are presented along with an appropriate set of statistical measures. The hypotheses presented in Chapter I are com- pared against the survey result, along with the statistical and empirical reasons for their acceptance or rejection. Finally, the results of applying the SCRAM model to the data base is presented. Survey Results The results of the survey are based on the 195 valid responses received as of December 9, 1974. A degree Of judgment was used to exclude those questionnaires which contained insufficient information to allow their response to be tabulated. For instance, ten of the respondents failed to answer the second page of the questionnaire. As stated in Table 4.1, 18 responses of this nature were not included in the data base. It may be recalled from Chapter IV that the ques- tionnaire was tabulated by use of the CODEBOOK option of the Statistical Package for Social Science library program. 106 107 In order to simplify tabulation, questions 1 and 3 were treated as binomials. Since the instructions for both questions called for placing a check mark in the response blank for those services used and those services insisted upon, it was deemed appropriate to assume that a blank response was a negative and a check was a positive. Table 5.1 details the results of the customer response to services used or potentially desired. It may be noted that response rate has been converted to a percentage based on the 195 total respondents. A 95 percent confidence interval is presented which provides an indication to the quality of the data. Table 5.1 Services Used Number 95% Confidence Service of Users Percent SD Interval Q9 Market contact 144 73.8 .032 67. 5 < 1! < 80. 1 Odd lots 128 65.6 .034 58.9<1r<72.3 Office contact 122 62.6 .035 55.7 < N < 69. 5 Security custody 117 60.0 .035 53.l: mommm>< “mucoocommwu mmH so owmmnv eo.mvuveuvom.> mm.H mm.ma mm.oa mcoauOMmcouu no nonesz cmmz mo Hm>umucH mm om msam> moa>umm mocoofimcoo wmm new: oncommom coauocmcmue m.m manna 2.3 v n v 3.5 m~.m omen 3.3 E. uomucoo vexed: mm .mm v n v wmém 006. m~.~N 3.5 mm uomucoo muwmmo en.mm v n v 3.3 SN 8.x.” 22mm om 832585 nonmmmmm mm .3. v a V cm .mv mmé 3.3 v~.mo hm ucmEmoecms Onomuuom “mOOH>umm Hmscc< avéavnvmqoa ~04 no.3 Elma v: muOH coo 31m." v a v mm.m :4 mat: ma .3. am mucouo Seameommm 2.3 v :v 8...“ :4 85 mod me 8326 sueéumm ”mmow>umm cowuommcmua 3v A3 IS AS new: no Hm>uoucH mm mm osam> mucopcommom moa>umm occupaucou «mm can: mo Honssz uncommom 09Hm> umaaoo m.m manna 113 It should be noted that the means specified in Table 5.6 may overgeneralize the true nature of the pop- ulation. For instance, 67 (or 34.4%) of the respondents reported making approximately two transactions per year. Furthermore, 33 (or 16.9%) and 37 (19.0%) of the respondents reported aVerage transactions of $1,000 and $2,000, respec- tively. In order to clarify this response, Tables B-9 and B-lO of Appendix B present a complete tabulation of the responses to these questions. Several notes Of caution are worth reviewing at this point. First, it must be recalled that the responses to this survey are perceptual in nature, especially in the case where customers were asked to project what prices they would be hypothetically willing to pay for various services. When the customer must actually make the decision of whether to pay a given price for a service or lose it altogether, it is possible that a price much higher than his preference response could be charged without a loss Of patronage. Second, customer preferences are subject tO change. Their interest in the stock market waxes and wanes. They may vary their market activity with their amount of leisure time, current income, and past and present success. Third, customer responses are in part a product of the current environment. Especially in times of poor stock market conditions, fewer services may be requested or 114 insisted upon. If the market were to reverse itself, the results of a duplicate survey might appear very different. However brokerage firms are now painfully aware that their price/service configurations cannot be geared only for favorable markets. Requests for Additional Services One of the most interesting set of responses to the questionnaire came from the open-end question on the last page of the survey which asked the customers to state which other services they would like to receive. It was hoped that this question might yield some useful suggestions for potential new services. Many of the respondents to the first portion of the questionnaire left this question blank, put a dash through it, or simply stated "none." However, 79 respondents (40.5%) did take the Opportunity to suggest additional services or complain about the present service configuration. The range of suggestions for additional services was as wide as the imaginations of the respondents. For instance, one customer requested periodic information on Oil lease filing. Other more realistic requests included a desire for information about real estate investment, stock futures, and "foreign investment opportunities." Three customers wished to purchase gold through their broker. 115 Two customers suggested the addition of a tax and estate planning service. One respondent requested a series of investment lectures by a qualified investment counselor. Another group Of customers was interested in receiving more information about stocks and bonds from the brokerage house. Two high-volume customers requested more information about bonds and high-yield government securities. A similar customer wanted more information about utilities. Another customer wished to receive daily information about institutional buy and sell orders. Two customers felt that they would like to have their monthly statement annotated with additional information about recent price changes and news items related to stocks held in their accounts. Two customers were interested in more information concerning the options market. Finally, two customers requested some form of an after-hours quotation system. A group of eight customers stated a desire for the brokerage firm to function as a secondary information dis- seminator. Two of these respondents wanted their local brokerage firm to provide a library of advisory services, investment periodicals, and other printed materials. Several customers wanted the firm to send some of these materials directly to them. In particular, a desire was expressed by two respondents to receive a monthly copy of the Standard and Poor's Stock Guide. Several other 116 respondents suggested an in-house monthly or semi—monthly research letter. Suggestions for Improved Broker Performance Other brokerage customers were concerned with factors related to improvement in the performance of the brokerage function and used the open-end question to present their views. For instance, eight customers made reference to a desire for some form of a quarterly or semiannual port- folio review by their broker. The results of this review would be to provide a set of buy-sell-hold recommendations and, at least for one of these customers, a set Of reasons for the proposed actions. Two other respondents reported frustration with brokerage clearing procedures. One re— spondent was concerned with the time it took to obtain certificates once the securities were purchased. The second respondent was concerned with the length of time between the sale of a security and the receipt of payment. One respondent wrote a pleading letter suggesting that brokerage firms should become diversified investment counselors. Thirteen customers responded that no additional services were necessary, but added reasons. Eight of the customers felt themselves to be more qualified to perform all other functions themselves. Five others felt that the 117 brokerage house should execute orders only, and that no other services are necessary. At the other end of the spectrum were those customers who felt that they had not received enough attention from their brokers. At least four respondents made reference to a desire for the broker to call when he feels a stock should be sold or when the market appears ready to turn downward. Although three of these respondents were small, a fourth was a heavy trader. A more prolific customer was concerned with reshap— ing the nature of the brokerage function altogether. He suggested that: Brokers supply too much unnecessary information on stocks and most of it is not up to date. I would like all of this cut down to an office operated by one person. When a person wants information on a stock the broker should have a complete record . . . and supply the infor- mation to the caller for a fee of $.25 charged to sale of stock unless the broker knows the caller is a purchaser and not one seeking cheap information. In a similar light were the responses Of eight other customers who made pleas for "reduced fees," discount brokerage, and "competitive rates just like the rest of us in business have." One of these customers further added that he would not reenter the market until brokerage rates are reduced to their 1968 levels. 