National Academies Press: OpenBook

Conflict of Interest in Medical Research, Education, and Practice (2009)

Chapter: Appendix C: Conflict of Interest in Four Professions: A Comparative Analysis

« Previous: Appendix B: U.S. Public Health Service Regulations: Objectivity in Research (42 CFR 50)
Suggested Citation:"Appendix C: Conflict of Interest in Four Professions: A Comparative Analysis." Institute of Medicine. 2009. Conflict of Interest in Medical Research, Education, and Practice. Washington, DC: The National Academies Press. doi: 10.17226/12598.
×
Page 302
Suggested Citation:"Appendix C: Conflict of Interest in Four Professions: A Comparative Analysis." Institute of Medicine. 2009. Conflict of Interest in Medical Research, Education, and Practice. Washington, DC: The National Academies Press. doi: 10.17226/12598.
×
Page 303
Suggested Citation:"Appendix C: Conflict of Interest in Four Professions: A Comparative Analysis." Institute of Medicine. 2009. Conflict of Interest in Medical Research, Education, and Practice. Washington, DC: The National Academies Press. doi: 10.17226/12598.
×
Page 304
Suggested Citation:"Appendix C: Conflict of Interest in Four Professions: A Comparative Analysis." Institute of Medicine. 2009. Conflict of Interest in Medical Research, Education, and Practice. Washington, DC: The National Academies Press. doi: 10.17226/12598.
×
Page 305
Suggested Citation:"Appendix C: Conflict of Interest in Four Professions: A Comparative Analysis." Institute of Medicine. 2009. Conflict of Interest in Medical Research, Education, and Practice. Washington, DC: The National Academies Press. doi: 10.17226/12598.
×
Page 306
Suggested Citation:"Appendix C: Conflict of Interest in Four Professions: A Comparative Analysis." Institute of Medicine. 2009. Conflict of Interest in Medical Research, Education, and Practice. Washington, DC: The National Academies Press. doi: 10.17226/12598.
×
Page 307
Suggested Citation:"Appendix C: Conflict of Interest in Four Professions: A Comparative Analysis." Institute of Medicine. 2009. Conflict of Interest in Medical Research, Education, and Practice. Washington, DC: The National Academies Press. doi: 10.17226/12598.
×
Page 308
Suggested Citation:"Appendix C: Conflict of Interest in Four Professions: A Comparative Analysis." Institute of Medicine. 2009. Conflict of Interest in Medical Research, Education, and Practice. Washington, DC: The National Academies Press. doi: 10.17226/12598.
×
Page 309
Suggested Citation:"Appendix C: Conflict of Interest in Four Professions: A Comparative Analysis." Institute of Medicine. 2009. Conflict of Interest in Medical Research, Education, and Practice. Washington, DC: The National Academies Press. doi: 10.17226/12598.
×
Page 310
Suggested Citation:"Appendix C: Conflict of Interest in Four Professions: A Comparative Analysis." Institute of Medicine. 2009. Conflict of Interest in Medical Research, Education, and Practice. Washington, DC: The National Academies Press. doi: 10.17226/12598.
×
Page 311
Suggested Citation:"Appendix C: Conflict of Interest in Four Professions: A Comparative Analysis." Institute of Medicine. 2009. Conflict of Interest in Medical Research, Education, and Practice. Washington, DC: The National Academies Press. doi: 10.17226/12598.
×
Page 312
Suggested Citation:"Appendix C: Conflict of Interest in Four Professions: A Comparative Analysis." Institute of Medicine. 2009. Conflict of Interest in Medical Research, Education, and Practice. Washington, DC: The National Academies Press. doi: 10.17226/12598.
×
Page 313
Suggested Citation:"Appendix C: Conflict of Interest in Four Professions: A Comparative Analysis." Institute of Medicine. 2009. Conflict of Interest in Medical Research, Education, and Practice. Washington, DC: The National Academies Press. doi: 10.17226/12598.
×
Page 314
Suggested Citation:"Appendix C: Conflict of Interest in Four Professions: A Comparative Analysis." Institute of Medicine. 2009. Conflict of Interest in Medical Research, Education, and Practice. Washington, DC: The National Academies Press. doi: 10.17226/12598.
×
Page 315
Suggested Citation:"Appendix C: Conflict of Interest in Four Professions: A Comparative Analysis." Institute of Medicine. 2009. Conflict of Interest in Medical Research, Education, and Practice. Washington, DC: The National Academies Press. doi: 10.17226/12598.
×
Page 316
Suggested Citation:"Appendix C: Conflict of Interest in Four Professions: A Comparative Analysis." Institute of Medicine. 2009. Conflict of Interest in Medical Research, Education, and Practice. Washington, DC: The National Academies Press. doi: 10.17226/12598.
×
Page 317
Suggested Citation:"Appendix C: Conflict of Interest in Four Professions: A Comparative Analysis." Institute of Medicine. 2009. Conflict of Interest in Medical Research, Education, and Practice. Washington, DC: The National Academies Press. doi: 10.17226/12598.
×
Page 318
Suggested Citation:"Appendix C: Conflict of Interest in Four Professions: A Comparative Analysis." Institute of Medicine. 2009. Conflict of Interest in Medical Research, Education, and Practice. Washington, DC: The National Academies Press. doi: 10.17226/12598.
×
Page 319
Suggested Citation:"Appendix C: Conflict of Interest in Four Professions: A Comparative Analysis." Institute of Medicine. 2009. Conflict of Interest in Medical Research, Education, and Practice. Washington, DC: The National Academies Press. doi: 10.17226/12598.
×
Page 320
Suggested Citation:"Appendix C: Conflict of Interest in Four Professions: A Comparative Analysis." Institute of Medicine. 2009. Conflict of Interest in Medical Research, Education, and Practice. Washington, DC: The National Academies Press. doi: 10.17226/12598.
×
Page 321
Suggested Citation:"Appendix C: Conflict of Interest in Four Professions: A Comparative Analysis." Institute of Medicine. 2009. Conflict of Interest in Medical Research, Education, and Practice. Washington, DC: The National Academies Press. doi: 10.17226/12598.
×
Page 322
Suggested Citation:"Appendix C: Conflict of Interest in Four Professions: A Comparative Analysis." Institute of Medicine. 2009. Conflict of Interest in Medical Research, Education, and Practice. Washington, DC: The National Academies Press. doi: 10.17226/12598.
×
Page 323
Suggested Citation:"Appendix C: Conflict of Interest in Four Professions: A Comparative Analysis." Institute of Medicine. 2009. Conflict of Interest in Medical Research, Education, and Practice. Washington, DC: The National Academies Press. doi: 10.17226/12598.
×
Page 324
Suggested Citation:"Appendix C: Conflict of Interest in Four Professions: A Comparative Analysis." Institute of Medicine. 2009. Conflict of Interest in Medical Research, Education, and Practice. Washington, DC: The National Academies Press. doi: 10.17226/12598.
×
Page 325
Suggested Citation:"Appendix C: Conflict of Interest in Four Professions: A Comparative Analysis." Institute of Medicine. 2009. Conflict of Interest in Medical Research, Education, and Practice. Washington, DC: The National Academies Press. doi: 10.17226/12598.
×
Page 326
Suggested Citation:"Appendix C: Conflict of Interest in Four Professions: A Comparative Analysis." Institute of Medicine. 2009. Conflict of Interest in Medical Research, Education, and Practice. Washington, DC: The National Academies Press. doi: 10.17226/12598.
×
Page 327
Suggested Citation:"Appendix C: Conflict of Interest in Four Professions: A Comparative Analysis." Institute of Medicine. 2009. Conflict of Interest in Medical Research, Education, and Practice. Washington, DC: The National Academies Press. doi: 10.17226/12598.
×
Page 328
Suggested Citation:"Appendix C: Conflict of Interest in Four Professions: A Comparative Analysis." Institute of Medicine. 2009. Conflict of Interest in Medical Research, Education, and Practice. Washington, DC: The National Academies Press. doi: 10.17226/12598.
×
Page 329
Suggested Citation:"Appendix C: Conflict of Interest in Four Professions: A Comparative Analysis." Institute of Medicine. 2009. Conflict of Interest in Medical Research, Education, and Practice. Washington, DC: The National Academies Press. doi: 10.17226/12598.
×
Page 330
Suggested Citation:"Appendix C: Conflict of Interest in Four Professions: A Comparative Analysis." Institute of Medicine. 2009. Conflict of Interest in Medical Research, Education, and Practice. Washington, DC: The National Academies Press. doi: 10.17226/12598.
×
Page 331
Suggested Citation:"Appendix C: Conflict of Interest in Four Professions: A Comparative Analysis." Institute of Medicine. 2009. Conflict of Interest in Medical Research, Education, and Practice. Washington, DC: The National Academies Press. doi: 10.17226/12598.
×
Page 332
Suggested Citation:"Appendix C: Conflict of Interest in Four Professions: A Comparative Analysis." Institute of Medicine. 2009. Conflict of Interest in Medical Research, Education, and Practice. Washington, DC: The National Academies Press. doi: 10.17226/12598.
×
Page 333
Suggested Citation:"Appendix C: Conflict of Interest in Four Professions: A Comparative Analysis." Institute of Medicine. 2009. Conflict of Interest in Medical Research, Education, and Practice. Washington, DC: The National Academies Press. doi: 10.17226/12598.
×
Page 334
Suggested Citation:"Appendix C: Conflict of Interest in Four Professions: A Comparative Analysis." Institute of Medicine. 2009. Conflict of Interest in Medical Research, Education, and Practice. Washington, DC: The National Academies Press. doi: 10.17226/12598.
×
Page 335
Suggested Citation:"Appendix C: Conflict of Interest in Four Professions: A Comparative Analysis." Institute of Medicine. 2009. Conflict of Interest in Medical Research, Education, and Practice. Washington, DC: The National Academies Press. doi: 10.17226/12598.
×
Page 336
Suggested Citation:"Appendix C: Conflict of Interest in Four Professions: A Comparative Analysis." Institute of Medicine. 2009. Conflict of Interest in Medical Research, Education, and Practice. Washington, DC: The National Academies Press. doi: 10.17226/12598.
×
Page 337
Suggested Citation:"Appendix C: Conflict of Interest in Four Professions: A Comparative Analysis." Institute of Medicine. 2009. Conflict of Interest in Medical Research, Education, and Practice. Washington, DC: The National Academies Press. doi: 10.17226/12598.
×
Page 338
Suggested Citation:"Appendix C: Conflict of Interest in Four Professions: A Comparative Analysis." Institute of Medicine. 2009. Conflict of Interest in Medical Research, Education, and Practice. Washington, DC: The National Academies Press. doi: 10.17226/12598.
×
Page 339
Suggested Citation:"Appendix C: Conflict of Interest in Four Professions: A Comparative Analysis." Institute of Medicine. 2009. Conflict of Interest in Medical Research, Education, and Practice. Washington, DC: The National Academies Press. doi: 10.17226/12598.
×
Page 340
Suggested Citation:"Appendix C: Conflict of Interest in Four Professions: A Comparative Analysis." Institute of Medicine. 2009. Conflict of Interest in Medical Research, Education, and Practice. Washington, DC: The National Academies Press. doi: 10.17226/12598.
×
Page 341
Suggested Citation:"Appendix C: Conflict of Interest in Four Professions: A Comparative Analysis." Institute of Medicine. 2009. Conflict of Interest in Medical Research, Education, and Practice. Washington, DC: The National Academies Press. doi: 10.17226/12598.
×
Page 342
Suggested Citation:"Appendix C: Conflict of Interest in Four Professions: A Comparative Analysis." Institute of Medicine. 2009. Conflict of Interest in Medical Research, Education, and Practice. Washington, DC: The National Academies Press. doi: 10.17226/12598.
×
Page 343
Suggested Citation:"Appendix C: Conflict of Interest in Four Professions: A Comparative Analysis." Institute of Medicine. 2009. Conflict of Interest in Medical Research, Education, and Practice. Washington, DC: The National Academies Press. doi: 10.17226/12598.
×
Page 344
Suggested Citation:"Appendix C: Conflict of Interest in Four Professions: A Comparative Analysis." Institute of Medicine. 2009. Conflict of Interest in Medical Research, Education, and Practice. Washington, DC: The National Academies Press. doi: 10.17226/12598.
×
Page 345
Suggested Citation:"Appendix C: Conflict of Interest in Four Professions: A Comparative Analysis." Institute of Medicine. 2009. Conflict of Interest in Medical Research, Education, and Practice. Washington, DC: The National Academies Press. doi: 10.17226/12598.
×
Page 346
Suggested Citation:"Appendix C: Conflict of Interest in Four Professions: A Comparative Analysis." Institute of Medicine. 2009. Conflict of Interest in Medical Research, Education, and Practice. Washington, DC: The National Academies Press. doi: 10.17226/12598.
×
Page 347
Suggested Citation:"Appendix C: Conflict of Interest in Four Professions: A Comparative Analysis." Institute of Medicine. 2009. Conflict of Interest in Medical Research, Education, and Practice. Washington, DC: The National Academies Press. doi: 10.17226/12598.
×
Page 348
Suggested Citation:"Appendix C: Conflict of Interest in Four Professions: A Comparative Analysis." Institute of Medicine. 2009. Conflict of Interest in Medical Research, Education, and Practice. Washington, DC: The National Academies Press. doi: 10.17226/12598.
×
Page 349
Suggested Citation:"Appendix C: Conflict of Interest in Four Professions: A Comparative Analysis." Institute of Medicine. 2009. Conflict of Interest in Medical Research, Education, and Practice. Washington, DC: The National Academies Press. doi: 10.17226/12598.
×
Page 350
Suggested Citation:"Appendix C: Conflict of Interest in Four Professions: A Comparative Analysis." Institute of Medicine. 2009. Conflict of Interest in Medical Research, Education, and Practice. Washington, DC: The National Academies Press. doi: 10.17226/12598.
×
Page 351
Suggested Citation:"Appendix C: Conflict of Interest in Four Professions: A Comparative Analysis." Institute of Medicine. 2009. Conflict of Interest in Medical Research, Education, and Practice. Washington, DC: The National Academies Press. doi: 10.17226/12598.
×
Page 352
Suggested Citation:"Appendix C: Conflict of Interest in Four Professions: A Comparative Analysis." Institute of Medicine. 2009. Conflict of Interest in Medical Research, Education, and Practice. Washington, DC: The National Academies Press. doi: 10.17226/12598.
×
Page 353
Suggested Citation:"Appendix C: Conflict of Interest in Four Professions: A Comparative Analysis." Institute of Medicine. 2009. Conflict of Interest in Medical Research, Education, and Practice. Washington, DC: The National Academies Press. doi: 10.17226/12598.
×
Page 354
Suggested Citation:"Appendix C: Conflict of Interest in Four Professions: A Comparative Analysis." Institute of Medicine. 2009. Conflict of Interest in Medical Research, Education, and Practice. Washington, DC: The National Academies Press. doi: 10.17226/12598.
×
Page 355
Suggested Citation:"Appendix C: Conflict of Interest in Four Professions: A Comparative Analysis." Institute of Medicine. 2009. Conflict of Interest in Medical Research, Education, and Practice. Washington, DC: The National Academies Press. doi: 10.17226/12598.
×
Page 356
Suggested Citation:"Appendix C: Conflict of Interest in Four Professions: A Comparative Analysis." Institute of Medicine. 2009. Conflict of Interest in Medical Research, Education, and Practice. Washington, DC: The National Academies Press. doi: 10.17226/12598.
×
Page 357

