ABSTRACT ACCOUNTING STRATEGIES FOR DEFENSE OF DISCRIMINATORY PRICING UNDER THE ROBINSON—PATMAN ACT by Ronald M. Copeland Statement of the Problem Section 2(a) of the Robinson-Patman Act provides that, given the requisite conditions, it is unlawful‘to discriminate in price between purchasers of commodities except where the price difference can be justified by a demonstration that cost differences larger than the price differences exist. Although no mention of accounting is made in the Act, this cost provision surely has accounting implications. Little has been written about all the broad implications for accounting of the Robinson-Patman Act. In the first place, the statistical record of hobinsoinatman cost justification cases is entirely inadequate for determining the scape of Section 2(a) enfbrcement, the relative importance of the cost defense, and the types of practices which have been successfully'justified. hbt only is the existing record woefully outdated, but no public evaluation has been made of’the nonadjudicated Section 2(a) proceedings, even though the existence and probable significance of these cases has long been acknowledged. Partly'because of this lack of data, controversy has arisen over the "workability” of the cost.defense. One group of respected critics claims that cost justification is not workable due both to inherent limitations in the nature of accounting and to the administrative pro- cedures used by the Federal Trade Commission, while another group of Ronald M. Copeland ardent and prominent supporters claims that the cost defense is work- able. An operational definition of a "workable" cost defense has not been produced by either detractors or supporters. In addition, no comprehensive approach to the problem of cost justification has appeared in the literature as a guide to those persons interested in the use of accounting as a tool in solving Robinson-Patman problems. very little has been written about what must be done to justify price differences, how it is to be done, when it is to be done, and whether or not it should be attempted. No decision process has been described which would integrate all these pertinent aspects of the cost defense. Findings The statistical record of adjudicated cases was updated and evalu- ated. Evaluation of cost justifications' significance was made by con- sidering both the relative use of the cost defense and the relative suc- cess of the defense when it was used. In this regard, a successful defense is one in which the Federal Trade Commission accepted the ac- counting procedures employed in the study. In terms of both criteria, the cost defense has been significant in the adjudicated FTC proceedings. The lack of information concerning the nonadjudicated cases stems primarily from the Commission's refusal to allow'researchers access to their files. Nevertheless, it is evident from the scant data available that a large proportion of all FTC restraint of trade investigations are terminated without formal action being taken.against the respondent and that a higher than average dismissal rate applies to Robinson-Patman investigations. It is further evident that cost justification has played a large part in the dismissal of Section 2(a) investigations. Ronald M. COpeland In the past, the cost defense has not played a significant role in private antitrust litigation or in Justice Department proceedings. Never- theless, the few cost justifications presented to suggest that the cost defense does have a potential significance for respondents in these pro- ceedings. The cost defense has been used in attempts to justify many differ- ent types of discriminatory practices. It is impossible to generalize from the adjudicated record the relative success of these defenses for particular practices, partly because of the unique circumstances of each case and partly because of the way in which the Commission reports its findings. Yet it is evident that where the discriminatory practice is in fact justified, a cost defense can succeed. From these statistical data it is apparent that the use of a cost defense stands a high probability of winning acceptance by the FTC. In addition to a high probability for success, a workable cost defense must also be economically rational. In this regard, a workable cost defense can be defined as one in which the expense expected to be incurred in making the defense is less than the expected benefits derived from having made the defense. A cost defense could be prepared by a seller to meet the threat of impending or existing legal involvements brought in response to his price discrimination. The defense may be made to meet a particular con- tingency or it could be designed to provide data for whatever action may arise. In preparing the defense, the seller must consider the objectives of the defense, the methodology of data gathering, the timing of the data collection effort, the timing of the defense preparation, Ronald M. Copeland and the financial implications of the defense. Many alternatives of method, timing, and financing may be combined to produce a workable cost defense. Thus, the price discriminator is faced with a possibility of many different threats and has available many alternative responses to each threat. A wise discriminator will prepare a strategy of defense. In this regard, a strategy is a predetermined course of action deliberately focused on winning an advantageous position over an opponent and which is calculated to counteract actions taken by the opponent. A strateg would integrate all the considerations pertinent to the decision to implement a cost defense and would have as one of its outputs a decision whether or not the proposed defense was workable. A determination and evaluation of alternative strategies involves the integration of existing knowledge of accounting, financial manage- ment, and statistics. Accounting provides a general methodology which underlies all cost defenses. Generally accepted principles have been derived for accounting for Robinson~Patman.matters. Following these principles, a program or alternative programs for accomplishing the study should be prepared. The alternative programs available depend upon all the instant circumstances of each case. Each program will list the laternative procedures to be followed in.making the study. An approximation of’the physical input required for each study is made and coated out at the expected rate of renumeration for comparable functions. Probabilities that a given event will occur can be estimated.by reasonable inference to historical records or intuitive judgements. A decision tree provides a framework by which alternatives Ronald M. Copeland may be more clearly visualized and evaluated. It is proposed in this dissertation that the many alternative accounting strategies concerned with Robinson-Patman accounting be enum- erated and evaluated by use of recognized capital-budgeting techniques. If this proposal were implemented, the estimates of variables would be explicit and capable of refinement, the decision making process could be audited, and the effects of changes in variables could be evaluated. As a result, the business decision would be more rational. ACCOUNTING STRATEGIES FOR DEFENSE OF DISCRIMINATORY PRICING UNDER THE ROBINSON-PATMAN ACT By Ronald M; Copeland A Thesis Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Accounting and Financial Administration 1966 ACKNOWLEDGMENT I wish to thank the members of my thesis committee, Chairman, Dr. Gardner M. Jones, Dr. Floyd lfindal, and Dr. Richard J. lewis, for their helpful suggestions and guidance in the preparation of this thesis. I am deeply greatful to other faculty members and fellow graduate students for their patience with me when I used them for sounding boards as my ideas developed. Without the cooperative attitude of the Department of Accounting and Financial Administration, and especially of its chairman, Dr. James Don Edwards, this study would not have been the interesting and enlight- ening experience that it has been. I also wish to acknowledge the financial assistance that the American Accounting Association and the National Association of Accountants provided through their respective doctoral fellowship programs. I am indebted to my wife, Elizabeth, and.family for their encourage- ment and sacrifice which made this work possible. ii ACKNOWLEDGMENTS. . . . . LIST OF TABLES . . . . . LIST OF EXHIBITS . . . . LIST OF APPENDICES . . . Chapter I. INTRODUCTION. . Purpose . . . OF CONTENTS &c1cgr°urde e e e e e e Accounting Implications The PrObIQM‘e e e e e e Contributions of the Study. I‘thhOdOlogy e e e e e e e e Qualifications of the Scope e . e e e O f t II. SIGNIFICANCE OF THE COST DEFENSE: Significance of the Robinson-Patman Act Commerce........ Discriminate. . . . . . PurChaSGrSe e e e e e e COMOdj-ties e e e e e e like grade and quality. Lassen competition. . . Changing conditions . . Significance of the Cost Defense. Federal Trade Commission Enforcement of the PamnACteeeeeeeeeeeeeeeee Accounting complexities and expense . . . . The cost defense's “win" record . . . . . . Incidence of cost justification efforts . . Mluationeeeeeeeeeeeeeeeee O O O O 0 he 'séuéyIZII ADJUDICATED RECORD Rob Private Enforcement of the Robinson-Patman Act. Historical record . . . . . . Contrast with FTC enforcement M3805 e e e e e e e e e e e e The COSt defens.e e e e e e e Evaluatione e e e e e e e e e Justice Department Enforcement of the Robinson-Paman Act.........o.............. Evaluationeeeeeeeeeeeeeeeeeeeee Summary....................... iii ins eeeeeeeeeeeOeeeeeeeee vii vii III. SIGNIFICANCE OF THE COST DEFENSE: NONADJUDICATED RECORD Nature of NOnadjudicated Defense. . . . . Administrative procedure. . . . . . . . Advantages of Nonadjudicated Settlements. Significance of Informal Cost Defenses. . Preliminary investigations. . . . . . . Application for complaint . . . . . . . mSPOSition or complaint. e e e e e e e Robinsoinatman legal investigations. . Robinson-Patman accounting investigations erCtefldenceeeeeeeeeeeeeee ANewInformalProcedure.. .. ....... Potential Significance of Cost Justification. W O O O O C O O O C C O O I O O C O O 0 IV. THE NATURE OF COST JUSTIFICATION. . . . . Types of Practices Justified. . . . . . General Approach to Cost Justification. PTO’jUStification TGSte e e e e e e e e Eboemple of Cost Justification . . . . . ThOBituationeeeeeeeeeeee StudyprogI‘QMeeeeeeeeeeee findings. 0 O O O O O O O O O O O O 0 Principles of Robinson-Patman Accounting. Differences in price. . . . . . . . . . Segregation of groups . . . . Allocation of costs to groups Other principles. . . . . . . FTC advice. . . . . . . . . . SW3” 0 O O C I O O O O O O O O O C O O V. STRATEGIES FOR ROBINSON-PATMAN ACCOUNTING Strategy for a Robinson-Patman Defense. Timing of the Cost Justification Effort Financial Considerations. . . . . . . . Strategy of a workable Defense. General Framework of Analysis Alternative EventS. e e e e e Spedflc Analy813 e e e e e e BenefitSeeeeeeeeee Corditional mats e e e e e Pertinent historic events . Coat CSSiSmont e e e e e e The conditional probability Pertinent historic events and Probability assignment. . . Evaluation of the decision. Limitations of the example. Condusioneeeeeeeeee Oeeeeeee iv estimates 0 O O O O O O . 0 O O O O O 0 O O O O O O O f uncertain events of fact 0 O O O C O O O O O O O O Page VI. SUMMARY AND CONCLUSIONS . . . Restatement of the Problem. Statistical Findings. . . . A Workable Cost Defense Framework for Analysis. APPENDH O O O O O O O O O O O O O Bmumm O O O O O O O O O O O O 0 O O O O O O Page 112 112 113 115 117 119 122 TABLE II-l II-Z II-3 II-lJv II-S III-1 III-2 III-3 III-h III- 5 III-6 LIST OF TABLES Adjudicated Section 2(a) Cases, 1936-1965 . . . . . . . Cost Justification in Adjudicated Cases, 1936-1965. . . Private Suits Initiated, Reported, and Settled Out of Court, September l936-August, 1961. . . . . . . . . . mmages Awarded in Private Triple Damage Suits, 1936-1961eeeeeeeeeeeeeeeeeeeeee Justice Department Enforcement of the Robinson-Patman Aet’1936-1958eeeeeeeeeeeeeeeeeeee Preliminary Restraint of Trade Investigations for Vhich Application for Complaint was not Docketed, 1937-19L‘8eeeeeeeeeeeeeeeeeeeeee Disposition of Applications for Complaint, all Restraint OfTradOCases,l937-l950.. e e e ee eee ee ee Disposition of Complaints, all Restraint of Trade Cases, 1937-1950eeeeeeeeeeeeeeeeeeeeee Robinson-Patman legal Investigations for Mich no Formal Complaint was issued, 1937-1954 . . . . . . . . . . . Robinson-Pam Accounting Investigations for much no Foml Complaint was Issued, 1955-1962e e e e e e e e Number of Times Each Defense Contributed to Dismissal of Investigations in Nonadjudicated Proceedings, 9/ 36- 7/37. 0 O O O O O O O O O O O O O O O O O O O O O O O Practices Defended by Attempted Cost Justifications: Adjudicated C3393, 1936.1965e e e e e e e e e e e e e Page 20 21 29 33 36 47 49 51 52 62 LIST OF EXHIBITS EXHIBIT Page IIIel Procedural Steps for Processing A Robinsoinatman Action, 1966 O O C I C O C C C C O O O O O O O C O O O 141 IIIeZ Outline of Informal Procedures in Cases Defere the FTC, ”101' to 1914’9 Reorganization. e e e e e e e e e e e e Iv;1 Fixed and variable Costs Per Order, Order Processing amallingeeeeeeeeeeeeeeeeeeeee 73 IV-Z Comparison of Price Differences with Cost Differences Based on Class III Mirdmmn Quantities . . . . . . . . 73 V-1 Alternative Events, Hypothetical Ebcample. . . . . . . . 94 V-2 comj-tiom mats O O O O O O O O O O O O O O C O O O O 101 V-3 Conditional Probabilities of Uncertain Eyents . . . . . 106 LIST OF APPENDICES APPENDIX Page 1 Pertinent Sections of the Robinson-Patna Act . . . . . 119 vii CHAPTER I Introduction We The purpose of this dissertation is to provide a guide to those persons interested in using accounting as a tool to help them selve RobinsOinatman problems, such as determining future pricing policy and defending present price structure against charges of price dis- crimination. In order to use accounting for this purpose, it is neces- sary to understand the nature of the factors which create the problems, to know'what alternative actions may be taken to solve the problems, to know how to implement each alternative, and to know how to decide which alternative to use. This dissertation will be concerned with determining alternative accounting strategies which integrate all per- tinent factors and establishing a framework with which to evaluate the alternatives strategies. Bagggggung . One way in which business men expand sales volume is by stratifying their market and charging different prices for their product to differ- ent segments of the market. As a result of such pricing policies, some customers are discriminated against and harmful effects to competition may result. Therefore, the Federal government has regulated price dis- crimination through the provision of the Robinson-Patman Act. This act provides ”...that it shall be unlawful for any person engaged in comp merce...to discriminate in price between different purchasers of com- modities of like grade and quality" except for "differentials which make only due allowance for differences in the cost of manufacture, sale or delivery."1 Three different sanctions may be applied against alleged violators of Robinson-Patnan provisions. The Federal goverment, through the Federal Trade Commission, may file a civil suit which can terminate with an order directing that the discriminatory practices should cease and desist.2 Also, the party injured by the discrimination may file a suit seeking an injunction against the alleged practices.3 In addition, the Federal goverment, through the Justice Department, may file a criminal suit which can terminate with fines of not more than $5,000 and/or prison sentences ranging up to one year.“ Finally, the customer injured by the discrinnatory practices may file a private anti-trust suit which can terminate with awards for treble damages plus reasonable attorneys' fees.5 These sanctions may be applied singly or cumulatively. In light of these sanctions, it is entirely natural to suppose that the Robinson-Patman Act has exerted considerable pressure on the lSec. 1, Robinson-Patnan Antidiscrimination Act (Public-No. 692. 74th Congress) 49 Stat. 1526. This Act amends Section 2 of the (Dayton Act. Therefore, other provisions of the Clayton Act are applicable to Robinson-Patnan violations. Se below, footnotes 3, 5, Appendix I. 2.2“. Teeth were added to this provision by a 1959 law which provides that the respondent is subject to a $5000 per day penalty for each day he continues the practices forbidden by the order. Amendment to Sec. 11, W11 Act; 73 Stat. 245. 35cc. 16, Clayton Act, 38 Stat. 731; 15 USCA. Sec. 15 “Sec. 3, Robinson-Patman Act, #9 Stat. 1526. 5Scc. a, Clayton Act, 38 Stat. 731; 15 USCA. Sec. 15. executives who determine price policy; That is, the Robinson-Patman Act has directly intervened into the pricing decisions of those firms subjected'to either public or private litigation. Undoubtedly there are additional instances where price managers have been coerced by threat of legal action pursuant to Robinson-Patman provisions. In addition, selfhrestraint, motivated either by fear of legal action or by earnest desire to obey the law, probably accounts for more situations in which the Robinson-Patman Act impinged upon management's prerogatives with regard to pricing decisions. Here is the dilemma: sound business reasons for engaging in price discrimination, such as expanded sales and profits, exist. 0n the other hand, management is bound to the law, either by threat of sanction or by willingness to comply; The resulting conflict will generate pres- sure which, in turn, will initiate a search for an escape, i.e., pat- terns of action which will resolve the conflict. Some obvious escapes are provided for in the wording of the Act. That is, the law'exempts (as legal) price discrimination under certain conditions. Although there are at least six conditions under which price discrimination is legal for Robinson-Patman purposes,6 the most practical and available is the cost defense:7 discriminatory price differentials can be justified by corresponding cost differentials. 6T'aggart, Herbert F., ggst Justification (Michigan Business Studies, vc1. XIV, NO. 3) Bureau of Business Research, Ann Arbor, 1959, p. 5h8. The defense provided under Section 2(a) of the Clayton Act, i.e., Sec- tion 1 of the Robinson-Patman Act, are. . ."the price differences must (1) occur in interstate commerce; (2) relate to commodities of like grade and quality; (3) have the requisite effect on competition; (4) be other than price changes occurring because of market conditions...and (5) not be justified by cost differences of equal or greater magnitude." The sixth condition is stated in Sec. 2(b) of the Clayton Act which allows the use of price differentials to meet the lower price of a competition. 7Ibid., p. 5&9 Here, then, is a way in which firms may enjoy the fruits of price discrimination without incurring the risk of being found in an inde- fensible position in antitrust action. "All" that must be done is to collect the pertinent cost data, justify the discrimination by these data, and then, at will, expand sales and profits. In addition, the new'cost data might even allow'management to make better decisions with regards to resource allocation, sales emphasis, or product development. Therefbre, an incentive exists for the development of a Robinson-Patman (R—f) accounting system. A R-P accounting system would provide cost justification data. It would entail all those methods, procedures, and devices needed to col- lect data, before or after charges of price discrimination are leveled, to support a firm in a successful defense of charges if and when they are leveled. The "system" would include only those incremental methods, procedures,.and devices needed to provide data not already being pro- vided. That is, a R-P system is the new, extra accounting effert needed to produce the new, extra data. This type of accounting can be con- trasted with cost determinations made in the normal course of the ac- counting procedure and used primarily fer purposes other than cost justification. Tb provide a successful defense, however, a R-P accounting system would have to provide data which meets the requirements of the law. In this regard, the Robinsoinatman Act does not prescribe any particular set of data, procedures to collect data, specific accounting records, ‘or methods of presenting data fer a cost justification. Nevertheless, ‘the statutory provisions make it imperative for the firm to have ade- <1uate and dependable cost data to support such a defense. At a minimum, this would require that costs be determined fer each product and be classified by customer groups for which circumstances of sale and de— livery are substantially the same.8 EVen though the Act does not indicate the requirements of a R-P system, materials published by the Federal Trade Commission and liti- gated court cases do provide insights into the matter. These sources have been extensively researched by other investigators who have derived generalizations concerning the nature of required cost data. In addi- tion, the accounting staff of the Federal Trade Commission is available for counsel on matters pertaining to the type of data useful for a cost defense. Studies of actual cost-defense cases suggest that the best time to prepare a cost defense is befOre the complaint is issued.9 This posi- tion stems from several conditions. First, the fact that data were prepared before charges were filed serves as evidence of the good faith of the discriminator. Additionally, data gathered as an event occurs is often more accurate and credible than reconstructed data. Further- more, the availability of‘data before charges are filed may be suffi- cient to convince the complaining party that the discrimination was entirely justified, and thus preclude the impending litigation. co li one The cost of goods delivered to customers, as determined by contem- jporary cost accounting procedures, is an amalgamation of data combined 8Wlxon, Rufus, ed. Agoountants' Handbook, Fburth Edition. New York: Ronald Press 00., 195 a p. 9—55. 9Teggart, op.cit., p. 5#9. under arbitrary assumptions and subject to estimates of varying magni- tude.10 That is, the costs of original inputs are classified at ac- quisition and are subsequently reclassified and combined in an attempt to parallel the physical flow'through the productive process. The actual classifications employed depends upon.the purposes for which the data are collected, i.e., whether they are collected for planning, control, or product costing purposes. Many assumptions must be made as to the manner in which costs "attach" to a product, especially those joint costs which can only be allocated to a product instead of being directly identified with it. Often no attempt is made in the formal accounts to allocate joint costs, although such allocations are both theoretically correct and practically feasible. For example, distribution costs are rarely allo- cated to individual products or customers. 'In situations where charges of price discrimination might be made under the Robinson-Patman Act, cost data adequate for a defense might be available if the respondent has knowledge of relevant costs and utilizes appropriate accounting procedures to collect, summarize and report them. The amount of new'accounting data that would have to be gathered in order to support a defense of price discrimination charges would depend entirely upon the amount and nature of accounting data already being generated by the existing accounting system. In this regard, generated data refers to information which has been recorded and which is available or retrievable at no extra cost. The term‘ggtg includes all bits of information derived from internal sources, such as sales 10Anthony, Rebert N. Management Accountigg. Homewood: Richard D. M, Inc” 1960’ pe 6e orders, freight bills, etc. The more pertinent the data already being generated, the less would be the new effort needed. Since some firms already have elaborate cost accounting systems which contain apparati for evaluating distribution costs, it is highly likely that the pro- duction of cost justification data may be accomplished with a minimum of new effort. Modern data processing methods and improved statistical techniques for cost analysis may help overcome some of the accounting problems involved in collecting and preparing cost justification data. The emu- puter has made possible greater flexibility in arranging and rearranging common data for many uses. matrix algebra and related techniques offer a logical framework for manipulating that data. These innovations provide more infermation, in a wider variety of arrangements, in a minimun of time, and with minimal incremental costs. Regardless of the data processing methods and tools used, it is necessary that firms employ people who recognize the various require- ments of the law, and who can assure that the required information is provided by the accounting system. Information must be at hand with which to establish the costs related to sales of certain products or to certain customers while, at the same time, fulfilling management's other needs for data. Accounting classification and summarization must be structured in such a way as to allow data reclassification for such special purposes as defense of particular prices. The incremental accounting effort required for a RpP system to be designed, installed and operated will probably necessitate that the firm incur expenses for material, personnel, and facilities above those expenses previously incurred. At the least, resources already employed will be diverted by the operation of a R-P accounting system. The magnitude of these new expenses will depend upon the amount and type of new effort needed for the R-P accounting system. A firm employing a R—P system can expect to derive certain bene- fits from the additional information at its disposal. Among these bene- fits are the following: larger gross profits from increased sales re- sulting from justifiable discriminatory pricing; penalties avoided by successful defense of a private anti-trust suit; cost savings through increased efficiency resulting from the availability of additional data; etc. In any circumstance, the firm must decide if the expected costs of getting the new Robinsoinatman information is justified by the benefits to be expected from having the information. Our concern, in this paper, is to develop a framework which will provide an ordering to the per- tinent cost and benefit information, so that alternative actions may be effectively evaluated. W It has been argued that Robinson-Patman accounting, or other sys- tems which produce information of a similar nature, are useful manage- ment tools and should be implemented, and that the information derived from these systems is worth more than the cost of the system. One writer even suggests that the derived information produces operating cost savings alone which are so large as to warrant the installation of the system, regardless of the anti-trust implications of the data.11 That is, the anti-trust cost savings resulting from the availability of additional data can be considered a bonus. 