Division of Research Graduate School of Business Administration The University of Michigan March 1980 THE IMPACT OF REGULATION ON CONTROLS: FIRMS' RESPONSE TO THE FOREIGN CORRUPT PRACTICES ACT Working Paper No. 203 Michael W. Maher The University of Michigan FOR DISCUSSION PURPOSES ONLY None of this material is to be quoted or reproduced without the express permission of the Division of Research.

A —

THE IMPACT OF REGULATION ON CONTROLS: FIRMS' RESPONSE TO THE FOREIGN CORRUPT PRACTICES ACT ABSTRACT This paper presents a conceptual and an empirical analysis of the way coalitions of managers (i.e. firms) respond to regulation. The conceptual analysis provides a model of managerial behavior based on the assumptions that these coalitions make cost/benefit analyses of activities in their interest. The model is tested by examining how firms have responded to the internal accounting controls provisions of the Foreign Corrupt Practices Act of 1977. The hypotheses that the regulation would not directly affect the level of internal accounting controls, but would affect the expenditure of resources for efforts to "prove" compliance with the law were generally supported by the findings.

i4.

THE IMPACT OF REGULATION ON CONTROLS: FIRMS' RESPONSE TO THE FOREIGN CORRUPT PRACTICES ACT I. INTRODUCTION This paper deals with two emerging areas of interest: the internal economics of organizations and the impact of regulation on firms. Increasing interest in the way individuals in firms behave as economic decision makers is evidenced by a growing body of literature dealing with the economics of internal organizations. In many respects, this literature is a continuation of research into the theory of the firm, where the focus of attention is on the firm, itself, rather than on markets in which firms are only a part of the system (in somewhat the same sense that households are part of the system). In examining the resource allocation mechanisms of the firm, the firm is viewed as a nonmarket institution which exists as an alternative to markets.2 A number of authors have indicated that empirical evidence of the decisionmaking processes can enrich our understanding of the internal economics of organizations. This paper provides such evidence by examining the response of firms' managers to government regulation. The passage of the Foreign Corrupt Practices Act of 1977 (FCPA) provides an opportunity to observe mtznagers' strategies as they respond to regulation. Thus, an outcome of this paper is a better understanding of firms'4 response functions in the face of regulation. Government regulation imposes penalties on enterprises in two ways: The enterprise incurs both (1) costs of complying with the regulation and (2') costs of failing to comply with the regulation. Managers of the enterprise are expected to expend resources to meet the legal requirements; however, it is unlikely to be in their best interests to commit so many resources that

-2 - total compliance with the law is achieved. In short, regulation is expected to impose both costs of compliance and costs of noncompliance on enterprises. The question of how managers of firms balance those two costs in determining what they believe is an optimal response to regulation is the subject of this paper. In section two of the paper, some details about the nature of the regulation (FCPA) are presented to provide a background for understanding how managers are expected to respond to it. Of particular importance is the application of the provision that the costs and benefits of internal accounting controls are to be balanced. Possible strategies of the Securities Exchange Commission (SEC) in enforcing the law will be considered here as these are important for predicting how firms will respond to the law. Section three provides the conceptual basis for the investigation. The thrust of this model is that the internal accounting control provision of the FCPA should have no impact on the level of controls because firms are assumed to be at an equilibrium level of controls independent of the FCPA. Alternative explanations of possible managerial behavior are also presented in section three. Sections four and five respectively present the methodology and findings of the empirical study. These findings provide some support for the hypothesis that the FCPA has had no direct impact on the level of controls in the firms studied. Support is also shown for the hypothesis that, under conditions of uncertainty about interpretation and enforcement of the FCPA, in nearly all of the firms studied, managers were found to invest resources in documentation, formalization, and other "defense mechanisms" to prove compliance with the law.

-3 - In the final section, implications of these findings for understanding how managers allocate the resources of firms in responding to regulation are discussed. II. NATURE OF THE REGULATION To provide the necessary background for a conceptual and empirical analysis of firms' response to the FCPA, a brief description of the history, provisions, and interpretations of the law are presented in this section. History of the Regulation. As a result of investigations by the Office of the Watergate Special Prosecutor, the SEC became aware of several instances in which corporations had used funds for illegal domestic political contributions. Because the SEC believed these activities involved matters of significance to public investors, and the nondisclosure of these activities constituted violations of the federal securities law, the SEC published a statement calling for the public filing of questionable or illegal payments, both domestic and foreign, under a voluntary disclosure program. This program, as part of the total SEC investigation, revealed several 6 hundred instances of questionable payments. Further, the SEC reported: (M)ost of the instances of reported abuse also involved some falsification of corporate records or the maintenance of records that appear to be inadequate. In many of the reports submitted voluntarily by corporations, the description of the payments and their documentation appears to have been inadequate to permit ready identification or verification of the purpose of the payments....Many of the defects and evasions of the system of financial accountability represented intentional attempts to conceal certain activities. Not surprisingly, corporate officials are unlikely to engage in questionable or illegal conduct and simultaneously reflect it accurately on corporate books and records.7

-4 - Provisions of the FCPA Congress responded to these investigations and disclosures by enacting the Foreign Corrupt Practices Act of 1977 (FCPA). The provisions of the Act deal with questionable payments to foreign officials, record keeping and internal accounting controls. Antibribery provisions. The major impetus for the legislation was the revelation of questionable or illegal payments particularly in the context of the dealings of multinational corporations with foreign officials.8 The FCPA makes it unlawful for virtually any officer, director, employee, or agent acting for any company, whether or not registered with the SEC, as well as individuals who are citizens, nationals or residents of the United States to use the mails or interstate commerce corruptly for the purpose of paying foreign officials to obtain or retain business. "Grease payments" paid to government employees, whose duties are essentially ministerial, or clerical, to expedite routine matters (e.g. payments to customs officials to expedite shipments of goods) are apparently excluded from the definition of "foreign official."9 Accounting and control provisions. The revelation that records were falsified and off-the-books funds were maintained to facilitate the questionable payments prompted to SEC to propose legislation dealing with recordkeeping and controls. As subsequently passed, this legislation required all companies having stock registered with the SEC (including companies having only domestic operations) (1) to make and keep accurate books and records, and (2) to devise and maintain a system of internal accounting controls meeting the objectives stated by the American Institute of Certified Public Accountants (AICPA) in Statement on Auditing Standards (SAS) No. 1, Section 320.28.. Thus according to Senator Proxmire, "It (the accounting provisions of the FCPA) requires that

the businessmen of the country must be responsible to set up an accounting system that will inform them of what happens to their assets so that, if a bribe is paid, they will know."1 In a subsequent release, (Release No. 34-155570, February 15, 1979) the SEC noted that while the accounting provisions of the FCPA are intended to operate in tandem with the antibribery provisions of the law, the accounting provisions may themselves provide a basis for enforcement action. The SEC further noted in the release, "It bears emphasis that the accounting provisions of the FCPA are not exclusively concerned with the preparation of financial statements. An equally important objective of the new law,..., is the goal of corporate accountability."12 Despite the original legislative intent contained in the Senate Banking Committee Report of April 7, 1977 that the accounting provisions of the FCPA were intended to operate in tandem with the antibribery provisions,3 statements by the SEC14 and its enforcement actions5 indicate that the Commission does not intend to use the accounting provisions only in conjunction with the antibribery provisions. (See Siedel [1980].) Thus, firms registered with the SEC that have no dealings with foreign officials must still respond to the internal accounting control provisions of the FCPA. Ambiguities of the Law. The response of firms to the regulation may be affected not only by uncertainty about SEC enforcement strategy, but also because the law may be too vague to be enforced. In a letter to the Senate Banking Committee, the Auditing Standards Committee of the AICPA stated: "We do not believe that the existing auditing literature, prepared to guide auditors in determining the scope and nature of examinations of financial statements, provides sufficient