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

C Conflict of Interest in Four Professions: A Comparative Analysis Michael Davis* and Josephine Johnston** This paper presents a selective survey of the ways in which important professions other than medicine understand and regulate conflicts of inter- est. The professions evaluated here—law (lawyers), accountancy (certified public accountants [CPAs]), architecture, and engineering—each differ from medicine in having clients or employers rather than patients as the focus of concern. The difference is not simply one of terminology. A client or an employer is not necessarily human. Many are corporations or governments. Even the human clients differ from patients. With some exceptions (e.g., clients accused of crimes), they are typically healthy, calm, and relatively well-informed about the service to be provided; they are seldom as vulner- able as a physician’s patient typically is. A client or employer simply asks that something be done (a building put up, a machine designed, a contract drawn, or a company audited). Emergencies are much rarer in these pro- fessions than they are in medicine, and time to think through a problem is more plentiful. Because of their relative sophistication and bargaining strength (compared both with patients and with the professional in ques- tion), clients or employers need not readily consent to accept the conflicts disclosed to them; they are more likely to insist that a conflict be avoided or resolved or to use the conflict to better the bargain. In other words, law, ac-   Michael Davis, Ph.D., is a senior fellow within the Center for Study of Ethics in the * Professions and a professor of philosophy in the Humanities Department, Illinois Institute of Technology.   * Josephine Johnston, L.L.B., is a research scholar at the Hastings Center, Garrison, New * York. 302

APPENDIX C 303 counting, architecture, and engineering are professions in which one might expect much less concern with conflicts of interest than in medicine. Although these are the chief differences between medicine and the pro- fessions discussed here, they are not the only ones. These other professions differ substantially in size from medicine—and from each other. Physicians outnumber architects in the United States by about 10 to 1, engineers out- number physicians by about 3 to 1, and the numbers of individuals in the other professions fall somewhere in between. Importantly, only one profes- sion, engineering, does much that physicians would recognize as scientific research. The professions evaluated here were chosen because none is a close an- alogue of medicine. Medicine tends to be the model for adjacent professions (osteopathy, dentistry, pharmacy, nursing, and so on). The comparison of medicine with an adjacent profession would provide less contrast and there- fore less understanding of conflict of interest as a general problem for pro- fessions. All of the professions discussed here have substantial experience with employment in large organizations. Two of the professions—engineer- ing and accounting—have a long history of employment in such organiza- tions. Only a small minority of engineers has ever been self-employed in the way that most physicians, except those in research and teaching, were until recently. Even self-employed architects, lawyers, and accountants of- ten work for and in large organizations in a way that physicians have only recently begun to do in large numbers. Looking at how these nonmedical professions respond to the conflicts of interest that are more likely to arise in large organizations should help physicians both look critically at present arrangements and anticipate the future. Finally, these professions all recog- nize conflicts of interest as posing a threat to the integrity of the profession and have developed ethics rules to address the threat. TERMINOLOGY “Conflict of interest” is not an old term. The first court case to use it in something like the sense that is now standard occurred in 1949. Federal legislation first addressed conflict of interest in the late 1950s. The Index of Legal Periodicals had no heading for “conflict of interest” until 1967; Black’s Law Dictionary had none until 1979. No ordinary dictionary of English seems to have had an entry for “conflict of interest” before 1971. The term also began to appear in codes of ethics in the 1970s, although related terms, such as “adverse interest,” “conflicting interest,” “bias,”   In re Equitable Office Bldg. Corp., D.C.N.Y., 83 F. Supp. 531.   Staffreport of the Antitrust Subcommittee (Subcommittee No. 5) of House Judiciary Com- mittee, 85th Cong., 2d sess., Federal Conflict of Interest Legislation (Comm. Print 1958).

304 CONFLICT OF INTEREST “prejudice,” and the like appeared in codes much earlier. This short history may explain, at least in part, the variation in how the term is used among professions. We are all trying to keep pace with the usage. The term “conflict of interest” is not self-explanatory but is an idiom or term of art (a term designed to pick out a phenomenon until then lack- ing a suitable name). For the professions discussed here, the term groups together a range of scenarios in which the professional judgment of the individual in question risks being compromised. These professions do not use explicit definitions of “conflict of interest” but instead describe in their codes a variety of situations that fall under the heading “conflict of inter- est” and that must either be avoided or managed in specified ways. For example, the definitions section of the American Bar Association’s (ABA’s) Model Rules of Professional Conduct includes definitions of “informed consent” and “fraud” but not “conflict of interest.” Instead, situations labeled conflicts of interest are described in the Model Rules. Similarly, the American Institute of Certified Public Accountants’ (AICPA’s) Code of Professional Conduct includes a definition of some of the terms used in its conflict of interest rules, such as “immediate family” but not “conflict of interest.” In fact, AICPA’s Code does not use the term “conflict of interest” at all, speaking instead of various threats to “independence” and “objectiv- ity,” including the threat posed by certain financial interests. The major concern uniting the professions’ use of the term is to protect the judgment of individual professionals from undue influence, whether the risk arises from gifts or kickbacks; an individual’s personal (generally fi- nancial) interests; or the interests of family members, colleagues, or current and former clients. In some of the professions, a situation—for example, the representation of both plaintiff and defendant in the same legal case—is labeled a conflict of interest when it would be described as a conflict of ob- ligations or responsibilities in the report to which this paper is an appendix (see Chapter 2) because neither obligation would be considered secondary to the other. For all of the professions discussed in this paper, a certain sort of expert   Neil R. Luebke, “Conflict of Interest as a Moral Category,” Business and Professional Ethics Journal 1987; 6 (Spring): 66–81.   Michael Davis, “Conflict of Interest Revisited,” Business and Professional Ethics Journal 1993; 12 (Winter): 21–41.   American Bar Association, Model Rules for Professional Conduct: Rules 1.0. www.abanet. org/cpr/mrpc/rule_1_0.html.   American Bar Association, Model Rules for Professional Conduct: Rules 1.7–1.10. www. abanet.org/cpr/mrpc/mrpc_toc.html.   The American Institute of Certified Public Accountants, AICPA Code of Professional Responsibility: Section 100—Independence, Integrity, and Objectivity. www.aicpa.org/About/ code/sec100.htm.

APPENDIX C 305 and trustworthy judgment in individual situations (the judgment character- istic of the profession) is what makes members of the profession useful. A conflict of interest makes that judgment unreliable just when reliability is needed. A conflict of interest is therefore always considered a threat to the good that the profession seeks to achieve and is often also a threat to the profession’s reputation. That is what makes having a conflict of interest a serious concern in professional ethics. The next four sections of this paper consider in detail how the four professions on our list understand conflict of interest, respond to it, and why. The final section summarizes and compares the professions and iden- tifies approaches from which medical research, education, and practice might learn. LAWYERS The main legal professions—lawyers and judges—have traditionally taken conflict of interest very seriously. Because justice is to be fairly meted out, interests that might cause a judge to be or appear to be partial are also generally prohibited. Although lawyers owe obligations to the legal system and the public, their primary obligation is to their clients; interests that might interfere with this obligation are generally to be avoided. The legal professions have some of the strictest rules about conflict of interest and have a long history of examining and enforcing those rules. In the interests of space, the focus here is on lawyers because their work, in most respects, is more like that of physicians than is the work of judges. (Judges are more like physicians serving on drug and device approval panels or as authors of review articles, whose charge is to weigh all the evidence and reach a reasoned and impartial decision.) In legal practice, conflicts of interest are conceptualized in the context of the attorney-client relationship, which is protected by very strong obliga- tions of loyalty and confidentiality. Broadly speaking, two kinds of conflict are understood to arise in that relationship: first, conflicts between the in- terests of two or more clients (whether they be current clients or a current client and a former client) and, second, conflicts between the interests of one or more clients and the personal interests of the attorney. The first kind of conflict is created by the act of entering into a certain attorney-client relationship. For that reason, lawyers routinely conduct “conflicts checks” before taking on a new client or a new file from an existing client. The second kind of conflict can be created by either enter-   For an example of rules of professional conduct for judges, see New York State Commission on Judicial Conduct, Rules of Conduct, at http://www.scjc.state.ny.us/Legal%20Authorities/ rgjc.htm. Judicial ethics emphasizes independence.