118 A total of fourteen respondents reported varying forms of distrust or lack Of confidence in the brokers themselves. As one customer stated: I'm convinced most brokers are inept and self serving in their recommendations. I've talked to a lot of other small investors ($5,000 to $20,000 portfolio), and each have had a dis- tasteful loss experience with incompetent salesmen with disastrous recommendations. I'd rather be fully responsible for my own losses than be "set up" by a crooked office recommendation. Three Of the same group of customers were concerned with brokers pushing house underwritings which may or may not be in the customer's best interest. One of these respondents further suggested that the problem might be solved by putting brokers on straight salary. Three other customers simply stated that they would like the services Of more qualified stock brokers. Two respondents accused the brokers of being mere "order takers." Finally, three customers responded with a simple plea for honesty on the part of the brokers and in all phases of the brokerage industry. Such ideals would improve confidence in the brokerage industry. As may be expected in declining markets, several respondents made reference to their present discouragement. Indeed, a portion of the negative comments heretofore pre- sented are certainly a partial function of the losses experienced by many investors at the time of this survey. 119 One customer stated that "market improvement" was more important than any service the broker could provide, and there are certainly few brokers or customers who would not agree with this thought. Perhaps one respondent summarized it best when he suggested that brokerage firms "begin selling crying towels." Cross Company Comparisons Question 9 of the survey requested that each respondent indicate the name of the brokerage firm with which he does most of his business. It was hoped that a large enough number of respondents would be obtained for bigger companies in order that cross company comparison could be made. But 37 of the 195 survey respondents either failed or refused to respond to this question, and the remaining 158 respondents were spread over 75 different companies. The response rate to only four companies was high enough to warrant a report of their findings. Tables 5.7 through 5.10 provide a summary of the survey findings from Merrill Lynch, Pierce, Fenner, & Smith, Dean Witter and Company, Hornblower & Weeks- Hemphill Noyes, and E. F. Hutton and Company. The total number of responses for each of these companies was 36, 14, 11, and 11, respectively. Although these returns are too small to draw statistical comparisons, they do lend 120 .mommucwoumm one mononucmumm cw mumneszm ma.mm om.N Av.¢vv we Am.mhv mm uomucoo umxuoz om.mm mm.m Am.huv OH Am.mmv an uomusoo ouammo nn.ae om.~ Aa.mmv ma Ab.oov em nouamnom -.~e mm.m xm.~e a AH.HH. v ucmSommcms oaaouuuom ov.aa mm.m Av.v¢v ma AN.~nv mm muOH poo mv.HH ma.m m.hmv 0H Av.HvV ma mucouo auamwoomm ms.m oo.m 1m.mmv «a exe.mee mN stormso suensomm Ame mmcommom mow>umm mo umwmcH on: own 0&3 mow>umm new: xcmm new: mucwocommom mucmpcommom mo Honssz mo nonasz “mm I Z HMHOBV :uflEm a .uossom .mouoam .nusun Hawuumz h.m magma 121 .mommucmoumm who mononucoumm ca muonsszm om.- mm.~ Am.eoe m xm.~me ma uumucoo boxer: He.m~ om.~ xv.H~e m Av.HeV oH nonueou monumo oo.m~ sm.~ Am.mmv e Ae.mme «a suspense mm.me oo.m xo.ov o AH.AV. H parameters oeaouuuom oo.ma mm.~ xe.mme e Aa.eme m muoH poo ea.ma om.m Im.eae N xe.mme m mumpuo seemeummm oo.e om.~ xe.mme m exe.aee o” spoumeo sueusumm E , wmcommwm woun>H0m MO umflme—H 0&3 0mD 0&3 00w>uwm new: room can: mucoocommmm mucoocommmm mo nonssz mo Honssz lee n 2 Hence. .08 are aware: coon m.m «Heme 122 .moomucmoumm mum mommevcoumm cw muonsszm oo.mv mm.a Am.e~v m Am.emv o uumueoo serum: oo.mm om.m Am.emv m Am.vme m uomuaou moaewo oo.mm ew.m “v.0mv e Am.mee m nonemmmm oo.o~a mm.m xo.ov o 1H.me a parameters oeaomuuom mm.- om.~ xe.~ev m .m.ame m who” one oo.v~ mm.m Am.e~. m xe.eme v mumeuo spaneommm mm.om so.m x~.mae m exm.mee m seoumso sueueumm Ame uncommon oua>nmm mo umfimcH on: on: 0:3 mu«>umm new: xcmm cmmz mucoocommom mucmocommmm mo umnssz mo umnssz AHA n z fiancee mmzoz Haanmsmmlmxmoz a nozoansuom m.m edema .mmmeucmouom one mononucmuem cw muoneszm 123 oo.vH om.~ xm.n~e m Am.omv oa uomueoo umxumz no.0m m~.m A~.mav N Am.vmv o uomueoo ouammo oo.om om.m Aa.mv H Am.vmv m nonmemom oo.ov oo.H Aa.mv H AH.mV a ucmsmmmcme Oflaouuuom oo.mH ma.~ Am.m¢v m Ah.~>v m mead poo mv.eH mm.~ Im.e~e m Im.eme o mumeuo seameoomm no.m oo.m Aa.mv a “v.0mv v >o0umso augusomm Awe oncommmm ouw>uwm m0 unwmcH 0:3 omD 0:3 mow>uom :00: rear new: mucoocOQmom mucmocommmm m0 nonssz m0 nonesz Add u z fiancee .oo a coupe: .m .m oa.m menus 124 themselves to some tentative observations which may form the basis for future research. As expected, the results of the responses from these leading companies generally parallel the findings obtained from the total sample. However these four groups of cus- tomers tended to be higher users of some Of the services, and were willing to pay more for others. Market contact tended to lead the list of services most used and most insisted upon, as well as highest ranked. Over 90 percent of the respondents in two Of the firms reported use of the service, which infers a superiority Of service quality and value. Portfolio management again tended to be ranked last, with the exception of the one respondent from E. F. Hutton who ranked the service first and insisted upon the offering. It is noteworthy that E. F. Hutton is one Of the few firms which formally offers portfolio management service. Other statistics presented in these statutes generally tend to follow the primary sample with the exception Of the willingness to pay. In general, the respondents from Hornblower & Weeks were willing to pay more than the sample population for most of the services, whereas the customers from E. F. Hutton appeared willing to pay much less for several services. The responses from Dean Witter parallel the primary population, and the Merrill 125 Lynch respondents were generally willing to pay slightly more for four of the services. Consideration Of Hypothesis 1 Hypothesis 1, as it was set forth in Chapter I, states: American brokerage firms are currently Offer- ing a service configuration which is not of optimum value to brokerage customers in the sense that most customers are either (1) en- gaging in transactions without using certain services or (2) unwilling to pay for some of the services they are using. It may be recalled that necessity for testing this hypothesis hinged around the facts that (1) all secondary services are currently included in the brokerage fee sched- ule, and (2) as a result of the SEC action, secondary ser- vices must soon be disaggregated from the primary brokerage service and be individually charged to the customers who use them. A casual look at Table 5.3 reveals that the SEC action, when it takes place, will have a profound impact on the revenue levels Of the industry. Under the current fixed fee schedule, all of these services (except portfolio management) are included in the price of the brokerage