111cm. , p. 5&7. Other writers contend that the cost justification provision of the Robinson-Patman Act is not very useful to businessmen who use a dis- criminatory pricing policy. These writers believe that RPP accounting is too expensive, the resulting data are too uncertain, and the legal acceptance is too much in doubt to warrant the implementation of a R—P accounting system.12 Additionally, few firms have implemented cost accounting systems which are sound and adequate for the defense of price discrimination charges and which, at the same time, are suitable and practicable for everyday use by the individual business firm in spite of the alleged benefits of such a system.13 Perhaps this indicates. that company managers are too lazy, ignorant, or disinterested in the needs of their marketing men and the wellbeing of their stockholders to implement R-P accounting. Alternatively, it may indicate that managers recognise good economic reasons for not implementing R-P accounting: that the advocates of RpP accounting have incorrectly evaluated the alleged benefits of the derived data. A third alternative is that management has incorrectly evaluated the benefits to be derived from R—P accounting. The first contention can be rejected on an‘g‘pgiggi basis: Ameri- can business could not have achieved its present level of development if the allegation were true. Adoption of the second or third conten— tions, however, leads one to wonder about the type of analysis used by management in arriving at the decision to reject the preposal to imple- ment R-P accounting. Probably, more often than not, management has used an informal, intuitive approach. That is, in making the decision to 12For a detailed exposition of these conflicting views, see Chaptfir II below; 3Heckert, J.B. and R.B. Miner. Distribution Casts, New Yerk: Ronald Press co., 1953, p. 365. 10 forego R-P accounting, the managers have either: 1. never felt the need to investigate the potentialities of R—P accounting; 2. mulled over all the previously acquired facts and fallacies concerning RpP accounting and has reached a decision on that basis; or 3. decided to collect additional information (which may never be collected) and reached the decision by default. Hewever, the decision to implement or to forego R—P accounting is subject to a more fomal analysis than just described. That is, there exists an analytical framework in which this type of problem can be staidied. The decision to implement a RPP accounting system may be re- defined so that it becomes analogous to a capital budgeting problem and subject to all the known capital budgeting techniques. The problem then becomes: "Should X dollars of cost be incurred (to install and operate the system) in order to derive Y dollars of benefit." An application of the capital budgeting framework requires an equlicit determination of the costs involved in installing and operating the system, the benefits to be derived from the system, and the method .601- comparing the costs and benefits. When the costs are compared with the benefits, a rate of return can be computed. This rate can then be conquered to the expected rate of return from alternative investments, "id/0r with the cost of capital. If the rate of return of the 3-? pro- 3°¢t is higher than the cost of capital and the expected return on dumtive projects, and if funds are available, then the system should b’ ilnplemented. This formal analysis has several advantages over the informal, even u“"1811 it may lead to the same conclusions. In order to prepare the “MI analysis, an explicit statement of the basis of the decision is ll produced. This allows the analytical powers of several people to be combined for an attack on the problem. In addition, the explicit basis of the decision is subject to review and audit at a later date, provid- ing a sound basis for control of the decision making process. Sensi- tivity analysis can be performed on the formal decision model to esti- mate the impact of changes in a variable on the conclusion. In addition, an estimate of the magnitude of each variable may be refined by the application of more sophisticated mathematical procedures such as Bayesian statistics. With such an analysis, many different problems may be evaluated. For example, the framework can be used to determine answers to questions such as: 1. Should a given firm institute a Robinson-Patnan accounting system? 2. If so, what is the maximum amount the firm can afford to spend for such a system? 3. If not, how large must the probability that an unfavorable event (such as lost business, or lost anti-trust suits) will occur get before the firm can profitably institute a R—P accounting system? 1+. If not, how large must the conditional cost of an unfavorable event get before the firm can profitably institute a R—P system? Contribgtions of the Stug The contributions of this dissertation will be threefold. First, the study will produce a statistical summary of all the cases relevant to Rabinson-Patman cost justification purposes. This summary will up- d‘te the previously compiled record of the adjudicated Robinson-Patman “398 and will also contain a new compilation of data concerning the nonadjudicated Robinson-Patnan cases. These statistical data help determine the scope of Robinson-Patman enforcement, the relative im- pxxrtance of the cost defense, the type of practices which have been successfully justified, and the types of costs which have been used for justification purposes. In addition, the study will focus on the academic argument concern- irug the "workability" of the cost defense. An operational definition of “workable" will be produced which states the conditions necessary for such a defense to exist. Then the historical record of all cost de- fenses will be examined to resolve the argument. Finally, this study will integrate the specifics of the Robinson- .Painman cost defense with the generalities of accepted business-decision making techniques to produce an approach to the solution of Robinson- Pataman problems. The framework developed in the study will not only be a conceptual model for ordering the pertinent data of the problem but ‘wiJfl also be an operational decision making model. Thus, the material developed in this dissertaion may have practical uses. yeti-togglogy 1. Mine the provisions of the Robinson-Patnan Act, related commentary, and relevant court cases, to determine what estab- lishes a cost defense, which accounting data are useful, and what presentation form is most likely to satisfy "cost justification" requirements. 2. Suggest, based upon (1) above, the alternative accounting procedures for a hypothetical situation which will provide the required data. Alternative accounting strategies will be developed. 3. Approximate the incremental expenditure required to implement 13 each strategy, i.e., the additional cost of gathering and preparing the additional accounting data needed to defend dis- criminatory pricing. h. Approximate the incremental benefit to be expected as the out- come of each strategy. 5. Develop a framework with which to evaluate the alternative accounting strategies for providing data necessary to defend price discrimination. Qgélifigtion of the Scog of the Study 1. Whenever a price differential is attributed to a cost savings in this dissertation, it is assumed that the cost savings really exists, and the accounting problem is to identify the savings. It is not the accountant's task to manufacture cost differentials where none exist in fact. Thus, the accounting procedures discussed in this dissertation do not allude to ones used to manipulate data in an effort to create the appearance of cost savings. 2. Often a violator of the Robinson-Patman Act also violates other antitrust laws. However, the problems and suggestions discussed in this dissertation refer only to Robinsoinatman price discrimination situations. Conflict with other laws and inconsistencies in United States Antitrust policy is assumed awmy. i3. This study is primarily concerned with the accounting aspects of cost Justification. Little attention will be given to the legal intricacies although they are inherently related to accounting matters by the very nature of’the problem. That is, i ll.| Ill-Illa,“ 14 even though an accounting to produce data for legal purposes cannot be separated from the law, in this dissertation the ac- counting aspects will be treated from an accounting point of view rather than from a legal point of view. likewise, little of no attention will be given to the economic aspects of cost justification although economics is also inherently related to the problem. CHAPTER II SIGNIFICANCE OF THE COST DEFENSE: ADJUDICATED RECORD Significang of the Robinson-Patman Act It is generally agreed that the Robinson-Patman Act has had a sig- nificant impact on the pricing habits of American businessmen. The law has practically eliminated certain types of pricing arrangements, such as base-point pricing, and has caused others to be greatly modified, 81102:]: as discounts based upon yearly cumulative purchases. It is evident that businessmen have responded to the law, either under the impetus of legal sanctions or the earnest desire to obey the law. The number of firms affected by the law is determined partly by the marketing methods used by firms and by the restrictions of the law. To determine which fims are affected, the limiting words and phrases of Section 2(a) will be discussed.1 In excerpted form, Section 2(a) states: It shall be unlawml for aw person engaged in commerce... to discriminate in price between different purchasers of comedities of like grade and quality...where the effect of such discrimination may be substantially to lessen compe- tition... : Providdd, That nothing herein contained shall prevent differentials which make only due allowance for differences in the cost of manufacture, sale or delivery resultirg from the differing methods or quantities in which such commodities are to such purchasers sold or 1The ensuing discussion only indicates the simplest implications Of the key words and phrases of the Robinson-Patman Act. Each word or phrase has an extensive history of legal interpretation, the bulk of which is beyond the scope of this dissertation. For a more complete, detailed, annotated discussion of the legal ramifications of the words ofthe act, see Fredrick M. Rowe, Price Discrimination Under The Robinsorh-Patxnan Act. 15 16 delivered...: And provided further, That nothing herein contained shall prevent persons engaged in selling goods, wares, or merchandise in commerce from selecting their own customers in bona fide transactions and not in re- straint of trade: And provided further, That nothing herein contained shall prevent price changes from time to time where in response to changing conditions affecting 2 the market for or the marketability of the goods concerned. Commerce. To be subject to the provisions of the Robinson-Paman Act, the alleged discrimination must occur in interstate commerce. That means that the seller must be in a different state than the state in which at least one of the buyers resides. For discrimination to occur, sales must be made to at least two customers with each being charged a different price. Either the two customers may reside in two different states, or they may be in the same state if the seller is in a differ- ent state. Thus, situations where the seller and both of the buyers reside in the same state are not subject to the provisions of the law. ascriminate. For purposes of the law, discrimination only means that each customer was charged a different price. Am time two prices are charged for substantially identical goods, discrimination occurs. This concept can be contrasted with the economic concept of discrimina- tion which holds that discrimination occurs whenever a seller receives different returns from selling identical goods to two different cus- tomers. Under the legal concept, a uniform price to all customers, regardless of the costs of servicing them, would be nondiscriminatory end not subject to the Act. ‘- 2h9 Stat. 1526. For the complete text of the act, see Appendix I, below, p. 119. Following common usage, this provision will be called "Section 2(a) of the Robinson-Patman Act" although it is really part of Section 1 of the Robinson-Patman Act and Section 2(a) of the Clayton Act. l7 Purchasers. A purchaser is one who acquires an article in a bona fide sales transaction. Sales to different customers are requisite criteria fer legal coverage. Non-sale transactions, such as consignment deliveries, gifts, or loans are not subject to the Act. 92mmoditie . The Robinson-Patman Act is only concerned with the sale of tangible commodities or goods. Intangibles such as futures, stocks, bonds, advertising, or services, are not subject to the law. Like grade and Quality. -Until recently, the phrase "like grade and quality" referred only to physical differences in the commodities sold to each price-differentiated customer. However, recent interpre- tation of this provision broadens it to include intangible differences, such as brand names, in a more realistic understanding of the environ- ment of the market place. Thus, goods which are physically and/or intangibly differentiated are not subject to the Act. Lassen cogpgtition. Competition may be injured if a particular person is injured or if the discriminating practice may tend to create a monopoly fer the seller. The injured parties who might affect compe- tition are the unfavored buyer, who may always be hurt to the extent of the price differential; the customer of the unfavored buyer if the un- favored buyer passed the price differential down to him; or the com- petitor of the favored buyer who may now lose customers to the favored buyer. No sure and fast rule determines whether competition has been sufficiently lessened to become subject to the provisions of the Robinson- Patman Act. Changim audition . The changing conditions referred to in the Act are always a function of the passage of time. The amount of time that must pass between subsequent sales at differentiated prices is a 18 matter relative to the nature of the commodities sold and to the market in which they are sold. The Robinson-Patman act ferbids only relatively simultaneous sales at differentiated prices. If a significant market event has occurred between sales, the transaction is not subject to the provisions of the Act. In light of these general qualifications and limitations of the law, only finns which _s_e_l_l_ substantially identical commodities in _i_n_t_e_1_'; ‘gtgtg commerce, charging different‘pgiggg to different buyers in near siggltaneous transactions, need be concerned with Section 2(a) viola- tions. However, firms having these marketing characteristics constitute the bulk, in annual sales dollar terms, of the business transacted in the United States. Therefore, the Rebinson-Patman Act should be of significance to the bulk of American commerce. unfortunately, the exact significance of the Robinson-Patman Act cannot be measured. The effect of public policy has been obscured by the broad changes in the environment in which business is done. Added to this is the fact that the most significant consequences of the law are probably indirect, and therefore untraceable. Measurability is further impaired by the wide gaps in available information concerning what occurred both before and after legal proceedings were initiated.3 Signifiggngg of the ggst Defense It would be entirely natural to assume that if the Robinson-Patman Act is considered a significant factor by interstate sellers, escape clauses to the act would also be considered significant. Any time there exists a force which frustrates the free actions of’men and at the same time there is a way that this frustration.may be circumvented, one would 3Edwards, Corwin D., The Price Disggimination Law, pp. 617, 618. l9 expect,‘g riori, that this circumventing way would be considered sig- nificant. A first approximation of the significance of the cost defense may be made by'examining the historical record of adjudicated Robinson- Patman proceedings. The only proceedings which are pertinent to the discriminating seller and the only ones which effect him are either private or government actions under Section 2(a) and government action under Section 3.“ On the following pages, discussion will be separated into adjudicated proceedings of the Federal Trade Commission, private triple damage suits, and Justice Department enforcement activity. Eggegal 22399 Commission Enforcement of the Robinsoinatman Act An enmination of adjudicated Robinson-Patinan proceedings admin- istered by the Federal Trade Commission reveals that for a period from September, 1936 until December 31, 1965, 360 complaints were issued for suspected violations of Section 2(a). Of these, 218 terminated by orders to cease and desist; 131 proceedings were dropped, closed, or dismissed; and the remaining are still in process.5 Thirtyaseven respondents used cost justification as a defense against the charges of price discrimina- tion. Although the accounting techniques employed were acceptable to the Federal Trade Commission in about half the instances, only 11 respon- dents succeeded in staving off a cease and desist order (see Table 11-2). Evaluations of the significance of cost defenses used in adjudicated proceedings has been contradictory. Authorities taking opposite views “Rowe, Fredric M5,.22‘git., p. 467. "While the text of Section 3 contains no aggress provision for these defenses, the justifications available to exculpate a pricing practice under Section 2 are implicit in.Section 3." See Appendix I for full text of Section 3. 5300 Table 11-1, below, 9. 20. 20 Table 11-1. Adjudicated Section 2(a) Cases, 1936-1965 Orders to Proceedings Cost defenses, Khmber of’ cease and dropped, closed by year of Year complaints desist or dismissed complaint 1937‘ 16 - O O h 1938 13 9 3 o 1939 Zn 5 1 2 1940 20 8 2 3 1991 11 9 12 2 19h2 8 7 5 o 1943 6 2 O 2 19th, 8 3 3 l l9h5 1 5 2 O 1946 3 3 2 0 19b? 3 1 o o 1948 13 5 8 0 198 31 1 1 2 1950 15 5 37 l 1951 8 h 2 1 1952 11 h 5 3 1953 9 10 3 0 1954 5 11 5 0 1955 12 5 h 0 1956 2 11 1 O 1957 18 h 9 3 1958 24 15 2 2** 1959 22 19 o 7 1960 #8 7 O 2 1961 12 15 1 0 1962 2 10 h 0 1963 8 8 2 l l96h h 27 8 O 1965 3 5 9 1 Total 360 218 131 37 Sources: Rowe, Frederic M. Ezice Disggimingtion Under thg fighinson— W. Boston: Little, Brown 8: Co., 19 2, p. 537 for years 1937-1961. Supplement to Rowe, op,cit., for years 1962, 1963. Mr. John V. Buffington, Assistant to Chairman, Federal Trade Commission, for years 1964, 1965. Igade Regulation Repgrter. New Yerk: Commerce Clearing House, 1 . Notes: *Starting 9/36. 1"Includes one Section 2(c) charge which was defended as if it were a Section 2(a) charge, FTC Dkt. 7273. 21 Table 11-2. Cost Justification in Adjudicated Cases, 1936-1965 Disposition Techniques D- Dismissed“ FTC U- Unacceptable 0- Order to cease Docket Ap Acceptab1e* and desist Number Name of Respondent N- Not determined 0- Consent decree 2935 Kraft-Phenix Cheese Corp. A D 2937 Bird & Son, Inc. A D 2986 Standard Brands, Inc. U C 3224 E. B. Mueller & Co. U O 3685 United States Rubber Co. A O 3977 Champion Spark Plug Co. U 0 4307 International Salt CO. U 0 4319 Morton Salt Co. U 0 4389 Standard Oil Co. U D 4556 Curtiss Candy CO. U 0 4636 Bissell Carpet Sweeper Co. A D 4920 Minneapolis-Honeywell Regulator Co. A O 4972 United States Rubber Co. A O 5253 National Lead Co. U 0 5701 Horlicks Corp. A O 5728 Sylvania Electric Products U D 5768 C. E. Niehoff & Co. U O 5872 Thompson Products, Inc. A O 5989 Fruitvale Canning Co. U 0 6043 B. F. Goodrich Co. A c 6044 Goodyear Tire & Rubber Co. N C 6721 Hamburg Brothers, Inc. A D 6816 Airtex Products, Inc. A C 7018 National Dairy Products Corp. U 0 7129 The Borden Co. U D 7273 Thomasville Chair Co. U D 7357 American Motors Corp. A O 7365 American Metal Products Co. A D 7475 Foremost Dairies, Inc. U 0 7494 Cannon Mills Co. A D 7514 Mueller Co. N 0 7559 Sperry Rand Corp- A D 7635 Philadelphia Carpet 00. U o 7815 Chemway Corp. U 0 7850 Purolator Products U 0 8599 William H. Rorer Inc. U 0 8663 Beatrice Foods, Co. ‘** ** Source: Trade Regglation Repgrter. New York: Commerce Clearing House, Inc. 9 19 e Notes: I"At least one technique acceptable at some stage of the adminp istrative process. I"“‘By FTC or the courts. I"M'Pending. 22 on cost justification's significance can be found; some hold that it is significant while others that it is not. For example, a widely quoted rejection of the cost defense is: "However, the cost defense has proven largely illusory in practice."6 The opposite viewpoint, just as widely quoted, has been expressed as: "The conclusion to be derived...is in- evitably that the cost defense, with all its complexities and uncertain- ties, is the most practical and available'ZSf all the defenseé7."7 These conflicting conclusions are based upon observations concerned with three aspects of the cost defense: the accounting problem and its related expense, the "win and loss" record of adjudicated cases, and the incidence of attempts at cost justification. These aspects will be con- sidered separately below. Aggguntigg ggmglexities and expense. One opinion is that the accounting effort required to gather and present information sufficient fer cost justification purposes would be prohibitively expensive. Fbr example, this idea has been stated as: A thorough cost justification can be a prohibitively'eXh pensive project, demanding extravagant investments in time of company executives and professional talent. Only the most prosperous and patient business firm could afford pursuit of an often illusory defense.9 hath rate exceptions the cost defense has proven expen- sive and time consuming. Hundreds of thousands of dol- lars have been spent by several respondents to no avail... Most respondents simply cannot afford the lavish expendi- tures which have been made by some, or at least are 6Repgrt of the Attogggz General's National Committee to Study the i at we, p. 171. 7Thggart, Herbert F., Cost Justifi tion, p. 549. 830‘”: mu 1" 307° 9 Report of, op,cit., p. 173. 23 unwilling to spend the money on a project so question- able in its prOSpective results.10 On the other hand, N . W. J. Warmack, former accountant for the Federal Trade Commission and currently a consultant on Robinson-Patman accounting problems believes that: As far as price discrimination.matters are concerned, a prohibitive amount of work ordinarily is not required in order to develop accurate costs.ll Mr. Whrmack further elaborates this theme in an article titled "Robinson- Patman.Costing Not too Difficult or Expensive."12 He believes that "most interstate sellers can Obtain 'adequate Robinson-Patman costing at an annual expense that should seldom exceed the salary of a good clerk.”13 Taggart also believes that exploratory cost studies which may be sufficient for justification purposes could be initiated at little out of pocket expense. This effort may be considered, philosophically, "a relatively cheap form of insurance."14 Further, he has found that cost justifications which were not too expensive to launch have prevailed.15 The accounting problems (to be discussed in Chapter 4, below) have also been considered as an obstacle to cost justification. Consider the following statements: Consider...some of the intricacies inherent in the attempt to show'costs in a Robinson-Patman Act proceeding. The 10 Thggart, 02,9it., p. 546. ll Whrnack, William J., "Cost Accounting Problems Under the Robinson- Patman Act,” Robinson-Patman Act Sygpgsium, p. 107. l21damuack, William J ., "Robinson-Patnan Costing Not Too Infficult or Expensive," 1256 Trade Practice Annual. 13Whrmack, William.J., as quoted in Edwards, op,git., p. 599. 1""I‘bggart, og,cit., p. 548. l élgég-v P0 5“6° 24 elusiveness of cost data, which apparently cannot be obtained from ordinary business records, is reflected in proceedings against sellers....Whenever costs have been in issue, the Commission has not been content with accounting estimates; a study seems to be re- quired, involving perhaps stop-watch studies of time spent by some personnel such as salesmen and truck drivers, numerical counts of invoices or bills in some instances of the number of items or entries on such records, or other such quantitative measurement of the operation of a business...And "general knows ledge of the trade", to use the commission's phrase, unsupported by foctual analysis has as yet been far from acceptable...as the basis for cost showings in other proceedings before the Commission. No doubt the burden...to show...seller's costs...is heavy...It is not a question of obtaining information in the seller's hands. It is g matter of studying the seller's business afresh.1 Proof of a cost justification being what it is, too often no one can ascertain whether a price is cost- justified.17 However, accounting technology is such that certain costs can be identified without undue burden, and some methods have achieved "general acceptance" as valid techniques for identifying costs and their relation to products and customers. These techniques are described in Chapter IV “love Controversy even exists as to the focus of the accounting effort. Rowe believes that the accounting effort should be directed by the legal counsellor. He believes that "Cost justification under the Robinson- Phtman Act is an exercise in legal semantics rather than a scholastic pursuit of accounting verities."18 On the other hand, Taggart believes: EVery cost-justification study should be aimed at con- vincing the Commission accounting staff...This means that study results should be couched in accountants' 16 1 346 U. 5., pp. 68. 