306 CONFLICT OF INTEREST ing into a new attorney-client relationship; taking on a new file from an existing client; or taking on a personal interest, including but not limited to a financial interest. In part, because the conflict of interest of one lawyer is, as a general principle, imputed to all lawyers working in the same firm or group practice, procedures, internal databases, and software have been developed to assist large firms in identifying possible conflicts of interest. So, for example, a lawyer who marries will immediately report that change of status to his firm along with the spouse’s investments, family connec- tions, employer, and the like. In some cases, law firms have staff dedicated to monitoring conflicts of interest, including, according to one New York City law firm partner whom we spoke with, general counsel whose risk management responsibilities include conflict of interest issues. Lawyers also use letters of engagement to carefully specify which of the firm’s lawyers will be working for the client on the particular matter to avoid or make more manageable any future conflict of interest. For the first kind of conflict, the analogy with medicine is not particu- larly strong: physicians are not generally constrained from taking on new patients because of their loyalty to other (current or former) patients, al- though analogous problems concerning confidentiality may arise when one physician serves several members of the same family. The analogy becomes somewhat stronger if one thinks of a pharmaceutical company or a device manufacturer with whom a physician has a financial relationship as one “client” and one or more patients as the other “client.” The legal profession’s management of the second kind of conflict—a conflict between a lawyer’s personal interests and the interests of a client— could provide more direct instruction to medicine, insofar as there is a concern that physicians’ financial and other relationships with industry might lead physicians to make clinical decisions that they would not have made but for those financial or other relationships. Furthermore, the legal profession’s general attitude toward conflict of interest might be instructive for medicine. Conflicts of interest are under- stood to be a common feature of legal practice for which the profession has developed norms, rules, and procedures. Censure attaches not to finding oneself in a position in which agreeing to represent a client would create a conflict of interest (all lawyers are in this position from time to time, even though they try to avoid it) but to agreeing to represent that client without properly addressing the conflict of interest. Lawyers share some similarities with physicians. Until recently, many lawyers worked alone or in small group practices. Lawyers are under a strong obligation of fidelity to their individual clients; and although some clients are large companies or sophisticated and powerful individuals, many are vulnerable individuals, including people who are in trouble with the law, are victims of physical harm or abuse, or are making major decisions

APPENDIX C 307 that can have a lasting impact on their lives and the lives of their fami- lies (e.g., buying or selling property, making a will, or adopting a child). However, although some lawyers represent clients with a compromised decision-making ability, lawyers do not routinely deal with clients whose decision-making ability may be medically impaired. In the United States, as in most other common-law jurisdictions, a lawyer (often called an attorney) may conduct all aspects of litigation (in- cluding court appearances); may represent clients in negotiations; may give legal advice; and may prepare contracts, wills, and other legal documents. The specific criteria for admission to the bar are set by each state: candi- dates must generally hold a law degree (J.D.) from an accredited law school; pass that state’s bar examination; and in all but three jurisdictions, pass the Multistate Professional Responsibility Examination, which is a 2-hour-long multiple-choice test that includes questions about conflict of interest. To be accredited, American law schools are required to provide sub- stantial instruction to all J.D. students in the values, rules, and responsibili- ties of the legal profession, including instruction in the identification and management of conflicts of interest.10 Law professors themselves are subject to any conflict of interest policies of their own institutions. In keeping with the Association of American Law Schools’ 2003 Statement of Good Prac- tices by Law Professors in the Discharge of Their Ethical and Professional Responsibilities, law professors are obligated in publications and presenta- tions to “disclose the material facts relating to receipt of direct or indirect payment for, or any personal economic interest in, any covered activity that the professor undertakes in a professorial capacity.”11 To maintain the license to practice law, 41 U.S. states require comple- tion of a prescribed number of hours of continuing legal education (CLE), and 36 of these states mandate the inclusion of professional responsibility (also called “legal ethics”), including in some states “elimination of bias.”12 Providers of mandatory CLE, which can be law schools, law firms (which offer CLE only to lawyers in-house or to outside lawyers as well), or private companies, must individually be accredited by each state’s CLE accrediting authority. Mandatory CLE can be funded in a number of ways: it may be provided for a fee or it may be offered for free by the ABA, by state bar   American Bar Association, Bar Admissions Basic Overview. www.abanet.org/legaled/ baradmissions/basicoverview.html. 10  American Bar Association, 2007–2008 Standards for Approval of Law Schools, Interpre- tation 302-9. www.abanet.org/legaled/standards/standards.html. 11  Association of American Law Schools, Statement of Good Practices by Law Professors in the Discharge of Their Ethical and Professional Responsibilities, 2003. www.aals.org/ about_handbook_sgp_eth.php. 12  American Bar Association, Summary of MCLE Jurisdiction Requirements. www.abanet. org/cle/mcleview.html.

308 CONFLICT OF INTEREST associations, or by law firms. (One lawyer with whom we spoke noted that large malpractice firms sometimes sponsor CLE.) The lawyer or the lawyer’s employer pays the fee. Providers are required to offer tuition assistance to unemployed attorneys, attorneys working in the public sector, and those in a financial hardship situation. Lawyers seem unconcerned about the prospect of commercial interests being involved with CLE—they even allow corporate sponsorship at CLE events. Of course, in general, commercial providers of CLE are not potential clients or adverse parties but simply the makers of the tools or the providers of the services that lawyers use in the course of their work. As a result of these requirements, virtually all, if not all, U.S. lawyers have received instruction on the identification and management of conflicts of interest, and many continue to address this issue in their CLE. Because the legal profession has developed considerable case law, detailed rules (described below), and legal scholarship (two or three dozen articles per year) addressing lawyers’ conflicts of interest, there is much for American law students and lawyers to learn. Canons, Model Codes, and Model Rules Although local bar associations began to appear in the United States in the late 19th century, most U.S. lawyers at the time were only informally controlled by reputation and peer pressure. The ABA was founded in 1878; and one of its first major initiatives became, in 1908, the Canons of Professional Ethics, developed in response to a perceived need to promote and vouch for the integrity (or reliability) of lawyers generally.13 Initially there were 32 canons, and the number of canons eventually expanded to 47. The individual canons were fairly brief (the briefest is one sentence of two dozen words, whereas the longest is a few paragraphs). They were not accompanied by further guidance or detailed explanation, as is found in modern codes of legal ethics. Nevertheless, they were influential. By 1924, virtually every state and local bar association had adopted the canons.14 A number of the canons are relevant to conflict of interest, although it was the sixth canon (titled Adverse Influences and Conflicting Interests) that addressed the issue directly. Canon 6 consisted of three short paragraphs, and although it was fairly unsophisticated and incomplete, it captured the major conflict of interest issues that attorneys face even today. It attempted 13  Ted Schneyer, How Things Have Changed: Contrasting the Regulatory Environment of the Canons and the Model Rules. www.abanet.org/cpr/schneyer.pdf. 14  James M. Altman, “Considering the ABA’s 1908 Canons of Ethics,” Fordham Law Re- view 2003; 71: 2395–2524, at 2396, quoting Report of the Standing Committee on Profes- sional Ethics and Grievances, American Bar Association Report 1924; 49: at 466, 467.

APPENDIX C 309 a definition of client-client conflict of interest as a conflict of obligation (“when, in behalf of one client, it is his duty to contend for that which duty to another client requires him to oppose”), identified client loyalty and confidentiality as two values threatened by conflict of interest, identi- fied the particular problems of concurrent and subsequent representation of conflicting clients, and proposed one remedy for concurrent represen- tation (disclosure followed by informed consent).15 Other canons dealt with related conflict of interest issues, including a prohibition on a lawyer purchasing an interest in the subject that is a matter of litigation and a requirement that contingency fee arrangements be supervised by the court to prevent unjust charges. Fifteen amendments to the canons (none related to Canon 6) were adopted before the ABA developed a new code, the Model Code of Profes- sional Responsibility, in 1969.16 The Model Code was a far more detailed document than the canons. It contained nine “canons” (described as “axi- omatic norms”). These doubled as section titles. Each canon was followed by a series of Ethical Considerations and Disciplinary Rules. The Ethical Considerations were described as “aspirational in character,” representing the objectives toward which all lawyers should strive. The Disciplinary Rules were, unlike the Ethical Considerations, mandatory and set “the minimum level of conduct below which no lawyer can fall without being subject to disciplinary action.” Conflict of interest was mainly addressed in Canon 5 of the Model Code, which states: “A Lawyer Should Exercise Independent Professional Judgment on Behalf of a Client.” The first of the 24 Ethical Considerations for Canon 5 explains that [The] professional judgment of a lawyer should be exercised, within the bounds of the law, solely for the benefit of his client and free of compro- mising influences and loyalties. Neither his personal interests, the interests of other clients, nor the desires of third persons should be permitted to dilute his loyalty to his client. The seven Disciplinary Rules describe a mix of situations in which a conflict of interest is prohibited outright and situations in which the interest is per- missible only after the conflict of interest is fully disclosed to and informed consent is received from the client or clients in question. Although the states more or less uniformly adopted the Model Code, it was soon abandoned.17 In 1983, the ABA adopted a replacement. One theory about why the Model Code was so quickly replaced is that it mixed 15  American Bar Association Canons of Ethics, Canon 6. 16  American Bar Association, Model Code of Professional Responsibility, 1983. www.law. cornell.edu/ethics/aba/mcpr/MCPR.HTM. 17  Charles W. Wolfram, Modern Legal Ethics, West Publishing, 1986.

310 CONFLICT OF INTEREST minimum standards of conduct permitted under the law with ethical rules intended to set higher standards and therefore was perceived as confusing ethics with law (and, in the process, reducing ethical standards).18 In its report to the ABA recommending the adoption of a new code of profes- sional responsibility, the Kutak Commission on Evaluation and Professional Standards cited a steady increase in concern about professional ethics, including Supreme Court cases, statutes and regulations, opinions of the ABA’s Committee on Ethics and Professional Responsibility, and reports and articles, as leading it to reconsider the Model Code.19 The commission ultimately concluded that amendments would not suffice to address this increased concern, and so, just 15 years after adopting the Model Code, the ABA replaced it with the Model Rules of Professional Conduct. The states were slow to adopt the Model Rules, although today all states but Califor- nia, Maine, and New York have professional conduct rules that follow the format of the Model Rules. (New York still follows the Model Code, and California and Maine have developed their own rules.)20 The Model Rules are fairly detailed, clustered under eight headings, and accompanied by lengthy comments. The Model Rules most closely related to conflict of interest fall under the heading “Client-Lawyer Relationship” and are described in the following sections. Rule 1.7. Conflict of Interest: Current Clients Rule 1.721 begins with a strong statement of general principle, fol- lowed by a description of the circumstances in which the general principle does not apply. The general principle is that a lawyer “shall not represent a client if the representation involves a concurrent conflict of interest.” A concurrent conflict of interest is defined as the situation in which the representation of one client will be directly adverse to that of another cli- ent (conflict of obligation) or there is a “significant risk” that representing one client will be “materially limited” by the lawyer’s responsibilities to another client, a former client, or a third party or by the personal interest of the lawyer (true conflict of interest). However, the lawyer may proceed 18  Robert P. Lawry, “The Law and Ethics of Lawyers’ Conflicts of Interest,” in Thomas Murray and Josephine Johnston (eds.), Ethical Issues in Financial Conflicts of Interest in Biomedical Research, forthcoming. 19  American Bar Association Commission on Evaluation of Professional Standards (Robert J. Kutak Chairman), “Model Rules of Professional Conduct: Discussion Draft,” January 20, 1980. www.abanet.org/cpr/mrpc/kutak_1-80.pdf. 20  American Bar Association website, ABA Model Rules of Professional Conduct: State Adoption of Model Rules. www.abanet.org/cpr/mrpc/model_rules.html. 21  American Bar Association, Model Rules for Professional Conduct: Rule 1.7. www.abanet. org/cpr/mrpc/rule_1_7.html.