service. When the charges for secondary services are disaggregated, many fewer patrons will be paying for these services and the revenue for the industry as a whole will 126 decrease unless customer preferences change or other new services are provided. In general, the usage of these services by brokerage customers appears fairly high as demonstrated by Table 5.1. By contrast, their willingness to actually pay for several of these services appears quite low. Perhaps the most significant discrepancy appears in the market contact service, where 73.8 percent of the respondents use the service and 42.0 percent insist upon its presence. It also received the highest mean preference rank of 2.33 (see Table 5.3), but only 39.5 percent of the respondents were willing to assign any dollar value to it at all when asked how much they would be willing to pay for the service. It may therefore be concluded that this ser- vice is not one to be considered Of value by most brokerage customers, according to the evaluation criteria established by the hypothesis. A somewhat different situation exists with the service of office contact. A willingness to pay for the service was indicated by 28.2 percent of the customers, and it was insisted upon by 27.7 percent. But it was reported used by 62.6 percent. Even though the local Office is visited by many customers, it would appear that they do not attach very much value to their ability to do so. Furthermore, the removal of the local office would not 127 sufficiently Offend more than half of the users of the service enough to cause them to change firms or otherwise discontinue patronage. According to the evaluation criteria, a serious question must be raised as to whether this service is of value to most brokerage customers. The service most insisted upon was odd lots. Con- currently, it was also the service for which more patrons (58.5%) were willing to pay than any other service. It was noted that 65.6 percent of the customers make use of the service. It may therefore be concluded that this ser— vice is considered to be of value to a major portion of the brokerage customers. It may be recalled that portfolio management service was included in this study as a service which is ngp currently offered by most brokerage houses. The re- sponse from Table 5.1 indicates that 11.3 percent of the respondents currently use or would potentially use the service. These respondents were also willing to pay a dollar fee of $63.24 per year--nearly twice the level of any other service in the survey. Although this service may not be considered to be Of value to most brokerage customers, it appears that a market does exist for at least a portion Of the firms to Offer this service on some basis commensurate with the willingness of customers to pay. 128 The three remaining services Of security custody, research, and specialty orders are all notable for the fact that an almost equal number Of respondents are willing to pay for the services as make use of them. (It may again be concluded that these services are Of value to brokerage customers. Arguments Favoring Rejection of Hypothesis 1 From the previous discussion, it may be noted that four of the six services currently being offered by the brokerage industry are of significant value according to the evaluation criteria established by the hypothesis. It may be further argued that the other two questionable ser- vices Of office and market contact are at least of some value to the patrons or they would not make use of them at the indicated rates of 62.6 percent and 73.8 percent, respectively. When comparing the usage rate of the six services currently being offered against the 11.3 percent rate of the proposed new service Of portfolio management, it may be argued that an insufficient number of patrons expressed a sufficient interest to warrant its addition by the entire industry. Although there may be room for a selected number of firms to begin handling the service, it may not be Of significant value to most brokerage customers. 129 Arguments Favoring Acceptance of Hypothesis 1 The arguments favoring acceptance of the hypothesis correspond to the arguments for rejection. Although four of the services appear to be of value to brokerage custom- ers, two of the most critical services--Office and market contact--do not hold very much value based on the unwill- ingness of the customers to pay for these services. Conclusion Subject to the previously mentioned limitations associated with the results of this study, it should be concluded that the weight of the evidence indicates that Hypothesis 1 should be accepted and that brokerage firms should admit to themselves that a major part of their service Offering requires adjustment. The primary reason for acceptance of the hypothesis is based on the evaluation Of the local Office and broker contact services. The brokerage industry is organized around a series of local Offices and brokers scattered in every population center in the nation. Clearly the alter- ation or elimination of the local Offices and/or market contact with customers will change the character Of the brokerage industry more than the addition or subtraction Of many other services, including the other five tested by this survey. 130 For these reasons, it must be concluded that American brokerage firms are not currently Offering a service configuration which is of Optimum value to brokerage customers. Consideration of Hypothesis 2 Hypothesis 2, as it was set forth in Chapter I, states, An analytical cost/revenue model developed for the determination of optimum price/service mix is superior to accounting methods which consider profit on a service by service basis and do not take into account the interaction of the demands for various services. It may be recalled from Chapter IV that the analytical cost/revenue model as set forth in Figure 4.2 was computerized (see Appendix D) in order to simplify tabulation. It may also be recalled that the model utilizes the input of the aforementioned survey, as well as certain cost data. Cost Assumptions The limitations section of this dissertation described a problem that existed in Obtaining valid cost information and noted that it would be necessary to make estimates of the fixed and variable costs associated with each service. Table 5.11 presents the results of an attempt to estimate these costs. 131 Table 5.11 Cost Assumptions Fixed Variable Service Cost Cost (5) (5) Office contact 3,000 5 Research 1,500 10 Market contact 1,000 50 Portfolio management 1,000 40 Specialty orders 300 5 Odd lots 300 5 Security custody 300 10 Because of the sample sized input being used, it was necessary to assume that the estimated costs must be adjusted to match the size of the sample. This adjustment allows for a fair comparison of the costs associated with the revenues for each service. It should be kept in mind that the output of the model will therefore also be a sample output and reflect the nature of a firm with 195 customers. However, simple multiplication of the output would provide an estimate of the costs, revenues, profits, and potentials for a firm Of a large size. It was estimated by the researcher that the sample of 195 customers would generate a total commission revenue of about $200,000. The researcher further estimated that the seven services under discussion might account for about one-fourth to one-third of total revenue. 132 The reconstruction Of estimated costs was augmented by the use of an income and expense report produced by the New York Stock Exchange. The report quotes average expense ratios (as a percentage of gross revenues) for the 225 mem- bers of the Exchange who execute buy and sell orders on the NYSE trading floor.2 Since the report noted total occupancy expenses to be an average of 7.7 percent of revenue for the reporting firms, it may be judged that about 6 percent con— stitutes the cost of occupying the branch offices. Fixed expenses for the sample population were therefore estab- lished at $3,000 per year, and variable costs were estimated by the researcher to be an additional $5 per year per customer. Fixed costs for research were estimated in a similar manner. The average cost of research for the industry is equal to approximately 3 percent of revenue.3 Therefore, a fixed cost estimate to match the sample would equal $1,500. A subjective variable cost estimate Of $10 per customer was assigned to cover the cost of printing and dissemination of the information. 