69 *1953)- 71b1ge, po 790 18Rowe, ep,e1t., p. 312. 25 language, based on the use of accountant's techniques. while statisticians, engineers, lawyers, executives, and others can and should contribute to the final re- sult, the cost-justification study is and necessariby must be by and for accountants.19 It appears that every argument concerning the complexity and expense of accounting effort used to condemn cost justification has an opposite argument used to justify the cost defense. In any event, the degree of accounting complexity and.the amount of expense required to implement a cost justification are not useful criteria with which to evaluate the significance of the cost defense. For all business situations, the size of an expenditure is only meaningful when compared to the benefits expected to be gained by having made the exp penditure: expense is a relative matter. The larger the expected gain is, the larger can.be the expense employed to derive the gain. Thus, only a relative measure of expenditure would be a useful criterion with which to judge the usefulness of the cost defense. The egg; defense's ”win" record. The cost defense has been evalup ated in tems of the number or relative number of times its use has helped the respondent "win" his case. For example, the Attorney Gen- eral's Committee to study the anti-trust laws concluded that the cost defense was ”illusory" by the use of such a measurement. The committee stated: However, the cost defense has proved largely illusory in practice. After one successful cost defense before the Commission in.l937,‘ not until...seventeen.years later did an accused seller in a fully contested proceeding succeed in establishing a complete cost defense to de- feat every element in.a discrimination charge.’ While in two other litigated and fully adjudicated proceedings a cost defense partially prevailed, the Federal Trade Commission in seven other recorded instances rejected 19 Taggart, °2aét0 p. 542. 26 attempted cost justifications out of hand.“20 The comittee's score keeping set the record at two wins, two draws, and - seven losses. However, Taggart looks at the same basic data, the adjudicated pro- ceedings, and redefines the outcomes. He contends that a "successful" defense is one in which the accounting technique employed was acceptable by the Commission and which did in fact find the existing cost differ- ences.21 This definition of "success" is contrasted with the Committee's which, by implication, defined success as the finding of cost differences greater than the price differences. On this basis, Taggart's score is four to seven. But Taggart goes one step farther. He contends that the base of the comparison should be broader than just the "litigated and fully adjudi- cated proceedings" and should also include these proceedings in which the "complaints were dismissed because the respondents presented cost figures adequate to place them within the promised“?2 Three additional sales cause Taggart's score to become seven to seven. Applying T‘aggart's measure to the updated record, the score is sixteen to twenty-one. Ingidengg of cost justification efforts. The cost defense has been evaluated in terms of the number of times it has been used. For example, Taggart says, Perhaps the most convincing evidence that "illusory" is an appropriate adjective is the simple fact that. . .only 21 respondents...have pleaded cost defense in Commission proceedings since 1914 and only two defendents in treble damage suits. If the figures are limited to the Robinson- 20Report of, op.cit., p. 171. *Asterisk replaces footnote in the 21 { original. Taggart, my, p. 544. 22'.'.‘o:l.d., p. 544. 2? Patman era, only 22 sellers have pleaded the cost defense in 22 years. This statistic speaks more eloquently than the numbers of favorable and unfavorable decisions.23 These statistics do not indicate that cost justification is playing a very important role in price-discrimination cases at the present time or that it can'be called an inviting loophole in the law.2 At the time Taggart made these statements, there had been 283 complaints issued by the Foderal Trade Commission for suspected 2(a) violations. Cost justification was attempted in about 8% of these cases. However, this type of computation assumes that all of the suspected 2(a) violations were in fact cost justifiable; that in each case the price differences could be accounted for by corresponding cost differ. ences. Obviously this is unrealistic. A.more realistic assumption would be that every time the discrimination was in fact justifiable, the respondent would make some effort to defend himself. If this assumption is accepted, then the computation would be as follows: Of the 311 orders issued by the Commission in Robinsonp Patman Cases, a majority were concerned with violations regarding which no cost defense was possible. There are 118 cases involving violations of Section 2(a), and only in these might the respondents have been acquitted if they had presented a satisfactory showing of cost dif- forences.* In the majority of these cases, however, the respondents chose not to contest the complaint. So general was this acquiescence to the Commission's orders that the respondp ents invoked their rights to a trial in only 32 of the cases in which orders were issued under Section 2(a); and in 7 of’these they waived a part of their rights to sub- mit evidence, briefs, or argument. The remaining 25 cases constitute the entire group in which a cost defonse was relevant to the charges, and the respondents fully used their right to defend themselves. Cost defonses were offered in 12 of these 25 cases, plus 1 case in which a study of costs was included in a stipu- lation of facts.25 2“Ibid., p. 546. 25Edwards, op,o1t., pp. 588, 589. *Asterisk replaces footnote in the original. 28 The pertinent statistic derived from this calculation is that the cost defense was used about 50% of the time it could have been used. EValuation. EValuations of the significance of the cost defense used in adjudicated Federal Trade Commission cases have considered the problems of accounting complexity and expense, the cost defense's win and loss record, and the incidence of cost justification effort. The degree of accounting complexity and amount of expense required to achieve a defense are relative matters which become meaningful only when contrasted to the benefits expected to be gained by the effort expended. As such, the absolute size of such effort is not a useful criterion with which to evaluate the cost defense. The win and loss record, when rightly defined in broad terms, indicates that the cost defense has been successful a significant portion of the time. Finally, the cost defense has beam employed a significant portion of the times it could have been employed. Thus, it may be concluded that the cost defense is a significant factor in adjudicated Federal Trade Commission cases 0 mivate figment of the figbinson-Patman Act Private enforcement of the Robinson-Patnan Act is authorized by Section ’4 of the Clayton Act. Specifically, this provision holds that a party injured by a discriminatory act may sue the discriminator for triple damages plus a reasonable attorney's fee. Two purposes are served by this provision: to assist in the implementation of the law and to recompense the party injured for his injury. It was believed that the triple damages sanction would effectively accomplish these Objectives because it placed the largest financial burden of all RDblnson—Patman sanctions on the discriminator and transferred the proceeds to the plaintiff. Histogiga; record. A review of the historical record of private enforcement is pertinent to an evaluation of triple damage suits under the Robinson-Patman Act. For a period from September, 1936 until Aug- ust, 1961, about 330 private suits were initiated.26 Of these, about 250 were for alleged violation of Section 2(a) provisions.27 One hun- dred eleven cases are reported, which means that about 220 suits were either dropped, settled out of court, or otherwise terminated before coming to trial. 0f the 111 reported cases, 51 alleged violations were for Section 2(a) provisions only, while another 34 were for joint vio- lations of 2(a) provisions along with other Robinson-Patman provisions.28 Thus, in the first 25 years of the enforcement of the Robinson-Patnan Act, 85 cases, 77% of’the total number of private Robinson-Patman suits were for alleged Section 2(a) violations. Table II-3. Private Suits Initiated, Reported, and Settled Out of Court, September 1936-August, 1961 Settled Out Initiated Reported of Court Total Robinson-Patman Private Suits 330 111 219 Section 2(a) Private Suits 250 85 165 Source: Richard J. Barber. "Private EnfOrcement of'the Antitrust Laws: The Robinson-Patman Emperience," George washiggton Law Review, XXX (December, 1961), pp. 193-l9h. 26Barber, Richard J., "Private Enforcement of the Antitrust Laws: The Robinson-Patman Dcperience," George Washigton 1.33 Mew, December 19619 P- 1930 27Calculated by using ratio of 2(a) violations to reported cases. 590 Table II-B. ZBBarber, 02,355., pp. 19h-95. 30 Final judgment has been recorded in 65 of these cases. In 57 cases the decision was favorable to the defendant, with 33 resulting in out- right dismissals. The remaining eight decisions were for the plaintiff: damages were awarded in six cases; judgment has been entered for the plaintiff in one case but damages must still be proven; and the plaintiff received declaratory relief in the eighth case.29 Contrast with FTC enfbrcement. In a comparison of private and gov- ernment enforcement of the Robinson-Patman Act, private suits initiated represent about 32% of government complaints issued:30 private 2(a) suits initiated represent about 73% of government 2(a) complaints is- sued.31 But private 2(a) suits terminating with decisions unfavorable to the defendant represent only 5% of government 2(a) proceedings ending 32 with the issuance of orders to cease and desist. 0r seen another way, 6% of private suits have been favorable to the plaintiffs while 61% of the government suits have ended favorably for the government.33 This disparity in effectiveness of private and public Robinson- Patman Act enforcement may be due to several causes. First, private suitors have encountered more difficulty in procuring evidence than 34 encountered.by the government. The government has greater power to 29Ibid., p. 192. 30Rowe, op.git., p. 536, reports that there were 1,0#0 complaints issued for all Robinsoinatman violations up to June 30, 1961. (330/1040 = 32%) ‘318ee Table I151, Table 11.3 (250/3u3 = 73%) 3%;pgg.. (8/168 = 5%) ”limo. (168/276 - 61%) 3“Tomlin, w: 0., ”Private Recovery Under the Robinsoinatman Act," Tens Law Review, December, 1961;, p. 177. 31 force the respondent to produce records, making dispositions, etc., than has the private suitor. In addition, the goverment is only interested in preventing discrimination while the private suitor is interested in punishing the discriminator. Thus, the defendant may be more 'willing to acquiesce to governnent demands for data than to the private suitor. Secondly, the goverment is better able to withstand the costs and other burdens of litigation than the private suitor. legal costs and use of executives' time for antitrust litigation may be overwhelming for the private suitor while the goverment has specialized employees hired particularly to handle this activity. Additionally, the private suitor is motivated by profits and will only pursue the suit if the expected gains are larger than the expected costs, while the goverment will pursue a case as a matter of principle even if the costs for prosecuting the matter are larger than the benefits to be derived. Generally the burden of proof is harder for a private suitor than for the governnent. A private suitor must prove the defendant's viola- tion of the Act, the causal relation between the violation and the claimedinjury, and the measurement of the injury sustained. Each part of his case must be explicitly proven. 0n the other hand, the govern- ment may only have to demonstrate the possibility for an injury to exist and does not have to determine the exact extent of injury sustained.35 Finally, the private plaintiff has a narrower concept of injury to prove than has the government. The Government must prove general injury either a particular competitor or to competition in general or show that the respondents practice will tend to create a monopoly. However, the ¥ 3 5 Section 5 of the Clayton Act may provide an exception to this Statement. Section 5 states that a judgment from a contested gavel-ment Proceeding is prima facie evidence of an antitrust violation in subse- quent private suits. 32 private suitor must prove specific damages sustained by a causal func- tion of the discriminator's actions. Recent court decisions have not been content with a general showing of damages (i.e., the difference between the per unit price paid by the plaintiff and the price charged the most favored buyer multiplied by the number of units purchased by the plaintiff). Rather, these court decisions require explicit demon- strations of the specific decline in assets or similar evidence. Damages Plaintiffs have received final awards fer damages due to injuries arising out of Robinsoinatman violations in six cases. The judgments (as trebled) have ranged from a low of $3,030 to a high of $180,000; the attorneys', fees have ranged from no specified amount to $40,000. Table 211;“ shows the awards granted in the six cases where plaintiffs prevailed. The cost defense. Little information exists concerning the extent tn: which cost justification played a role in settling private Robinson- Ffiatman litigation. No information exists about the reasons or terms of ‘tlae 220 out-ofhcourt settlements. In resolving these cases, the cost cieefense may or may not have played a significant part. However, it is known that the cost defense was used in at least 1311ree of the 85 reported Section 2(a) private triple damage suits. Two Of these actions terminated with settlements favorable to the defendant "iiile the third defense failed to convince the judge or the jury that ‘tlae price differences were justified by cost differences.36 However, even in this last case it is unknown whether the accounting techniques ‘ 36The two successes are: American Can Go. vs. Russelville Canning (33., 191 F. 2d 38, and Reid vs. Harper, 235 F. 2d 420. The failure was Iiruce's Juices vs. American Can Co. See Table II-h. 33 Log cgemncoucgom one .mma .0 .Aaema .uoDEoooav xxx 020a>0m_zma : Eng o .390 c.0050.“ 3qu umgfiufi 05 mo unosoouomnm ougnnm: .uenwem A. nausea “eonsom .00 .338 as ”3de commences 000.00 00.000.0aa Ams0a .aa0 0000 0N.a 00H maxoepcoe .> goons new .0.0 Nam .Apa0 3000 00.0 00H .00 :00 neoaaosa 000.0: 00.000.00H A030H .0Ha .0.00 000 .0000 .a am .> .02H .000000 0.000nm Aoemav 000 .0.0 mom .00 .meoaa .meam weave 000.0 00.000.0m Amnma .uao anode 000 0m.0 000 .s .00 H0>0n0 a xooam ecosaao Amemav mam .m.0 0mm .00 umeam 0:0 . 00.000.m A.un0 £000 000 00.0 0mH .> .0a00 moaam noea< eponaeaam A000H0 awn .m.0 man .00 senom pogone 00m.m mm.m:m.ma “.0n0 game ~00 0~.a mna .0 0.000 asaom 00.00 caoaaos< - Aemmflv mad .0.0 000 .00 000nm 0:00 0.0002 000.HH0 00.000.00 0 .Amnmfl .uao anode was 0~.a mom .> oaooz moor.” 032.53.: nemeada weapons c0338 musewfipfl no 332 33.003 .330 003.8 :05 3:3 5 223‘ £23 35. 03a 34 employed were acceptable and the cost differences were inadequate or whether the techniques themselves were inadequate. Evaluation. In spite of the small number of times it has been used in private litigation, one authority believes that cost justification is likely to be more persuasive in private litigation than in government actions "because the court, and especially the jury, may be willing to accept rough estimates of the savings, or make a partial allowance for its role, even though this would not be true in the case of the Federal Trade Cbmmission.”37 Although this may be true, it is evident from the statistics on cases settled for the defendant that other defenses must 'be more available than the cost defense. That is, the cost defense was Iised in less than h% of the private cases contended while it was used in 114 of the Federal Trade Commission proceedings contended.38 Thus, ciefendants in private litigation are more prone to use defenses other than cost justification to a greater degree than they would in govern- naent actions. It does not appear as if the cost defense has played a significant Inole in private antitrust litigation, i.e., its actual employment was insignificant relative to its potential employment. However, cost Justification is potentially significant for defense in private liti- gation; it has been successful 66% of the few times it has been employed. .Jusggg Department Morgent of the Robinson-Patman Act Section 3 of the Robinson-Patman Act added criminal prohibitions to the civil remedies for price discrimination already covered by the g. 37Barber, 02,01t., p. 210. 38 Edwards, op.cit., p. 589 reports that the cost defense was used in 17 out of 121 contested 2(a) proceedings. 35 Act. These prohibitions are solely enforceable through actions taken by the Department of Justice. Specifically, Section 3 "prohibits three kinds of trade practices, (a) general price discriminations, (b) geo- graphical price discriminations, and (c) selling 'at unreasonably low prices fer the purpose of destroying competition or eliminating a com- petitor.'"39 The Justice Department may also bring charges of violation of Section 2(a) against a respondent, although this is an unusual occur- rence which only occurs when the Robinson-Patman charges are accompanied by charges of violation of other antitrust laws as well. A list of the suits filed by the Justice Department and their ultimate disposition appear in Table II-5. As may be seen from the exhibit, Robinson-Patman enforcement by the Justice Department has been sporadic. This type of effert is generally considered to be insignificant as a law enforcement tool.”° In any event, the cost defense has been.used in Justice Department proceedings. "The Borden case...culminated in dismissal of the price “1 It is not known whether discrimination charges as cost justified." the cost defense could have been used in other cases, i.e., whether the discriminatory practices were cost justified. It is probable to suppose that if the discrimination was in fact justified, the cost defense would have been used. Egalggtio . Although the cost defense was successful the one time it was used in Justice Department proceedings, the total number of pro- 39Nashville Milk c0. vs. Carnation 00., 355 0.3. 373-377 (1958). 4030‘”, 02:01te’ Po 469e “I'M" p0 297e Table II-5. Justice Department Enforcement 36 or the Robinson-Patnan Act, 1936-1958 Year Name Violation Disposition l9h0 American Petroleum Institute Sec. 2 Clayton Case dropped 19% General Motors Corporation Sec. 2 Clayton Case dropped 19u5 libbeyanens-Fbrd Glass Co. Sec. 2 Clayton Consent order 1948 Bounan Dairy CO. Section 3 R-P Acquittal 1948 Borden Co. Section 3 R-P Dropped 1951 Borden Co. Sec 2(a) Clayton 1955 Safeway Stores Inc. Section 3 R-P (convicted) Fine & sent. 1955 Naryland Cooperative Milk Producers, Inc. Section 3 R—P Acquittal 1958 Feirmont Fbods Co. Section 3 R—P Convicted, Fine Source: C. D. Edwards, The Price Discrimination Law, washington: The Brookings Institute, 1959, pp. 2-84. ceedings is so small that the potential use of the cost defense for these proceedings is insignificant. Summagy Although it is commonly believed that the Robinson-Patman Act has had a significant effect on the actions of interstate sellers, it is not clear how one of the exceptions to the restrictions, the cost defense, is regarded. On one hand it has been called the "most practical and available" defense, and on the other hand it has been termed merely "illusory”. However, both conclusions stem from observations of the same basic set of data; the adjudicated Federal Trade Commission proceed- 37 ings, the reported private suits, and the initiated Justice Department suits. Evaluation of the significance of cost justifications used in ad- judicated Federal Trade Comission proceedings can be made by considering the relative use of the defense and by considering the relative success of the defense when it is used. In terms of both criteria, the cost defense has proven significant. In the past, the cost defense has not played a significant role in private antitrust litigation. That is, its relative use has been insig- nificant in terms of its potential use. Nevertheless, cost justification has been significantly successful in terms of the times it has been em- ployed. Thus, it does have a potential significance for defense in pri- vate litigation. his potential use of the cost defense for Justice Department pro- ceedings is extremely limited. However, there is every reason to believe that if the discriminatory practice challenged by the goverment is in fact cost justified, then the defense could be used successfully. In summary, it is noted that.the cost defense has been‘used success- fully a significant number of times relative to the number of times it has been attempted in every type of proceeding where it could have been employed. CHAPTER III SIGNIFICANCE OF THE COST DEFENSE: NCNADJUDICATED RECORD Nature of Nonadjudicated Defense The statistical record of adjudicated Federal Trade Commission proceedings does not adequately reflect "the substantive significance and practical effectiveness of the cost defense.”1 This occurs because a complaint under Section 2(a) is issued only after a thorough investi- gation by the Commission's staff, and many of the cases investigated are never adjudicated. If the respondent believes that he has a valid cost defense he probably will make it available on an infernal basis to the investigators. These infbrmally presented cost defenses are considered by the commission in deciding whether or not to issue a for- mal complaint. If the Commission thinks that the defense is adequate, no complaint is issued.2 In order to understand the nature of a nonadjudicative defense, it is necessary to know something about the procedures followed by the Federal Trade Commission to process a Robinson-Patman complaint. An examination of these procedures will reveal how and when an instance of price discrimination which is alleged to violate the Robinsoinatman Act may be settled. againistrative procedure.3 The starting point for the administra- lEdwards, Corwin D., The Price Discrimination Law, p. 588. 2Ibid., p. 587. 3Federal Trade Commission, OrganizationI ProceduresI Rules of Practice, and Statutes, pp. 9-55. 38 Kg.) \0 tive procedure followed in processing a Robinson-Patman action is a sales transaction by one seller to two or more customers. Any party interested in this transaction may request the Federal Trade Commission to institute a Robinsoinatman proceeding by simply submitting to the Commission a signed statement setting forth the alleged violation of the law and the name and address of'the perpetrators of the alleged viola- tion. The parties who may be interested are a competitor of the seller, an unfavored purchaser, a competitor of the favored purchaser, a cus- tomer of the unfavored purchaser, a governmental agency, or the Commis- sion under its own initiative. No special form or procedure is required for the requested action. The Commission may not (and does not) act on all complaints. Hows ever, when it does have sufficient information indicating a violation of the law, it may proceed in one of three directions. First, the com- mission may afford the respondent an opportunity to dispose of the matter on an informal, nonadjudicatory basis. This procedure is used only when it appears that the public interest will be fully safeguarded thereby. Secondly, the Commissionrmay provide the respondent with an opportunity to dispose of the matter by the entry of a consent order. If the respon- dent accepts this order, he must admit the jurisdictional facts, stop the alleged practices, and waive his rights to further procedural steps. However, even if the respondent does accept the consent order, the Com- mission.may, at its option, proceed with an adjudicatory examination. This third step is conducted by a hearing examiner appointed by the Commission. After hearing testimony, taking dispositions, and examining evi- dence, the hearing examiner will reach an initial decision. This no decision will automatically become final within thirty days if the respondent does not appeal or if the Commission does not elect to review the case. However, if the respondent thinks that he is justified in his discrimination even though the initial decision was unfavorable, he will probably appeal. The Cbmmission reaches a decision after weighing the evidence pre- sented and issues a statement of its findings and an appropriate order. If the respondent is still dissatiSfied, he may appeal the decision to the courts. In fact, some Robinson-Patman cases have gone all the way to the Supreme Court. The alternative procedures may better be visualized by reference to the diagram in Exhibit III-l. Of the thirteen steps shown, six repre- sent terminal action. That is, once a complaint for alleged price dis- crimination has been made to the Federal Trade Commission, the matter may be disposed of in six alternative ways: 1) no action taken by commission, 2) infermal settlement accepted by commission, 3) consent decree settlement accepted by commission, 4) initial decision of com. missiOn.becomes final, 5) final decision of commission becomes final, 6) court decision becomes final. The first two dispositions constitute nonadjudicated settlements: the last four are adjudicated settlements. Advantage of Nonadjudicateg Settlements There is every reason to believe that an absolute advantage, both for the government and for the respondent, lies in an early adjudicated disposition of the alleged charges, i.e., that the investigation is dropped before formal charges are filed or that the respondent signs an assurance of discontinuance. For the government, these procedures economically and expeditiously 41 35:6 3 cameos“: noanwoon posefibem .m 0H 50.3.3 madvndhm ucofihoboo .m .D To d 53333.: AIIIIIIIIIMHA Hung meaooem c0338 efiofiom NH » m333m use 3.3er He madam 3.26 gm “agendas .5 50.3358 mouse daemon "season 93.88 £3280 9E hm no 5.3.33 hm 00.»er 0:00:00 033303 00.88 pesefirem 00."er human wcopaamem HH 4? m A1 w ._‘ /:fiasoo E a PE 3 pgogm 225 made: undead—poo gem meuawaanEH endeamsoo couwfieon 1.33 song nonunH PE /o.E Hesogg S «1‘ a A m A «h H H 363 1333 Honda meaooem Essvsnoaoz :33er :33 co megvem ohm m e ”333. oz women. 