-6 - guidance to enable registrants to determine the adequacy of a system of internal accounting control as contemplated by the proposed rule." 15 In addition, Mautz, et. al. (1980) found considerable disagreement among corporate managers about such terminology as "internal control" and "assets". In short, it appears that corporate strategies for responding to the law must be made under uncertainty about the specific provisions and intent of the law. Cost/Benefit Provisions. The key issue in this paper is whether and (if so) how the FCPA will have an impact on the way resources are used in firms. The Senate Banking Committee Report of June 22, 1976, noted the importance of reasonableness and cost considerations in establishing and maintaining a system of internal controls. "The committee recognizes that no system of internal controls is perfect,... management must necessarily estimate and evaluate the cost/benefit relationships of the steps to be taken in fulfillment of its responsibilities under this paragraph" (i.e. paragraph (2) of section 13(b)). The SEC reiterated this view:.,."it is not in the interest of shareholders for the cost of internal accounting control to exceed the benefits thereof. Such benefits, and in many cases such costs, are not likely to be precisely quantifiable. Therefore, many decisions on reasonable assurance will necessarily depend in part on estimates and judgments by management which are reasonable under the circumstances."17 From these statements, it appears that company officials are asked to do no more than they would as prudent economically motivated managers; that is, to invest in an optimal level of controls after weighing both their costs and benefits. Of course, the antibribery provisions of the law may

-7 - have an impact on this evaluation for some companies, as discussed in the next section. Further, the difficulty in applying cost-benefit analysis to controls cannot be ignored, particularly when, according to the SEC, "... the benefits to be considered often may include not only quantitative benefits, such as reduction in exposure to theft of assets, but also qualitative benefits, such as the reputation of the company and its management. 18 During the testimony, disagreement was revealed about the expected impact of the FCPA on the use of firms' resources. Witnesses testifying that the problems revealed by the disclosures of companies reporting questionable payments did not represent a breakdown in controls included 19 Roderick Hills, Chairman of the SEC at that time, and Thomas Holton, Chairman of the AICPA Committee on SEC Regulations who stated: "There is no indication that it was the lack of adequate systems of internal accounting controls of these companies that resulted in the abuses and prevented their 20 detection and disclosure." An implication of these statements is that the FCPA would not affect current systems of internal accounting controls because (these) firms already have adequate controls; the FCPA merely makes law what is already being done in firms. This view was supported by Senator Proxmire who noted that the law... "would merely codify the requirement that a corporation keep honest records, a requirement that is at least implicit in the entire system of corporate accountability."2 However, Representative Moss testified that the questionable payments disclosures indicated that companies had problems with their internal controls. "These accounting gimmicks point to part of the underlying problem, the ineffectiveness of internal corporate accounting and auditing controls."22

-8 - The courts' interpretation of these views could become important in later litigation and thus have an impact on the way firms respond to the law. "For instance, if a court determines that the legislation was not intended to restructure existing internal accounting control systems, a company might argue that its system should be judged by the state of the art which existed when the FCPA was passed. But if a court determines that the legislation was directed toward reform of existing systems, a company's state of the art defense, based on 1977 systems, would carry much less weight" (Siedel, 1980, pp. 19-20). If companies have ineffective internal accounting controls, will the FCPA force a greater investment in controls that has been made in the past? In the next section, a conceptual analysis of the way managers are expected to respond to the regulation is presented, followed by empirical results which provide a partial answer to the above question. III. CONCEPTUAL ANALYSIS OF FIRMS' RESPONSE TO THE FOREIGN CORRUPT PRACTICES ACT Government regulation imposes costs on enterprises in two ways: (1) the cost of complying with the regulation and (2) the cost of not complying with the regulation. The first category includes the cost of taking actions required by the law. The second category of costs includes fines and losses due to damaged reputation of the firm if it is found to be in violation of the law. The expected cost,,of course would take into account the probability of being found in violation of the law, and the time periods over which the losses would occur. Firms, (i.e. coalitions of managers for purposes of this study) are expected to minimize the total costs of regulation, taking into account the costs of compliance, as wll11 as the probability and consequences of being caught if not in compliance, all on a risk-adjusted and present-value basis. As shown in Figure 1, there is no reason to expect the optimal compliance effort to be total compliance. It may make sense for firms

-9 - to accept some risk of incurring costs for not complying with the regulation. Similar reasonsing can be applied to an analysis of the optimal level of internal accounting controls under the FCPA. For the moment, the development of controls to deal with the antibribery provisions of the FCPA is ignored, and our focus is on determining the optimal level of controls to deal with operations unaffected by the antibribery provisions (i.e. domestic operations). The key assumption to this analysis is that managers of firms, acting in their own best interest, had developed or were developing what they believed to be the optimal level of controls independently of the FCPA. This optimal level of controls is that level at which total control costs are minimized, where the total control costs are defined to include both the costs of controls and the cost of behavior variant from that desired by those establishing objectives for the firm. In the context of the principalagent model (e.g Ross [1973], Jensen and Meckling [1976]) principals (managers) are presumed to develop and maintain control systems, taking into account various means of controlling agents (subordinates) (e.g. bonding, incentive schemes, monitoring systems) of which internal accounting controls are a part. Taking all these factors into account, an internal equilibrium is assumed to be reached which considers the costs of various types of control (e.g. alternative monitoring systems and incentive schemes ) and the cost of variant behavior.4 As demonstrated by the simple diagram in Figure 2, this is the point at which total agency cost is minimized. The impact of cash flow timing, uncertainty, and risk preferences are assumed to be factored into the analysis as well. Although it is assumed that an optimal level of controls is reached in each firm, the results of that analysis may vary. In some firms, it may be optimal to tolerate a high level of variant behavior, while in others it may not.