2Security Commission BusineS§_(New York: The New York Stock Exchange, 1971). This report was difficult to produce because of the incongruity of the accounting systems between firms. The year 1971 was therefore the last year it was published. 3Based on a telephone conversation with Mr. Michael Casella of the New York Stock Exchange, September 26, 1974. 133 By inspection of the aforementioned report, the researcher estimated a fixed cost of $1,000 for market contact service on the assumption that this service Should share a portion of the communication equipment costs for a firm. A variable cost of $50 per customer was subjectively estimated to cover a portion Of the cost of having the broker discuss market conditions with the customer. Portfolio management has only recently been intro- duced by several firms, and it will not be known for some time exactly what the service will cost to Offer in the long run. A fixed cost of $1,000 per year was estimated by the researcher to cover the cost of facilities and new personnel, and a variable cost of $40 per customer was estimated to cover the annual cost inspecting the customer's portfolio and sending him reports. Based on the researcher's estimation, the three remaining services of specialty orders, security custody, and odd lots were all assigned a nominal fixed cost Of $300 each. Specialty orders and odd lots were assigned an addi— tional variable cost of an estimated $5 per transaction to defer paperwork and other costs. Security custody was assigned a variable cost of an estimated $10 per transaction to cover paperwork costs and costs of disseminating annual and quarterly stockholder reports. It should be noted that these costs for security custody assume that there is no .134 advantage to the brokerage firm in holding securities for customers except to provide a service. That is, it assumes that the firms do not borrow against their customer's stocks held in account. Examples Of SCRAM Model Applications The computerized version Service Cost Revenue Analysis model was utilized on the previously described 195 survey responses and cost assumptions. A subjective 25 percent price tolerance coefficient was utilized. The following output was Obtained.“ The maximum potential revenue for this group of customers was calculated to be $10,589.38. The previously described "package" output proved to yield the most profit- able configuration, although it called for offering only the services of odd lots and specialty orders. Total revenue for the configuration was $1,427.13, total cost was $1,180.00 and total profit was a scant $247.13. In view of the lack of profitability in the industry itself, these figures may represent a fairly accurate representation of the real world. A The second most profitable configuration called for Offering the services of security custody, Odd lots, l'A photocopy of the complete computer output may be found in Appendix C, Figures C.1 through C.4. 135 and specialty orders on an individual basis. The total revenue for this configuration was $2,444.14, total cost was $2,230.00, and total profit was $214.14. It should be noted that a shift in assumptions from package to individual offerings resulted in a completely new configuration with a completely different set of costs. Because of the lack of high profitability in the package output, it was not unexpected to find that none of the package "plus" outputs showed a profit. The best configuration utilized the package output of odd lots and specialty orders and added security custody as an individual service to those who desired it. This configuration yielded a net loss of $83.45. It may therefore be concluded that under the foregoing set of assumptions the typical brokerage firm would do well to consider developing an independent service configuration Of security custody, specialty orders, and odd lots. All other services Should be discontinued. Such a conclusion is obviously subject to many limitations. First, the model is intended as an aid for managerial planning and control--not a substitute for managerial judgment. Many other factors must be considered before undertaking such a drastic change. Second, the results of the model output is strongly predicated upon the results of the survey input. If the 136 survey were conducted at almost any other point in time, the results of the output from the model would most probably show some variation. In general, customers would Show a greater willingness to pay for services under better market conditions. Third, the inputs Of costs and the price tolerance coefficient were subjective. If a firm were actually going to make a decision based on the use of this model, it would be necessary to obtain solid cost data. In order to test for the change in optimum config- uration which might result from a change of assumptions, it was decided to rerun the SCRAM model after assigning a hypothetical lesser set of fixed costs. From Table 5.12 it may be noted that the variable costs were left the same, but fixed costs were reduced by about half. Under a reduced set Of costs, it was not surprising to see that the profitability level of the output increased. The list of services within the configuration also increased, as with the package output which called for the offering of odd lots, specialty orders, research, and portfolio manage- ment and yielded a net profit of $1,572.23. A profit of $1,523.06 was Obtained from a configuration Of the above package output plus an independent Offering of security custody. The most profitable independent configuration called for Offering security custody, odd lots, specialty 137 Table 5.12 Second Cost Assumptions Fixed Variable Service Cost Cost ($) (5) Office contact 1,600 5 Research 800 10 Market contact 400 50 '- Portfolio management 400 40 Specialty orders 200 5 M Odd lots 200 5 Security custody 200 10 orders, and research, but the net profit for this output was only $823.16.5 Table 5.13 presents a standard accounting analysis of the results of the survey. It was again assumed that the services would be Offered at a mean average price. A price tolerance Of 25 percent was assumed, and all respondents who indicated usage of a particular service and a willing— ness to pay within 25 percent of the mean average price were considered to be probable customers of the service. By multiplication, the third line of the table computes gross profits for each service. sA photocopy of the complete computer output under this set of cost assumptions may be found in Appendix C, Figures C.5 through C.8. 138 Ammo. Imme.ee .mm mme men are Immme Ame Immoee gauche umz oon.Hu om: 0mm: omml own: cum: omen Amy mumoo mannawm> oovn ooo.H| oomu ooen com: com: com: Ame mumoo omxwm "coflumEdmmm umoo ocooom eowu mumoo "mmmq Imem.ae Imam.~e lemme Amoco mom .mm Immme Ame Impose venous umz oon.an can own: 0mm: can: own: omen Ame mumoo manoeum> ooo.an ooo.m| oom.al ooo.an com: com: com: Awe mumoo omxam "coaumssmme umoo umwmm scum mumou "mama mmH.H «me me~.H mmv.a mam mme mam Ame undone mmouu em me mm mm on em me mumsoumeu mo umnsez mm.mm mm.h~ H>.Nm vm.mm mv.~a wa.