0.5 m 82 .532 essaméonfinqm a 32825 .5 macaw assessors .HuHHH finesse III III II lull-Ii Ill, Illalll Ill .lllll. #2 accomplish the'objectives of the Act. In this regard, the Government is only interested in compliance rather than punishment. Informal settlement "also expedites and simplifies the Commission's legal pro- cedure, saves the Government...the greater expense of litigation incident to trial of a complaint, and makes possible the handling of a large num- ber of cases in an expeditious way." Nonadjudicated settlements save the respondent the time and effort of extensive accounting and legal involvement incident to a trial of formal charges. The respondent also benefits because these informal settlements can not be used as admissions of defacto antitrust violations by private litigants nor do they toll the statute of limitations for private antitrust litigation.5 In addition, three subjective factors enter into consideration of the advantages of early settlement. The first has been succincthy stated as: Apparently it is frequently possible to convince the Commission's accountants at the investigation stage that somq_price differences are justified by differences in cost: however, once the Commission's preliminary investi- gation.- in the judgment of the Commission's accountants - suggests the lack of satisfactory proof of cost justifi- cation and the Commission proceeds to issue a complaint, successful defense on the basis of cost justification is well-nigh impossible because the Commission's accountants have already reached adverse judgment. Probably this is the reason for the notorious failures of many expen- sive a elaborate attempts to justify differences in costs. In addition to diSbelieving a cost defense which has failed in an informal settlement, the Commission may well disregard cost defenses “Annual Report of the Federal Trade Commission (1936),p.50. Although the quote referred only to stipulations, it is also apropos for all in- formal settlements. 5Barber, Richard J., "Private Enforcement of the Antitrust Laws: The Robinsoinatman Experience," George washiggton Law Review, December 1961, p. 187. 6Parkany, John, F.T.C. Enforcement of the Robinson-Patman Act, 1946-52, p. 261. 43 which have been prepared after the investigation stage. A former direc- tor of the Bureau of Economics of the Federal Trade Commission, of which the Bureau of Accounting was a part, stated: The issuance of a complaint under Section 2(a) means either that no cost defense was submitted to the Commis- sion during the preliminary investigation or that the Commission's accountants fOund what was placed before them less than convincing. It follows as a matter of course that the cases thus initiated were often decided without cost information having been submitted and that, where such information was submitted during the trial, the Commission usually rejected it as inadequate.7 Finally, the preparation of cost data prior to an investigation by the Commission may be an indication of the respondent's "good faith" and may cast a halo over the study. Although the residual radiance of "good faith" hasn't been explicitly demonstrated in any of the adjudicated cases, the lack of it has. In one case the Commission stated, "Perhaps most damaging to the respondent's position is the fact that the responp dent had not made any cost studies prior to the investigation of this matter."8 Signifiggnce of Informal Cost Defense There have been indications that a significant number of cost de- fenses have beennmade on an informal basis. For example, Edward F. Hawrey, former Commissioner of the Foderal Trade Commission, stated, ”There have been, of course, a large number of cases in which the cost defense was explored on an informal basis."9 In this same vein, Corwin D. Edwards wrote, 7Edwards, 93433., p. 588. <8Thomasville Chair 00., F. T. C. Dkt. 7273 (March 15, 1961), p. 6. 9A5 quoted in H. F. Taggart, Cost Justification, p. 5h5. an Since the Commission does not announce the results of investigations that terminate without complaint, no record is available to show the frequency with which costs have been successfully offered as defense in this informal way. While the writer was director of the Commission's Bureau of Industrial Economics and thus in general charge of the accounting staff concerned with problems of price discrimination, a considerable number of investigations were terminated in this way.”10 And again, another investigator reports that he was told by the Commission's chief accountant in 1952 that more than half of the time of the Commission's accounting staff is spent on cases that do not result in legal proceedings.11 Since all informal settlements are wrapped in the cloak of con- fidentiality, neither the names of the respondents nor the number of such cases are public information. However, the Commission has reported some data from which inferences concerning the numbers of informal cost defonses may be drawn. The following sections will present the historical record of infor- mal settlements, first classified by the three major procedural steps when informal settlements could have been.made, then by the indirect evidence only relevant to Robinson-Patman investigations, and finally ‘by direct evidence of cost justification. The statistics presented for the first classification, procedural steps for settlement, refer to cases subject to Section 5 of the FTC Act and to Sections 2, 3, 7, 8 of the Clayton Act. In addition, the administrative procedures followed during the period covered by the statistics reported in Tables III-l, III-2, III-3 differ from those presently followed because the Federal Trade Commission was administratively organized differently than at loEdwards, op.cit., p. 58?, Asterisk replaces footnote in original. llParkarV, OBeCite, pe 260a ‘45 present. Therefore, the procedural location for settlements recorded in the first three tables are marked respectively 1, 2, and 3, on Exhib- it III-Z, a flow chart of the procedural steps followed during most of the period for which statistics are available. Preliminary investigation. Most of the matters brought to the Com- mission's attention come "principally in the form of letters from the public complaining of business skulduggery or from indignant competitors of business firms that have been cutting too sharply the corners of fair competition. "12 These requests for corrective action are called appliggtions for complaint, as distinguished from formal complaints issued by the Commission. Requently the initial request for action contains insufficient informationand so a preliminary investigation of the charges must be conducted before deciding whether or not to docket the appli cation’ for complaint. In simple cases the preliminary investigations were handled by mail. In more complex cases the preliminary investigation was assigned to an attorney examiner in a field office. His report was reviewed by a branch manager who submitted recommendations to the Commission. Some alleged charges were dropped at this procedural stage. As can be seen from Table III-l, a significant proportion of field investigations are terminated without an application for complaint being docketed. The percentage of investigations not leading to a docketed application for various years ranges from 3% to 91%, and averages 68%. A significant trend toward more lenient docketing restrictions or less discriminating preliminary investigations is evident. Statistics are available only through 1948. ¥ ”Annual Report of the Federal Trade Commission (1955), p. 25. W com 0“ a 000 wndvnwnm vcosnue>oo .m .D "novwcdnmmz .:0dmmflssoo opens Hemoppm on» me pummem Hesnz< Hoocwoo “menu on ammo mnemom .vcflmfidsoo whopno scammflsfioo cowmmfiEEoo ca nonopeooao asshom He>omo.m new cowmmflsEou op moo. sneezedmen an pecmfim noapeafidfipm .m eSfimem Heaps 09 :OHmmHEEou D coapmfivflCfiIMHem comma>fio m.nesfisdxm meanu ho nowpemfipme>nfi huenfisdaemm .H \mmmWmem awash so capmoCeEEooeh use seabem m goasasaaamaooa mean an wowed .oea can ouoeom mouse as owaeooopm HaeuoncH no aaaasao .NnHHH sasaaam 47 Table III-l. Preliminary Restraint of Trade Investigations for Which Application for Complaint was not Docketed, 1937-1948 A DockeEed as % Investigations Fiscal Inquiries Closed Applications not Leading to June 30 After Investigation Complaint Application* 1937 583 275 68 1938 453 110 81 1939 379 36 91 1940 519 49 91 1941 400 78 84 1942 238 34 88 1943 140 29 83 1944 69 17 80 1945 5 209 23 1946 32 175 15 1947 5 163 3 1948 24 118 . 17 Source: Report of the Federal Trade Commission. Washington: U. S. Government Printing Office, 1930-1948. Note: * Calculated by dividing the number of inquiries closed by sum of columns 13 and _E_3_. Application to; complaint. Inquiries which survived the preliminary investigation, plus additional initial inquiries containing sufficient information are docketed as applications for complaint. When an appli- cation has been docketed, it is assigned to an attorney for investiga- tion, in which the facts regarding the matter are developed. He inter- views the party complained against, advising him of the charges, and 48 requesting the submission of such evidence as may be desired or in justification.1 The examining attorney develops the facts from all available sources, summarizes the evidence in a report, and makes recommendations as to what action the Commission should take. This report is then reviewed by the Chief Examiner before being sent to the Commission. The alleged charges may be dropped at this procedural stage, marked "2" on Exhibit III-2. As can'be seen from Table III-2, a significant proportion of the docketed applications for complaint are terminated without a complaint having been issued. These cases terminated by a stipulation to cease and desist do not include any Robinson-Batman violations because the stipulation procedure was not made available to Clayton Act violators. The applications closed for other reasons are mostly made up of cases dismissed for lack of merit. The percentage of applications closed for various years for other reasons ranges from 27% to 78% and averages 46%. A trend toward more closures in recent years complements the inverse trend of Table III-l. Since fewer cases were weeded out in the preliminary investigation, more cases are weeded out at this stage of the procedural process. Statistics are available only through 1950. Dispgsition of complaint. The Commission may then decide that the case should be closed without further action, that the respondent should ‘be given the opportunity to sign a stipulation of facts and agree to icease and desist from the practices charged, or that a formal complaint Should be issued. The Commission only issues a complaint after careml cmonsideration of the facts and evidence developed by the investigations. _-‘ 13Annual Report of the Federal Trade Commission (1937), p. 38. 49 Table 111.2. Disposition of Applications for Complaint, all Restraint of Trade Cases, 1937-1950 A B C D Applications % of Total Fiscal Stipulation to Closed for Finalized Applications June 30 Cease and Desist Other Reasons"l as Cbmplaint Closed“I 1937 641 350 293 27 1938 599 383 310 30 1939 510 i 405 371 31 1940 608 419 329 31 1941 512 _5h5 332 39 1942 508 526 233 42 1943 264 446 170 ' 50 1944 272 281 173 39 1945 248 257 142 40 1946 86 227 94 56 1947 129 38? 41 69 1948 114 508 66 74 1949 129 794 90 78 1950 147 778 104 76 Source: Annual Re rt of the Federal Trade Cbmmission. Washington: U. S. Government Printing Office, 1933:1950. Netes: * This classification includes such reasons as death, business or practice discontinued, private controversy, controlling court decisions, and dismissal for lack of merit. This last reason accounts fer more than half the dismissals. ‘* Calculated by dividing column‘g by sum of columns'g,.§,.g. 50 As may be seen from Table III-3, a large proportion of cases is dismissed for lack of merit even at this late stage of procedural devel- opnent (marked ”3" in Exhibit III-3). The percentage of complaints dis- missed for lack of merit ranges from 3% to 43% and averages 10%. A trend toward greater dismissal is evident in the most recent years and complements the inverse trend of fewer dismissals in the preliminary investigations. As with the trend for applications for complaint, this also indicates a shifting of the weeding out process from the early stages of procedural involvement to the latter stages. Statistics are available only through 1950. Robinson-Patnan legal investigations. The Federal Trade Commission has released statistics hearing more directly on the significance of cost Justification than can be implied from the data relevant to all restraint-of-trade cases processed by the Comission. However, due to the way the data are reported, it is impossible to pinpoint the location of the administrative process at which the alleged charges were dismissed. Table III-4 presents data concerning the disposition of legal in- vestigations for alleged violations of Robinson-Patman provisions. As can be seen in the table, a sizable proportion of all Robinson-Patnan legal investigations completed did not result in the issuance of a com- plaint. The percentage of investigations not leading to a complaint ranged from 39$ to 91% and averaged 74%. No appreciable trend was dis- cernable over the years for which data were available. In addition, the legal Robinson-Patman investigations up through 1948 represent 10% or total investigations conducted by the Comission.1u ‘ mm, Tables III-l, III-2, III-4. 51 Table III-3. Disposition of Complaints, all Restraint of Trade Cases, 1937-1950 A B C D E Orders to Stipulations Dismissed Closed Fiscal Cease & to Cease & for Lack for Other % Closed “ June 30 ' Desist Desist of Merit Reasons* 1937 306 17 13 41 3 1938 253 5 13 16 5 1939 391 4 12 21 3 1940 289 1 26 21 8 1941 350 2 32 24 8 1942 258 3 22 23 7 1943 168 1 ll 7 6 1944 125 1 14 15 9 1945 142 4 15 11 9 1946 90 2 19 13 15 194? 58 2 l9 9 21 1948 74 9 34 27 24 1949 47 5 33 5 37 1950 77 2 60 2 43 Source: M Remrt of the Federal Trade 25mission. Washington: U. S. Government Printing Office, 19 1950. Notes: '* Includes such reasons as death, business or practice dis- contimed, private controversy, controlling court decisions. ** Calculated by dividing column C by sum of columns 3;, g, 9, LL 52 Table III-4. Robinson-Patnan legal Investigations for Vhich no Complaint was Issued, 1937-1954 R—P legal REP 51 of Oogpleted Fiscal Investigations Complaints Investigations not June 30 Completed Issued leading to Complaint“ 193? 169 36 78 1938 152 19 88 1939 134 35 74 1940 156 ' 43 72 1941 145 58 60 1942 79 19 76 1943 193 18 91 1944 53 I ' 22 58 1945 ‘ 60 23 62 1946 67 12 82 1947 106 12 89 1948 96 43 55 1949 190 115 39 1950 140 22 84 1951 In 18 as 1952 n 17 as 1953 . 132 14 89 1954 104 15 86 Sources: Re rt of the Federal de Comission. Washington: U. S. Goverment Printirg Office, 19351954. Rowe, Roderick M. Price Discrimination Under the Robinson- m. Boston: little, Brown 8: 00., 19 2, p. 537. Notes: "' Calculated by dividing remainder of columns A less g by colurm A. " Statistics unavailable. I." in!!! I lilo {I III I]! II. III III 53 It is evident from these data that a majority of Robinson-Patman legal investigations have exonerated the respondent of any infraction of the law and that these proceedings have a higher average dismissal rate than any of the investigations previously reported in this chapter. The lack of‘a trend in the number of Robinson-Patman cases or in the per- centage of'dismissals can be contrasted to the noticeable trends for data concerning all restraint of trade cases. This indicates that Robinsoinatman cases were receiving relatively more attention than other cases in recent years, in spite of the fact that Robinson-Patman cases are relatively harder to administer and utilize more time,effcrt 15 Increasing emphasis and resources than other types of investigations. placed cn.Robinsoinatman investigations can not help but to increase the significance of the cost defense. Rebinsoinatman agcounting investigations. After an administrative reorganization of the Federal Trade Oomission, the Accounting Bureau started releasing data concerning the number of Robinson-Patman account- ing investigations it had conducted. These investigations included accounting analyses and studies of pricing policy, which implies that they were concerned only with Section 2(a) and Section 2(f) violations.16 Data about these studies were reported in such a way that it is impos- sible to determine the stage in the administrative process to which the investigations relate. Table III-5 presents data concerned with Robinson-Patman accounting investigations for which no formal complaint was issued. It is evident that a significant proportion of all investigations did not lead to a 15Annual Repgrt of the Foderal Trade Commission {1222}, p. 8. 16 rt of'the de a1 Tr de ssion l 60 , P- 35- 514 Table III-5. Robinston-Patman Accounting Investigations for Which no Fbrmal Complaint was Issued, 1955-1962 R-P Invegtigations B % Investigations Fiscal by Accounting Section 2(a), not loading to June 30 Bureau Complaints 2(r) Complaint‘ 1955 33 13 61 1956 30 2 93 1957 29 20 31 1958 #3 30 30 1959 #2 24 u3 1960 65 52 20 1961 61 1h 77 1962 78 8 89 Sources: Annual Report of the Epderal Trade Commission. Washington: U. S. Government Printing Office, 1955-19 2. Rowe, Frederick M. Price scrimination U er the obinscn- Patman Ag . Boston: Little, Brown & Co., 1532, p. 537. Trade figgg§ation Reporter. New Yerk: Commerce Clearing House, 19 . Note: * Calculated by dividing remainder of columns‘g less‘g by columnIA. complaint, the percentage ranging from 20% to 93% and averaging 57%. It appears as if there was an increasing trend in the number of investi- gations conducted, although the eight year period for which data were available is not long enough to provide a basis for inference. The statistics about the number of accounting investigations not leading to a complaint tend to corroborate Parkany's observation about the proportion of time spent by the commission's accounting staff on 55 cases for which no legal action followed the investigations.17 However, the statistics developed in Table III-5 are not exactly comparable to other statistics developed in this chapter. TherefOre, no further con- clusions can be drawn from them. Direct Ezidence. The previous exhibits and tables related to statistics from which the significance of the cost defense in infermal settlements could only be inferred. However, there are some data, the cnfly'pgblig disclosure to this writer's knowledge, which directly re- flect the significance of cost justification. In July, 1937, the Honorable wright Patman, co-sponsor of the Robinson-Patman Act, requested from the Federal Trade Commission in- formation concerning its informal settlement activities. In reply to this request, Mr. William A. Ayres, then Chairman of the Commission, submitted synopsized studies which detailed the causes for terminating 6# price discrimination investigations which had been initiated during the period from September, 1936 until July, 1937. Mr. Patman intro- duced the letters and the studies into the Congressional Record.18 Analysis of the studies reported discloses that 58 of the 6k inp vestigations involved suspected violations of Section 2(a). However, 21 of'these 58 allegations were unsupported by evidence that discrimina- tion actually occurred. The reasons for the dismissal of the remaining 37 investigations are detailed in Table III-6 below. As can be seen in the table, the cost defense was completely sufficient for dismissal in four investigations and contributed to the dismissal in an additional five investigations. Thus, in a ten month period, the cost defense was 17Su re, p. M. 1881 Congressional Record Appendix, pp. 2336-2341 (193?). 56 Table III-6. Number of Times each Defense Contributed to Dismissal of Investigations in Nonadjudicated Proceedings, 9/ 36-7/3? Defense Defense Muclly A Partially Total Times Reason for Dismissal Sufficient Sufficient Defense Used Sale to U.S. Governnent 2 2 Meeting Competition 5 2 7 No injury to Competition 2 l; 6 No Commodity Sold 1 1 rufferent Grade or Quality 3 3 Intrastate Comerce ll 11 Other ‘ 4 L; Cost Justification h 5 9 Source: 81st Comessional Record Qgendix. Washington: U.S. Govern- ment Printing Office, 1937, pp. 2336-23“, letter by W. A. Ayers, Chairman, Federal Trade Commission. successfully used to justify nine instances of price discrimination in nonadjudicated proceedings. Some interesting contrasts may be drawn from these observations. First, 90% of the total investigations dropped were for suspected 2(a) violations.19 However, in the adjudicated cases,2(a) complaints repre- sent 33% of all Robinson-Patman complaints issued. Thus, it appears that the Federal Trade Commission is more interested in suspected 2(a) violations than is apparent from the adjudicated record. Secondly, the nine successful informal cost justifications during ten months may be 19'Ihis statistic may not be representative of the current rela- tionship because the cases from which it arose occurred just after the passage of the Robinson-Patman Act when there still was widespread un- familiarity with its provisions. 57 contrasted to 16 successful formal justifications (out of 37 attempts) during a 30 year period. Only looking at this first year's experience, it can thus be seen that the real significance of cost justification must be related to the informal settlement procedures rather than to the formal procedures. A New Informal Procedure As part of the 1963 administrative reorganization of the Commission, a new procedure was instituted by which a respondent could settle a vio- lation of the law on a nonadjudicated basis. In this procedure, the respondent signs an "assurance of voluntary compliance,” and the matters are closed on the basis of the parties' assurances that the law will be complied with and that the questioned practices will not be engaged in in the future. This new procedure differs from the old stipulation-to- cease-and-desist procedure in that now Section 2(a) violations may be settled in this manner. An assurance of voluntary compliance requires no formal wording or form; the respondent, with the aid of the Commission, simply states in his own words what he has done in the past but will no longer do. How» ever, if he had been discriminating in price between different purchasers, but part of the discrimination was cost justified, he would only have to agree not to discriminate to the extent of the unjustified portion. Cost justification may therefore play a part in this informal procedure. These assurances are only allowed when the Commission determines that the public interest will be fully safeguarded by this indormal action. However, the making of an assurance in no way constitutes an admission by the parties that the practices intquestion were illegal. At least one Robinson-Patman 2(a) case was settled by an assurance 58 20 Unfortunately, the public record does not of voluntary compliance. disclose.if cost justification played any part in the final disposition of the matter. Potential Significance of Cost Justification ' The limit of the potential use of the cost defense is the actual number of Robinson-Patman violations which occurs each year, i.e., the maximum number of potential cost justifications is determined by the number of actual violations. The statistics about actual violations is of academic interest and could be used to evaluate the public awareness of the law (by comparison with applications for complaint), public and private enforcement of the law (by comparison with number of legal actions initiated), effectiveness of the law (by comparison with cease and desist orders and judgments awarded), effectiveness of the cost defense (comparison of actual cost defenses initiated with potential usage), etc. One method for estimating the number of violations which occur would be as follows: a violation requires the near simultaneous sale of like commodities at different prices to different customers in interstate commerce. The 1963 census reported that there were 2,320,1U7 businesses in the United States engaged in the manufacture and/or sale of commodities.21 Just assuming that 1%»of these discriminate in price, 23,000 potential violations of Section 2(a) occur. (It should be rememp bored that those situations where competition is not affected, where the price difference just meets a price of a competitor, or where the 20W, No. 27, Federal Trade Omission, November 11, 1965, p. 3. 21 1261 ansus of Business, Department of Commerce, 1965. 