-10 - Of course, it is unrealistic to assume that the results obtained in firms is the same as that which would be obtained in the absence of information and analysis costs. Further, the process is dynamic, the development of controls is evolutionary, and the behavior that the controls 25 are designed to control is always changing. Controls may lag the recognition of a need for controls -- they are not implemented instantaneously. However the point, in the words of a legal group in recommending a program for compliance with the FCPA, "... we start with the assumption that you have already spent all the money you think appropriate from a business point of view for internal control systems...." (Foreign Corrupt Practices Act Reporter [1979.], p. 127). Hypothesis 1: Impact of the FCPA on Internal Accounting Controls The assumption of an internal (dynamic) equilibrium coupled with the cost-benefit provisions of the FCPA lay the groundwork for the first hypothesis: H(l): The internal accounting controls provision of the Foreign Corrupt Practices Act has no direct impact on resources used to improve internal accounting controls for firms' operations that are unaffected by the antibribery provision of the law. In essence, H(1) predicts that the FCPA does not affect the control equilibrium which was arrived at independently of the law. This hypothesis deals with the correction of specific internal accounting control defects. That is, whether the FCPA leads to the recognition of defects for which the cost of controls is less than the benefits derived from repairing those defects. Without the cost/benefit provisions of the law, many firms would be expected to invest in internal accounting controls because the costs of variant behavior would be higher due to increased expected litigation costs, penalties, etc. for violation of the law. However, with

-11 - the cost/benefit provisions of the law, H(1) predicts that for firms' operations unaffected by the antibribery provisions of the law, there will not be investment in internal accounting controls because of the law. (The FCPA may lead to the consumption of resources for other reasons,, as predicted by the second hypothesis, below.) Alternative Explanations. Three explanations for rejecting this hypothesis are considered here. First, as previously indicated in the Nature of the Regulation section, the provisions of the FCPA appear ambiguous to many commentators. The measurement of costs and benefits, particularly the measurement of such qualitative benefits as company reputation, is very difficult. Thus, the assessment of costs and benefits made in a firm may be very different from an assessment made by an adversary attempting to "prove" that costs were less than benefits in a specific situation. Consequently, managers who are risk-averse to such adversarial cost/benefit analyses may have incentives to "overinvest" in internal accounting controls; that is, to invest more than they would for business purposes. A second reason for expenditures on controls, even if firms were at equilibrium independent of the FCPA, relates to the attempts of certain coalitions in firms to gain power at the expense of other coalitions. The FCPA focuses attention on coalitions of financial personnel in firms (e.g. controller's department, internal audit staff) who may have incentives to use the FCPA as a leverage to obtain more funds and staff for their coalition. If the financial managers successfully convince other managers of the need to invest more resources in controls, thereby obtaining more resources for the financial coalition, the firm may "overinvest" in controls.

-12 - (Of course, using the FCPA as a leverage to argue for more controls could constitute an admission by financial managers that they had not performed their duty in maintaining an optimal level of controls.) Both of the above alternative explanations are consistent with the assumption that managers form a coalition (i.e. a firm) and perform rational cost/benefit analyses in their (the coalition's) best interests. A third explanation for findings contrary to H(1) is that the kind of analysis 27 had not been performed. That is, managers had not considered the costs and benefits of controls, and with enactment of the FCPA they realize that control benefits exceed control costs, thus they invest in controls. In short, the FCPA imposes on firms a type of analysis that was not otherwise done. Impact of antibribery provisions on controls. A logical extension of H(l) is the impact of the antibribery provisions of the FCPA on the optimal level of controls. Although the einpirical tests are beyond the scope of this study, it follows from the analysis leading to H(1) that if the FCPA had an impact on the costs of variant behavior, the equilibrium level of internal accounting controls would change. The antibribery provisions of the FCPA increase the expected cost to the firm of those questionable payments which come under the jurisdiction of the FCPA. Thus, the level of controls may be affected by the antibribery provisions of the FCPA, as demonstrated in Figure 3. Hypothesis 2: Impact of the FCPA on Other Compliance Efforts. Even in companies in which controls are initially unaffected by the FCPA, the regulation may require firms to consume resources. The basis for this argument is that it is not enough to be in compliance with the law, firms must be prepared to "prove" they are in compliance. In short, firms are

-13 - expected to establish a record which can be used if their systems and environment of internal accounting controls are ever challenged in court. These costs are analogous to those incurred by firms (and others) to support their self-reported tax returns. Just as the process of record-keeping may provide information that changes self-reported tax obligations, the review and documentation of internal accounting control procedures and practices may reveal defects for which the cost of correction is exceeded by the benefits. In this case, the FCPA would have an indirect effect on the level of controls, but not a direct effect as predicted by H(1). Each of the "Big 8" CPA firms, as well as various legal authorities (e.g. Foreign Corrupt Practices Act Reporter [1979] and professional associations (e.g. American Bar Association [1978], AICPA [1979]) have recommended strategies for compliance programs. In general, these strategies include formal communication of an interpretation of the statute by legal counsel, seminars and training programs for financial and operating personnel, and a formal statement of ethical policy in the firm's code of conduct.29 "The reason for all of the above is, of course, to show (emphasis added) that the company does have a good system and program" (Foreign Corrupt Practices Act Reporter, p. 122). Finally, a formal evaluation of the current system, including cost/benefit analysis, is recommended. "A company may or may not have to make changes in its internal control system, but it will definitely have to make a specific decision on this point and document it" (Foreign Corrupt Practices Act Reporter, p. 124). t olbY~~f~ -.-zr- rg.

-14 - The second hypothesis predicts that managers, acting in their selfinterest, will incur costs of reviewing existing controls and providing evidence that their firms have systems of internal accounting controls which meet the requirements of the FCPA. H(2): The internal accounting control provision of the Foreign Corrupt Practices Act results in the expenditure of resources to provide evidence that firms are in compliance with the law. These expenditures are other than those used to improve internal accounting controls. For this study, H(2) applies only to those operations of firms unaffected by the antibribery provisions of the law. Alternative explanations. Managers of firms may not believe their firms will be affected by the FCPA. Interpretations of the FCPA indicate that "It does not appear that either the SEC or the Department of Justice is going to be enforcing the accounting and record keeping sections by themselves. Instead they will be pursuing their normal enforcement of all improper payments and bribes and will be using the accounting and record keeping sections in order to make their cases easier to win and to obtain consent decrees against companies which incorporate these provisions." (Foreign Corrupt Practices Act Reporter, p. 104). SEC enforcement actions to date have not been brought under the internal accounting controls provisions of the FCPA, alone. If managers believe the probability of action brought by the SEC against their firm for other.violations of securities laws is very low, they many not believe an expenditure of resources to "prove" compliance is justified. IV. METHODOLOGY Data to test the above hypotheses were collected in tandem with a much more extensive study of internal control practices in American firms during May-August, 1979 (17-20 months after enactment of the FCPA). (See Mautz, et al [1980].) In that study, a research team (including the author) collected

-15 - data using questionnaire, interview and document examination methods. A subset of the interview questions from that study, as well as additional documentation of company practices, provide the data base of this study. (The specific interview questions for this study are-listed in the Appendix.) Managers in the firms studied were assured that the firms would remain anonymous, and the information would be kept confidential so that the reported responses could not be traced to a firm. Documents. Written documents in the firms provided the major source of data for this study. All documents developed as a result of the FCPA, including legal interpretations, corporate policy statements, and codes of conduct internal control questionnaires, materials for seminars, and instructions for evaluating the firm's internal control system, were examined. In addition, company policy and procedure manuals and the external auditor's evaluation of internal controls were examined to determine the nature and cause of changes in internal controls, if any. The latter were particularly useful for determining whether changes in a firm's internal accounting controls were prompted by the FCPA or for other reasons (e.g. external auditor's pre-FCPA comments). Finally, published documents by the firms, including annual financial statements, 8-K and and 10-K reports, and minutes of stockholders' meetings were part of this data base. Interviews. Interviews with managers and staff in the participating firms provided additional information. Interviews were held with the chief financial officer and other top financial managers (e.g. controller, if separate from