~a mo.m Ame wowwm mcawmmmo pomucoo nomucoo codumsuomcH paramount: muoq mwmowo Snowman vogue: mowmmo soumomom OAHOmuuom poo apamaoomm unannomm memsamea meeuasouoe ma.m manna 139 The next two sets of computations subtract the respective fixed and variable costs for each set of cost assumptions. The resulting net profits and losses indicate a pattern for determining which services should be offered utilizing standard accounting procedures. For the most r profitable configurations, the two results agree. But for the secondary results, the service of security custody appears in several configurations, although Table 5.13 shows this service to be a substantial money loser under ' both sets of cost assumptions. Conversely, the service of portfolio management is profitable under the second set of costs, but it is not included in the configuration output for independent services. Herein lies one of the major points of this dis- sertation: simple accounting procedures may fail to take into account the customer's perception of the entire service package. The reason for including the unprofitable service of security custody in certain configuration offerings is that some customers will not purchase the configuration at all unless security custody is included. Therefore the total revenue generated by offering this service in some configurations may far exceed the cost of offering it. Conversely, the service Of portfolio management which proves to be profitable in Table 15.13 may be excluded from some configurations because of certain customers who may have 140 requested the service and also insisted upon other services that were not offered, resulting in a loss of the patronage of those customers. Therefore portfolio management offered in an independent configuration would have actually gener- ated a loss under the second set of cost assumptions, despite its apparent profitability in Table 15.13. Conclusion From the foregoing discussion, it may be noted that certain accounting methods may fail to consider the inter— action Of demands for various services. The use of such methods may overlook other more profitable alternatives which, when considered separately, may appear to be un- profitable. When combined with other services, it was shown above that certain unprofitable services may generate revenue which exceeds their cost, as illustrated by the SCRAM model. i For these reasons, the evidence favors the acceptance of Hypothesis 2. CHAPTER VI SUMMARY, CONCLUSIONS, AND RECOMMENDATIONS This chapter summarizes the dissertation and states several conclusions resulting from analysis of the data. Recommendations are made to the brokerage industry con- cerning certain actions which should be taken, and other possible uses and extensions of the Service Cost Revenue Analysis Model are discussed. Summary This dissertation began by describing the prof- itability problem of the American brokerage industry. In addition to these problems, an action by the Securities and Exchange Commission has dictated that brokerage fees, which are currently fixed in most categories by an industry-wide rate scale, must now become competitive. In addition, the firms must begin making separate charges for special ser- vices beyond order execution--services that were previously available at no extra charge. Through the use Of the four P's managerial marketing model, an analysis is made of the current marketing strategy 141 142 of the industry. Factors discussed included the industry's use Of advertising, the functions of the broker as a sales- man, the location of brokerage offices, the dilemma of pricing services in light of the SEC action, and the nature Of the services provided by most brokerage houses. The elements of price and service were selected for further analysis. In order to form a basis for the further exploration of prices and services, an examination is made Of the cur- rent state Of the art in economic price theory as it relates to the pricing of groups of goods and services. A short discussion is made Of the physical distribution concept of level of service. A description is made of the nature of marketing in the service industries. Finally, the theory of cost/revenue analysis was presented as a potential means for solving the problem of price/service mix determination in the brokerage industry. The dissertation then develops an analytical model for the optimization of price/service mix. The model is designed to consider the costs, revenues, and profits associated with all possible combinations of services, and to consider the interactive effects of service offerings. The model outputs the four most profitable configurations of services under the assumptions that (1) all services may be purchased independently, (2) all services must be purchased together 143 together in a package, and (3) all services not offered in the package may be added on. A final output yields the maximum potential revenue obtainable if all customers were served and all services were Offered. Finally, the model is tested by using the input Of a survey of the customers of American brokerage firms. Based on certain assumptions about the costs associated with seven key brokerage services, an optimum price/service configuration is recommended. The survey input itself is also analyzed for the purpose of providing useful informa— tion about customer attitudes toward brokerage services. Recommendations for the Brokerage Industry Based on the results of the survey presented in this dissertation, a serious question must be raised as to whether all firms in the brokerage industry should continue to Operate branch offices in local communities. Besides having saturated many markets, it appears that brokerage customers may not consider the presence Of local Offices of high enough value to continue to justify their existence for all firms. It is therefore recommended that those firms in the industry which are interested in improving their efficiency and effectiveness should survey the attitudes of the customers of their various branch offices and, either through the use of the model presented in this dissertation .‘ A. 144 or by some other means, decide upon which offices should be closed. The closing of various branch offices does not mean that these customers should be forgotten, but rather that they should be offered a reduced service package by the same firm through, for instance, the use of telephone P: contact by an 800 WATTS line number. This recommendation relates to the normative nature of the results of survey as described in Chapter V. It must r be recalled that these results and the output of the model b reflect a cross-sectional evaluation Of the brokerage population. This result should not preclude but perhaps enhance a program of market segmentation on the part of industry. The unbundling of services, if enforced by the SEC, will Obviously reduce the revenue of the industry as a whole, as may be noted in Figure 5.3. It will simply not be possible or desirable for all firms to Offer all services to all customers. As indicated by the preliminary results Of the individual company responses presented in Tables 5.7 through 5.10, a variation in preferences may exist between the customers of different firms. It may be wise for the firms to concentrate on those services felt to be most profitable by their customers, while at the same time attempting to search for other profitable services (such as portfolio management) which are not Offered by most other firms. 145 A second recommendation for the industry relates to performance of the brokerage function itself. At least some evidence has been presented to suggest that a revised system of compensating brokers may be necessary. Based on the response to the Open-end question discussed in Chapter V, it appears that many of the brokers of the future may be forced to take on a more comprehensive role in the service which they provide their individual customers. This implies that salesmen must place less emphasis on trading volume and more emphasis on customer welfare. If certain brokers take on the role of portfolio managers, then they must be com— pensated on some other basis besides transaction volume. If brokers are to be expected to spend additional time with individual customers discussing buy and sell recommendations, then the compensation system must reflect his expanded role. Indeed, it may be in the best interest of many firms to consider salary and bonus or straight salary compensation. This action, when combined with better trained brokers, may make the service of the broker appear more valuable in the eyes of the brokerage customer and eliminate the need for discontinuing this service of office contact. To those firms who find it necessary to change their service configuration as a