APPENDIX C 311 despite this conflict with the written informed consent of each client and provided that so proceeding (1) is not prohibited by law and (2) will not involve representing two opposing parties in litigation and (3) provided that the lawyer “reasonably believes” that he or she can provide competent and diligent representation to both clients. (The Model Rules define “reasonably believes” in both subjective and objective terms: the lawyer must actually believe, and the belief must be reasonable.) Rule 1.7 is accompanied by a comment, which is 35 paragraphs long.22 Its second paragraph describes the process that lawyers must go through under the rule: they must 1) clearly identify the client or clients; 2) determine whether a conflict of interest exists; 3) decide whether the representation may be undertaken despite the existence of a conflict, i.e., whether the client’s consent could be an appropriate cure; and 4) if so, consult with the clients affected under paragraph (a) [any clients affected by a concurrent conflict of interest] and obtain their informed consent, confirmed in writing. Determining whether a conflict of interest exists often involves some judgment (although the comment casts the net fairly widely to include cases both of “direct adverseness” and of “significant risk that a lawyer’s ability [to act for the client] will be materially limited”). The key judgment here, however, is whether the conflict is “consentable”—bearing in mind that the presumption is that a lawyer must not represent opposing parties in litiga- tion or where prohibited by law. The rationale for the division between consentable and nonconsentable seems to be that some conflicts of interest are too risky for the client or profession—for example, the lawyer might ap- pear to a reasonable outsider to be taking egregious advantage of the client for the lawyer’s personal benefit (even though the lawyer is not). Paragraph 14 of the comment for Rule 1.7 begins by noting that al- though clients may ordinarily consent to representation notwithstanding a conflict, “some conflicts are nonconsentable, meaning that the lawyer can- not properly ask for . . . agreement or provide representation on the basis of the client’s consent.” Although it is long, the comment provides little additional guidance on how to determine whether a conflict is consentable other than to note at Paragraph 15 that representation is prohibited if the lawyer “cannot reasonably conclude that [he or she] will be able to provide competent and diligent representation.” The next two rules provide more specific guidance. 22  AmericanBar Association, Model Rules for Professional Conduct: Rule 1.7 Comment. www.abanet.org/cpr/mrpc/rule_1_7_comm.html.

312 CONFLICT OF INTEREST Rule 1.8. Conflict of Interest: Current Clients: Specific Rules Rule 1.823 adds to the general principles contained in Rule 1.7 10 classes of conflict of interest situations, some of which may be resolved with the consent of the client (and sometimes subject to other protective mea- sures) and some of which cannot be resolved even with consent. Rule 1.8 therefore helps lawyers to determine when a conflict may be consentable. Conflict of interest situations that are not consentable include • soliciting or preparing an instrument to receive a substantial gift from a client unless the client is a relative, • negotiating literary or media rights that would substantially rely on information relating to the representation, • providing financial assistance to a client for litigation except where the client is indigent or where litigation costs will be repaid under a contin- gency agreement or lien, and • having sexual relations with a client unless the sexual relationship preceded the attorney-client relationship. Conflict of interest situations that are in principle consentable (provided that other conditions are met, such as advising the client of the desirability of seeking independent legal counsel) include • entering into a business transaction with a client, • knowingly acquiring an ownership or other interest that is adverse to a client, • using information about one client to another client’s disadvantage, • accepting compensation for representing a client from a third party, and • representing two or more clients in an aggregated settlement or agreement (for example, both parties in a friendly divorce). In such situations, provided that full disclosure is followed by valid informed consent, a lawyer might reasonably be able to argue that an “arm’s-length transaction” took place (something not possible in the non- consentable situations), that is, that the client was fully able to look after its own interests without relying on the lawyer. 23  AmericanBar Association, Model Rules for Professional Conduct: Rule 1.8. www.abanet. org/cpr/mrpc/rule_1_8.html.

APPENDIX C 313 Rule 1.9. Duties to Former Clients Rule 1.924 provides that a lawyer should not agree to represent a per- son if the lawyer has previously represented a client in the same or a related matter and the interests of the new person are materially adverse to the interests of the former client, unless the lawyer has the written informed consent of the former client. (The requirement of “written” consent assures both a record, in case of a later dispute, and more formality at the time that consent is given.) The same rule applies when a lawyer knowingly takes on a new client whose interests are materially adverse to the interests of a former or a current client of that lawyer’s former firm and about whom the lawyer has acquired protected information. This rule is designed to respond to lawyer mobility and to ensure that lawyers do not bring with them conflicts of interest from their previous firms without the consent of the former client. In contrast to the first two parts of Rule 1.9, which allow the lawyer to represent the new client with the informed consent of the former client, the third part of Rule 1.9 (which concerns loyalty rather than conflict of interest) provides that information about former clients cannot be used to the disadvantage of the former client unless use of that information is otherwise allowed in the rules (that is, loyalty to the client continues after the representation ends). Like Rule 1.8, therefore, Rule 1.9 distinguishes be- tween conflict of interest situations that are low risk enough to be resolved by informed consent and those that are nonconsentable because the risk is too high to be resolved by disclosure and consent or the problem is more serious than conflict of interest (conscious disloyalty). Rule 1.10. Imputation of Conflicts of Interest: General Rule Rule 1.1025 is extremely important today when so many lawyers prac- tice in large firms rather than as sole practitioners. Under the rule, the conflicts of interest of one lawyer in a firm are imputed to all lawyers in the firm: “a firm of lawyers is essentially one lawyer for purposes of the rules governing loyalty to the client.”26 The rule prohibits a member of a law firm from knowingly representing a client that any one of them practicing alone would be prohibited from representing under Rules 1.7 and 1.9, un- less the prohibition is based on a personal interest of the lawyer and “does 24  American Bar Association, Model Rules for Professional Conduct: Rule 1.9. www.abanet. org/cpr/mrpc/rule_1_9.html. 25  American Bar Association, Model Rules for Professional Conduct: Rule 1.10. www. abanet.org/cpr/mrpc/rule_1_10.html. 26  American Bar Association, Model Rules for Professional Conduct: Rule 1.10. comment, www.abanet.org/cpr/mrpc/rule_1_10_comm.html.

314 CONFLICT OF INTEREST not present a significant risk of materially limiting the representation of the client by the remaining lawyers in the firm.” Disqualification under the rule may be waived with the consent of the affected client (and subject to the prohibitions contained in Rule 1.7). Rule 1.10 is strict and can be very burdensome for large law firms. Law firms have devised methods for “screening” lawyers within firms as a way of managing conflicts of interest (discussed in more detail below), but it is important to note that screens (which in other professions or arenas might be described as “firewalls”) are not discussed in the Model Rules, except in limited situations involving former government lawyers. In addition to these codes of ethics, case law has developed over several centuries to deal with lawyers’ conflicts of interest (under various names). In the United States, this case law is helpfully summarized in the American Law Institute’s Restatement of the Law Governing Lawyers.27 Chapter 8 of that document analyzes conflicts of interest in general, including conflicts between a lawyer and a client, among current clients, between a lawyer and a former client, and because of the lawyer’s obligation to a third per- son. This case law complicates matters for lawyers, since they must follow both the rules (explicitly or implicitly) laid down in those cases and the ABA’s ethical rules. At the same time, the case law effectively addresses the malpractice liability of lawyers who fail to resolve a conflict of interest ad- equately and provides civil remedies for this malpractice, including damages paid to harmed clients or third parties and disqualification from continuing to represent a client in a particular matter. The combined effect of the case law described in the restatement of the law governing lawyers and the ethical rules of each jurisdiction is that lawyers who fail to avoid conflicts of interest or to manage them adequately can be sued for malpractice, forced to pay monetary damages, disqualified by a judge from representing a client, or some combination of these. They can also lose their fee or receive only a reduced fee and face disciplinary action by the state bar (including disbarment). Although conflicts of interest are, in the first instance, to be identified by individual practitioners, local bars and the courts can become involved at later stages if those conflicts have not been properly managed. In this way, lawyers rely on self-regulation backed by the threat of professional and legal sanctions. In fact, few lawyers are disciplined and even fewer are disbarred. A 2006 ABA survey found that of over 120,000 complaints filed against law- yers on any issue, only 3.5 percent led to formal discipline and less than 0.5 27  Charles W Wolfram (ed.), Restatement of the Law Third, The Law Governing Lawyers, American Law Institute, 2000.

APPENDIX C 315 percent led to disbarment.28 Although data on the numbers of allegations or findings of improper management of conflict of interest are not available at the federal level, some data are available at the state level. For example, between 2002 and 2006, sanctions for conflict of interest were imposed in only 11 percent of 530 cases, only 1 of which resulted in revocation of the license to practice (other sanctions were suspensions and reprimands).29 Conflict of Interest Management: Key Issues for Lawyers Although the bar’s system of conflict of interest management is well developed, it is not without its critics or its thorny issues. For example, despite the large number of cases, articles, and the detailed codes or rules, the legal profession’s management of conflict of interest is still described as “abstruse,” “arcane,” and “intractable.”30 According to law professor Kevin McMunigal, however, the primary problem in the law’s conflict of interest doctrine is a failure to recognize the regulation of conflict of inter- est as a kind of risk management and not prevention of direct harm. This failure leads the legal profession to use harm rules, which punish lawyers who harm clients, and risk rules, which aim to prevent harm to clients indi- rectly by keeping lawyers out of risky situations or by otherwise managing conflicts of interest but without engaging the all important question: How much risk is too much? McMunigal gives the example of a lawyer in a high-profile criminal case who early on accepts a lucrative book deal to write about the case. When the prosecution offers to settle the case (an option, McMunigal adds, that would clearly be in the client’s best interests), the lawyer advises her client to reject the settlement. McMunigal argues that the risky situation of the book deal (with the temptation being to take the case to trial to ensure publicity and probably help future book sales) probably led the lawyer to give bad legal advice (which would be harmful if it was followed). However, as McMunigal sees it, the legal profession uses the language of conflict of interest to describe and address both the risky situation and the harmful action. If the legal profession could more clearly separate harm rules—for example, rules against lawyers providing bad legal advice or against lawyers entering into unfair business deals with their clients—from risk rules—for example, rules prohibiting a lawyer from preparing an instrument by which 28  American Bar Association, Survey on Lawyer Discipline Systems: 2006. www.abanet. org/cpr/discipline/sold/home.html. 29  Michigan Attorney Discipline Board, Annual Reports 2000–2006, Appendix B (in each annual report). www.adbmich.org/ANNUALRPT.HTM. 30  Kevin C. McMunigal, “Conflict of Interests as Risk Analysis,” in Michael Davis and Andrew Stark (eds.), Conflict of Interest in the Professions, New York: Oxford University Press, 2001.

316 CONFLICT OF INTEREST the lawyer will receive a gift from the client or rules against accepting book deals based on cases—then, McMunigal argues, the legal profession would also avoid confusion about the goals of and justification for each kind of rule.31 McMunigal suggests that conflict of interest rules should be under- stood as being restricted to governing risky situations, leaving rules against breaching confidentiality or against providing incompetent advice to deal with situations in which actual harm has resulted (and in which the causal chain can be proved). In the medical context, conflict of interest rules could therefore focus on identifying and governing situations in which the risk of tainted judgment is considered unacceptably high. The rules invoked to deal with actual harm to patients (other than harm to trust in their physi- cians and the profession in general) would allege not conflict of interest but inappropriate practice, bias, breach of confidentiality, or the like. Another consequence of sharply distinguishing between risk and actual harm would be to make unnecessary the distinction between “actual” and “potential” conflicts of interest, which McMunigal considers to be a distinction of little practical use. In his analysis of the legal profession’s management of conflicts of inter- est, law professor Robert Lawry focuses on a different kind of problem.32 He sees a gradual lessening of professional standards to allow for the greater mobility of lawyers, which is at least partially justified by an appeal to the increased sophistication of some clients. One reason for this reduc- tion in standards is, as mentioned above, the mixing over time of ethics (a guide to good professional behavior) and law (minimum standards by which to police the profession). Another reason is the reality of modern legal practice, where lawyers move from town to town and firm to firm. Yet another reason that exceptions to conflict of interest rules and methods for managing conflicts of interest have developed is to allow medium to large firms to serve many clients, some of whom will, from time to time, have both opposing interests and an interest in relying on lawyers who know them. 31  Consider the debate over financial conflicts of interest in biomedical research: there is some confusion about whether the goal of conflict of interest rules is to identify cases of actual harm (usually bias) or to reduce the risk of harm (Shira Lipton, Elizabeth Boyd, and Lisa Bero, “Conflicts of Interest in Academic Research: Policies, Processes, and Attitudes,” Accountability in Research 2004; 11(2): 83–102). The parties can end talking past each another, with one side asking for proof that harm was caused in this or that case by a conflict of interest before agreeing to the rules and the other side appealing to intuitive ideas about risk or to data showing correlations between conflicts of interest and bad outcomes in aggregates to justify prohibitions or other measures. 32  Robert P. Lawry, “The Law and Ethics of Lawyers’ Conflicts of Interest,” in Thomas H. Murray and Josephine Johnston (eds.), Ethical Issues in Financial Conflicts of Interest in Biomedical Research, Baltimore (MD): Johns Hopkins University Press, forthcoming.