-16 - the chief financial officer), the chief internal auditor, the manager of the data processing, and the managers of two-three major functional areas in the firm (e.g. purchasing, marketing) and/or major operating units (e.g. division manager). Interviews with other personnel responsible for implementing the firm's compliance program (e.g. "FCPA auditors") and for managing the firm's documentation and manuals rounded out the interview schedule. In total, 336 managers were interviewed in 49 companies. The interview guide included a combination of open-ended and highly specific questions to ascertain the strategies that firms have used in responding to the FCPA. These questions were refined through pilot testing at four firms not participating in the data collection phase of the study. Interviews were purposely held with a variety of managers in a company. While the firm's documents generally provided the "company policy," various managers in the firm were responsible for implementing it. In particular, the chief financial officer, the controller (if a separate position), the chief internal auditor, and a few other financial staff were primarily responsible for implementing corporate policy regarding internal controls. Operating managers, who were on the receiving end of controls, were important to interview to determine whether the procedures claimed by the financial staff were, in fact, being followed. Also, operating managers' assessment of the costs and benefits of controls over their operations provided an additional check on the statements made by financial managers. The chief legal counsel was an important source of information about the legal interpretation of the FCPA for the firm, as well as the steps required for the firm to legally comply with the law.

-17 - Different managers in a firm may be assumed to have different perspectives and information sets, thus some variation in responses was expected. In fact, one reason for interviewing multiple managers was to obtain data from managers with as many perspectives as possible. For purposes of this study, these responses were checked against documentation, as well. Inconsistent responses, and claims which could not be supported by documentation, in a firm resulted in treating some or all of the firm's responses as "unusable"because of ambiguity (see Tables 3 through 5). To avoid "prepared" answers, the interviewees were not given information about the questions that would be asked until the interview team arrived at the firm for the day-long series of interviews. Sample of Firms. A random sample of 135 firms from the Fortune 1300 (1000 industrials plus the top 50 banking, diversified financial, insurance, transportation, retailing and utility companies) was drawn. This list was turned over to representatives of the Financial Executive Research Foundation, the sponsoring organization of the empirical study, to contact firms. Firms were asked if they would participate in a one-day interview program with two interviewers who were part of a research team doing a study of the state of the art of internal controls. Of the 135 companies selected, 91 agreed to participate. Of these 91 firms, 50 were selected by the interviewers on the basis of mutual availability of personnel. However, during the intial interviews in one firm, the managers being interviewed would not answer the questions put to them. Consequently, the interview was cancelled. For an analysis of the sample, as shown in Table 1, that firm is classified as a "firm refusing

-18 - participate," leaving 90 firms classified as "firms agreeing to participate." Of the remaining firms, 34 refused to participate, citing a variety of reasons, including business developments (e.g. merger, bankruptcy, strike) other scheduling problems (e.g. the firm was highly decentralized with the personnel needed for interviews in distant locations), and lack of interest in the topic. In many of the firms refusing, managers believed the cost of the research to the firm was quite high with no apparent benefits. Finally, in 11 firms, no response was made to the initial inquiry by the representatives of the Financial Executives Research Foundation. Follow-ups were not made with these firms because at that time more than enough firms had agreed to participate in the study, considering the resource 32 and time constraints of the interviewers. Sample bias. The 34 firms in which managers refused to participate pose a potential for sample bias, if incentives for nonparticipation are examined. Managers who believed the level of controls in the firm was less than optimum may have had incentives to reject participation to avoid possible embarrassment. Of course, these same managers may have had incentives to participate if they believed participation would signal "good faith" in compliance with the FCPA, or if they believed participation would help assure that standards of control for enforcement by the SEC are not set unrealistically high because the interviewers have seen only companies with "strong" controls. In short, refusal to participate may have been done for a variety of reasons which were not verifiable by this author. If firms that had relatively "poor" controls (in a cost/benefit sense) were more likely to refuse than agree to participate, then the sample of firms studied is biased in support of H(1).

-19 - Some evidence, admittedly indirect, that this occurred was found by comparing the categories of firms according to whether they had reported questionable payments to the SEC. As shown in Table 2, firms that refused (and those not responding to the request for participation) had reported proportionately more questionable payments than the firms in which agreement to participate was obtained. If "poor" controls are correlated with (the reporting of) questionable payments, then the sample of firms studied may have proportionately more firms with "good" controls than does the total sample selected. This point should be kept in mind when interpreting the findings reported in the next section. V. FINDINGS For each company, responses from all members of management and staff that were interviewed were compared with each other and with the documentation to obtain a consensus response. If a consensus response could not be obtained, the response is reported as "unusable response" in Tables 3 through 5. While the interview data were obtained between May and August 1979, the documentation obtained and reviewed precedes those dates. The expenditure of resources after the data collection period (May-August 1979) is not considered in these findings. For purposes of this study, resources consumed and other actions taken only in response to the SEC's proposed internal accounting control reporting requirements (SEC Release No. 34-15772, April 30, 1979) were excluded from the findings because the proposed reporting requirements were not part of the regulation, at the time these data were collected.

-20 - As shown in Table 3, the FCPA had a direct impact on resource consumption in 40 of the 45 firms in which usable responses were obtained. Firms in which the only response has been a summary of the FCPA and/or a letter from top management regarding the provisions of the FCPA were not considered to have consumed resources because of the FCPA. The FCPA was claimed to "heighten interest" or "increase awareness" of controls in many companies. This was not considered prima facie evidence that the FCPA had a direct impact on the consumption of resources. At the other extreme, the implementation of new controls and/or the initiation of new programs for reviewing internal control systems were regarded as evidence that the FCPA has had a direct impact on the consumption of resources. Attempts to determine a dollar value of the direct cost of the FCPA were unsuccessful. Although specific transactions could be observed in several firms that had hired personnel or obtained the services of CPA firms; for the most part, the costs were the unquantified opportunity costs of financial staff time. Another cost of regulation, as previously indicated, is the cost of noncompliance. Specific measures of expected costs of noncompliance could not be obtained in the firms studied. Hypothesis 1: Impact on Controls. As shown in Table 4, the FCPA had a direct impact on controls (all improvement) in only 5 of the firms studied. These were firms in which a concensus response was obtained and supported by documentation that controls had been increased over those operations not affected by the antibribery provisions of the FCPA. These results support Hypothesis 1. For the most part, firms were at an equilibrium, albeit a dynamic equilibrium, independent of the FCPA;

-21 - thus, the FCPA did not have a direct impact on their level of controls. The following rationale was given in the 5 firms that had increased controls because of the FCPA: (1) The cost of controls was recognized as less than the benefits and the FCPA prompted actions to add controls or increase the rate at which controls were added (4 firms). (2) In one firm, management believed the company was a potential "target" because of its visibility. Therefore, a conservative approach was being taken and controls had been added even though the level of controls was believed to be at an optimum independent of the FCPA. The reader should not imply from these findings which strongly support H(1) that managers of firms have, in general, made a formal analysis of the costs and benefits of controls which is the basis for their belief that controls were at an optimal level, independent of the FCPA. In fact, very little evidence of formal ex ante analysis was found. Instead, assessments of the need for controls appeared to follow external auditors' comments in letters to management and (more frequently) observedoccurrences of variant behavior. Hypothesis 2: Impact on Other Compliance Efforts. All 5 of the firms that responded to the FCPA by increasing internal accounting controls consumed resources to demonstrate that they were in compliance with the law. Of greater interest, however, is the impact of the FCPA on the remaining 44 firms. As shown in Table 5, resources were consumed to provide evidence that the firms were in compliance with the law in 35 of the. 41 firms in which usable responses were obtained. Managers in the remaining 6 firms providing usable responses either believed they were unaffected by the FCPA, so they took no action, or they believed that the firm's controls were strong enough, given the cost/benefit stipulation in law that no further "proof" of compliance would be necessary.