result of the SEC action, a final recommendation relates to the degree to which ser- vices should be offered independently. Although the SEC 146 has directed that some of the services must be priced and offered independently and that brokerage rates must become competitive, this does not dictate that all services must be independently offered. As indicated by the various outputs of the SCRAM model presented in the previous chapter, it may indeed be more profitable to leave some of the secondary services in a package with the primary service and Offer others independently. Indeed, the decision to Offer and charge for some of the services independently may so anger certain customers that they will discontinue their patronage of the firm. Of course all decisions to package or unpackage services must relate to the current rulings of the SEC. Other Possible Model Tests One of the most difficult problems in the utiliza- tion of cost/revenue analysis as a tool for the planning and control of any phase of marketing is the problem of segmenting the costs and revenues in such a manner that they can be correspondingly matched and compared. Once this task is accomplished, it is possible to perform a variety of tests on the data, as discussed in Chapter III. It may be recalled that several of the most common break- downs were by product, customer, sales territory, and order size. 147 Since this dissertation was concerned with the brokerage industry, the primary segmentation of costs and revenues was by the services Offered by most brokerage firms. Since the model called for an output of the most favorable price/service configurations, the results of this disaggregation may be found in Figures C.1 through C.4 of Appendix C. An attempt to make a secondary breakdown of the data by individual companies was discussed in the previous chapter. The results of breaking down the survey findings of four companies was presented in Tables 5.7 through 5.10, but a company by company analysis utilizing the model was not used because of the small number of responses. This breakdown nonetheless illustrates how a secondary breakdown of the data could be made. An individual firm may wish to make use of the model for managerial control by making other secondary breakdowns of the data base. In this instance, the survey may be conducted only on the customers of that company. It may then be useful to separate customers by regions, terri- tories, or local offices for the purpose of ascertaining the differences in costs, probable revenues, potentials, and Optimum service configurations between Specific subsegments. With this information in hand, segments which show consider— able variation could be given additional managerial attention. 148 It may prove profitable to phase out certain services (like Office contact) from certain highly competitive territories, but retain the same service in others. With other services which Show variations in profitability, it may be possible to explore the reasons for the variation. For instance, k the quality level of the services may vary between the ; . offices or territories because poor personnel training, lack of facilities or equipment, or other problems which u might be going unnoticed. I Besides control, another purpose for testing addi- tional breakdowns is for the value of this information in planning for market segmentation and hence target marketing. Assuming that the company is utilizing a cross-sectional survey Of the entire population of brokerage customers, similar secondary breakdowns of the data besides the pri- mary breakdown by services could be made to provide useful information for the entry into a new market. It is essen— tial that a target market be defined before a marketing mix can be determined. By observing the data contained in this dissertation, it can be seen that it would be a simple procedure to sub- divide the respondents by the annual number of transactions they make or the average dollar value of their transactions. The marketing mix could then be adjusted to meet a particu— larly profitable market segment. In a similar manner, the 149 survey results from a city where the firm is not currently located could be compared, by use Of the model, with the results from current locations, and the potentiality of entering that market could be assessed. A final use of the SCRAM model relates to the fact rt that the dissertation chose only to test seven services. It would be a simple matter to add other services to the survey (such as those suggested by the respondents in the ‘1?:"-;; r- , Open-ended question) in order that the potential for offering other configurations might be tested. Possible Extensions of SCRAMpModel At the onset Of this dissertation, it was stated that the analytical model to be presented should be of universal applicability for the determination of price/ service mix in many service industries. In order to make the model more useful to all service industries, including the brokerage industry, it may prove useful to make certain changes to the model itself. It may be recalled that price determination in the model was based on average prices. The disadvantage of this somewhat arbitrary determination of price is that the resulting revenues (and hence profits) may potentially be higher if prices were set at various other levels. It might therefore be useful to include a subsegment in the model 150 which would test other prices above and below the computed averages. In accordance with the theory of cost/revenue analysis, the overall revenue gained by offering certain services at lower prices may exceed the revenues lost by lowering the prices. Similarly, higher prices in some services may add to overall revenue. The addition of price variation to the model would include the economist's concept of price elasticity, which would greatly improve the model's performance. The addition of other factors could also make the model more useful. For instance, it may be possible to include subsegments for ascertaining the impact of other factors in the marketing mix, such as advertising. Coef— ficients could be computed for customer evaluations of the effectiveness of certain services, which could in turn be compared with the efficiency of offering those services. All of these factors would require additional survey input, as well as much more theoretical consideration, testing, and analysis prior to their inclusion. The reliability of the model would be greatly improved if the data base were solidified through the use of a more effective survey of price preferences and costs. The survey method used in this dissertation was adopted for use in a mail questionnaire and therefore lacks the sophis- tication of technique which might be used in some other 151 survey method. For instance, if the data had been collected by personal interviews, it may have been possible to use questions which probe deeper into the customer's evaluation of this willingness to pay. Limitations It is in order at this time tO reiterate the limitations Of this dissertation. First, this study is preliminary rather than conclusive in nature. It is not intended that any of the firms in the industry will take any direct action based on the results of this survey, but rather that they will use the foregoing results as a base for further research on which action can be taken. Second, it must be recalled that the survey con- tained herein was conducted during a period of extreme market decline, and the negative attitudes of the respond- ents toward brokerage firms in general are reflected in the results. If the survey were conducted in a riSing market or at any other point in time, the results might appear quite different. Third, the survey contained in this study calls for subjective evaluations by the questionnaire respondents. They are requested to estimate the prices they migpp be willing to pay for various services. Whether they will actually pay more or less than the stated price when the services are individually offered is unknown. 