APPENDIX C 317 Screens, Chinese Walls, and Cones of Silence One major mechanism for managing conflicts of interest with con- current clients is through the use of screens (also known as “firewalls,” “Chinese walls,” or “cones of silence”). Screens are mechanisms by which lawyers working on one matter are prohibited from certain kinds of com- munication with lawyers in the same firm working on a conflicting matter. The prohibition is sometimes augmented by placing the lawyers in separate locations (on different floors or in different buildings), controls on e-mail and file access, and the like. Some screens are simply matters of honor; some involve real walls. Screens cannot change a nonconsentable conflict into a consentable one. Instead, screens are used “to encourage clients to consent to a loyalty conflict.”33 Lawry sees screens as further evidence of erosion in lawyers’ conflict of interest standards. Lawyers were introduced to the idea of screens in 1975 by Formal Opinion 342 of the ABA’s Standing Committee on Ethics, which argued that former government lawyers should be permitted to work for a firm doing business with the government if they are screened within the firm from files that they worked on while they were in government. With- out such screens, it was argued, “good lawyers would avoid government work, to the detriment of the common good.” This principle, which was developed for government lawyers and which is discussed only in the Model Rules (and comments) in reference to government lawyers, is now routinely extended to private lawyers. Lawry reports that 22 of 51 jurisdictions al- low screening without the consent of the former client.34 Screens have also received some recognition in the courts and are endorsed in the restatement of the law governing lawyers as a way of dealing with conflicts created by lawyers switching firms. Reasonable people will likely continue to disagree about whether screens are a sign of eroding legal ethics or evidence that le- gal practitioners are both committed to legal ethics and capable of creating effective management systems. Nevertheless, the debate shows that lawyers are engaging with the fundamental question of how to balance the risks of conflict of interest in such situations against the benefits of tolerating the conflict if it is properly managed. ACCOUNTANTS As in other professions, much work in accounting can be conducted by uncertified (or unlicensed) individuals, but some accounting functions 33  Susan R. Martyn, “Visions of the Eternal Law Firm: The Future of Law Firm Screens,” South Carolina Law Review 1994; 45(1): 937–959. 34  Thomas D. Morgan and Ronald D. Rotunda, Selected Standards on Professional Respon- sibility, New York: Foundation Press, 2006.

318 CONFLICT OF INTEREST can be carried out only by a certified accountant. Certified accountants in the United States are: (1) CPAs, who are licensed by their state to provide auditing and attestation services; (2) certified internal auditors, who mostly provide their services directly to their employers; or (3) certified manage- ment accountants and certified business accountants, who, although they deal with the public, cannot audit public companies. Each of these certi- fications is issued by a professional body that maintains a code of ethics and that examines applicants on the basis of that code of ethics as well as on the basis of their technical skills. Individuals may carry more than one certification, but only licensed CPAs can perform the mandatory audits of publicly traded U.S. companies. CPAs are the focus of this discussion because of their prominent public role and their recent struggles to manage conflicts of interest. Like lawyers, CPAs have clients—the companies that hire them to prepare their financial statements—but like architects and engineers and unlike lawyers, CPAs are wary of going too far in acting in their clients’ interests. Lawyers, particularly during litigation, must primarily attend to their client’s interests, leaving it to opposing counsel or the judge to find the flaws in their argument or weaknesses in the client’s case. CPAs, in contrast, are obliged to put the public interest first when they perform an audit or at- testation; they are not allowed to withhold or ignore negative information; indeed, part of their job is to seek out such information.35 The rationale for privileging the public interest is that shareholders and other investors rely on the work of CPAs when they make decisions about whether and how to invest their money. Thus, although a company will engage and pay a CPA (often through an accounting firm) to perform its audits, both the company and the public are the beneficiaries of the CPA’s work. The public benefits from having financial information that it can rely on. The company benefits from the public’s ability to trust the company’s financial reports. Unlike a physician’s patients, the accountant’s clients are frequently sophisticated individuals or businesses; few are physically or mentally com- promised. Although the matters entrusted to the accountant are seldom trivial, they are not literally life and death. 35  That said, they are not generally required to blow the whistle on their clients by reporting fraud to outside agencies (Leonard J. Brooks, “Conflict of Interest in the Accounting Profes- sion,” in Michael Davis and Andrew Stark (eds.), Conflict of Interest in the Professions, New York: Oxford University Press, 2001). They are only required by law to report fraud to the client’s senior management and its audit committee, a subcommittee of the client’s board of directors that itself is under strict reporting requirements (Section 10A(1)(b) of the Securities Exchange Act of 1934 and American Institute of Certified Public Accountants, Statement on Auditing Standards: No. 99, Considerations of Fraud in a Financial Audit Statement).

APPENDIX C 319 Licenses and Professional Membership As in other professions, professional accounting societies were devel- oped to ensure clients that people holding themselves out as accountants met minimal levels of education, competence, and ethical conduct.36 Today, professional accounting societies do not license accountants—that function has been taken over by the state-designated accountancy boards—but the professional societies provide guidance on many issues, from technical ac- counting standards to ethics. Clients, who rely on accountants to provide specialized services and advice, may find some reassurance in the imprima- tur of good standing in a professional society. AICPA is a voluntary association. Although all CPAs must be licensed by their state boards of accountancy (or the equivalent), they are not obliged to be members of their state or national CPA organizations. Mem- bership in good standing of the state or national CPA organization can, however, enhance the reputation of the CPA. (AICPA provides marketing tool kits to its members.) The specific requirements for the CPA license vary somewhat from state to state, but all states require that individuals pass the Uniform Certified Public Accountant examination, which was developed and which is main- tained by AICPA and which is administered by the National Association of State Boards of Accountancy. Questions about professional ethics are included in the Uniform Certified Public Accountant examination and are based on AICPA’s Code of Professional Conduct. Many state licensing boards require continuing professional educa- tion (CPE) that includes ethics or professional conduct. These state boards prescribe CPE course requirements, but the courses are offered to CPAs by approved companies (sometimes called “sponsors”) for a fee. Some atten- tion is paid to the independence of the CPE programs and their sponsors. In New York, for example, CPE can be offered only by sponsors that have been approved by the New York Department of Education, which requires sponsors to have a “direct interest in offering courses on a regular basis” and will not approve “programs devoted to the promotion of particu- lar products or services” or “[i]nsurance, pension, investment, software and other offerings primarily promotional or informational in nature.”37 36  LeonardJ. Brooks, “Conflict of Interest in the Accounting Profession,” in Michael Davis and Andrew Stark (eds.), Conflict of Interest in the Professions, New York: Oxford University Press, 2001. 37  The University of New York, The State Education Department, State Board of Public Accountancy, Instructions for Completing Application for Continuing Education Sponsor Agreement. www.op.nysed.gov/cpa-mcesponsorapplication.pdf.

320 CONFLICT OF INTEREST CPE is offered by a range of sponsors, including universities and private companies.38 Accountancy After Enron In addition to understanding and following AICPA’s Code of Profes- sional Conduct (described in more detail below), CPAs need to be aware of and follow the rules of their state board of accountancy; the ethics stan- dards of their state CPA organization (if they are members); any applicable state laws; and any applicable federal laws, notably, the Public Company Accounting Reform and Investor Protection Act of 2002.39 This act, com- monly known as the Sarbanes-Oxley Act, was passed in response to a number of corporate and accounting scandals in the late 1990s and early 2000s. Before passage of the act, the largest accounting firms had diversified their practices to the extent that audits were a small part of the services that they provided to their clients. Considered key to gaining insight into the client’s business, audit prices steadily declined while other services increased in profitability. As a business school professor at the University of Saskatch- ewan puts it, “Clients became sophisticated purchasers, shopping around for the best deal and putting intense pressure on audit prices, and thus on profits . . . some companies [clients] began to resort to a practice known as ‘opinion shopping.’”40 Accounting firms were soon offering consulting services to audit clients that brought in far more than the audit fee, and therefore, “the auditors did not want to do anything to rock the boat with clients, potentially jeopardizing their chief source of income.”41 This tension between auditing and consulting was identified and cri- tiqued before the Enron scandal, but it was only after the collapse of Enron and WorldCom that practices and codes of conduct changed to address it. Boyd calls Enron “the ‘smoking gun’ evidence, indicating that the profes- sion had reached a stage where commercial interests simply overwhelmed allegiance to professional integrity.” Policy makers were not content to leave it to accounting firms or AICPA to address the issue. They chose to pass legislation—the Sarbanes-Oxley Act—to restore the reliability of the public company audit process. Subject to certain preapprovals, Section 201 of the Sarbanes-Oxley Act prohibits auditors from providing a number of other services contem- poraneous to the audit, including bookkeeping, management functions, 38  See the registry of the National Registry of CPE Sponsors. http://registry.nasbatools. com/display_page. 39  Sarbanes-Oxley Act of 2002. http://thomas.loc.gov/cgi-bin/query/z?c107:H.R.3763.ENR. 40  Colin Boyd, “The Structural Origins of Conflicts of Interest in the Accounting Profes- sion,” Business Ethics Quarterly 2004; 14(3): 377–398. 41  Arthur Levitt, Take on the Street, New York: Pantheon Books, 2002.

APPENDIX C 321 investment advice, investment banking services, and legal services. The act also creates the Public Company Accounting Oversight Board (PCAOB) to oversee CPA practice in relation to public companies. Under Section 103 of the act, PCAOB has established standards and rules on a variety of issues, including ethics, that apply to registered public accounting firms preparing and issuing audit reports as required by the act or the rules of the Securities and Exchange Commission. PCAOB’s conflict of interest rules are designed to preserve the inde- pendence of the accounting firm.42 Rule 3520 states: “a registered public accounting firm and its associated persons must be independent of the firm’s audit client throughout the audit and professional engagement period.” Rules 3521, 3522, and 3523 describe situations in which an accounting firm cannot be considered independent, for example, if the firm provides a service or product to the audit client for a contingent fee or a commission, if the firm provides assistance in planning or tax advice on certain types of potentially abusive tax transactions to an audit client, or if the firm provides any tax services to certain persons in a financial reporting oversight role at an audit client or to immediate family members of such persons. Violation of the PCAOB rules can lead to an investigation by PCAOB. Following a hearing, sanctions can be imposed, including (1) revoking a firm’s registration; (2) barring an individual from participating in audits of public companies; (3) monetary penalties; and (4) remedial measures, such as training, the implementation of new quality control procedures, or the appointment of an independent monitor.43 PCAOB’s website reports that 3 of the 17 disciplinary proceedings before PCAOB over the past 3 years have found “independence” violations. There has been some criticism that the post-Enron measures are too burdensome,44 with others countering that the measures fail to do enough to end auditor-client “coziness.”45 Either way, this legislation and accom- panying rules and sanctions serve as a cautionary tale. They installed the external regulation of a profession that apparently had not sufficiently regu- lated its own conflicts of interest. Since the Sarbanes-Oxley Act was passed, AICPA and state CPA societies have strengthened their codes of conduct, an example of a change in law forcing a tightening of ethical standards. 42  Public Company Accounting Oversight Board, Bylaws and Rules—Rules—Professional Standards, Section 3. www.pcaobus.org/Rules/Rules_of_the_Board/Section_3.pdf. 43  Section 105(b)(4) of the Sarbanes-Oxley Act of 2002. 44  Jonathan D. Glater, “Here It Comes: The Sarbanes-Oxley Backlash,” New York Times, April 17, 2005. 45  Richard L. Kaplan, “The Mother of All Conflicts: Auditors and Their Clients,” Iowa Journal of Corporate Law 2004; 29: 363–383.