The primary form of response to provide evidence that firms were in compliance with the law consisted of an extensive review and documentation 34 of current practices. Despite the fact that managers in most firms believed their controls were at an optimal level for business purposes, formal evaluations and cost/benefit analyses of controls were rare before the FCPA was enacted. These analyses may, of course, reveal situations in which the addition of controls is cost/benefit justified. The data were collected for this study too soon after these reviews were started to determine whether they would reveal the need (in a cost/benefit sense) for more (or less) controls. Thus, the FCPA may have an impact on controls after these formal evaluations in firms in which the FCPA had no direct impact on controls. In effect, the FCPA has increased the benefits of making a formal cost/benefit analysis (by increasing the expected costs of not making a formal analysis to "prove" compliance with the FCPA). The notion that there are benefits from greater documentation and formalization was supported by the responses from the interviewees. In no firm did managers claim negative "benefits" from formalization and documentation while many claimed positive benefits. However, those claiming positive benefits believed the resources could have been spent elsewhere. There was also some support for the notion that greater awareness and interest in controls by directors, top management and operating managers would automatically lead to changes in controls. Thus despite the findings of this study, the FCPA may have an impact on controls in many firms. While the direct impact on internal accounting controls was apparent in only a few firms in this study, the potential for indirect effects is large, particularly in the firms in which an extensive review and documentation of controls is being performed.

-23 - VI. CONCLUSIONS As indicated at the beginning of this paper, the objective of the study was to better understand the response to regulation by managers, or coalitions of manangers, in firms. The major conclusions that can be drawn from the study follow. (1) As hypothesized, managers generally did not respond by immediately taking steps to increase the level of internal controls in the firms studied. (2) However, an expenditure of resources on activities designed to "prove" compliance with the law was observed in most of the firms studied. (3) Among the consequences of the compliance efforts is a move toward greater formalization and documentation of internal controls. (4) The law has increased the benefits of performing formal cost/benefit analyses of internal controls in many firms. As a result of these analyses, the level of internal controls may change; however, this study was performed too soon after these analyses were started to determine their impact on the level of internal controls. (5) The law has reportedly led to "increased awareness" by top management and operating executives of the importance of internal controls in many companies. This, in itself, could lead to improved controls according to the interviewees. (6) Finally, the FCPA has given additional leverage to the coalition of financial managers and staff in some of the firms studied which allows them to obtain more resources for their function. In many firms, the respondents indicated that they expected benefits to accrue to the firm, independent of the law, from the documentation, review and other compliance efforts undertaken. Thus, while the FCPA has led to costly compliance efforts, some benefits are expected to result from these efforts. Of course, if those beneifts were as greater, or greater than the costs, managers would be expected to expend resources on,4

-24 - them independent of the FCPA. In short, these findings support the notion that the FCPA has been both costly and beneficial to firms; however, it appears to have been more costly than beneficial, at this point, in the firms studied. While the full impact of regulation cannot be known so soon after it has been passed, this study has afforded the opportunity to observe initial firms' responses in the early years after enactment. As previously indicated, there is considerable leeway for SEC interpretation and enforcement of the law. As of this writing, very few cases had been brought under the FCPA, and none have been settled in the courts. Thus managers have been faced with considerable uncertainty in their response to the law. The strategy followed by most of the firms studied has been to follow a defensive strategy of review and documentation. It is left for future research (1) to determine whether there will be secondary effects of review and documentation on controls and (2) to determine the effects of specific SEC interpretation and enforcement actions on the way firms respond to the regulation.

-25 - FIGURE 1 TRADE-OFF BETWEEN COMPLIANCE COSTS AND THE COSTS OF NOT COMPLYING WITH REGULATION A Costs Total Costs of __ Regulation 1 I I I Compliance j Costs I I I I I Complying I I Optimal Compliance Effort Compliance Efforts I I I i

FIGURE 2 TRADE-OFF BETWEEN THE COST OF CONTROLS AND THE COST OF VARIANT BEHAVIOR Costs t Total Agency Cost Cost of Controls I I I I I I I I Cost of Variant Behavior Ilk~ Optimal Level of Controls Level of Controls

-27 - FIGURE 3 EXAMPLE IMPACT OF ANTIBRIBERY PROVISIONS OF THE FCPA ON THE OPTIMAL LEVEL OF CONTROLS Total Agency Cost After FCPA ncy re Costs \ Total Ac Cost bej FCPA ge fo I \I | I~~~I \ 'i I \'1 / I I I I~~~~~~~ I Cost of / Controls Cost of Varian Behavior _after FCPA I I 1I I I I I I I before I FCPA I 1 I I I I — - -- Optimal- Optimal Level Level of of Controls Controls before after FCPA FCPA Level of Controls

-28 - TABLE 1 ANALYSIS OF SAMPLE Firms agreeing to participate: Firms studied......................... 49 Other firms agreeing to participate, but not studied... 41 Total firms agreeing to participate.......... 90 * Firms refusing to participate.......... 34 ** Firms in which contact was not made to request participation.. 11 Total firms in sample 135 * Number includes one firm in bankruptcy, one firm on strike and one firm in merger (verified by reference to Wall Street Journal articles and other business publications). ** Number includes one firm on strike (verified by reference to Wall Street Journal articles).

-29 -TABLE 2 COMPARISON OF FIRMS ACCORDING TO REPORTED QUESTIONABLE PAYMENT Firms Not Firms Reporting Reporting Questionable Sensitive Payments33 Payments Total Firms Firms Agreeing to Participate...... 12 78 90 Firms Refusing to Participate....... 8 26 34 Firms Not Responding to Participation Request....... 4 7 11 Total Firms.. 24 111 135 The Ho that the proportion of firms reporting questionable payments is the same in each of the three groups (agreeing, refusing, not responding) is rejected at p <.001, using a chi-square test. The Ho that the proportion of firms reporting questionable payments is the same in the "firms agreeing" and "firms refusing" groups is rejected at p <.05, using a chisquare test.

TABLE 3 IMPACT OF FCPA ON RESOURCE CONSUMPTION No. of Firms FCPA had a direct impact on resource consumption........ 40 FCPA had no direct impact on resource consumption......... 5 Unusable response......................... 4 Total............49 Diffeences significant at p <.001 (chi-square) Differences significant at p <.001 (chi-square).

TABLE 4 IMPACT OF FCPA ON CONTROLS No. of Firms FCPA had a direct impact on internal accounting controls...5................. 5 FCPA had no direct impact on internal accounting controls.... 39 Unusuable response...............rnl ag.. 5 Total.............. 49 Differences signiicant at p < 001 (chi-square) Differences significant at p <.001 (chi-square).