59 price difference is just justified by cost differences are not unlawful under the provisions of the Act.) Few complaints relative to potential complaints are ever brought to the attention of the Federal Trade Commission or the courts. This may be accounted for by a number of possible reasons. For example, the injured party may be ignorant of the discrimination, since it is unlikely that the favored party would jeopardize his position by disclosing the discrimination.22 If the injured party is aware of the discrimination, he may be unaware of the legal remedies at his disposal or be disinclined to use them for fear of retribution from the discriminator or fear of not recapturing the large expenses required for private legal action. In addition, the injured party may always change suppliers and/or bar- gain, by threat‘of legal action, for a more favorable price. For these reasons, it is unlikely that the full potential of cost justification will ever be realized. EEEEEEI It is evident from the data released by the Commission and pre- sented in this chapter that a large proportion of all Federal Trade Commission restraint of trade investigations are terminated without formal action being taken and that a higher than average dismissal rate applies to Robinson-Patman investigations. It is further evident that cost justification.has played a large part in the dismissal of Robinson- Patman investigations. Therefore, it may be said that cost justifica- tion is significant in relation to formal settlements and also in rela- tion to informal settlements. It is only of academic interest that the cost defense has only achieved a small proportion of its potential. 22Section 2(f) of the Robinson-Patman Act makes it illegal for a buyer to knowingly buy at a discriminately low price. CHAPTER IV THE NATURE OF CCST JUSTIFICATION Types of Practices Justified The cost defense has been offered as justification for many differ- ent types of pricing practices which allegedly violate the provisions of the Robinson-Patman Act. Each type of practice requires different justi- fication procedures. Because of the variety‘of forms in which price discrimination appears, it is pedagogically necessary to classify and discuss the different forms in order to relate them to different justi- fication procedures. It is possible to classify the pricing practices which have appeared in the adjudicated proceedings under the following categories: 1. Quantity discounts in which prices are determined by the physical quantity or dollar magnitude of a given transaction, order, or shipment. 2. volume discounts in which the price charged varies directly with the aggregate physical volume or aggregate dollar volume of purchases over extended periods of time. This type of dis- count often takes the form of an end-ofeperiod rebate. 3. Functional discounts in which the price charged to each cus- tomer depends on his position in the distribution chain. h. Territorial discounts in which the price charged depends upon the location of the customer's place of business. 5. Offbscale discounts in which one or more buyers pays prices 60 61 that do not conform to the systematic patterns of prices appli- cable to other buyers. 6. Other discounts in which prices are determined by a combina- tion or variation of the above. The frequency with which attempts were made to cost-justify each type of pricing practice is shown in Table IVQl. As may be seen in the Table, attempts to justify some types of practices have been.more common than for other types. Quantity, volume, and functional discounts are most often cost defended. A cost justification for each type of discrimination involves peculiarities which may make implementing the defense more difficult than is at first apparent. 'Costs relevant for defense purposes are different for each type of discrimination, so few generalizations would be applicable to all justifications. However, some observations may be made about the difficulty inherent in defenses for particular discrimina- tory practices. Quantity discounts presuppose that a lump-sum cost or a cost varying unproportionally to the number of units in an order is incurred in making the sale and delivery, regardless of the size of the transaction. The fraction of such a lump which is applicable to each unit sold grows smaller with each increase in the number of units included in the trans- action. "Where this is the basis for cost justification, cost differ- ences will justify substantial price differences at the lower end of a quantityadiscount scale, but the possibilities of cost justification quickly become negligible as the number of units in the transaction increases."1 Thus, price differences among small customers who buy lEdwards, Corwin D., go Price Discrimination Law, p. 588. 62 Table IV-l. Practices Defended by Attempted Cost Justifications: Adjudicated Cases, 1936-1965 Discounts Total Successes‘ Failures Undetermined Quantity 3 3 O 0 volume 10 u 6 0 Functional 9 3 5 1 Territorial 1 O 1 O Off-scale h l 3 O Other 9 h h 1 Totals 36" 15 19 2 Sources: Thggart, Herbert F. Cost Justification. Ann Arbor: Bureau of Business Research (University of Michigan), 1959. Trade figgfiiation Reporter. New York: Commerce Clearing House, 19 . Notes: * At least one technique accepted by the FTC. "Not including FTC DKT 8663. slightly different quantities may be easily cost justified, but price differences between buyers of large or moderate amounts may be extremely hard to cost justify. A cost justification for annual volume discounts is particularly difficult if all individual orders are for similar quantities, because few of the seller's activities are eliminated or substantially reduced just because some buyers purchase large annual volumes. The Commission is predisposed to-reject cost defenses for volume discounts except where differences in volume reflect differences in the size of individual purchases.2 ?;§;g., p. 238. I! in.‘ |¢ll|.| .l I. III]... I! ll! ' III III II: IIIII'III J | i ’x 63 Functional discounts only give rise to cost justification problems when the requisite effect on competition is present, i.e., when differ- ent levels in the distribution.chainxcompete with each other. When this occurs and there are numerous functional classes or where there is much diversity in methods of doing business with different groups of buyers, development of'a successful cost defense becomes a formidable task: it would have to be shown that there were differences in the costs of doing business with each group and that these costs relative to the volume of business for each group produce differences larger than the discrimina- tory discounts between groups. The more groups existing, the larger will be the number of price combinations to be justified. Territorial discrimination is only cost justifiable when prices vary with distribution costs. That is, when prices differ in sales to customers who are located in different places, the costs that are rele- vant to the price differences are those incurred in distributing the goods. "Thus, in both basing-point systems and zone-price systems, the cost defense is automatically precluded by discrepancies between the price structure and the structure of transportation rates."3 Other territorial pricing schemes based on costs might be justifiable if in- formation.systems existed which could collect the costs. Until recently however, the field of physical distribution accounting has been largely ignored by accountants,“ even though some theoretical work has been done which lays the ground work for implementing practical systems for col- lecting data.5 3Ibig., p. 590. "lewis, Richard J., A §§§1ness legistics lnformation and Accounting §ystem for marketing Agglysis, p. 9. 51bid., Chs. IV, v. 6h Unsystematic price concessions made to individual customers can be cost defended only by a showing that costs differed concomitantly for each customer. If, in setting these special prices, the seller departed significantly from his own discount schedule, a showing that the dis- count schedule was cost justified would prove that the off-scale prices were not justified, and a showing that the off-scale selling was cost justified would prove that all the other sales at regular discounts were not cost-justified.6 General Approach to Cost Justification Cost justification involves the matching of differences in the prices paid by customers with cost differences as large or larger: the customer charged the higher price must have caused the incurrence of a more costly manufacturing, sales, or delivery effort. It is there- fore necessary to determine both the‘ppipg and thelpggp associated with both the favored and the unfavored customers. A first step in the process of cost justification is to determine the magnitude of the discrimination between customers. This requires that customers be grouped in classes which approximate the nature of the discrimination. For example, for functional discrimination the customers would be grouped by function; for volume discrimination they would be grouped into appropriate volume classes; and with quantity discrimination individual transactions would be grouped into quantity classes. A per unit price is determined for each group by dividing the aggregate dollar sales value by the average quantity purchases. Since discrimination did occur between groups, the per unit price will be different for each group. The magnitude of the discrimination is 6 Edwards, op,cit., p. 591. 65 determined by subtracting from the price charged the high-priced group the prices associated with the other groups. For example, if the per unit price for groups A, B, and C were $8, $7, $5, respectively, the discrimination which would have to be cost justified between A and B, and between A and C would be $1 and $3. The discrimination between B and C would be $2. The magnitude of the discrimination is often stated as a percentage of the higher price (by dividing the difference by the higher price). For the example above, B received a 12.5%vdiscount from the price charged A. In some instances, this descriptive form is most useful for comparing price differences with cost differences. Costs are also collected for each group. For justification pur- poses, not all costs must be analyzed: only relevant costs must be allocated to the groups. For each respondent, the instant circumstances, the type of discrimination, and particular methods of doing business determine which costs are pertinent. For example, manufacturing costs are irrelevant where goods are sold from stock; freight costs are ir- relevant where goods are sold F. O. B. shipping point. Most often costs relevant for justification purposes are related to order getting, order processing, and delivery efforts. First, costs that are directly associated with each group are assigned or charged to it. For example, territorial discrimination may be justified by differences in freight costs which are directly attri- butable to each territorial group. Another good example is where ad. vertising costs for branded goods are directly associated with groups who purchased those goods and not allocated to other groups who pur- chased physically identical but unbranded goods. Next, costs jointly incurred by all groups are allocated to each group on some appropriate basis. For example, salesmen's salaries can be allocated to the groups on the basis of the number of sales calls made to each group. Only costs which do not vary proportionately with the number of units sold should be allocated to each group. Proporb tionatelyavarying costs can not contribute to cost differences between groups. Total costs for each group are determined by summing all the direct and indirect costs assigned to each. For unit costs are derived by dividing total costs by the average number of units purchased by each group. The magnitude of cost differences is now calculated by sub- tracting from the cost of the high-cost group the costs associated with the other groups. These per unit cost differences may then be compared with the price differences which were previously calculated. If the'price discrimination was stated as a percent, the cost differences must also be stated in this form. The per unit costs for each group are divided by higher-price group's per unit price. This states cost as a percent of unfavored unit's price, i.e., the same base used to determine the extent of the discrimination. These relative costs for each group are now subtracted from the higher-cost group to determine the magnitude of cost differences. Returning to the previous example, assume that the costs associated with groups A and B are $6.h0 and $4.80 respectively. A cost difference of $1.60 results. Or stated as percents of price, A's cost is 80%, 8'5 60%, and the difference is 20%. In either case, the $1.60 cost differ- ence justified the $1.00 price difference, or the 20% cost difference justifies the 12.5% discount. 67 Care must be taken in the selection of costs to be included in the study because the inclusion of irrelevant costs may invalidate the study. For example, if in justifying quantity discounts, costs which are con- trolled primarily by the number of customers, by the trade channels through which sales are made, or by other factors unrelated to the num- ber of orders received are included in the study, costs would be inflated and so unit cost differences between orders for small and large quantities would be overstated. This would invalidate the justification. To illustrate,7 consider a quantityadiscount situation in which the orderbreceiving department has an annual budget of $20,000. Of this, $6,000 is for costs determined by the number of customers serviced during the year and unrelated to the size of any particular order. The depart- ment's activities are governed by a volume of about 1,000 orders received each year. All orders take about the same time to process. If the department order cost is computed by dividing the total $20,000 by the 1,000 orders, then the per order cost is $20. This, in turn, is 5%tof a $400 sales order and 2.5%'of a $800 order. On the other hand, if the $6,000 irrelevant cost is rightly eliminated, then the per order cost is $12.. This is 3.5; of the $400 order and 1.75% of the $800 order. The cost difference between orders which can legally be used to justify price differences is the 1.75% (3.5%L1.75%)s not the 2.5%(5%F2.5%) originally'calculated. ti on at A test exists by which the chances for success of a detailed cost justification effort may be predicted with a high degree of accuracy. 7Basedupon an illustration by Whiting, Herbert 0., "Quantity Discounts", Nana ement Controls, November, 1965, p. 237. 68 This test can demonstrate whether or not a proposed or existing price differential is supportable by cost differences without incurring the great expense of a detailed cost defense. If the test demonstrates that a cost defense would fail, then the detailed study may be avoided; if, on the other hand, it demonstrates that the defense might prevail, a comprehensive study must be made in order to achieve legal recognition. The pro-justification test has no legal standing. Basically the test involves the same procedures as the detailed study except that estimates of costs, rather than actual costs, are used. By estimating costs without actually gathering empirical data, the ex: peace of gathering and analysing the data is eliminated, thus allowing the test to be a rather inexpensive exercise. In making the estimates, costs should be purposely exaggerated to the extent that there is no question that actual costs are less than the ones used in the test. If the resulting cost differences, based on admittedly overstated estimated costs, are smaller than price differ- ences between customers or transactions, then ”it is a foregone con- clusion.that smaller and more precise unit costs will not produce the required cost differences.“8 That is, if the test determines that price differences still exceed cost differences, the prices can never be justi- fied by more precise cost calculations and a more elaborate cost defense is doomed to feil. Thus, use of the test will reduce the probability that a detailed cost justification will fail by eliminating those situations where the maximum obtainable cost differential is still too small. This should abiliting, Herbert 0., "Cost Justification of Price Differences," Mena ement Controls, November, 1965, p. 227. 69 not be taken to mean that in all other situations the cost defense would succeed; many situations will still occur where the government will not accept the cost study because of defects in its preparation. Thus, the test just indicates the potential inherent in any given situation. As an illustration,9 consider the case of a manufacturer who nor- mally sells his product out of premanufactured inventories for $300 per unit, f.o.b. shipping point, in annual volumes ranging from 50 to 200 units per customer. He is offered a contract for a guaranteed annual volume of 500 units if he will grant the purchaser a 6% volume discount. The producer does not want to reduce the price he charges his other customers. Since sales from inventory preclude justification by differences in.manufecturing costs and since the producer does not bear any delivery costs, the only source of cost differences is sales solicitation and service expense. It is estimated that each sales call on a customer actually costs about $50, regardless of the customer's annual volume. Fer purposes of the test however, the cost per lowhvolume customer is overestimated at $70 and per highpvolume customer at $80 per call. It is also estimated that salesmen typically see customers according to the following call schedule: Annual Number of volume Sales ls 20-80 10 81-130 20 131-and up 30 9Adapted from example by Whiting, Ibid., p. 227. 70 The following cost/volume relationships can be developed: Annual volume per Customer :2 L02 222 5.02 Sales value (list) $15,000 30,000 60,000 150,000 Number of calls 10 20 30 30 Cost per call $ 70 70 80 80 Total call cost $ 00 1 #00 2 400 2,400 Call cost as i of sales % 5.6 5.6 t.0 1,6 Cost differences vs. 100 vs. 200 - .6 3.0 2.# EEK 0 As may be seen by this test, a 6% price difference cannot be justi- fied in terms of costs, even.when they are purposely overstated. Thus, no full-blown cost justification effort could ever justify the discrimi- nation. Example of Qgst Justification10 The situation. A cost study is to be done for a company which manu- factures and sells a product in a highly competitive market. The product is sold to three classes of distributors who resell to dealers who, in turn, retail to the ultimate purchasers. A11 distributors perform the same marketing functions; they differ only in terms of minimum seasonal orders, minimum reorder quantities, and inventory maintenance policy. The distributor agreements provide that Class 1 distributors may order any quantity, that Class 2 distributors must buy at least 6 units, and that Class 3 distributors must buy more than 18 units. The discount schedule is based on Class 1 prices of $225 per unit. Class 2 and Class 3 distributors are charged, respectively, two percent and four percent less than Class 1 prices or $220.50 and $216.00. 10Adapted from illustration by Cohrs, James C., "A Cost Justifica- tion Study," Mana ement Controls, November, 1965, pp. 230-233. For more detailed actual studies, see Taggart, Herbert F., Cost Justification. 71 All production is made for stock and all shipments are f.o.b. ship- ping point. Consequently, manufacturing costs and differences in freight costs cannot be used to support price differences. The only costs that could be used for this purpose are those incurred to receive, process and ship an order. These activities involved the operations of the following departments: order (order entry and billing), traffic (pre- paration of bills of lading), accounts receivable, credit and collection, and shipping. Selling costs are incurred on a commission basis which differs in size‘for sales to each class of distributor. Study program. A.pretest, based upon the average quantity pur- chased by Class 3 distributors, indicated that a cost justification might succeed. Therefore a formal cost justification study was under- taken. The study proceeded along the following lines: 1. Analyze the organization of the company to determine which departments were involved in the order processing and filling functions, determining lines of communication and responsibility in order to establish bases for the allocation of overhead. 2. Flowhchart the order entry, processing, shipping, invoicing and collection procedures to ascertain the personnel involved and the amount of time they spend in the order processing and filling function. 3. Obtain volume statistics relative to orders, back orders and line items processed. h. Accumulate both direct costs and overhead for each department engaged in the order-processing procedure. 5. Study shipping department operations and determine the typical times required to ship various size orders. 72 6. Develop an hourly direct-cost rate and an overhead rate (based on projections of annual shipping volume) for shipping depart- ment operations. 7. Interview sales and marketing personnel to determine the extent to which their activities were identifiable with particular transactions. 8. Reallocate corporate overhead expenses originally accumulated by function to departments involved in the order-processing and filling functions according to equitable and generally accepted bases of allocation. 9. .Accumulate total costs of the order-processing and filling functions and compute the cost of filling various sized orders. 10. Analyze the orders received for a test period to determine whether distributors were ordering in specified quantities. jindigg_. It was determined that Class 1 distributors typically ordered three units, Class 2 distributors six units, and Class 3 dis- tributors 18 units. The analysis revealed that the fixed costs of processing each order was approximately $10, regardless of the number of units ordered. The time necessary to fill each order shipped during representative periods was determined by actual count.‘ Order-filling costs, based on the statistics thus accumulated, ranged from $1.8h on a 1 unit order to $2.50 on a 18 unit order (See Ekhibit IV-l). The price discrimination between each Class and the differential costs determined in the study are stated in terms of shipments of Class 3 distributor*quantities (Exhibit IV;2). As may be seen in the exhibit, the discrimination between Class 1 and Class 2 or Class 3 distributors has been.fu11y justified. However, the discrimination between Class 2 73 Ethibit Ivel. Fixed and Variable Costs Per Order, Order Processing and Filling Variable Costs Per Order Number Order 2253; Costs Per Order* of Units Shippipg Department Order department $h.6h Accounts Receivable 1.35 1 $1.8b $1.05 Credit and Collection 1.30 3 2.00 1.10 Traffic 2,21 6 2.25 1.20 $10.00 18 2,59 l.h0 Note: ‘ These costs are relevant to the determination of cost differ- ences in functional classes only becausem mopgeps deter- mine the functional classes. Exhibit IVeZ. Comparison of Price Differences with Cost Differences Based on Class III Minimum Quantities Class I Class II Class III $ $ f Aggregate Sales Value of 18 Units M M W Price Differences "' vs. Class II - - 81,00 Differential Costs Order Processing Fixed 60.00 30.00 10.00 Order Processing variable 6.60 3.60 1.40 Shipping . 12.00 6.75 2.50 Sales Commissions (3 7%, 5%. and 4%) .2249 we: 2.52.2 Totals fizgo 238,80 162.l+2 Cost Differences vs. Class I 123 30 192.68 vs. Class II - 62,38 Excess of Cost Differences Over Price Differences vs. Class I vs. Class II "I I "I I {1" N \o o to o O Ox oo 7b and Class 3 distributors has not been justified. Principles of Robinson-Patman Accounting Unquestionably, the existence of generally accepted "principles" of accounting for Robinson-Patman.matters would aid all parties involved with such proceedings. The respondents to such complaints or the large number of sellers who might desire to explore their costs in connection with a review of present or future price discount schedules would then have a guide to help them prepare their study, and the government would have a standard by which the study might be judged. Unfortunately, generally accepted principles have been hard to derive, for the Commission has not seen fit to publish such standards and the litigated cases do not easily lend themselves to analysis. Private investigators have examined the adjudicated record and have concluded that each case involved improvisation and that few clear precedents were established.11 Nevertheless, several investigators have derived accounting principles which have wide applicability even though they are very general. A great deal of creditability may be given to these principles because several investigators have each derived them, even though the investigators came from diverse business backgrounds and conducted their work independently, The-backgrounds represented by the investigators include the fields of law (Rowe), accounting (Taggart), economics (Edwards), and marketing (Lynn). The research of Rowe, Edwards, and Taggart was done over the same time-span with little cross-reference llEdwards, op,cit., p. 591. 75 or communication.12 For clarity, the principles of accounting for Robinson-Patman matters enumerated in this chapter will be numbered so that proper credit may be given to the investigator who derived them.13 In addition, an explana- tion of or setting of the proper context precedes some of the principles. Differenggs in price. The cost analyst must first determine what constitutes "differentials" in price in order to establish the boundariies for the cost differences to be justified. A "differential“ in price is the numerical difference in price paid by two customers who purchased at different prices. In general, the price of a commodity is measured by'the exchange value determined by arms-length transactions between ‘buyer and seller. This measurement pierces the form of the sales trans- action and includes adjustments for freight charges, discounts, allowances, 12Neither Taggart nor Edwards refer to Rowe in their books. Edwards states that he did not rely on Taggart, (Edwards, Ibid., p. 239.). Taggart does not mention this work of Edwards. Rowe started his project before Edwards' book or Taggart's book was published (Rowe, Fredrick M., Price Discrimination Under the Robinson-Patman Act, p. xv.). 