322 CONFLICT OF INTEREST AICPA Code of Professional Conduct The current version of the AICPA Code of Professional Conduct46 emphasizes independence and objectivity. The AICPA Code of Profes- sional Conduct is divided into principles and rules. The six principles are expressed in a sentence or two. Each principle is clarified by up to five sub- paragraphs. The dozen rules that follow the principles are also expressed in one or two sentences but are followed by more detailed guidance in the form of Interpretations of Rules of Conduct and Ethics Rulings (rather like the American Medical Association’s [AMA’s] Opinions). AICPA’s professional ethics committee adopted Interpretations of Rules of Conduct “after exposure to state societies, state boards, practice units and other interested parties.” The Interpretations of Rules of Conduct are intended “to provide guidelines as to the scope and application of the Rules but are not intended to limit such scope or application.” Ethics rulings are formal rulings made by AICPA’s professional ethics committee applying the rules and their interpretations to a particular set of facts. AICPA members who depart from ethics rulings in similar circumstances will be “requested to justify such departures.”47 Two of the Code’s principles bear on conflict of interest. The first, titled Integrity, states that “to maintain and broaden public confidence, members should perform all professional responsibilities with the highest sense of integrity” and includes a reference to observing the principles of objectivity and independence in its final subparagraph. The other relevant principle, titled the Public Interest, requires members “to act in a way that will serve the public interest, honor the public trust, and demonstrate commitment to professionalism.” Acknowledging that members may encounter conflicting pressures, this principle advises that, when resolving such conflicts, mem- bers recall that when they “fulfill their responsibility to the public, clients’ and employers’ interests are best served.”48 Although these two principles clearly bear on conflict of interest, it is the fourth principle (titled Objectivity and Independence) that addresses conflict of interest most directly. It states: “A member should maintain objectivity and be free of conflicts of interest in discharging professional re- 46  The American Institute of Certified Public Accountants, AICPA Code of Professional Responsibility. www.aicpa.org/About/code/index.html. 47  The American Institute of Certified Public Accountants, AICPA Code of Professional Responsibility: Introduction, Other Guidance. www.aicpa.org/About/code/othguid.htm. 48  The American Institute of Certified Public Accountants, AICPA Code of Professional Responsibility: Section 53, Article II, The Public Interest, and Section 54, Article III, Integrity. www.aicpa.org/About/code/sec50.htm.

APPENDIX C 323 sponsibilities. A member in public practice should be independent in fact and appearance when providing auditing and other attestation services.”49 The general rule, therefore, is that all CPAs should always be free of conflicts of interest. An even higher standard is set for CPAs in public practice. That is, when a CPA is performing certain services in public prac- tice, the CPA should maintain objectivity in appearance as well as in fact. Public practice is defined as the performance for a client of accounting, tax, personal financial planning, litigation support, and other professional ser- vices for which practice standards are promulgated by AICPA. (The closest medical equivalent to a CPA’s public practice is a physician’s publication of research or the testimonials or other public statements that an individuals makes as a physician.) Four subparagraphs follow this principle. The first discusses, in a fairly philosophical way, objectivity (describing it as “a state of mind” and impos- ing the obligation to be “impartial, intellectually honest, and free of con- flicts of interest”) and then independence (“precludes relationships that may appear to impair a member’s objectivity in rendering attestation services”). After noting the variety of roles that an accountant might play in society, including teaching, the second subclause states: “Regardless of service or capacity, members should protect the integrity of their work, maintain objectivity, and avoid any subordination of their judgment.” The principle is thus fairly strict: CPAs should at no time enter into relationships that might even appear to impair their objectivity. The third subclause focuses on accountants working in public practice, stating that to protect their independence (the appearance of objectivity), they should be continually assessing “client relationships and public responsibility” and “should be independent in fact and appearance.” The effect of this subparagraph is to require constant vigilance of the possible impact of interactions with clients on the CPA’s (actual and apparent) commitment to the public interest. The fourth subclause applies to members not in public practice (e.g., members employed by a company rather than acting as an external accountant or auditor). That subclass concedes that CPAs not in public practice “cannot maintain the appearance of independence” but nevertheless imposes on them “the responsibility to maintain objectivity in rendering professional services.” More detail in the form of rules, interpretations, and ethics rulings fol- low at Sections 101 (Independence) and 102 (Integrity and Objectivity).50 49  The American Institute of Certified Public Accountants, AICPA Code of Professional Responsibility: Section 55, Article IV, Objectivity and Independence. www.aicpa.org/About/ code/et_55.html. 50  The American Institute of Certified Public Accountants, AICPA Code of Professional Responsibility: Section 100, Independence, Integrity, and Objectivity. www.aicpa.org/About/ code/sec100.htm.

324 CONFLICT OF INTEREST Section 101 begins with the rule that a member in public practice shall be independent according to the standards set by state boards, state CPA insti- tutes, the U.S. Securities and Exchange Commission, and PCAOB, among others. The interpretation for this rule then describes the circumstances under which independence will be considered impaired. Situations of im- pairment include the following: when the accountant, during the period of professional engagement, holds or commits to acquiring a financial interest in the client; when the accountant has a loan to or from the client or in some circumstances when the accounting firm, one of its partners, or a part- ner’s immediate family members hold an ownership stake in the client; and when the accountant’s firm or a partner or employee of the firm was simul- taneously a director, officer, or employee of the client. (The last prohibition applies not simply during the period of the professional engagement but during the whole period covered by the financial statements being prepared by the accountant.) When in doubt about whether a particular circumstance might cause independence to be questioned, the section asks that members “evaluate whether that circumstance would lead a reasonable person aware of all the relevant facts to conclude that there is an unacceptable threat to the member’s and the firm’s independence” (emphasis added). Fourteen additional interpretations follow, and that number is far too many to cover in any detail here. It will be enough to note here that the net cast is fairly wide. Section 101 captures a number of situations involving the accountant; the firm; colleagues; family members; close relatives; or current or previous financial, employment, ownership, or management relation- ships with the client. The interpretations also differentiate between direct financial interests (such as ownership and investment interests) and indirect financial interests, including some holdings through mutual funds. In one of the Ethics Rules accompanying Section 101, the question of the CPA’s acceptance of gifts or entertainment from a client is posed. The answer is that the acceptance of gifts or entertainment from a client that the CPA is auditing will be considered to impair objectivity, unless the value is “clearly insignificant to the recipient.” The ethics ruling is less restrictive when the client is not an “attest client” (i.e., one for whom the CPA performs audit- ing or other attestation services), although even in such cases the CPA is required to assess whether accepting the gift is reasonable, given the nature, value, timing, and frequency of the gift. The overall aim of the independence rule and its interpretations and ethics rulings at Section 101 is to ensure that audits of companies will be carried out by accountants and accountancy firms who provide no other financial services to the company, are not investors in or directors of the company, do not and have not recently worked for the company, have no other financial or other ties to the company, and are otherwise free of any appearance suggesting to a reasonable person a loss of objectivity. Despite

APPENDIX C 325 its detail, this section has been criticized for failing to address one particu- lar threat to independence (and objectivity): the significance of the client company’s audit fees to the bottom line of the accountant or the firm.51 One observer points out that Arthur Andersen’s Houston, Texas, office re- ceived $1 million per week from Enron while it was auditing the company: “[The] livelihoods of several audit partners and several hundred audit firm employees depend[ed] on keeping a client happy.” Section 102 (Integrity and Objectivity) is far briefer than Section 101 and applies to all CPAs in all of their work (i.e., not simply to CPAs per- forming audits). It begins with the rule that “in the performance of any professional service a member shall maintain objectivity and integrity, shall be free of conflicts of interest, and shall not knowingly misrepresent facts or subordinate his or her judgment to others.” Although the rule is concerned with the misrepresentation of fact as well as conflict of interest, the inter- pretations for the rule offer a definition of conflict of interest: “a conflict of interest may occur if a member performs a professional service for a cli- ent or employer and the member or his or her firm has a relationship with another person, entity, product, or service that could, in the member’s pro- fessional judgment, be viewed by the client, employer, or other appropriate parties as impairing the member’s objectivity.” Note that the relationship need not be financial or familial, and it need not actually impair the CPA’s objectivity—it is enough that it could, in the member’s judgment, appear to impair his or her objectivity. In contrast to Section 101, which provides no way to manage the impairments to independence that it covers, Section 102 suggests that dis- closure and consent may be acceptable ways to manage conflicts of interest “if the member believes that the professional service can be performed with objectivity, and the relationship is disclosed to and consent is obtained from such client, employer, or other appropriate parties” (emphasis added). The interpretation explicitly states that concern about independence in certain professional engagements, such as audits and reviews, as expressed in the previous rule (Section 101), cannot be addressed by disclosure and consent. The discussion ends with an explicitly nonexhaustive list of situations that ought to raise independence concerns, including when the CPA has a sig- nificant financial interest in or is on the management of a company that is a major competitor of the client for which he or she performs management consulting services, when the CPA is asked to perform litigation services for a case filed against one of his or her clients, or when a CPA provides services for several members of the same family with opposing interests. 51  David Cotton, “Fixing CPA Ethics Can Be an Inside Job,” Washington Post, October 20, 2002. P. B2.

326 CONFLICT OF INTEREST These are all situations in which the usual public reliance on the auditor’s work is absent. Enforcement of Ethics Rules The AICPA Code of Professional Conduct focuses on the impact of existing and preexisting relationships (whether of the accountant, firm, colleagues, and sometimes, members of the family) on the accountant’s objectivity and independence. Objectivity and independence are particularly protected when the accountant is auditing a client for the public’s benefit. When offering any service except external auditing, a CPA might be able to disclose a conflict of interest to the client and proceed with the client’s consent, but this option is simply not available in a public audit (or other attestation) situation. The reason for the focus on independence from cli- ents (a matter of appearance), from the CPA’s personal interests, and from the interests of the CPA’s firm, colleagues, and family during audits (and other attestation services) in particular is that a CPA performing an external audit has a primary obligation to the public. Indeed, the second principle in the Code of Professional Conduct, titled The Public Interest, states that the public relies “on the objectivity and integrity of certified public accountants to maintain the orderly functioning of commerce.” The rules and practices of CPAs, therefore, are fairly strict when it comes to preserving independence and avoiding even the appearance of conflict of interest. It is important to note, however, that like architects and engineers but unlike lawyers, accountants are able to work for compet- ing clients, often but not always with the knowledge and consent of both parties.52 The rationale for this difference is that accountants are able to perform an audit of one client without disclosing (or even relying on) infor- mation about the other. It would be much harder for a lawyer to adequately represent two competing clients without using the confidential information of one against or in favor of the other. The preamble to the AICPA Code of Professional Conduct neatly describes how it is enforced: “Compliance with the Code of Professional Conduct . . . depends primarily on members’ understanding and voluntary actions, secondarily on reinforcement by peers and public opinion, and ultimately on disciplinary proceedings, when necessary, against members who fail to comply with the Rules.” Members of AICPA are on notice that they must be prepared to justify any departure that they make from the Code of Professional Conduct. 52  Leonard J. Brooks, “Conflict of Interest in the Accounting Profession,” in Michael Davis and Andrew Stark (eds.), Conflict of Interest in the Professions, New York: Oxford University Press, 2001.