-32 - TABLE 5 IMPACT OF THE FCPA ON OTHER COMPLIANCE EFFORTS No. of Firms FCPA had a direct impact on other compliance efforts.... 35 FCPA had no direct impact on other compliance efforts...... 6 Unusuable responses....................... 3 Total 44 * Differences significant at p <.001 (chi-square). ** Firms increasing controls as a result of FCPA were excluded here because it was impossible to differentiate increases in controls from other compliance efforts in those firms.

-33 - APPENDIX INTERVIEW GUIDE QUESTIONS3 (Excluding procedures for examining documents) A. INTRODUCTORY SESSION (Chief Financial Officer, Controller, Chief Legal Counsel) 1. What has been your company's response to the Foreign Corrupt Practices Act of 1977? If documentation, specify type and provide copies. If personnel, specify responsibilities (should be part of interviewee group). 2. What problems of compliance do you visualize? 3. Did your company have any sensitive payments problems? If yes, describe. 4. How do you believe the SEC will operationalize the FCPA? 5. What do you believe will be the costs and benefits of the FCPA? 6. (a) If you had a 20% increase in your budget for internal control purposes, how would you use it? (b) If you had a 20% reduction, what would you trim back? B. INTERNAL AUDIT INTERVIEW (Chief Internal Auditor) 1. Major changes in the size and composition of internal audit staff? Over past 5-10 years? Over next 5-10 years? 2. What is your budget? 3. Are adequate resources allocated to internal audit? (a) Where would/could you use more personnel? (Cost/benefit justified?) (b) Where would/could you use fewer personnel? (Cost/benefit justified?) 4. Changing role and activities of internal auditing (a) (How) Has internal audit's role and activities changed over the past several years? (b) What (if any) changes do you foresee in the future? (c) What are the causes of these changes?

-34 - 5. What do you view as the greatest exposure in your company in terms of internal control? (a) What is the single worst incident of "breakdown" in internal control that you can envision? (b) What major areas of gradual control erosion can you envision? (c) What has been done/could be done to minimize the probability or effect of that occurrence? 6. Are formal/informal controls uniformly applied? (a) In all locations? If not, please describe how controls vary. (e.g., Foreign, remote, small, joint ventures, partly owned subsidiaries.) (b) To all types of projects or activities? (e.g., R&D nonroutine activities.) If not, please describe how controls vary. 7. (How) Are controls changing in your company (e.g., scope, formality)? (a) In response to the Foreign Corrupt Practices Act? (b) For other reasons? 8. What is your overall evaluation of controls in your company? (e.g., In comparison to other companies with which you are familiar.) 9. What (if anything) could be done to improve the controls of activities in your company? (e.g., Documentation, formalization, personnel training.) C. INTERVIEW WITH OPERATING AND STAFF MANAGERS (e.g., President of Marketing, Director of Purchasing, Division Vice-President, Division Controller) 1. What do you view as the greatest exposure in your area of responsibility in terms of internal control? What has been done/could be done to minimize the probability or effects of that risk? 2. In general, (how) are controls of activities in your area of responsibility changin in your company (e.g., scope, formality)? (a) In response to the Foreign Corrupt Practices Act? (b) For other reasons? 3. What do you view as the greatest exposure in your company in terms of internal control? (a) What is the single worst incident of "breakdown" in internal control that you can envision? (b) What major areas of gradual control erosion can you envision? i ~(c) What has been done/could be done to minimize the probability or effect of that occurrence?

-35 - 4. Are formal/informal controls uniformly applied? (a) In all locations? If not, please describe how controls vary. (e.g., Foreign, remote, small, joint ventures, partly owned subsidiaries.) (b) To all types of projects or activities? (e.g., R&D nonroutine activities.) If not, please describe how controls vary. 5. How well are the activities in your area of responsibility controlled in your company? Please explain and discuss your evaluation. - In the discussion, determine how the respondent interprets the term "control" - what does it mean to him/her? - Also determine the relative emphasis placed on formal and/or informal means of control). 6. Does the control of these activities enhance or detract for their overall effectiveness? (Is there any feeling of redundancy in controls, "over-control", costs of control exceeding benefits? If enhancement -- why?) 7. Improvement in controls. (a) In your opinion, what (if anything) could be done to improve the control of activities in your area of responsiblity. (formal, informal, more outside review, etc.)? (b) How specifically would you spend an additional $100,000 allocated to you for internal control purposes? AM=;;:t an I. _ t.: C -a of

-36 - FOOTNOTES Partial reviews of this literature are provided by Leibenstein (1979) and Spence (1975). Behavioral theory of the firm researchers (e.g. Cyert and March [1963], March and Simon [1958] and Simon [1978] deal with the economics of internal operations of the firm using alternatives to the rational-profit-maximizing-man paradigm of neoclassical microeconomics. Machlup (1967) examined the controversy between neoclassical microtheory and the views of the "behavioralists." Baumol (1959), Williamson (1963), (1976), Morsen and Downs (1965), Mirrlees (1976), Williamson, Wachter and Harris (1975), Stiglitz (1975), Jensen and Meckling (1976), among others, examine internal organizational problems from an optimizing viewpoint in which managers are rational maximizers who operate in their self-interest (which may not be in the self-interest of owners, society, consumers or other constituents of the firm). The problem of multiperson decision making has received considerable attention in von Neuman and Morgenstern (1944) and has more recently been addressed by Mauschak (1955), and Mauschd and Radner (1972), who analyze multiperson decision processes when members of the organization constitute a team, sharing a common reward. Wilson (1968) analyzed the problem when the group is not a team and Ross (1973) extended Wilson's work to consider characteristics of payoff rules that provide an optimal sharing of risk among superiors and subordinates. The idea that firms emerge because they outperform the market is central to the works of Coase (1937) and Alchian and Demsetz (1972) who demonstrate transactions and monitoring cost advantages of firms. Williamson (1975) analyzes the advantages of shifting transactions from the market to organizations. 3For recent examples of empirical work in this area see Salamon and Smith (1979), who provide evidence consistent with the notion that managers in management controlled firms attempt to manipulate information in financial statements more than managers in owner-controlled firms do; and Maher, Ramanathan and Peterson (1979), who find an association between employee performance and the resolution of certain kinds of information and preference conflicts. The term "firm" is defined in this paper as the coalition of individuals within the firm who collectively determine how the firm will respond to the regulation. The study of conflicts among members of this coalition is beyond the scope of this paper however, as reported later in this paper, there is some evidence that the FCPA has given financial managers leverage in obtaining resources. (Williamson [1963] indicates some conditions under which managers place a positive value on certain expenditure.)