13Each of the listed principles has been defined by the following authors in the cited works: Rowe, F. M., Price Discrimination Under the fighinson-Patman Ac ; Taggart, H. F., ggst Justification; Edwards, C. D., fie Price Discrimination Iaw; Lynn, R. A. "Is Cost Defense Workable," Journal of Marketi , January, 1965, pp. 37-43. Principle Rowe Ihggart Edwards .EZEB p. p. p. p. 1 275 557 ~ #4 - 2 277 529 594 #0 3 280 - 595 #1 4 281 539 596 39 5 281 539 597 - 6 293 540 597 39 7 291 5&0 - - 8 293 438 - he 9 281 538 596 - 10 - 540 600 - 11 293 5#l 607 #1 12 — 5u2 - - 76 rebates, and additional services provided by the seller. Cash discounts are held to be credit costs and are not considered in determining the value of the transaction. The substance of each transaction determines the value of the basic consideration which passes from buyer to seller. Thus, Principle 1 may be stated as follows: 1. ”The price is the amount which the buyer agrees to pay the seller fer goods, including as 'goods' whatever transportation find additional services] the seller may furnish,” net of dis- counts, allowances, and rebates.ln Two problems which still exist in price determination occur when customers purchase a line of products at an aggregate or average price and where the goods purchased are substantially of like grade and quality but differ in minor respects. These problems may be resolved by adopt- ing pragmatic assumptions which are theoretically sound and are reason- able. Fbr example, the costs of producing the minor differences might be computed and used to adjust the prices of dissimilar goods to make them comparable. Segregation of group . Once the prices charged to customers, quantities, or volumes are determined, no problem exists for segregating the price-favored groups from the unfavored ones. Although this group- ing is the one a respondent would like to cost justify, care must be taken to insure that members of each group have similar characteristics and that all members having these characteristics are included in the group. For example, large orders or volumes cannot be included in groups containing small orders or volumes; customers being serviced by a given procedure may not be included in a group of customers being 1“Advisory Committee on Cost Justification, Report to the Federal Trade Commission, p. h. 77 serviced by another procedure. In addition, all customers who are serviced alike, and all orders or volumes of a given size must be in- cluded in the same groups. Two principles related to the segregation of groups are: 2. The characteristics of all members of a given cost-group must be homogeneous insofar as quantities purchased and methods of servicing are concerned. 3. All customers and purchases in given quantities and volumes having characteristics similar to those already segregated into a particular group must also be included in that group. Allocation of costs toggpppp_. Conventional accounting systems usually contain accounts titled and classified according to the nominal income statement basis rather than by the functional relationship be- tween input and output. Therefore, costs are usually recorded first on a nominal basis; a basis which is not too useful for cost justification purposes. If a firm ever does desire information about functional or departmental costs, they must be determined by allocating the aggrega- ted nominal costs to particular functions or departments. For justifi- cation purposes, the allocation of costs from nominal accounts to func- tional accounts must be both theoretically sound and based upon empirical evidence as to the creditability of the allocation. Four separate principles have been derived which relate to cost allocation. They are: u. Price differentials may be justified by cost differences which result from an allocation of the total costs of each operating function to customers or sales classes on appropriate bases. Sales classes must include all customers who receive like 78 services or who are otherwise of homegeneous character. Ap- propriate bases are those which have received general accept- ance in conventional cost accounting practice and are supported by a’ logical causal relationship. All costs, both fixed and variable, can be initially assigned to functions without dis— crimination. Price differentials may be justified by cost differences re- sulting from the allocation to commodities or territories of costs directly related to sales by using sales dollars as the allocation basis. For example, salesmen's commissions could be assigned in this manner. However, general overhead costs may not be assigned on the basis of sales dollars, partly be- cause price differences between customers will cause cost differences to appear in the items being analyzed. The costs assigned on this basis will be proportional to the price charged; therefore, the cheaper price will have an apparent cost savings. The cost of personal services can be satisfactorily allocated only by using time or a time proxy as the basis of the alloca- tion. Therefore, time studies must often be made. However, substitute bases which approximate time may be used in place of time studies. For example, clerical workers' activity can be measured in terms of the number of invoices or invoice lines or whatever the appropriate work unit may be. Salesmen's time may be proxied by the number'of calls made. A proxy for superb visory time may be that time is spent in proportion to the wage costs of subordinates for which the supervisor is respon- sible. 7. 79 Only minor cost elements may be assigned to customers or customer classes on the basis of management estimates which are unsupported by verified, tangible evidence. The assignment of’major cost elements must be so supported, irrespective of the fact that such assignments have long and honorably been used by the company in preparing internal accounting reports. Other Principle . Five other principles have been derived from the adjudicated cases. Since these are selfkexplanatory and have little in common, no attempt will be made to subclassify them. These other principles are: 8. 9. Only costs actually incurred can be used for justification purposes. This specifically excludes as unacceptable all figures imputed according to the opportunity cost concept, i.e., that a particular function may be charged with an amount re- sulting from a lost opportunity. Thus, implicit rents and the cost of capital tied up in particular inventories can not be used in the cost defense.15 A "total cost" approach must be used in determining cost difk forentials. That is, if particular business functions or business facilities benefit sales to certain customers or customer classes, the costs associated with the functions or facilities must be proportioned to the customers or classes, 15This principle was derived from two cases in which respondents attempted to justify prices by using a 21%.and a 8% cost of capital figure to determine the amount lost by having funds tied up in inven- tories associated with particular groups. In a personal interview, Dr. Willard Mueller, Director of the Federal Trade Commission's Division of Economics stated that his division objected more to the amount used by respondents than to the opportunity cost concept itself. He stated that if the respondents had used a "more Christian" figure, such as 6%, they may not have met such resistance. 10. ll. 12. 80 irrespective of the fact that the costs are fixed and remain unchanged by the acquisition or loss of an additional customer or order. A "differential" cost approach may not be used to justify price differences. However, costs associated with a function which are totally inapplicable to certain customers or customer classes need not be applied thereto. For example, sales effort and advertising devoted to branded goods need not be allocated to unbranded goods. Price differentials may be justified by uSing analytical methods and data other than the methods and data provided by the respon- dent's regular accounting procedures. The respondent has the option of using the data produced by his existing accounting system or he may ignore those data. likewise, the goverment may disallow conventionally generated data. Data generated by auxilliary methods may be wholly admissable. Statistical sampling is a completely acceptable technique for providing data at a reduction in the time and costs required for an analysis. Samples may be used to determine the time covered by the study, to selected representative customers, territories, orders, invoices, and other aspects of the dis- tribution or manufacturing process. Acceptable statistical techniques may not contain any built-in bias; Random sampling is preferred. The cost justification should be aimed at convincing the Federal Trade Commission's accounting staff. The Commission rarely ignores its own staff in matters of accounting, and the courts accept the Commission's judgments as an expert on 81 accounting matters. The Commission's accounting staff is available for consultation in preparation for a cost study. FTC advice. Advice from the Commission's accounting staff may be requested by any person interested in obtaining information related to a cost justification study. while this advice cannot take the form of the government directing the study, it may provide valuable insights into segregating customers, quantities, or volumes into acceptable groups, into allocating costs between groups, and into selecting time- periods, customers, territories, or activities as representative samples. In this regard, FTC advice differs from advanced tax rulings in that the advice does not bind the commission. However, if the advice is sought and followed, it is highly unlikely that the Commission will later find fault with techniques used at its suggestion. 3.2.1232 A.review of the record of adjudicated cases indicates that cost justification.has been used primarily to defend quantity and volume dis- counts. Analysis of these cases does indicate that generally accepted principles of accounting for Robinson-Patman.matters exist and that these principles must be considered if the respondent to charges of price discrimination has hopes of justifying his price differences. It is evident from the nature of the cost defense that a general approach to cost justification exists. Conceptually this approach offers no prOblem although many exist in practice. The use of a prejustification test which can predict the failure of a defense can eliminate some of the hardship of attempting to justify a nonjustifiable price difference. Anya one seriously considering a cost defense would be wise to perform the test before incurring the expense of a full-blown justification effort. CHAPTER V STRATEGIES FOR ROBINSON-PATMAN ACCOUNTING Although useful for cost Justification purposes, lists of accept- able principles, practices, and specific accounting procedures like those described in the previous chapter are insufficient in and of them- selves as guides for business managers interested in protecting policies of price discrimination. Undoubtedly, additional questions mmst be asked about the cost defense even though the principles are understood, be- cause, at best, the principles only answer a specific "what to do" ques- tion. Still unanswered are the "where, when, and why” questions. That is, a businessman concerned with Robinsoinatm n matters must consider a grander, more inclusive approach to the problem of a defense than is available in the current literature. St ate for a Robinsoinetman fense Once a businessman embarks upon an active policy of price discrimina- tion, he ought to be prepared to receive inquiries from the Federal Trade Commission. In order to derive information useful for answering these inquiries, a strategy of defense should be determined. In this regard, a strategy is a predetermined course of’action.deliberately focused on winning an advantageous position over an opponent and which is calcu- lated to counteract actions taken by the opponent. A Robinsoinatman strategy would provide information adequate for the defense of price discrimination charges. At a minimum, the strategy would have to take cognizance of the following interrelated aspects of a 82 83 Robinsoinatman defense: A. Objectives of the defense must be explicitly understood. 1. 2. 3. 4. 5. 6. The price differentials are justified by cost differentials. The price differentials do not injure competition. The price differentials are not related to goods of like grade or quality. The price differentials are not related to goods traded in interstate commerce. The price differentials are not related to goods in a deteriorating market situation. A defense may include a combination of the above. B. Methodology used to gather data must be implementable. 1. 2. 3. 4. Data are gathered as part of the regular routine using the existing information system. Data are gathered by Special analysis of existing informa- tion system. Data are gathered by special analysis using an auxiliary information system. Data are gathered using a combination of the above. C. Timing of effort must be predetermined. l. 2. 3. he 5. Data are gathered before charges are filed. Data are gathered after charges are filed. Data are gathered both before and after charges are filed. Data are prepared for use early in administrative procedure. Data are prepared for use late in administrative procedure. D. Financial aSpects of the defense must be considered. 1. 2. The expected benefits to be derived from.fighting the charges must be estimated. The expected costs to be incurred in fighting the charges must be estimated. Only by considering all of these critical variables will a businessman have a basis of understanding sufficient for making an intelligent busi- ness decision. In this regard, the criterion used most often for judging a decision is whether or not the decision will have a positive net ef- fect on the firm's earnings or on the stockholder's wealth. The decision to fight charges of price discrimination on "principle" is not a busi- ness decision but rather an ethical one and, as such, is not subject to a business decision analysis. 84 The limitation of the lists prepared by Taggart, Lynn, Rowe, Edwards, and others now becomes clear. These lists give detailed con- sideration only to the objective of the defense (cost justification) and the methodology. Little attention is given the timing of the effort and practically none is devoted to the financial considerations directly related to any course of action. Timing of the Cost Justification Effort There are two aspects to the timing of cost justification effort. The first is related to the time for collecting the justification data. Previous writers have pointed out that it is better to prepare the cost justification before charges of price discrimination have been filed by the Federal Trade Commission. This stems from the psychological value inherent in these studies, because they are an expression of the "good feith" of the respondent. A prior determination of cost differences‘méy be given greater consideration than a determination made after the charges are filed. However, this factor is extremely hard to evaluate. A more obvious reason for collecting cost data prior to the dis- crimination is to determine the extent or limits of price concessions which legally may'be granted. However, this point may be a naive one which ignores the realities of the business world, for it is most prob- able that price concessions are granted by line executives in the heat of an exchange transaction while cost justifications are prepared by staff unpleyees only on the direction of a line comander. Nevertheless, exploratory cost studies certainly would be beneficial tools for the price manager, and may be prepared even if no discrimination is con- templated. The time for gathering data has implications for the design of the .l I llvl I]! 85 data gathering information system. No prior consideration is needed only if the data is to be collected after charges are filed. In all other cases, the information system must be designed to record, sum- marize, and retain the required data until the time when it is needed. A second aspect of the timing of cost justification effort is con- cerned with the presentation of the defense. Timing, in part, is deter- mined by existing institutional characteristics of the law and the admin_ istration of the law. That is, the law determines the minimum require- ments of data needed to support a defense, and the law's administration determines the methodological procedures which must be followed in pre- senting the defense. The timing implications can only be understood in terms of the procedure used to process Robinson-Patman charges. Financial Considerations Any Robinson-Patman price discrimination defense will require that a commitment of human effert and material be expended in order to achieve a particular goal. The magnitudes of both the effort committed and the goal sought may be measured in.monetary terms. These two items repre- sent the financial considerations of the cost defense. Defense expenses may be conceptually subclassified into the effort required for data collection and the effort used in data presentation. For any terminal settlement to be achieved, a measurable amount of time, effort, and material may be estimated for both these functions. Ex- pense estimates may be made by approximating the physical effort re- quired for a particular function, assigning a price (cost) to each physical unit of the effort, and multiplying. Obviously, this is easier to conceptualize than to implement. However, an explicit statement of 1These procedures are described, Supra., p. #1. 86 these magnitudes will lead to sounder business decisions than could be made subjectively without estimates as long as some knowledge of the variability of the estimate is available. A third class of defense expense exists. Instead of being an "out of pocket" expense like previously discussed items, this expense is an opportunity cost concept. It pertains to the opportunities which are lost once a given course of action is decided on to the exclusion of other courses of action. An example would be the contribution margins lost due to the customer who will no longer purchase from you, given the decision not to grant him a price concession. These opportunity costs are probably most significant in magnitude and therefore may be a determining fector in any decision regarding the cost defense. The magnitude of the total defense expense is a function of the objective of the defense, the timing and method of data collection, and the timing of data presentation. However, the functional relationship between total expenses and these variables is not direct. Some of the expenses are an increasing function of data collection time; as when more time passes using a built-in system of data collection, more data collection expenses will be incurred. Some expenses are a decreasing function of procedural termination time: as when more time passes during an administrative procedure, which is expected to terminate un- favorably for the respondent, the less contribution margin will be lost. Some expenses are a constant function of the method used to gather data; such as the expense of special analysis based upon each inquiry. The total size of this financial consideration is partly within the control of management. That is, the reSpondent decides on the magnitude and nature of the data collection effort, and has the Option of settling 87 the administrative procedure at most of the terminal points. However, it must be realized that the functional relationship between defense expense and critical variable is so diverse that by minimizing one type of expense (say, data gathering costs), the firm may be maximizing another type of expense (say, lost contribution margins). The size of the defense may be either large or small. While some defenses have cost as much as one half a million dollars,2 others may be minimal, as for example, when the firm makes no prior cost study and accepts a consent decree. A chief advantage that cost justification has over other defenses is that if only a portion of the price differential is justified, that portion is legal and any subsequent order to cease and desist issued by the Federal Trade Commission will exclude the legal portion. It is often easier to prove that a portion of the price differential is justi- fied than it is to prove that all of the differential is justified. Therefore, in terms of alternatives, the firm may seek to achieve some- thing less than total justification. The benefits associated with a Robinson-Patman defense can only be determined by considering the expense of the alternatives to making the defense (using expense in the broad sense). That is, given the decision to discriminate in price, certain expenses can be expected to result once charges of discrimination are filed whether cost justifica- tion or some other defense is prepared. A different set of expenses can be expected for each of the several defenses (or for each alterna- tive way of preparing each defense). For example, consider the alter- 2Lynn, Robert A., "Is the Cost Defense workable?“ Journal of Marketi , January, 1965, p. 39. 88 native of cost justifying or of signing a consent decree, i.e., not preparing any defense. The benefit of using the cost justification defense is the expense savings which result from having used the de- fense. The incremental costs, benefits, and penalties associated with this alternative (consent settlement vs. cost defense) can be concept- ualized as follows: Estimated losses of consent settlement Business not obtained because concessions not granted*.....xxxxx Estimated costs avoided by decision Accounting analysis avoided, on present data.....xxxxx Legal fees avoided, present and future’..........xxxxx Executive and organizational energy'saved‘.......xxxxx Goodwill saved by no legal publicity*............xxxxx Future accounting system, data processing, and documentation procedures avoided‘..........xxxxx Net benefit of decision to sign consent decree (+ or -) HEB Note: * amounts are at present value of future streams of yearly cash flows. Since the firm has complete control over its response to a given challenge and since many alternatives exist, several benefits also exist. The benefit of choosing one of two alternative actions is the incremental cost savings'which result from having chosen that action. A pairing-off of many alternatives will produce many benefits. However, only one bene~ fit can be associated with a given alternative. For example, the alter- native of having either a cost defense or no defense will have a benefit different than the benefit associated with the alternative of'having either a cost defense or a no-injury-to-competition defense. The financial considerations of alternative defense decisions may be used as a guide for evaluating alternative decisions., In each case, the benefits may be matched with the corresponding expenses. This is nothing more than the application of conventional capital budgeting logic. 