APPENDIX C 327 AIPCA’s Professional Ethics Executive Committee interprets and enforces the AICPA Code of Professional Conduct. The committee investigates al- legations of unethical conduct by both its members and the members of almost all state CPA organizations through its Joint Ethics Enforcement Program (JEEP).53 JEEP has existed since the 1970s and was created in rec- ognition of the fact that the codes of many state CPA societies are identical or similar to the provisions of the AICPA Code of Professional Conduct and that it is common for a CPA to be a member of both AICPA and one or more state societies. JEEP therefore provides an efficient mechanism for enforcing ethics rules consistently across the United States. Violations of the Code of Professional Conduct can result in a CPA being expelled, suspended for a period of 1 or 2 years from AICPA or from the local CPA society, directed to complete specified CPE courses, or directed to take other action (e.g., submit subsequent work papers for continued monitoring). All decisions to expel or suspend a CPA are made public through publication on AICPA’s website. Very few of these decisions (just 4 of over 150) during the past 3 years included a finding of a breach of the rules regarding independence. AICPA can also publicly admonish a member who has violated the Code of Professional Conduct. AICPA’s ethics committee can also conclude, upon investigation, that there is no evidence of a violation of the Code of Professional Conduct and therefore dismiss the case or simply close a case for lack of evidence or some other reason.54 The enforcement mechanisms of suspension, expulsion, and public ad- monishment seem designed to place the public and colleagues on notice that the CPA does not comply with the Code of Professional Conduct and to publicly embarrass the CPA, whereas requiring completion of CPE courses or submitting reports and work papers seems to aim to reeducate the CPA. Because a CPA does not need to be a member of AICPA or a state CPA or- ganization, a finding of violation of AICPA’s Code of Professional Conduct is not itself sufficient to withdraw the CPA’s license to practice. Only the state licensing boards can suspend or revoke a CPA’s license. One important lesson from the recent history of public accountancy is that a failure to address conflicts of interest led to federal regulation of conflict of interest in one important aspect of CPA practice: the auditing of publicly listed companies. Investor confidence in the objectivity and inde- pendence of auditors and therefore in the truthfulness of public companies’ 53  The American Institute of Certified Public Accountants, Professional Ethics Executive Committee, Fact Sheet 2004–2005. www.aicpa.org/download/ethics/ethics-committee-fact- sheet.pdf. 54  The American Institute of Certified Public Accountants website, Ethics Enforcement. www.aicpa.org/About/code/sec100.htm.www.aicpa.org/Professional+Resources/Professiona+E thics+Code+of+Professional+Conduct/Professional+Ethics/Ethics+Enforcement/defin_sanction. htm.

328 CONFLICT OF INTEREST financial statements was considered an important enough goal—given the huge financial stakes involved—to warrant federal legislation and the es- tablishment of a federal oversight body, the PCAOB. A second important lesson is that maintaining objective and independent judgment is not easy. Accountants primarily maintain their objectivity by avoiding most situa- tions that present a conflict of interest. They also maintain their indepen- dence by avoiding most situations that could reasonably appear to present a conflict of interest. ARCHITECTURE During the decade and a half before World War I, AMA organized med- icine as a modern profession. Among the milestones in that process were not only the rethinking of medical education (set forth in the 1910 Flexner Report) but also the abandonment in 1903 of AMA’s mandatory Code of Ethics of 1847 for the “suggestive and advisory” Principles of Medical Ethics. That was followed in 1912 by the abandonment of the 1903 prin- ciples for another code (with the same name) binding on all physicians (and surgeons). At about that time (1909), the American Institute of Architects (AIA) adopted its first code of ethics. That code applied only to members of the organization and not to all architects, but the code (like AMA’s 1847 code) was binding on all architects. The AIA kept this feature when it adopted a new code in 1977, one organized—like the ABA’s 1969 code (which was to be abandoned soon after)—into canons (broad statements of principle), ethical standards (more specific goals that AIA members should aspire to), rules (mandatory standards, the violation of which would jus- tify formal discipline, including expulsion from the AIA), and commentary (when necessary to avoid a common misinterpretation of a rule).55 The 1909 AIA code reached all architects, not just AIA members, through its adoption by state licensing boards as local standards of practice. That simple arrangement ended in the 1970s, when the courts declared the AIA’s code to be an unreasonable restraint on trade. While the AIA was rewriting its code to avoid another lawsuit (a process that did not end until 1990), the National Council of Architecture Registration Boards (NCARB) wrote its own code.56 Because the states’ licensure of architecture is generally similar to the states’ licensure of accountants (and, in some states, lawyers), continuing education requirements are also similar. Courses must be accredited to sat- isfy the continuing education requirements. The AIA itself offers some on- line courses that satisfy continuing education credit. Some of these courses 55  See www.aia.org/SiteObjects/files/codeofethics.pdf. 56  See www.architects.org/emplibrary/NCARB.pdf.

APPENDIX C 329 are now prepared by suppliers.57 So far, supplier-prepared courses do not seem to be a problem. One reason that they are not may be that architects’ specifications (the equivalent of a physician’s prescription) typically either state a generic requirement or take the form “brand x or its equivalent.” Another reason that supplier courses are not a problem may be that they do not include a trip to some ideal location, lavish entertainment, or other gifts. The course itself must be valuable enough to repay architects for their time and for lost opportunities to take other courses. NCARB on Conflict of Interest Architects resemble physicians, lawyers, and accountants in not being able to practice (that is, advertise, sign drawings, or otherwise publicly present themselves as architects) without registering as one (that is, being given a state license to practice). Beginning in 1919, registration boards have maintained a nonprofit group to provide a number of services to the profession: a standardized test for admission into the profession, standards for work experiences that a new graduate of an accredited architectural program should have before licensure (Intern Development Program), self- administered continuing education courses, and so on. NCARB’s code of ethics (Rules of Conduct) is just one of these services. Adopted in 1977 (and amended since), the Rules of Conduct are designed to provide hard-edged rules for discipline (once a state board adopts them). Besides the nominal “Rules”—five titles numbered with Arabic numerals—the code includes (1) actual rules under each rule (numbered with a decimal), (2) a brief commen- tary after most of these rules, and (3) a long introduction (40 percent of the entire 10-page code). Although the NCARB code does set a somewhat lower standard than the (shorter) AIA code, it generally does so by silence rather than by providing a formal rule significantly different from the correspond- ing AIA rule. The AIA issues ethics opinions much as AMA does; NCARB does not. A state registration board may, however, issue an opinion as part of disciplinary action against a particular architect. Conflict of interest is plainly important in the practice of architecture. The second of the five major divisions in NCARB’s Rules of Conduct is titled Conflict of Interest; the third major division, although it is titled Full Disclosure, is in part (Rule 3.1) about responding to conflict of interest. The other divisions of the code—Competence (Division 1), Compliance with Law (Division 4), and Professional Conduct (Division 5)—have no connection with conflict of interest. The overall strategy in these provisions is clear. Conflict of interest 57 See, for example, www.gp.com/build/paperless/education.html (a course offered by Georgia-Pacific).

330 CONFLICT OF INTEREST should generally be avoided, but when avoidance is not possible or at least not reasonable, the conflict must be fully disclosed to all appropriate parties and their consent must be won before the architect can proceed. Interest- ingly, the term “conflict of interest” is not used in any of the specific rules; its definition is, in effect, the rules under that title. All of the relevant rules (including Rule 3.1) are (primarily) concerned with financial interests. There are four rules under Rule 2, Conflict of Interest. Rule 2.1 applies to ordinary compensation for services. An architect “shall not accept compensation for services from more than one party on a project unless the circumstances are fully disclosed to and agreed to . . . by all interested parties.” Both disclosure and agreement are to be “in writ- ing.” The commentary explains that architects may sometimes find it hard to avoid receiving payment from two parties—for example, when ordering a large number of windows from a supplier later produces a rebate check. The architect cannot simply accept the rebate (even if it comes as a surprise) but must first inform the client (and other interested parties) of the payment and the reason for it. The architect cannot accept the payment unless at least the client (and any other interested party) approves. The commentary explains that the “bifurcated loyalty” that such a rebate threatens is “unacceptable unless all parties have understood it and accepted it.” The commentary does not limit the “parties” to the client. This is because in many architectural projects several parties may be affected by the payment, such as the engineer- ing firm typically present at any large project, the developer (who may be the immediate client but who is, in fact, a stand-in for the ultimate owner), the ultimate owner (who may be one or more individuals or a legal entity), and even the contractor or subcontractor who must work with the rebated supplies. The commentary can even be interpreted as including the window supplier’s competitors among those who must be informed of the payment and the reason for it. They are certainly “interested parties.” They are at a competitive disadvantage if they are not also making such rebates. Behind Rule 2.1 is a conception of architects as having a relatively settled loyalty to the client that everyone dealing with the client relies on. An un- usual payment (such as the rebate described above) unsettles the situation. There is no question here of the supplier buying the architect’s loyalty with the rebate (as there would be if the payment were a bribe or kickback). The problem is that “money talks,” and even architects cannot gauge how much they will listen the next time that they place an order of that sort. Their judg- ment that their professional judgment will not be affected is not relevant. That, too, is now under suspicion. Disclosure of the payment makes it possible for all interested parties to redefine their relationship to the architect to take account of this unusual feature. The client may, for example, require the architect to hand over the entire rebate (as well as ask other suppliers whether they will meet the com-

APPENDIX C 331 petition). However, because the architect’s fee is often a percentage of the total cost of the project, this solution may not be the best. It would create a “perverse incentive.” The architect would, in effect, be punished for saving the client money. The architect would have an incentive to avoid suppliers who give rebates. The client might then prefer to split the rebate with the architect, or they might work out some more complicated arrangement—of which all interested parties should be made aware to ensure that their trust in the architect’s judgment is not misplaced. Rule 2.2 concerns financial interests apart from payments, for example, stock in a potential supplier or a loan to a contractor. The architect must assess whether the interest (direct or indirect) is “substantial enough to influence his or her judgment in the performance of professional services” (whether or not it does or would in fact influence it). Architects thus have some discretion under this rule (as they do not under Rule 2.1). The rationale for allowing some discretion (concerning whether an interest is substantial enough) is that avoiding all financial interests seems too much to ask. For example, an architect with money in a large investment fund that holds a few shares of stock in one of the companies she or he is dealing with has an interest in that company. Is revealing such an interest worth the trouble? Should architects be required to avoid investing in any fund that might (on a given day) invest in a potential supplier? That seems too much to require, so long as the architect reveals any interest substantial enough to affect her or his judgment. Of course, when in doubt, the architect should reveal the interest. Rule 2.2 seems to work because it governs only interests other than payments, because architects seldom invest in suppliers and because most architects work in a small world (mostly developers or builders rather than individual clients) in which a substantial investment in a supplier would soon be known. If the interest is enough to influence the judgment, the architect must fully disclose it in writing to the client or employer (thus creating a paper trail). If the client or employer objects to the business association or the fi- nancial interest, the architect must either terminate it or offer to give up the commission or employment. The client or employer may have good reason to accept the bifurcated loyalty that the business association or financial interest in question creates, but the decision is the client’s or the employer’s (or both, when an architect has both a client, the person who has hired the firm, and an employer, the architectural firm). That decision should be made only with all the relevant facts laid before the decision maker in a form that the decision maker can understand. If the architect is unwilling to make full disclosure, she or he must resign from the job. There is no middle way (no way to manage the conflict) without full disclosure and consent. Rules 2.3 applies to any payment made in return for specifying or en- dorsing a supplier. Strictly speaking, this rule does not concern conflict of