152 Fourth, it must be remembered that the cost data used in the application of the model was subjectively estimated. Any conclusions which might be drawn from the inspection of the model output presented in Chapter V are therefore mere estimates of the results which might be obtained if more reliable cost data were available. i Fifth, the model presented in Figure 4.2 is rigid rather than flexible. If the model were modified, it would be possible to test a spectrum of price assumptions in order to determine the best possible prices which might be used in relation to the best service configuration. Finally, it must be remembered that the sample used in the survey was selected from the subscription list to Barron's. The results of the survey only relate the characteristics of this population, which may or may not be representative of the universe of all brokerage customers . Final Conclusions NO study as modest in scope as this one can hope to fully reshape either the brokerage industry, the theory of cost/revenue analysis, or the theory of price/service mix determination. Such was not the purpose. Instead, this dissertation proposed to lay one more brick in the wall of knowledge composing these subjects. 153 It is hoped that a basis has been developed for further research by the brokerage industry into the significant points raised concerning brokerage firm service Offerings. Because of the preliminary nature of the research, it would be improper for any firm to under- take any Of the proposed actions without much more extensive research and consideration. It would be similarly incorrect to assume that the computerized cost/revenue model presented for the optimization of price/service mix is by any means fully refined. Models of this nature require much more testing, analysis, and revision before they can be considered ready for general use by the service industries. Indeed, this study has applied the model to only one industry and for only seven services. The future will find new applications. Finally, it is hoped that the subject of cost/ revenue analysis which formed the theoretical base for the analytical model has been augmented. The theory of cost/revenue analysis holds the key to improved planning and control of many aspects of marketing. This dissertation has presented only one such application. APPENDICES APPENDIX A SURVEY AND COVER LETTERS This appendix contains a sample of the cover letters used in the primary and follow-up mailings, as well as a copy of the questionnaire. In the actual mailings, both letters were individually signed by the author. As described in Chapter IV, the letters used in the primary mailing were individually addressed. 154 155 MICHIGAN STATE UNIVERSITY GRADUATE SCHOOL OF BUSINESS ADMINISTRATION EAST LANSING ° MICHIGAN ' 48824 DEPARTMENT OF MARKETING AND TRANSPORTATION ADMINISTRATION November I, I974 XXXXXXXXXXXXX XXXXXXXXXXXXX XXXXXXXXXXXXX Dear Mr. XXXXXXX: As you may already be aware, the Securities and Exchange Commission has recently stated that brokerage charges should no longer be fixed in the form of a uniform rate schedule. In the past, the payment of brokerage fees entitled the brokerage customer to a series Of additional services at no extra charge. As a result of the SEC action, brokerage firms will soon begin charging less for their basic order processing service. Extra charges will be made for special services such as printed information and special orders-- services that were previously available at no extra charge. As part of the necessary research in my doctoral program, I am conducting a survey of the impact Of the SEC action on brokerage customers. You have been selected as part of a scientific random sample of all brokerage customers in the U.S., therefore your reply to the enclosed questionnaire is essential to the success of this study. For your convenience, a return envelope is enclosed. To protect the confidentiality of your reply, please do not sign the questionnaire <3r list a return address. Although the results of the survey will be published, there will be no reference made to specific individuals or specific replies. It is through basic research such as this that policy decisions are reached. I thank you for your cooperation in this attempt to advance academic knowledge. Sincerely, Brian G. Long Project Director 156 MICHIGAN STATE UNIVERSITY GRADUATE SCHOOL OF BUSINESS ADMINISTRATION EAST LANSING ' MICHIGAN ' 48824 DEPARTMENT OF MARKETING AND TRANSPORTATION ADMINISTRATION Novenber l5, I974 Dear Brokerage Customer: You will recall that several days ago you received a survey concerning your perceptions of the service offerings of brokerage firms. If you have already returned this form, I thank you for your reply. But if you have not filled out the survey, I again beg your cooperation. As stated in my previous letter, you were selected as part of a scientific random sample, and each reply is essential to the success of the study. Once again, be assured that all replies are confidential. My only purpose is to advance knowledge in an industry which meeds help. For your convenience, an additional copy of the survey and a retrun envelope is enclosed. I again thank you for your cooperation. Sincerely, Brian G. Long 157 As a brokerage customer you receive the service of having your order acted upon by an exchange or other securities market. In addition your use of the brokerage firm entitles you to a number of services which are simply included in the firm's service package. The following ls a list of seven of the most important extra services provided by most firms: Service A. PORTFOLIO MANAGEMENT (Allows control of your account to be delegated to a knowledgable money manager who initiates buy and sell orders in accordance with invest- ment objectives speci- fied by you) Service 8. SECURITY (Allows stock certificates to be CUSTODY held in account by the brokerage firm for convenience and safe keeping) Service C. RESEARCH (The collection of stock recommendations provided on a regular basis by most firms) Service D. OFFICE CONTACT (The ability to visit the local office of your brokerage firm to see your broker, deliver securities, or watch the ticker) Service E. MARKET CONTACT (Discussion of current ’ market conditions, trends, and potential trading actions with a knowledgeable broker or other person) Service F. SPECIALTY ORDERS (Limit loss, short, and other similar orders) . Service G. ODD LOTS (Purchase and sale of quantities less than lOO shares) o O 0.: n Which of these services do you currently use or potentially plan Ig_use in the future? Please\/" Of those services which you are currently using or intend to use, please indicate the rank order of importance you would attach to each. (Please number: T:?,3 for first, second, third, etc.) Now look again at your ranking of services. Which of these services do you consider to be absolute "musts," l.e., which services, if ngi_offered would cause you to discontinue your patronage 9j_your resent firm and seek another firm of higher service? Please 158 4 The normal brokerage fee for a $4000 transaction is about $70 at current brokerage rates. Assuming that the firm with which you are now doing business begins offering services on an "a-ia-carte" basis. Only one primary service will be offered-- order execution. All other services will now be available on an individual subscription basis. Also assume that you are making a series of $4000 transactions and that the charge for the primary service-- order execution-- is now available at half the former rate, i.e., $35. it is obvious that a separate charge must now be made for the services which you have ranked on the previous page. Please indicate how much you would be willing to pay per transaction for each of the following services. (Remember: A $4000 transaction is being assumed) Pleaseur— over 30 $0 2 4 '6 8 l0 l2 l4 l6 I8 20 22 24 26 28 specify SECURITY CUSTODY SPECIALTY ORDERS ODD LOTS ........ Because of their continuous nature, some services will not lend themselves to individual billings and would require payment on some form of a monthly or annual basis. Taking into account the size and nature of your present portfolio, please indicate how much you would be willing to pay per year for the following services. PleaseV” over |20 $0 l0 20 30 4O 50 60 70 80 90 |00 ll0 I20 specify PORTFOLIO MGT... RESEARCH INFO... OFFICE CONTACT.. MARKET CONTACT.. Please check to box indicating the number of transactions (buy or sell orders) which you made last year. Pleasey/" |_y0 [I-4 I 5-8 [ 9-12[I3-16|I7-20|2I-24[25-28129-32133-36I0ver 36] I .I' l | J I I J J I I 2.] (Specify) What would you estimate to be the average dollar value of your transactions? PleaseL/" $500 I $1000I $2000I $3000 I $4000I $5000 I $6000I $7000 I | IL I I I I J [T I I _ $8000 I $9000Iy$lOOOO I$Ii000 [$I2000 I$I3000 I$i4000| Other I l I_ I l I l I I I II _(Specify) 159 With which brokera e firm do you presently do most of your business? Please<%j’ . Bache & Company Hornblower & Weeks-Hemphill Noyes E. F. Hutton & Company Loeb Rhoades & Company Manley Bennett McDonald & Company McDonald & Company Merrill Lynch Pierce Fenner & Smith Paine Webber Jackson & Curtis Roney & Company Smith Hague & Company Watling Lerchen & Company Dean Witter & Company Reynolds Securities Donaldson, Lufkin & Jenrette Other (please Specify) [:1 One final question: Which other services would you like your broker to offer? (Please Describe) APPENDIX B SUPPLEMENTAL SURVEY RESULTS This appendix presents several tables which were too cumbersome to include in the text portion of Chapter V of this dissertation. Table 3.1 refers to question 2 of the questionnaire (see Appendix A); Tables B.2-B.4 refer to question 4; Tables B.5-B.8 refer to question 5; and Tables 8.9 and 3.10 refer to questions 6 and 7, respectively. 160 161. .momwucmoumm mum mommzucmumm CH mumnfidzm Am.mv h Am.mv ma Am.mav vm o.amv av Am.hmv mm Aa.vmv he vomucoo umxumz an.hv ma Ah.mv ha Aw.mav mm Am.mav Hm Am.mav hm Am.mmv on uomusou deflmwo Am.vv m Ao.mv Ha Am.oav mm m.omv oq Am.HHV mm Am.mmv bu cowpmauowsfi suummmmm Ao.as m Ao.as m Aa.ms m AH.~v e Am.av m Aa.mmv moa ucmsmmmcme onHOMuuom AH.vV m hm.oav on Av.vav mm Am.mv ma Am.¢mv mv Am.vmv mm muoH woo AH.vv m A>.mv ma Am.oav Hm an.mv. 5H A~.bv va Am.mmv moH mumvuo auamwummm Am.mv Ha Am.HHv mm Am.mv ma “v.0HV mm Am.mav mm on.o¢V mu wwoumsu muHHSUmm mt mcaxcmm v: mcflxcdm mm mcwxcmm mt mcwxcmm aw mcwxamm mcwxsmm uoz mow>umm mucmvsommmm mo uwnasz mmcommwm mcwxcmm oocmummmum H.m manna 162 Table 5.2 Dollar Response-~Security Custody Dollar Number of Percent of Response Respondents Respondents ($) 2 13 6.7 4 14 7.2 6 16 8.2 8 5 2.6 10 21 10.8 16 l 0.5 20 3 1.5 22 l 0.5 81 1 0.5 Blank 120 61.5 Table 5.3 Dollar Response--Specialty Orders Dollar Number of Percent of Response Respondents Respondents ($) 2 4 6 8 10 12 16 18 20 24 26 28 35 4O 80 Blank H N l-‘ WOHHOOOO‘I—‘NHNwh-Imb 5...- bHNNI—‘I—‘HWNDNU‘O‘O‘O‘Q UU'IOOU'IU'IUIQOI-‘OQI-‘l-‘NO‘ I'-‘ 0 U1 163 Table 8.4 Dollar Response--Odd Lots Dollar Number of Percent of Response Respondents Respondents ($) 2 7 3.6 4 14 7.2 6 15 7.7 8 2 1.0 10 44 22.6 12 3 1.5 14 2 1.0 16 4 2.1 18 1 0.5 20 10 5.1 24 2 1.0 30 2 1.0 35 4 2.1 40 2 1.0 50 1 0.5 77 l 0.5 Blank 81 41.5 Table 8.5 Dollar Response--Portfolio Management Dollar Number of Percent of Response Respondents Respondents ($) 10 8 4.1 20 1 0.5 30 l 0.5 40 4 2.1 50 7 3.6 60 2 1.0 100 5 2.6 120 9 4.6 Blank 158 81.0 164 Table 8.6 Dollar Response--Research Dollar Number of Percent of Response Respondents Respondents ($) 10 35 17.9 20 23 11.8 30 2 1.0 40 2 1.0 50 21 10.8 60 4 2.1 80 1 0.5 100 7 3.6 120 1 0.5 Blank 99 50.8 Table B-7 Dollar Response--Office Contact Dollar Number of Percent of Response Respondents Respondents ($) 10 21 10.8 20 14 7.2 30 3 1.5 40 l 0.5 50 10 5.1 60 1 0.5 70 l 0.5 100 2 1.0 Blank 142 72.8 165 Table B.8 Dollar Response--Market Contact Dollar Number of Percent of Response Respondents Respondents ($) 10 25 12.8 20 18 9.2 30 8 4.1 40 2 1.0 50 15 7.7 70 1 0.5 100 7 3.6 120 1 0.5 Blank 118 60.5 Table 8.9 Transaction Response Reported Annual Number of Number of Percent of Transactions Respondents Respondents 0 19 9.7 2 70 35.9 6 31 15.9 10 26 13.3 14 17 8.7 18 13 6.7 22 2 1.0 26 6 3.1 30 l 0.5 34 2 1.0 36 3 1.5 60 1 0.5 80 l 0.5 100 2 1.0 200 1 0.5 166 Table 8.10 Transaction Dollar Value Response Reported Annual Value of Number of Percent of Transactions Respondents Respondents ($) 500 12 6.1 I 1,000 34 17.4 i 2,000 38 19.5 , “ 3,000 30 15.4 ' 4,000 15 7.7 5,000 15 7.7 6,000 4 2.1 7,000 1 0.5 8,000 3 1.5 9,000 1 0.5 10,000 13 6.7 11,000 1 0.5 12,000 2 1.0 14,000 3 1.5 15,000 1 0.5 20,000 1 0.5 25,000 2 1.0 30,000 1 0.5 100,000 1 0.5 Blank 19 8.7 APPENDIX C SCRAM OUTPUT Figures C.1-C.4 and C.5-C.8 present two outputs of the computerized version of the SCRAM model (see Appen- dix D) utilizing the two different sets of cost assump- tions referred to in Chapter V. The first page of output (Figures C.1 and C.S) presents the prices computed by the PREP program (see Appendix D) as well as the price tolerance coefficient. The output of potential revenue is presented on this page by listing the ten most profitable service configurations. The pages which follow present the ten most profitable configurations resulting from the three remaining output assumptions (see Chapter IV). The cost assumptions used for each SCRAM run are indicated on the second page of the outputs (Figures C.2 and C.6). A key to the service codes presented in the output is as follows: A Security Custody = Specialty Orders = Odd Lots = Portfolio Management — Research Information = Office Contact = Market Contact QWMUOIII I 167 168 co. co. co. co. co. co. co. co. co. co. whmoo o«.m:m: Noomom: omcmMNm em.Ncnm co.n~:m ««.mo«m :m.eNNm oo.«om~ nn.m«oa enomemaa wtoozu whmoo owwN mod uvwmzoamwa no mwmtaz ca.m:m: Naowcc: cmoonmm omomenm woosmam «a.o:do :ucohnw moowamh nn.m«0c on.momou hHuoaa Ilzaa mazw>wd aduhzwmhom .coflumesmmm umou umufimiioscm>mm Hmflucwuom H.U musmwm ouwoma ca mound w umwou< c oucoma b omomq o uwoumq m uwomq : umoomc m umwomd m oumcoma « onhnmm unoccwmmocH omoommm 6:.momm mmobon so.mo=« ouowsu mu.mom: om.oowa N~.aoo om.cc~ :«.:¢:N wtoozH aoocacm ::.«osi “moamsi m:.-mn mooamni Snoaami Scoosdi ::.a:«i okoosl oN.o«« :«.:«N hwuoma qumoz oa.=« aa.comw m coona oc.ccn < m.o masons uom cu coma o a o 04 n o o woma u m< : om m m N omq « ZOthdzuHuzco ¥z<¢ Imhmon wgqum<> Imhmoo owXHm Iwow>¢mm no hmou 170 oc.omma oo.ooom oo.oomm oo.on:m oo.oc~m oo.omam oo.om:m oo- oc.omm co.ocH« mpmcu mn.@m:a mc.:oos cm.«c«m «a.omnm om.moom rh.m0am mm.aonm co. om.mc~ m«.~m:« uroozh h 73m Nm.cmmi ma.mo«i cm.ck«i oc.@Mdi ab.~wi km.nmi Nm.o¢i so. c~.oa« ma.~:~ Huoed mwcxoqo .coflumadmmd umoo umuflhiiusmuso womxomm m.U muzmflm ma wcomq nova um 20HhaaDmeroo cu yzam 171 oc.chm¢ co.cmom oo.o~mm co.oomm oo.om¢: oc.o~«k oc.omom oo.ccmm oo.om:n ca.omHN mhmou oH.0N:M «:.hmsm mo.:ocH :m.mwom ma.ww:: mn.nmcm «N.«oc: h«.mnnw so.mmnm mm.mnow wroozw c@.o:«~l om.Nro«I moomomi waodami mr.¢m:a km.momu 2.5- ncomcwi Mo.«ou macho- meoad 23y mnba quxodd .coflumESmmm umou umuflmiiusmuso msam wmmxomm v.0 wusoflm ow ed as q o m on n um m w0< m ma 3 o r m N a « onquDwHuzou yzqm nqumom¢ Adthmuhoa .coflumEdmm4 umou @coommiimscm>mm H8flucmuom m.u wuzmflm oumoma ad wound a oumood o oucomc s wwumd o ouoomc m uwumq a uwccm¢ m uuwumq N unwound « onhqmaoHuzoo xzqm anon” whosm «sown :m.no n:.~« nuomd @aoc ummoHya mN. uwQZumm unoccwmmUcH cc.am: oa.omom oo.oaoo ou.omao oo.omm: oo.omma oooooo: ooaosoo oo.o«mm oo.oumn mhmoo oooum 90.003 9 oo.m oo.oo~ o um.oos «Hoooao nm.mdwc :m.o~no mogsmsm :«.:::N Nm.moo: ma.«mms anomomo o«.nmm: szQZH om.0«~ «H.0mm hm.mom :m.o:m no.553 :«.:«m No .mmm deoamm a h oo.m oaocooa u oo.m ao.oo~ m om.m~o oaommo HROMQ 23w Jatmoz oaood geoooo m cocoa acoaaw 4 8.6 museflm m o« omooma m ommumq o uwowq u owumq o pmq m mouma : umoumq m umwoumq m mama « zomhamooHuzoo xzdm Imkmoo UdeHm<> umhmoo omXHu iwuH>mmm no hmoo 174 aa.aoo ca.o:- on.oomn oaoocsw oo.oom« oo.onm~ 90.95:: oooacom ca.a=:o acoomc: mhmoo m«.-:« oconmos mocmnon mm.~nno mnowmom mocaonn «docnnm 4n.::@m meccaos n~.-om mIOOZH md.h:: 0:.mm: 09.05: om.~no on.~«s co.«no ««.:mo :n.:oo mooéo:d MN.Nsm« bHuomu 23M moqvodd .coflumEdmm< umou ccoommiiusmuso mmmxomm >.U musmflm om ad owoumc m coma o uwouma x 00m 0 mom m womd 4 wwoom m woomc N move « zouh