-37 - See Securities Act Release No. 5466 (March 8, 1974). The SEC reported its investigation of companies in its "Report on Questionable and Illegal Corporate Payments and Practices," submitted to the Senate Banking, Housing and Urban Affairs Committee. According to the House Report (House Report No. 95-640, 95th Congress, 1st Session, 1977) more than 400 companies had disclosed questionable payments when the FCPA was enacted. (See Foreign Corrupt Practices Act Reporter, [1979], p. 102). A review of corporations by Charles E. Simon, Inc. reporting questionable payments revealed that most companies have reported multiple instances of sensitive payments, Sensitive Payments by Corporations (1977). 7Securities Exchange Commission, "Report on Questionable and Illegal Corporate Payments and Practices" (May 12, 1976), pp. 41-42. 'More than 400 corporations have admitted making questionable or illegal payments. The companies, most of them voluntarily, have reported paying out well in excess of $300 million in corporate funds to foreign government officials, politicians and political parties." (H.R. Report No. 95-640, 95th Congress, 1st Session (1977). 9 A portion of the text of the law which applies to "issuers" is presented here. Similar provisions apply to "all domestic concerns." (See Section 104, Public Law 95-213.) "StC. 30A. (a) It shall be unlawful for any issuer which has a class of securities registered pursuant to section 12 of this title or which is required to file reports under section S1(d) of this title, or for any officer, director, employee, or agent of such issiter or any stockholder thereof acting on behalf of such issuer, to make use of the mails or any means or instrumentality of interstate commerce corruptly in furthe ranre of an offer, payment, promise to pay, or authorization of the payment of any mronet, or offer, gift, promise to give, or authorization of the airing of anything of value to(1) any foreign oficial for purposes of"(A) influencing any act or deciion of such foreign official in his official capacity, including a decision to fail to perform his official functions: or "(B) inducing such foreign official to use his influence with a foreign government or instrumentality thereof to affect or inftuence any act or decision of such government or instru- mentality, in order to assist uch issuer in obtaining or retaining business for or with, or directing business to, any person;..,

-38 - See Securities Exchange Act Release No. 34-13185 (January 19, 1977). The text of the accounting requirements of the FCPA, as enacted in December 1977, follows: S^'. 10. S10. ctionw!(1) of the Sec*ritics Exchatnge Act of 1934 (16 U.S.C. 78g(b)) is anwnded by inserting '"()" after "(b)" and by addinq at the end thereof the follmoingj: "(2) Every issuer which has a class of securities registered pursuant to section 12 of this title and evern/ issuer which is required to file reports pursuant to section _15(d) of this title shull"(A) make and keep hooks, records, and accounts, which, in \ reasonable detail, accurately and fairly reflect the transactions and dispositions of the assets of the issuer; and "(B) devise and maintain a system of internal accounting controls suficient to provide reasonable assurances that"(i) transactions are executed in accordance with management's general or specific authorization; "(ii) transactions are recorded as necessary (1) to permit preparation of financial statements in conformity with generalty accepted accounting principles or any other criteria applicable to such statements, and (11) to maintain accountability for assets; "(iii) access to assets is permitted only in accordance with management's general or specific authorization; and "(iv) the recorded accountability for assets is compared with the existing assets at reasonable intervals and approvriate action is taken with respect to any differences.... The FCPA does not specify penalties for violation of the two accounting provisions, but the general penalties under the Securities Exchange Act of 1934 would apply. For a comprehensive legal review of the history, meaning and possible interpretation of the law, see Foreign Corrupt Practices Act Reporter (1979) and Siedel (1980). Congressional Record (September 15, 1976), Vol. 122, p. 15858. 12Federal Register (May 4, 1979), Vol. 44, p. 26704. 13"In the past, corporate bribery has been concealed by the falsification of corporate books and records. Title I (of the bill) removes this avenue of coverup, reinforcing the criminal sanctions which are intended to serve as the significant deterrent to corporate bribery. Taken together, the accounting requirements and criminal prohibitions of Title I should effectively deter corporate bribery of foreign government officials." (Senate Committee on Banking, Housing and Urban Affairs, Foreign Corrupt Practices and Domestic and Foreign Investment Disclosure Acts of 1977, Senate Report No. 95-114, 95th Congress, 1st Session [1977], p. 3.)

-39 - 14 In SEC v. Marlene Industries Corp. (79 Civil 1959 [S.D.N.Y., 1979] and SEC v. Aminex Resources Corp. (CCH Fed. Sec. L. Repet. para. 96,352 [D.D.C. 1978]); the SEC brought action for violation of securities laws which involved general fraud and improper practices, but not the antibribery provisions of the FCPA. In each case, violation of the accounting and control provisions of the FCPA was included as an additional charge. 15Foreign Corrupt Practices and Domestic and Foreign Investment Disclosure: Hearings on Senate Bill No. 305 Before the Senate Committee on Banking, Housing and Urban Affairs, 95th Congress, 1st Session (1977), pp. 220-221. 1Senate Committee on Banking, Housing and Urban Affairs, Corrupt Practices by U.S. Business Enterprises, Senate Report'No. 94-1031, 94th Congress, 2nd Session (1976), p. 8. 17SEC Release No. 34-15772 (April 30, 1979), p. 26704. 18 Ibid. 19 Foreign Payments Disclosure: Hearings on H.R. 15481, S. 3664, H.R. 13870 and H.R. 13953, Before the Subcommittee on Consumer Protection and Finance of the House Committee on Interstate and Foreign Commerce, 94th Congress, 2nd Session (1976), p. 26. 20Ibid, p. 158. 21Prohibiting Bribes to Foreign Officials: Hearings on S.3133, S.3370 and S.3418 Before the Senate Committee on Banking, Housing and Urban Affairs, 94th Congress, 2nd Session (1976), p. 1. 2Foreign Payments Disclosure: Hearings on H.R. 15481, S.3664, H.R. 13870 and H.R. 13953, Before the Subcommittee on Consumer Protection and Finance of the House Committee on Interstate and Foreign Commerce, 94th Congress, 2nd Session (1976), pp. 153-154. 23 An extensive literature dealing with these alternatives is available. For example, see Demski and Feltham (1978), Harris and Raviv (1978), Hurwicz and Shapiro (1978), Atkinson (1979), and Baiman (1979). For a review, see Demski (1980).

-40 - 24The owner-manager agency problem is not addressed here. For this study, it is assumed that a coalition of managers are the source of controls. Analysis of the way this coalition sets objectives and examination of whose interests it serves are issues which are beyond the scope of this paper. 2The impact of the dynamic nature of controls on an internal equilibrium was recognized by the SEC: "Because of the dynamic nature of internal accounting control and the resultant continuing nature of the evaluation process, the extent and timing of the review and monitoring of a system of internal accounting control are among the cost/benefit judgments involved in the concept of reasonable assurance." (Securities Exchange Act Release No. 34-15570, April 30, 1979, p. 26706.) 2For example, see Williamson (1963): "The essential notion that we propose in order to connect motives with behavior is that of expense preference. That is, the management does not have a neutral attitude toward costs. Directly or indirectly, certain classes of expenditure with them. In particular, staff expense, expenditures for emoluments and funds available for discretionary investment have value additional to that which derives from their productivity" (p. 1034.) 27 This argument was made by a group of financial executives (who did not participate in the empirical study) who reviewed an earlier version of the conceptual analyses presented here. It should not be construed from this analysis that these questionable payments necessarily constituted "variant behavior" from firms' objectives. In fact, if the benefits from these payments are sufficiently high, firms may make not expend resources to "control" them, even though these payments are now more costly because of the antibribery provisions of the FCPA. 29 2The importance of "good faith" in demonstrating compliance with the FCPA was emphasized by Harvey Pitt, then General Counsel for the SEC, at an American Bar Association Meeting in August 1978 (see Foreign Corrupt Practices Act Reporter, p. 506). For example, he advised adoption of corporate codes of conduct and procedures to check compliance with those codes as evidence of such good faith. 3According to Siedel (1980), "The unchallenged view expressed during Congressional hearings and debate that the Act does not expand the Commission's authority has not been contradicted by SEC enforcement actions to date, for in each case the FCPA violation has been an "add-on" to alleged violations of other securities law provisions" (pp. 33-34).