89 Stgategz if a workable Defense As conceived by previous writers, a successful cost defense effert was one which ended in the complete justification of the price differen- tial.3 However, this concept has a severe limitation, for, is a suc- cessful defense successful at any price? Ordinarily not. This would be true only if the firm were in business to vanquish the government in Robinson-Patman litigation. However, since most firms are financially motivated, management must consider the implications of the defense on the firm's earnings. Victory at any price is obviously a problem of suboptimisation, and as such, is not a feasible alternative in light of major corporate goals. Since the portion of price differential which is justified becomes legal, not all cost justifications which ended with cease and desist orders having been issued were failures or were unworkable. That is, even if a cost justification effort terminates by entering a consent decree agreement, it still may have been a successful defense if the remaining hoped-for but unachieved benefits (that portion of the price differential that has not been justified) are smaller than the efforts that would have been required to achieve "total success". On the other hand, "success" at the hands of the United States Supreme Court may in fact be a failure if the benefits to be derived are smaller than the effort expended. Therefore, a strategy for a workable cost defense should be based upon all of the financial considerations (which in turn are based upon the method of data collection, the timing of data collection and pre- 3This‘idea is implied by Lynn, Ibid., p. 37. Also see Repgrt of the Attorney;Genera1's National Committee to Studz_the Anti-trust Laws U. S. Government Printing Office, Wishington, 1955, p. 171. ' 9O sentation, and the alternatives considered). A 10gical starting point would be to determine limits of the expected benefits by examining the expenses (opportunity costs) involved with the decision to provide no defense and to accept a consent decree. This would set the maximum amount of expenses to be considered fer a defense. In many cases, it may become obvious at this point that any cost justification effort would entail too large an expenditure. In this situation, the workable defense would be no defense. A cost justification strategy would involve a unique set of vari- ables related to the method fer gathering data, the time for gathering data, and the time for presenting data. Since each variable is indis- crete, the number of strategies which are possible is infinite. For example, even assuming that there are only three methods for collecting data, only three times fer collecting data, and five points for pre- senting the data, fortybfive alternative strategies could be used. Thus, it is impossible to consider all alternatives given a particular situation. Each decision maker would have to consider the major condi- tions of his particular situation (such as number of competing customers, information already provided by the infermation system, methods of dis- tribution) in isolating critical variables for consideration. Then, by considering the most obvious factors, discrete alternatives may be developed. General Framework of Analysis The factors facing each firm are so unique that few generalizations about particular accounting strategies would be meaningful. In this regard, it may be‘noted that no two firms are identical with respect to particular circumstances such as time, place, person, and process, and 91 therefore each firm would have to implement procedures in a unique fashion in order to achieve the ends desired. However, a general ap- proach to the problem can be followed for providing a guide to business- men desiring information about a strategy for a cost defense. In each situation, the first step is to determine the extent of the discrimination, while also taking cognizance of the type of discrimina- tory practice followed and the most probable sources of justification data. A prejustification test is then run (as described in the previous chapter) and, if the results are positive, a program or programs for accomplishing the study are prepared. These programs will list the alternative steps and procedures fer making the study and the approxi- mate physical input required for the study to be made. Financial con- siderations for a given alternative pair may then be estimated and capital budgeting techniques employed to evaluate the decision. To help explain the use of capital budgeting techniques, an extended illustration follows. In this illustration, all the effort involved in the cost justification study will be referred to as a Robinson-Patman (R—P) accounting system. An application of the capital budgeting framework requires an exp plicit determination of the costs involved in installing and operating the cost justification information system, the benefits to be derived from the system, and the method for comparing the costs and benefits. When the costs are compared with the benefits, a rate of return can be computed. This rate can then be compared to the expected rate of return from.alternative investments, and/or with the cost of capital. If the rate of return of the cost justification project is higher than the cost of capital and the expected return on alternative projects, and if 92 funds are available, then the system should be implemented. Alternative Events In order to determine the expected costs and benefits resulting from the adoption of a given decision, all of the uncertain alternative events which may result must be examined. Once a firm undertakes a policy of price discrimination, several different events may occur, notwithstanding other actions that the firm may undertake. For example, the following events may occur with or without the firm having imple- mented a cost justification study. a. The firm may enjoy the fruits of discrimination unimperiled be- cause the unfavored customers never detect or complain about the discrimination. Alternatively, the injured purchasers may complain to the Federal Trade Commission and/or initiate a private suit for treble damages. b. Once the FTC is involved, the firm may convince the FTC that it has not discriminated or that the discrimination is justi- fied and legal. Alternatively, the FTC may not be satisfied with the explanation and may file a formal complaint. c. One response to a formal complaint is that the firm may sign a consent decree and stop the alleged discrimination. Alter- natively, the firm may fight the charges, relying upon one or more of the six defenses to the charge of illegal price dis- crimination.4 d. If the firm wins its fight in the FTC and the decision is not appealed to the courts, it may continue to discriminate. Al- ternatively, the firm may lose its fight with the FTC and it h§u_£_ra’ ’ p. 30 93 must stop practicing discrimination or it may appeal the decision to higher courts. If all legal remedies fail and unfavorable decisions are rendered, the firm must stop dis- criminating through its pricing policy. e. The firm may win the private suit for treble damages that was instituted at any time after the customer discovered that he was being discriminated against. Alternatively, the firm may lose the suit and pay the damage plus legal costs. These uncertain alternative events can be visualized better by reference to the decision tree in Exhibit Vel. The two main branches of the tree represent the alternatives available to the firm by using R~P accounting (upper branch) and by not using R-P accounting (lower branch). The decision tree shows both the decisions that the firm may make and the uncertain events that may occur in response to the decisions. Each main branch of the tree contains a sequence showing all the events that can possibly occur. These sequences are arranged in an order parallel to the alternatives just enumerated. The first junction of each branch shows the four basic possibilities; a. The customer neither complains to the FTC nor files a private suit, b. the customer files a private suit only, c. the customer complains to the FTC only, d. the customer complains to the FTC and files a private suit. The remainder of each branch follows through the sequence to its natural conclusion. It should be notice? ihtt the alternative events facing both main branches are identical. Each of the forty-eight uncertain alternative events has some 94 seen» .0 I: a 8 3.5.: «’2: I: 833 unhunnuuuuulmmaluuepeo UdSG d I'loavlflfi- a1ummnwnuu- e anew euepwuu be flunk eeue queen-u: uwww.uue>« eaan_lugh ens-Ina aeouuesuonhl nausea e>uueaueuu< "non amnesia conditional probability of occurring, i.e., the probability that the event will occur, given that the event immediately preceding it has already occurred. These probabilities may be estimated objectively by statistical evaluation of historic incidence or subjectively from the intuition of the decision maker. In addition, each of the fortybeight uncertain alternative events has a conditional cost associated with it, i.e., if that particular event were to occur, X dollars of cost would be incurred. Generally, these costs will be incurred for the necessary legal and accounting analysis required for that particular event to occur. Although a con- tinuous range of costs may exist for each event, in this paper, an ar- bitrary discrete value will be assumed for each event. These costs, like the probabilities, can be estimated by reference to historical precedent and/or intuitive inclination. The decision tree now contains a description of all the uncertain alternative events associated with a decision to have a discriminatory policy (Exhibit V;l), an estimate of the conditional costs of each event (Exhibit V-Z), and an estimate of the conditional probability that each event will occur (Exhibit Ve3). The expected cost of each uncertain event and of each decision may now be calculated. The expected cost at any point on the decision tree is the product of the conditional cost at that point and the product of all the probabilities along the branch up to that point. The expected benefits to be derived from making any decision depends entirely upon which decision is being made; i.e., what alternatives are being considered. With these data, many different problems may be evaluated. The re- mainder of this paper will be devoted to an analysis of questions such as: 96 1. Should the (hypothetical) firm institute a Robinson-Patman Accounting system? 2. If so, what is the maximum amount the firm can afford to spend for such a system? 3. If not, how'large must the probability of an unfavorable event get befOre the firm can profitably institute a RP? system? h. If not, how large must the conditional cost of an unfavorable event get before the firm can profitably institute a R-P system? Spggific Agglygis An analysis of the proposal to install a RpP accounting system requires an explicit determination of the expected costs and benefits associated with the installation and operation of the system. Benefits. The expected benefits associated with a RF? accounting system are the cost savings which result from having implemented the system. That is, given the decision to have a discriminatory pricing policy, certain costs can be expected to result if an RpP accounting system is implemented, and another set of costs can be expected to re- sult if an RpP accounting system is not implemented. If the costs with RPP accounting are smaller than the costs without 89?, the difference between the two costs, the cost savings, is the expected benefit of adopting R—P accounting. If the costs with RP? are larger than without R-P, no benefits exist and RpP accounting should not be implemented. In terms of the decision tree, the expected benefit is the sum of products of the conditional cost and probability of each event in the upper branch less the sum of’the products of the conditional cost and probability of each event in the lower branch. Conditional Costs. In order for any one of the fortybeight uncer- . tain alternative events to occur, the firm must incur certain legal and accounting costs. Each event has its "conditional" cost; that is, that 97 amount of cost incurred solely because the event occurred. An‘gg‘pggt measurement of conditional costs would involve subtracting the costs incurred before the event occurred from the costs incurred after the event occurred. However, an'gg‘ggtg measurement requires an estimation of the additional effort and materials needed to achieve the event. For example, the firm incurs certain legal and accounting costs even if it does not discriminate in prices. The firm probably would incur more costs if it did have a discriminatory pricing policy. The additional amount is a measure of conditional costs. Fbr any event to occur, a given amount of time, effort, and mate- rial must be expended. Those conditional costs associated with the effert and material used for data collection, and incurred by anyone other than a firm's legal representatives, will be termed ”accounting costs." Similarly, the costs of effort and materials expended by the firm's legal representative will be termed "legal costs." As the event under consideration advances along the sequence of possible events, the conditional costs would tend to be larger. For example, it can'be expected that the costs associated with the event "fight-the-charges-ofhthe-FTC" would be higher than those of the event "No-formal-charge—brought-by-FTC". These higher costs can be expected because more advanced events require greater amounts of accounting and legal effort. In terms of the decision tree, costs increase as events are further to the right and lower on any one branch. Since no particular firm is being studied, the determination of costs for alternative events, although based upon historic occurrences, will be purely hypothetical. In any event, the real costs incurred by different firms in actual situations would not be identical, so the use 98 of hypothetical figures should be no more misleading than the use of values taken from an actual case. The magnitude of costs incurred in real situations would depend, to a large extent, upon the information system and legal representation already existing in the firm. The better the present system and representation are, the smaller will be the con- ditional costs of’uncertain events. Pertinent historic events. The magnitude of costs associated with losing a private anti-trust suit is indicated by an examination of fully litigated cases. The treble damages and fees awarded for Robinson- Patman violations pursuant to Section 4 of the Clayton Act, covering a period from 1936 to 1961, are shown in the following table:5 ggse Judgment Fees 8 $ 57,000 $119h00 b 13,307 2,500 c 3,030 a 30,000 8,000 e 180,000 40,000 __1_: 4.16 .361! 354 000 Average $ 76,62h $16,150 However, the impact of these costs was lessened by a 1964 Revenue Ruling which allows treble damages to be classified as "ordinary and necessary business expenses" for tax purposes.6 Taxes will approximately halve the effect of private suit damages on after-tax income. Public information regarding the costs of a cost-justification de- fense is scanty. Taggart gives four indications of the accounting ef- fort required for elaborate defenses. A cost study report...had been prepared by Goodrich's public accountants...A former Commission accountant... 53u re, p. 33. 6Rev. Rul. 64-224 IRS 196“. 99 has acted as a consultant. work on the study had oo- cupied six men practically continuously for five or 7 six months, and several other individuals part time. An additional study by the same firm involved the following: The new study occupied the time of two principals of the accounting firm plus as many as thirteen other individuals for a period of 5% weeks. These probably were at least 6-day weeks, so some #95 man-days were involved. The recomputations therefore gost a minimum of $25,000, and very probably much more. In another case, the primary study required some 1,200 hours of work by Sylvania's public accountants and 1,800 hours by Sylvania's own staff. The factory warehousing study required 400 more man-hours. In an effort to ascertain cost-justification data as a routine accounting operation, The burden involved in the‘zcontinuous cost accounting7 system is obvious. It was based on detailed time re- ports by hundreds of employees throughout the country, all of which had to be summarized, tabulated, and utilized in making the necessary cost determinations and alloca- tions. Some six tons of special records were accumulated during the 4-3/0 years the system was in operation.10 However, less elaborate cost justifications have been successful.11 Ex- amples of the legal effort required are: The complaint against Champion was dated December 16, l939...Hearings began in April, 1947...The hearings were concluded in April, 1950, after 4,055 pages of testimony and argument...but the Cemmission did not decide the case until July 10, 1953.12 7Taggart, Herbert F., Cost Justification, p. 301. é;p;g., p. 361. 2gpig.. p. 361. 10M. . p. l#73. ltgpgg., p. 38. 12 Ibid., 112. p. 100 The transcript of record, including oral arguments, covered 3,592 pages of whifh approximately 523 were devoted to the cost issue. 3 The Standard Oil case may well stand as the most voluminous cost-justification proceeding in history. 0f its 8,057 pages or record nearly 0,000 were con- cerned with the cost issue.l Cost justification absorbed 783 of a total transcript record of 2,865 pages. It occupied 11 of the 35 days of hearings.1 The complaint bore the date December 21, l949...l,692 spages of record...thg commissions decision...was dated September 23, 19514.1 It was very exhaustive, covering 56 pages of transcript and taking one and one-half hours to deliver.17 In addition, a recent legal publication suggested that a fee of $25 per hour should be the minimum charge of any lawyer.18 Cost assigggent. These observed data were used to estimate the conditional costs for the uncertain events in the decision tree shown in.Ehhibit VQZ. The estimates were made by assigning to each'event two values which represented the hours of needed accounting and legal effort required for the given event to occur. Each accounting hour was coated out of $10;'each legal hour at $25. Since the type of effort associated with a R-P accounting system differs from the effort required to gather data by other means, the hours assigned to events in the upper branch of the decision tree are different than the hours assigned to events in 13Ibid., p. 1&2. lnIbid., p. 187. lsIbide’ p. 2540 16Ibid., p. 365. 17Ib1d., p. 502. 18Committee on Professional Economics, "Suggested Minimum Fee Schedule," Michi an State Bar Journal, August, 1961}, p. 16. 101 m.mNH m.om mama o.o m.~m m.om m.ow 3.0..” mama/l, a...) n.0w3 m.o a.ms N.N+~ N.3H o.m.H Nfiwa Nfia N.N: N.Nm m.m~ N6QH/ ofia Nowhg m.mm «.ma Agog 0003 nvmoo dosages—30 .NI> “rang 102 the lower branch. In addition, efforts, and therefore, costs, accumulate as the events under consideration advances along the sequence of possible events. In the lower branch, the first reaction of the firm to notification of complaints or suits requires 10 hours of accounting and 8 hours of legal effort (above and beyond the normal work levels). The signing of a consent decree involves another four legal hours. Fighting federal charges involves an additional 1,000 legal hours and 2,500 accounting hours of effort. The costs for defending a private suit are assumed to be twice the average fees awarded for lost suits, or $32,000 (of this $2,000 are accounting costs). In addition, those events which repre- sent a lost private suit carry an additional $93,000 for the average damages plus fees. Opportunity losses assigned to the decision "sign- a-consent-decree" are arbitrarily set at $10,000. It is assumed that a R—P accounting system cost $1,000 to install and maintain, and requires two full time men ($8,000 each per year) to oper- l9 ate. Costs for each event in the upper branch are also shown in Ex- hibit vaz. The gonditional probability of uncertain events. Since no real case is involved in this study, the determination of'probabilities for this section will be made entirely by reference to actual historic occurrences and/or general estimates of fact, i.e., no refinement will be made for circumstances peculiar to a particular firm or industry. Nevertheless, it should be recognized that techniques are available which could incorporate special circumstances into the estimate of 19This estimate is consistent with the opinion held by Mr. warmack, cited Supra., p. 23. 103 probability.20 The ensuing discussion will parallel the flow of events as depicted by the decision tree, starting with an estimation of probabilities for events in the lower main branch. This arrangement better fits the actual data to the stimates because, in practice, few firms have implemented R-P accounting. In addition, it is likely that there will be a substan- tial difference in the probabilities attached to each event in the two main branches, even though the events in both branches are identical. This reflects the increased probability that the firm can successfully defend itself against alleged charges if it has data provided by a R-P accounting system. That is, to the extent that cost justification is used as a defense to the alleged charges, the probability of losing the case will be less if the cost data are available than if they are not. Pertinent historic events and estimates of fact. A vital first step in the.process of probability determination is to estimate the num- ber of cases of price discrimination that have actually occurred. A conservative arbitrary estimate made in a previous chapter is that 23,000 interstate purchasers are discriminated against by at least one of their 21 During 1963 and 1964, the FTC received 1,309 and 22 suppliers each year. 1,366 applications for complaint. Using averages derived from actual Federal Trade Commission exp perience, it can be estimated that the Commission acts upon only 32% of all letters of inquiry from the public. In addition, 61% of all fully litigated cases concluded with the respondent consenting to cease 0 2 For a discussion of this procedure, see Schlaifer, Robert, Introduction to Statistics for Business Decisions, Chapter 12. 21Supra., p. 58. 2?Annual Report of the Federal Trade Commission, 1964, p. 19. 104 and desist,23 while only 8% of the cost defended cases were terminated in this manner.29 From the inception of the Robinsoinatman Act until 1961, only about 330 private suits for alleged violations were initiated. Of these, one third have been fully litigated with the plaintiff prevailing in only eight cases. In addition, it has been estimated that cost justi- fication was successfully executed in half of the cases in which it was attempted.25 Up through 1965, only 37 defendants in federal suits have pleaded the cost defense in justification for their price discrimination. About a third of these defenses utilized techniques that were fully acceptable to the Federal Trade Commission. However, successful nonadjudicated cost defenses occurred 16 times more frequently than the total number of formal attem'pts.26 Probability assignment. These observed data and some deliberate guesses, for purposes of illustration, can be translated into conditional probabilities for the uncertain events in the decision tree. The prob- ability of each uncertain event will lie somewhere between 0.0, the certain probability that the event will not occur, and 1.0, the certain probability that the event will occur. The larger the probability, i.e., the closer to one, the more likely it is that the event will occur. Given a subset of mutually inclusive and exhaustive alternative events, 2 3Subcommittee No. 5, House Committee on the Judiciary, 86th Con- gress, lst Session, Repgrt on the Consent Decree Program of the Depart- ment of Justice, p. 7. 2“Supra” p. 21. 2 5.5%" P90 299 320 26 Supra., pp. 21, 54. I f I1} I... If! IIIII I] ..1’l.l|‘! ll..!-. I 105 1.9., all the next events immediately following a junction in the decision tree, the sum of their probabilities will always be one.27 The historic relative frequency of an event occurring has been translated into percentages which serve also as probabilities. In addi- tion, the intuitive estimations are fitted to the events directly as stated. These conditional probabilities are assigned to the uncertain events in the lower main branch of the decision tree shown in Exhibit V-3. The probabilities assigned to events in the upper branch have been subjectively modified to reflect the higher probability of a successful defense of alleged violations resulting from the availability of defense data. The impact will be most evident in the higher probabilities that the FTC will not file a formal complaint, and that the firm will win both the government and private suits. These modified probabilities are assigned to events in the upper main branch of the decision tree in Exhibit V¥3. Evaluation of the decision. Expected cost for each terminal act (an act at the end of each branch) was computed by multiplying the condi- tional cost of the terminal act by the product of all the conditional probabilities along the branch of the tree up to the act. The sum of these expected costs fer all acts in the upper branch is the expected cost of the decision to use a discriminatory pricing policy backed up with data supplied from an RpP accounting system. Similarly, the exe pected cost of the decision to implement discriminatory pricing without any back-up data is found by adding the expected costs of terminal acts in the lower branch. 27Schlaifer, og,cit., pp. 203, 210. 106 mm ow. unheard :ddvhooca Ho unavadnndnowm Eogaoo .m...> “Snag For the hypothetical firm being considered in this paper, the ex- pected costs of the decision without R—P accounting is $7,223 and with an R-P system is $17,909. There is an absolute advantage gained by not implementing a R-P system, and so, given a proposal to implement such a system, any rational profit maximizing manager would render a negative decision. This result has supported the evidence of real situations: few firms have in fact implemented Robinsoinatman accounting systems or have made cost justification studies before becoming involved in anti-trust litigation. But now, because of the explicit statement of assumptions, it is possible to determine exactly why a Robinson-Patman accounting system is not a desirable project to undertake. It may be readily observed that the expected costs of the upper branch are hardly more than the sure costs of implementing an R—P system. This implies that the costs above the value added are small. that is surprising, however, is the small size of expected costs fer the lower branch. A fuller explanation of the negative value of a R~P system will separate those factors re- lated to expected cost magnitude from those related to expected proba- bility'magnitude. The first major reason that the expected costs of an unsupported discriminatory pricing policy was so small is due, in part, to the small size of the conditional cost fer terminal events. To a large extent, conditional costs are in the control of the firm: management can choose the lhmits fbr acceptable expenditures. In addition, some accounting effert and legal representation already exists; therefere, the condi- tional costs for moving to some alternatives is small. And finally, those few high cost events rarely occur. 108 Mereover, the introduction of a R—P system has the effect of lower- ing the extremely high conditional costs (the legal costs and penalties of losing in litigation) and raising the low conditional costs (the paper and procedural work of "keeping track".) However, since the low cost events are more likely to occur than the high cost events, the net effect is to increase the expected costs to the firm. In order for the RPP system of the example to pay fer itself, the high cost events, i.e., the costs of losing government and private actions, would have to be about 5 times larger than indicated in the model. Costs of those magnitudes are not unheard of in Robinson-Patman litigation, and occur quite frequently in other types of antitrust liti- gation. In this same vein, if the opportunity costs associated with signing a consent decree (lost business or profit margins from once favored customers who would no longer trade with the seller) were only $80,000, a not unreasonable figure, a R—P system might be feasible. However, even if the conditional costs of unfavorable events were five times larger or if the opportunity costs were $80,000, Robinson-Patman accounting still may not be a profitable project to undertake: the rate of return on this investment may still be smaller than the return on alternative investments. If resources are limited, available funds should be allocated to projects yielding the highest rate of return. The other major reason that the expected costs of an.unsupported discriminatory pricing policy was so small is because the likelihood of an expensive act occurring was so small. Price discrimination creates three important junctions in the sequence of possible events where major probability implications are determined; where the customer reacts to. the discrimination, where the firm settles with the government, and where 109 the firm settles with the customer. The large probability that the firm will sign a consent decree in settlement of the goverment action reflects the fact that that choice is the least expensive appearing alternative. This option limits the probability of a costly legal action, irrespective of the actual justi- fications involved. The legal costs are real, out-of-pocket costs. In addition, it precludes the possibility of preparing a defacto antitrust case for unfavored customers.28 On the other hand, opportunity costs are rarely used as criteria by which to judge managerial efficiency, and so management tries to minimize real costs, not opportunity costs. The probability that the plaintiff will win a private suit is so small that the expected damages and fees are negligible. This reflects the fact that he must gather information about the discriminating policy, must prove discrimination, and must determine the extent of damages; all hard tasks.29 In addition, he must be willing to incur heavy legal expenses. Few customers will be willing or able to sumount these obstacles. Each of these probabilities help make the expected costs of an unsupported price discrimination policy a relatively small amount. Homer, the fiml determination is rather insensitive to changes in these variables. Even if each unfavorable probability were changed, one at a time, to the least favorable value of 1 and turned into a certain fact, a negative value would still be determined for the decision 28mm loss of a goverment suit is accepted as defacto proof of illegal discrimination by the courts in private suits. 