-41 - In several firms, managers who refused to participate cited these reaasons. An examination of the business press during that period revealed one firm on strike, one firm going through bankruptcy, and one firm going through a merger. 32 The sample of 49 firms studied covers 22 (two-digit) SIC industry and special groups, all ten deciles of firms ranked by size and profitability in the Fortune 1000 plus special group listings, and all major geographic regions of the United States. A detailed analysis of the sample is provided in Mautz, et al (1980). "Firms reporting questionable payments" were identified from the following three sources: (1) Companies reporting through December 31, 1976, as reported in Sensitive Payments by Corporations (Washington; Charles E. Simon, Inc., 1977); (2) Companies reporting questionable payments through December 1977 which were headquartered in New York, New Jersey and Connecticut, as reported in Simon and Kennedy (1978); and (3) Firms that had filed with the SEC, as of July 1979, lengthy reports detailing the findings of investigations of company affairs in light of questionable payments. These sources included about two-thirds of the (approximately) 600 companies which had revealed questionable payments by the time of the data collection. Identification of the remaining one-third of the firms disclosing questionable payments required considerable cost in searching SEC files which the author believed was not justified, considering the use of the data for this study. 34 These findings are consistent with a recent questionnaire study of managers in 674 firms, which found that the primary response to the FCPA was review and documentation of current internal control practices, which less than 10% of the respondents indicated a response of strengthening specific internal control measures. See Mautz, et al (1980) for more details. Questions listed are a subset of the total questions asked. Although the total set of questions were used for this study, only those that are most pertinent are included here to save space. age z ',; ". -,..J ' - ' ' - -.

-42 - REFERENCES AICPA (1979), Report of the Special Advisory Committee on Internal Accounting Control (New York: AICPA, 1979). Alchian, A., and H. Demsetz (1972), "Production, Information Costs, and Economic Organization," The American Economic Review (December, 1972). American Bar Association (1978) (The Committee on Corporate Law and Accounting, Section of Corporation, Banking and Business Law), A Guide to the Section 13(b)(2) Accounting Requirements of the Securities Exchange Act of 1934 (Washington: American Bar Association, 1978). Atkinson, A. (1979), "Information Incentives in a Standard-Setting Model of Control," Journal of Accounting Research (Spring, 1979). Baiman, S. (1979), "Optimal Conditional Variance Investigation for a Class of Principal-Agent Problems," Unpublished Working Paper, CarnegieMellon University (1979). Baumol, W. (1959), "The Revenue Maximization Hypothesis," in Business Behavior, Value and Growth (New York: Macmillan, 1959). Coase, R. H. (1937), "The Nature of the Firm," Econometrica (November, 1937). Cyert, R. M., and J. G. March (1963), A Behavioral Theory of the Firm (Prentice-Hall, 1963). Demski, J. (1980), "Lecture: Application to Managerial Accounting," in G. Lobo and M. Maher (ed.) Information Economics and Accounting Research. (Ann Arbor: The University of Michigan, Graduate School of Business Administration, 1980). Demski, J., and G. Feltham (1978), "Economic Incentives in Budgeting and Control Systems," The Accounting Review (April, 1978). Foreign Corrupt Practices Act Reporter (1979) (Cleveland: Business Laws, Inc., 1979). Harris, M., and A. Raviv (1978), "Some Results on Incentive Contracts with Applications to Education and Employment, Health Insurance and Law Enforcement," The American Economic Review, Vol. 68 (1978), 20-30. Hurwicz, L., and L. Shapiro (1978), "Incentive Structures Maximizing Residual Gain under Incomplete Information," The Bell Journal of Economics, Vol. 9 (1978), 180-191. Jensen, M. C., and W. H. Meckling (1976), "Theory of the Firm: Managerial Behavior, Agency Costs and Ownership Structure," Journal of Financial Economics, Vol. 3 (1976), 305-360.

-43 - Kennedy, T., and C. Simon (1978), An Examination of Questionable Payments (Washington: Charles C. Simon, Inc., 1978). Leibenstein, H. (1979), "A Branch of Economics Is Missing: Micro-Micro Theory," Journal of Economic Literature (June, 1979). Machlup, F. (1967), "Theories of the Firm: Marginalist, Behavioral, Managerial," American Economic Review (March, 1967). Maher, M. K. Ramanathan and R. Peterson (1979), "Information Accuracy, Preference Congruence and Employee Performance: A Field Study," Journal of Accounting Research (Autumn, 1979). March, J. G., and H. A. Simon (1958), Organizations (Wiley, 1958). Marschak, J. (1955), "Elements for a Theory of Teams," Management Science, Vol. 1 (1955), 127-137. Marschak, J., and R. Radner (1972), Economic Theory of Teams (New Haven: Yale University Press, 1972). Mautz, R., W. Kell, M. Maher, A. Merten, R. Reilly, D. Severance, and B. J. White (1980), Internal Control in U.S. Corporations, A Report on the State of the Art, Volume I (New York: Financial Executives Research Foundation, 1980). Mirrlees, J. (1976), "The Optimal Structure of Incentives and Authority within an Organization," Bell Journal of Economics (Spring, 1976). Monsen, R. J., and A. Downs (1965), "A Theory of the Large Managerial Firm," Journal of Political Economy, Vol. 73 (1965), 221-236. von Neuman, J., and 0. Morgenstern (1944), Theory of Games and Economic Behavior (Princeton: Princeton University Press, 1944). Ross, S. A. (1973), "The Economic Theory of Agency: The Principal's Problem," The American Economic Review, Vol. 63 (1973), 134-139. Salamon, G., and D. Smith (1979), "Corporate Control and Managerial Misrepresentation of Firm Performance," The Bell Journal of Economics (Spring, 1979). Senstive Payments by Corporations (1977) (Washington: Charles E. Simon, Inc., 1977). Siedel, G. (1980), "Legal Dimensions of Internal Accounting Control and the Foreign Corrupt Practices Act," Unpublished Working Paper, University of Michigan, forthcoming in Mautz, et al. (1980). Spence, M. (1975), "The Economics of Internal Organization," The Bell Journal of Economics (Spring, 1975).

9 -44 -Stiglitz, J. (1975), "Incentives, Risk and Information: Notes Towards a Theory of Hierarchy," Bell Journal of Economics (Autumn, 1975). Williamson, 0. (1975), Markets and Hierarchies: Analysis and Antitrust Implications (New York: The Free Press, 1975). Williamson, 0. (1963), "Managerial Discretion and Business Behavior," American Economic Review (December, 1963). Williamson, 0. (1976), "The Economics of Internal Organization: Exit and Voice in Relation to Markets and Hierarchies," American Economic Review (May, 1976). Williamson, O., M. L. Wachter, and J. E. Harris (1975), "Understanding the Employment Relation: The Analysis of Idiosyncratic Exchange," Bell Journal of Economics (Spring, 1975). Wilson, R. B. (1968), "The Theory of Syndicates," Econometrica, Vol. 36 (1968), 119-132.