29Barber, Richard J ., "Private Mercement of the Antitrust Laws: The Robinson-Patman Dcperience," George Washingt_gn Law Review, December, 1961, pp. 203, 210. 110 to install a R-P accounting system. That is, even if all unfavored customers complained to the FTC and/or sued for treble damages, the decision to install an R-P system would still be rejected, all other things held constant. This same conclusion would result if, singly, all goverrment actions were fought, or all plaintiffs won their private suits. Only a simultaneous change in several probabilities and/or a change in conditional costs would alter the situation sufficiently enough to make the proposal to install an R-p system desirable. limitations of the example. The preceding analysis has several shortcomixgs, most of men, however, could and would be remedied if the model were applied to a real situation. Although some of these shortcomings have already been enumerated, the most glaring will briefly be discussed below. In the study, historical events were used as a base for determining the conditional costs and probabilities. However, data from past events are pertinent to the analysis only to the extent that future events follow the past. In addition, the data refer to average activity which may not be pertinent to arw one particular firm. A proper amlysis would only consider future costs and probabilities relevant to the particular firm under consideration. time value of money was not considered in the analysis. A con- siderable time difference may separate the cost flows associated with different alternative events, thus causing them to have different utilities to the firm. However, common present-value techniques will allow all flows to be equated at a partiwlar point in time, thus facili- tating a more meaningful comparison and analysis.30 30m... J. "DecisiOn Trees for Decision Making" Harvard Business Review, July-August, 1964, p. 133. 111 All costs and benefits associated with the decision to price dis- criminate may not have been included in the study. Some of these may, in fact, be larger and more pertinent than the items that were considered. For example, no mention was made of the ability for a R-P system to pro- vide data useful for managerial control which generates additional cost savings.31 In addition, the ill effects of lost prestige, goodwill, etc. might be quantified and considered. Qaglusion This discussion describes a technique for evaluating proposals to collect information, and shows that such proposals are amenable to recog- nized capital budgeting evaluation and analysis. As it happened, the decision produced by the model was similar to that derived from an in- tuitive analysis. However, use of the formal analysis allows the esti- mates of variables to be refined, the decision maidng process to be audited, and the effect of changes in variables to be evaluated. 3J'T'aggart believes that these alone are sufficient to mrrant the implementation of R-P accounting, op,cit., p. 547. CHAPTER VI SUM1ARY AND CONCLUSIONS Restatement of the Problem Section 2(a) of the Robinsoinatman Act provides that, given the requisite conditions, it is unlawful to discriminate in price between purchasers of commodities except where the price differences can be justified by'a demonstration that cost differences larger than the price differences exist. Although no mention of accounting is made in the Act, this cost provision surely has accounting implications. Little has been written about all the broad implications for accounting of the Robinson. Patman Act. In the first place, the statistical record of Robinson-Patman cost justification cases is entirely inadequate for determining the scope of Section 2(a) enforcement, the relative importance of the cost defense and the types of practices which have been successfully justified. Not only is the existing record woefully outdated, but no data is publicly avail- able about the nonadjudicated Section 2(a) proceedings, even though the existence and probable significance of these cases has long been acknowb ledged. Partly because of'this lack of data, controversy has arisen over the "workability” of the cost defense. One group of respected critics claims that cost justification is not workable due both to inherent limitations in the nature of accounting and to the administrative pro- cedures used by the Federal Trade Commission, while another group of 112 113 ardent and prominent supporters claims that the cost defense is work- able. An operational definition of’a "workable" cost defense has not been produced by either detractors or supporters. In addition, no overall approach to the problem of cost justifica- tion has appeared in the literature as a guide to those persons interested in the use of accounting as a tool in solving Robinsoinatman problems. very little has been written comprehensively about what must be done to justify price differences, how it is to be done, and when it is to be done. Practically nothing has appeared which would help one decide whether or not a cost defense should even be attempted. No comprehensive approach or decision process has been described which integrates all these pertinent aspects of the cost defense. Statistical Findings The updated statistical record of adjudicated cases shows that 360 Section 2(a) complaints were issued by the Federal Trade Commission from September, 1936 until December, 1965. These are the only violations to which the cost defense is pertinent. Of’these complaints, 218 termin- ated with the issuance of'orders to cease and desist, and 131 were dis- missed because a valid defense had been presented. The cost defense was employed in 37 cases of which 23 ended with the issuance of an order to cease and desist. Evaluation of’the significance of cost justifications used in adju- dicated Federal Trade Commission proceedings can be made by considering both the relative use of the cost defense and the relative success of the defense when it was used. In this regard, a successful defense is one in which the Comission accepted the accounting procedures employed in the study. Whether or not the actual cost differences were larger than 11h price differences is irrelevant for judging the success of a technique because all that can be asked of accounting is that it describe actuality. The expense or difficulty of a cost defense also is not a valid criterion by which to judge the defense because this measurement is meaningless unless compared to the expected benefits accruing because the defense had been offered. In terms of both valid criteria, the cost defense has been signifi- cant in the adjudicated Federal Trade Commission proceedings. On the average, the cost defense was employed in lOflmcf all adjudicated Sec 2(a) cases and in at least a third of these it was successful. The lack of information concerning the Commission's nonadjudicated cases stems.primarily'from the Commission's refusal to allow'researchers access to their files. Nevertheless, it is evident from the scant data released by the commission that a large proportion (68%) of all Federal Trade Commission restraint of trade investigations are terminated with- out fermal action'being taken against the respondent and that a higher than average dismissal rate applies to Robinson-Patman investigations. It is further evident that cost justification has played a large part in the dismissal of Section 2(a) investigations. These conclusions stem from the available data which showed that on an average, 7h§ of all Robinson-Patman legal investigations and 57% of .11 Robinson-Patman accounting investigations terminated without a formal complaint having been issued. In terms of frequency, the nonadjudicated cost inwestiga- tions dropped are more than 13 times as numerous as the cost defenses attempted in adjudicated proceedings (for comparable time periods). Undoubtedly, the greater relative attempts at and success of the cost defense in nonadjudicated cases is due to the feet that the government 115 had already informally rejected the cost defenses which appeared in the adjudicated proceedings. In the past, the cost defense has not played a significant role in private antitrust litigation. That is, its relative use has been low in terms of its potential use. Nevertheless, cost justification has been significantly successful in terms of the times it has been employed. Thus, it does have a potential significance for defense in private liti- gation. ‘ The potential use of the cost defense in Justice Department pro- ceedings is extremely limited because the Justice Department is not very active in its pursuit of Robinson-Patman.violators. However, there is every reason to believe that if the discriminatory practice challenged by the government is in fact cost justified, then the cost defense could be used successfully. The cost defense has been used in attempts to justify mam differ- ent types of discriminatory practices. It is impossible to generalize from.the adjudicated record the relative success of these defenses for particular practices, partly because of the particular circumstances of each case and partly because of the way in which the Commission reports its findings. Yet it is evident that where the discriminatory practice is in fact justified, a cost defense can succeed. A Wbrkgble Qgst Defense From these statistical data it is apparent that the use of a cost defense stands a better than even chance of'winning acceptance by the Federal Trade Commission. In addition to a good probability for success, a workable cost defense must also be economically rational. In this regard, a workable cost defense can be defined as one in which the 116 expense expected to be incurred to make the defense is less than or equal to the expected benefits derived from having made the defense. A cost defense could be prepared by a seller to meet the threat of impending or existing legal involvements brought in response to his price discrimination by a customer, a competitor, a customer's customer, or the goverment. The defense may be made to meet a particular contingency or it could be a general one designed to provide data for whatever action may arise, whether it be a private antitrust suit, a Federal Trade Com- mission investigation, or other threat. In preparing the defense how- ever, the seller must consider the objectives of the defense, the metho- dolog of data gathering, the timing of the data collection effort, and the timirg of the defense preparation, and the financial implications of the defense. Marv alternatives of method, timing, and financing may be combined to produce a workable cost defense. Thus, the price discriminator is faced with a possibility of may different threats and has available many alternative responses to each threat. A wise discriminator will prepare a strategy of defense. In this regard, a strategy is a predetermined course of action deliberately focused on winning an advantageous position over an opponent and which is calculated to counteract actions taken by the opponent. A strategy would integrate all the considerations pertinent to the decision to implement a cost defense, and would have as one of its outputs a decision whether or not the proposed defense was workable. The only time when all of the alternative strategies are open to the seller is before the proposed price discrimination schedule is put into effect. At any time after this, some alternatives have been fore- gone and are forever precluded. Only a limited number of strategies 117 exist when the defense is considered in response to a challenge or com- plaint from the Federal Trade Commission. A workable cost defense is thus harder to find at this point. Framework for Anal is A determination and evaluation of alternative strategies involves the integration of existing knowledge of accounting, financial manage- ment, and statistics. Accounting provides a general methodology which underlies all cost defenses. This methodology involves the separation of the discriminated elements into classes, the determination of costs applicable to each class, and the comparison of cost differences between classes with the price differences between classes. If the cost differ- ences are larger than the price differences, then the discrimination has been justified. This methodology offers no conceptual problem although many practical ones exist. Generally accepted principles have been de- rived for accounting for Robinsoinatman matters. These principles must be considered in preparing the defense if the respondent to price dis- crimination charges has hopes of Justifying his price differences. A test exists by which the assured failure of‘a detailed cost study can be predicted before the justification effort is expended. This pre- justification test follows the same general procedures as the detailed study except that estimates of costs are used instead of actual empiri- cally derived costs. The use of this test can eliminate some of the frustration inevitably encountered in the pursuit of a non-justifiable cost defense. Anyone seriously considering a cost defense would be wise to perfbrm the test before incurring the expense of a full justi- fication effort. Once the possibility for a cost defense has been demonstrated, a 118 program or alternative programs fer accomplishing the study should be prepared.. The alternative programs available depend upon the time in the procedural process when the study is contemplated, the existing circumstances, etc. These programs will list the alternative steps and procedures to be followed in making the study. An approximation of the physical input required for the study may be made, and costed out at the going or expected rate of remuneration needed for comparable func- tions. Care should be taken to estimate the opportunity costs associated with each decision. Probabilities that a given event will occur can be estimated by reasonable inference from historical or intuitive judgments. A decision tree provides a framework by which all pertinent alter- natives may be more clearly visualized and evaluated. This technique is not only a conceptual model but, given the facts peculiar to a given situation, it actually may be used to compute the costs and benefits associated with the alternative. It is proposed in this dissertation that several programs fer ac- complishing cost studies be worked out in advance, that a program.or programs be assigned to each of the alternative events which may occur in connection with a policy of price discrimination, and that recognized capital budgeting techniques be applied to evaluate the alternatives. The implementation of such a proposal will allow'the estimates of varia- bles to be explicit and capable of refinement, the decision making pro- cess to be audited, and the effects of changes in variables to be evalu- ated. As a result, the business decision can not help but to be more rational. APPENDIX I PERTINENT SECTION OS THE ROBINSON-PATMAN ACT Public--No. 692--8hth Congress H. R. 81412 AN ACT To amend section 2 of the Act entitled "An Act to supplement existing laws against unlawful restraints and monopolies, and for other purposes", approved October 15, 1914, as amended (U.S.C., title 15, sec. 13), and for other purposes. Be it enacted by the Senate and House of Representatives of the United States of America in Congress assembled, That Section 2 of the Act entitled "An Act to supplement existing laws against unlawml re- straints and monopolies, and for other purposes", approved October 15, l9lh, as amended (U.S.C., title 15, sec. 13), is amended to read as follows: ”Sec. 2 (a) That it shall be unlawful for arm person engaged in commerce, to discriminate in price between different purchasers of comp modities of like grade and quality, and where the effect of such dis- crimination may be substantially to lessen competition or tend to create a monopoly in any line of commerce, Provided, That nothing herein con— tained shall prevent differentials which make only due allowance for differences in the cost of manufacture, sale, or delivery resulting from the differing methods or quantities in which such commodities are *0 such purchasers sold or delivered: mvideg, however, That the Federal Trade Commission may, after due investigation and hearing to 119 120 all interested parties, fix and establish quantity limits, and revise the same is it finds necessary, as to particular commodities or classes of commodities, where it finds that available purchasers in greater quantities are so few as to render differentials on account thereof un- justly discriminatory or promotive of monopoly in any line of commerce; and the feregoing shall then not be construed to permit differentials based on differences in quantities greater than those so fixed and established; And provided further, That nothing herein contained shall prevent persons engaged in selling goods, wares, or merchandise in com- merce for selecting their own customers in bona fide transactions and not in restraint of trade:. Agd provided further, That nothing herein contained shall prevent price changes from time to time where in re- sponse to changing conditions affecting the market fbr'or the market- ability of the goods concerned, such as but not limited to actual or imminent deterioration of perishable goods, obsolescence of seasonal goods, distress sales under court process, or sales in good faith in discontinuance of business in the goods concerned. ”(b) Upon proof being made, at any hearing on a complaint under this section, that there has been discrimination in price or services or facilities furnished, the burden of rebutting the prima facie case thus made by showing Justification shall be upon the person charged with a violation of this section, and unless justification shall be affirma- tively shown, the Commission is authorised to issue an order terminating the discrimination: Provided, however, That nothing herein contained shall prevent a seller rebutting the prima facie case thus made by shows ing that his lower price or the furnishing of services or facilities to any purchaser or purchasers was made in good faith to meet an.aqually 121 low price of a competitor, or the services or facilities furnished by a competitor. Sec. 3. It shall be unlawful for any person engaged in commerce, in the course of such commerce, to be a party to, or assist in, any transaction of sale, or contract to sell, which discriminates to his knowledge against competitors of the purchaser, in that, any discount, rebate, allowance, or advertising service charge is granted to the pur- chaser over and above any discount, rebate, allowance, or advertising service charge available at the time of such transaction to said compet- itors in respect of a sale of goods of like grade, quality, and quantity; to sell, or contract to sell, goods in any part of the United States at prices lower than those exacted by said person elsewhere in the United States for the purpose of destroying competition, or eliminating a com- petitor in such part of the United States; or, to sell, or contract to sell, goods at unreasonably low'prices fer the purpose of destroying competition or eliminating a competitor. Any person'violating any of’the provisions of this section shall, upon conviction thereof, be fined not more than $5,000 or imprisoned not more than one year, or both. BIBLIOGRAPHY Public Documents Co ress and the Mono 1 Problem: Years of Antitrust Development, . House Document No. 2%: 5th Congress lst Session, H. Res. 11“, 1956. Federal Trade Commission. Annual Regrt of the Federal Trade Commis- sion. Washington, D.C.: U. 3. Government Printirg Office (separate editions) l936-196h. . fights Summagz. Washington, D. 0.: Federal Trade Com- mission (published irregularly) 1958-.1965. . Organization, Procedures, Rules of Practice, and Sgatutes. Washington, D.C.: U. 3. Government Printing Office, 19 3. House Committee on the Judiciary. trust Laws: April 1 , 1955. Remt of the Attorney General's National Qmmittee to 51:29: the Anti- trust Laws. Washington, D.C.: U. S. Government Printirg Office, 1955. Select Committee on Small Business, Pripe fiscrimination, the Robinson- Patman Act, and the Attopnez General's National Committee to Study the Aptitrust Laws. House of Representatives, th Congress, 2nd Session. Washington,D.C.: U. S. Government Printing Office, 1955. . Statistics on Federal Antitrust Activity Tb the Select anenittee on mall Business 02 the House of Representatives. Washington, D.C.: Committee print, 1955. Sub-committee No. 5. Remrt on the Consenp Decree Progam of the papal-talent o; Justipe. House Committee on the Judiciary, th Congress, lst Session. Washington, D.C.: U. S. Goverment Printing Office, 1955. U. S. Department of Commerce. stribut on st countin for Whole- gglim. Bireau of Foreign and Domestic Commerce, Domestic Com- merce Series #106. Washington, D.C.: U. S. Governnent Printing Office, 1939. Eoks agg Mono gpapg Anthomr, Robert N. Management Accounting. Homewood: Richard D. Irwin, Inc., 1960e 122 123 Cost Justification Committee. Report to the Federal Trade Commission. thhington, D.C.: Federal Trade CommissionTYmimeograph), 1955. Edwards, Corwin D. The Price Discrimination law. Washington, D.C.: The Brookings Institution, 1959. Heckert, John B., and R. B. Miner. Distribution Costs. New York: Ronald Press Co., 1953. Lewis, Richard J. A Business To stics Information and Accounti ten for Marketi Anal sis. East Lansing: Michigan State University (Doctoral missertation), 1965. Parkary, John. F.I,Q. Enforcement of the Robinson-Patman Act, 12146-52. New York: Columbia University Doctoral Dissertation , 1955. fibinson-Ppmp Apt mum, New York: Commerce Clearing House, Inc., 19 7. Rowe, Frederick M. Price Escrimination Under the Robinson-Patman Act. Boston: little, Brown and Co., 19 2. Schlaifer, Robert. Introduction to Statistics for Business Decisions. New York: McGraw-Hill Book Co., 19 1. Taggart, Herbert F. Cost Justification. Ann Arbor: Bureau of Business Research (University of Michigan), 1959. Trade Reggtion Repgrter. New York: Commerce Clearing House, Inc. , l . Wixon, Rufus, ed. Accountants' Handbook, hth ed. New York: Ronald Press Co., 19 . Periog cals Abrams, J. "New Methods for Testing Pricing Decisions,” Journal of Mketyxg XXVI (Jay, 1961*), 6-9e Barber, Richard J. "Private Enforcement of the Antitrust Laws: The Robinson-Patman Ebcperience," eorge Washipgton Law Review, XXX Cohrs, James C. "A Cost Justification Study,” Mana ement Controls, walla, N. K. ”Art of Product Pricing: Abstract," Management Review, LIII (June, 1961;), 63-66. Fennelly, D. J. "On the Judging of Mince Pies," Harvard Business Review, XLII (November, 196%, 77-86. 12b Falks, M. "Total Cost Approach to Physical Distribution," Business Mane ement, XXIV (August, 1963), 5-14. Frommer, A. M. "Determining Compensating Sales Quantities for Price and Cost Changes," Egg Bulletin, XLXL (May, 1963), 35-h2. Gargiulo, G. R. "The Use of Statistical Methods in Determining Cost Differentials,” Management Services, II (May, 1965), h5-52. Gray, A. Wt "Broad Scope of the Robinson-Patman Act," Purchasi INI (March, 196“), 90-91. Green, P. E. "Bayesian Decision Theory in Pricing Strategy,“ Journal of Marketi , XXVII (January, 1963), 5-1h. Hinkle, C. L. "Strategy of Price Deals," Harvard Business Review, XLIII (July, 1965): 75‘85- Horton, A. J. "Need for Distribution Cost Accounting," N55 Bulletin, XIV (April, l96b), n1-42. Johnson, L. L. ”Joint Cost and Price Discrimination,” Journal of Business, XXXVII (January, 1964), 32-h6. Kemker, H. "Price Discrimination Under the Robinson-Patman Act," Unive sit of'Florida Law Review, XIV (Summer, 1961), 155-175. Lynn, R. MA. "Is the Cost Defense workable?," Journal of marketi AXIX(January, 1965), 37.42. Magee, J. F. "Decision Trees for Decision marking," Harvard Business Rgziew, XLII (July, l96h), 126-139. Manes, R. P. "Using Computers to Improve Distribution of Service Costs," Journal of Accountan , CXV (March, 1963), 57-60. McAnly, H. T. "Administrative Expense and Profit in Product Pricing," gem; of |Agcountang, CXVI (August, 1963), 33-38. Hellman, M. "Marketing Cost Analysis: Its Relationship to Factory Costing Methods," NAA Bulletin, XLIII (January, 1962), 25-33. . Marketing Cost Analysis: Development and Current Practice," Apggunpipg Review, XXXVIII (January, 1963), 118-123. Neul, M. C. "value, Not Cost, Determines Price," Iron Age, CXCIV September, l96h), 116-117. Simon, S. I. "Cost Accounting and the Law," Acppuntipg Review, XXXIX (October, 1964), 884-889. Smder, R. E. ”Physical Distribution Costs," Distribution kc, LXII (IbeOl‘, 1963), 35-1420 125 Tarpey, L. "Indirect Price Discrimination and Robinson-Patman," Journal of Marketing, JCCJII (January, 1963), 68-71. Tomlin, W. D. "Private Recovery Under the Robinson-Pabnan Act," Texas Law Review, XLIII (December, 1961;), 168-193. whiting, H. G. "Cost Justification of Price Differentials," Manage- ment Controls, XII (November, 1965), 225-229. . "Quantity Discounts," Tana ement Controls, XII (November, 1965), 236-2140.