332 CONFLICT OF INTEREST interest but concerns bribes, kickbacks, and other side payments that buy the architect’s judgment. Architects are simply forbidden to solicit or accept such payments. The brief commentary notes that this rule is “absolute”; that is, it admits of no exception, even when all the relevant parties would agree to the payment after full disclosure. So, for example, an architect cannot have an agreement with a supplier that she or he will recommend a certain window frame even if she or he fully informs the clients of that agreement and the clients say, “Fine.” Why? Although many of the payments in ques- tion are in fact illegal, the rule is indifferent to their legality. Even legal pay- ments for specifying or endorsing a supplier (say, lending one’s name to an advertising campaign) are forbidden. What explains this striking departure from architecture’s standard strategy of allowing conflict of interest when the relevant parties consent after full disclosure? The answer seems to be this: conflict of interest threatens professional judgment. It makes it less reliable than it would otherwise be. Sometimes such threats cannot be avoided or cannot be avoided at reasonable cost. Those relying on the architect’s judgment then have the right to weigh the costs and benefits and decide whether to take the risk. In contrast, an agreement to specify or endorse a product does not threaten professional judgment. It does something much more dramatic. The architect has, in this respect, signed away judgment. By the agreement, the architect gives up future judgment of the appropriateness of the product in question. The agreement with the sup- plier prejudges the matter. The architect cannot both claim the power of an architect in that respect (the right to use her or his judgment to decide what is appropriate in that case) and follow an agreement prejudging the case. Side payments for endorsement are also, in one respect, unnecessary. The client or employer derives no benefit whatsoever from them, and (generally) the architect does not need them to survive or prosper. They are simply not an essential part of practicing architecture. This explanation of Rule 2.3 treats it as something other than a rule concerned with conflict of interest. Selling one’s judgment does not, in gen- eral, create a conflict of interest (that is, it does not threaten professional judgment). However, sometimes it does. For example, if Person A is paid to endorse a product as part of an advertising campaign, Person A will have a greater tendency to specify that product than he or she otherwise would. That tendency is what makes Rule 2.3 in part a rule concerned with conflict of interest. Forbidding endorsements for pay eliminates one sort of conflict of interest. Rule 2.4 concerns the architect acting as adjudicator, that is, as the interpreter of building contract documents or the judge of contract per- formance. When acting in this role, an architect is to “render decisions impartially, favoring neither party in the dispute.” The commentary makes clear that it is customary in the construction industry for the architect, even

APPENDIX C 333 though he or she is paid by the owner and owes loyalty to the owner, to settle disputes between the owner and a contractor, subcontractor, or sup- plier concerning whether work has been performed as the contract requires or whether the contract requires this or that. When acting in this capacity, the architect must (according to NCARB) act impartially. If the architect does not believe himself or herself to be capable of acting in that way, he or she “may appropriately decline to act in those two roles” (as the agent of the owner and as a judge between the owner and an adversary). The architect’s role in such circumstances has a threat to independent judgment built into it (an interest but not a “special” interest). Both architects and those they work with are aware of that threat to independent judgment. They have traditionally tolerated it since the alternative is whatever delay is necessarily consequent on seeking a truly impartial judge far from the work site. Nonetheless, the architect must at least believe himself or herself to be able to render impartial judgment. If the threat to impartiality is significant enough that the architect doubts his or her own judgment, the architect may (and, indeed, should) decline. Interestingly, the rule is not satisfied if the ar- chitect merely believes himself or herself to be impartial; the architect must actually render an impartial decision. If the decision is obviously biased, the architect would be subject to discipline under the rule, even though the architect believed himself or herself to be impartial. Like Rule 2.3, Rule 2.4 is an absolute rule (although the commentary does not say that explicitly). The rationale for its absoluteness is also much the same as that for Rule 2.3. The point of asking the architect to judge between the owner and those working on a site is to receive quickly (something approaching) impartial judgment (a judgment informed by the architect’s knowledge of construction, the documents, and local custom). If the architect were known to be partial, his or her value as a judge would be much reduced. The rule preserves the usefulness of architects in settling such disputes (an efficiency serving everyone’s interests in the long run). Like Rule 2.3, Rule 2.4 is (primarily) concerned not with conflict of inter- est, strictly speaking, but with something closely related, that is, the typi- cal outcome of judgment free of conflict of interest (as well as of bias and prejudice): an impartial decision. The last of NCARB’s conflict of interest rules is Rule 3.1. It requires an architect making a “public statement on architectural questions” (that is, speaking publicly in a professional capacity) to “disclose when he or she is being compensated for making such statement or when he or she has an economic interest in the issue.” So, for example, an architect paid by a developer to testify on behalf of a project would have to state that she or he is being so paid. An architect writing a journal article on behalf of a certain manufacturer’s product would have to disclose ownership of even a single share of stock in that company. For public statements, the standard

334 CONFLICT OF INTEREST of disclosure is more demanding than for statements to client, employer, or other private person. (The term “substantial enough” in Rule 2.2 has no counterpart in Rule 3.1.) The commentary explains why the standard is so demanding: to preserve “the probity which the public expects of the archi- tectural profession,” architects are “not allowed under the circumstances described in the rule to disguise the fact that they are not speaking on the particular issue as an independent professional but as a professional en- gaged to act on behalf of a client” or with a judgment perhaps arising from the wrong sort of interest (a private interest rather than the public interest). The public is entitled to know that the architect might have a certain bias (or even that, from the public’s perspective, might seem to have a certain bias), a legitimate bias if it is disclosed but otherwise an illegitimate bias. If architects routinely made public statements in the service of clients without acknowledging that service or in the service of a private interest (however small) without acknowledging that service, their public statements would eventually lose the power that comes from their being thought to be inde- pendent. The public statements would be regarded as unreliable (as, indeed, they would be). This rationale is as interesting for what it leaves out as for what it includes. Like most professions, architecture recognizes itself as having an obligation to serve the public interest (an obligation that may not belong in hard-edged rules but appears, for example, in Canon II of the AIA code). The NCARB commentary might therefore have appealed to this obligation in support of a rule governing public statements (protecting the public). Instead, the commentary appeals to the interest that the profession itself has in maintaining its reliability (“the probity the public expects”) both to explain and support the rule. AIA on Conflict of Interest The NCARB rules just discussed are the hard-edged rules concerning conflict of interest that state registration boards use to decide whether to discipline a licensed architect. We turn now to the AIA’s code. Like AMA, the AIA is a voluntary organization. Also like AMA, it no longer is an orga- nization to which a majority of the profession belongs. Yet, just as no AMA member wants AMA to discipline her or him, so no AIA member wants the AIA to discipline her or him. An AIA member charged with wrongdoing will generally hire a lawyer to present her or his side at the National Ethics Council (NEC) and, if the AIA member loses there, may seek redress in the courts. At least as much as physicians, architects live by their reputations. For that reason, an NEC-appointed hearing officer collects evidence and the full NEC (minus the hearing officer) decides the case in secret. However, for any serious discipline (censure, suspension, or expulsion), the ultimate deci-

APPENDIX C 335 sion is made public (the architect’s name disappears from the membership role, and the architect can no longer be listed as an AIA member). The NEC publishes its decision in the form of a judicial opinion, stating the facts found, the penalty, and the rationale for it, without identifying the parties. The NEC also issues interpretations of the code (Advisory Opinions).58 The AIA code (2007) is about half the length of NCARB’s and devotes proportionally much less space to conflict of interest. Canon III (Obliga- tions to the Client) provides the overall framework for conflict of interest. AIA members should “exercise unprejudiced and unbiased judgment when performing all professional services.” Ethical Standard 3.2 (titled Conflict of Interest) simply states the general strategy for avoiding tendencies to bias and prejudice. Members should “avoid conflicts of interest in their professional practices and fully disclose all unavoidable conflicts as they arise.” There are only two disciplinary rules under this ethical standard. The second rule, Rule 3.202 (render decisions impartially), merely restates NCARB’s Rule 2.2 (with a briefer commentary), but the first rule, Rule 3.201, adds something new. Rule 3.201 prohibits AIA members from rendering professional services if their “professional judgment could be affected by responsibilities to an- other project or person, or by [their own] interests.” The only exception to this prohibition is (the usual) “unless all those who rely on the Member’s judgment consent after full disclosure.” Rule 3.201 understands “interest” as including more than financial interest. Any “responsibility” to another project or person that could affect a member’s judgment is an interest for the purposes of this rule (as is any self-interest, even if it is not financial or familial). The commentary underscores the point. The rule is, it says, “in- tended to embrace the full range of situations that may present a Member with a conflict between his interests or responsibilities and the interests of others.” The commentary goes on to give an equally wide reading of “all those who rely.” Those entitled to disclosure “may include a client, owner, employee, contractor, or others who rely on or are affected by the Member’s professional judgment.” An AIA member who cannot appropri- ately disclose a “conflict directly to the affected person must take steps to ensure that disclosure is made by another means.” (Direct disclosure may not be possible because the client is, for example, an individual who is out of town or an organization whose officers are hard to reach; sending notice is not equivalent to “appropriate disclosure.”) If a member cannot make adequate disclosure of a conflict of interest (directly or indirectly), he or she cannot render the professional services in question. The member must decline or withdraw. In addition to the rules under Rule 3.2, there are at least two rules in 58  For either, see www.aia.org/about_ethics#nec.

336 CONFLICT OF INTEREST Canon II (Obligations to the Public) that are (at least in part) concerned with conflict of interest. Rule 2.103 forbids an AIA member from serving in a public capacity to “accept payments or gifts which are intended to influ- ence their judgment.” This rule covers bribes (payments made in exchange for some future illegal service) but not kickbacks (a payment for a referral or some other favor already done). The rules cover more than bribes, for example, a dinner or a gift of theater tickets of whatever value given with the intention of influencing judgment. This additional coverage is what justifies discussion of this rule as concerned with conflict of interest. This is another absolute rule. It is nonetheless unusual in one respect. It is the intention of the payer or giver, not its likely consequence or the recipient’s intention, that determines whether the payment or gift is pro- hibited. The rationale for this approach to payments and gifts is obvious. An architect serving in a public capacity will, in the ordinary course of life, receive many payments and gifts. Prohibiting them all would be unreason- able, but some should be prohibited. For example, no one wants to forbid a gift from the architect’s mother or brother-in-law that is part of the nor- mal exchange of gifts among family members. (Such gifts would seldom be given with the intent of influencing the architect’s professional judgment.) In contrast, the AIA would, presumably, want to prohibit a gift from a potential developer hoping to reduce the hostility of an architect toward a project that he or she has in mind when that architect is a member of the local planning commission. Rule 2.301 is concerned with public statements on architectural is- sues. It is (almost) identical to NCARB’s Rule 3.1. Although there is no commentary, its placement under the canon concerned with obligations to the public suggests that its rationale is a bit different. Architects per- form a useful service whenever they inform the public of their judgments on architectural issues. They perform a useful service whether they speak disinterestedly or on behalf of a client or interest. However, the service performed is different. Rule 2.301 requires architects to make clear which service they are performing so that their audience, the public, can evaluate it using the appropriate criteria. The underlying rationale is not so much to protect independent judgment as it is not to mislead the public concerning what it may reasonably expect of the judgments offered. The public’s trust in what architects say depends in part on knowing who they are working for when they say it. ENGINEERING Engineering and medicine have historically been very different profes- sions. Engineers have, for example, generally worked in large organizations, beginning with the army; physicians (like architects) have, until recently,

APPENDIX C 337 generally worked alone or in small practices (with or without an affiliation with a nearby hospital). Those who employ engineers tend to be the rich and the powerful, not the sick or the wounded. When a work of engineer- ing fails, the result may be hundreds or even thousands of deaths—typically of people of whom the engineers knows little—not, as in conventional medicine, just one person, a patient, known to the physician. (Of course, when physicians advise the Food and Drug Administration [FDA] or a drug manufacturer, the analogy with engineering is much closer.) For these reasons (and others), many of the conflicts of interest that engineers have thought about over the last century lack an exact analogue in medicine. They are nonetheless worth considering in detail because they illustrate how a profession can work from a basic understanding of conflict of interest to a system of detailed rules likely to be of use to practitioners in what would otherwise be situations hard to navigate. To understand the system of rules, it is important to understand something of the institutions in which they are embedded. Background Institutions Engineering is divided into four major disciplines (as well as many smaller ones): civil, mechanical, electrical, and chemical. These are more closely related to each other than medicine is to such other health care dis- ciplines, such as dentistry or osteopathy. That is, they are generally taught in departments of the same school; the curricula are similar, especially in the first 2 undergraduate years; the schools have the same accreditation body (ABET, Inc.);59 and students receive the same first degree upon graduation, a B.S. (with different majors). Nonetheless, engineering has never created the equivalent of AMA. Instead, there are five major societies. One each for the major disciplines: the American Society of Civil Engineers (ASCE), the American Society of Mechanical Engineers (ASME), the Institute of Electri- cal and Electronic Engineers (IEEE), and the American Institute of Chemical Engineers (AIChE). The fifth major society, the National Society of Profes- sional Engineers (NSPE), cuts across these four. Its members are (primarily) Professional Engineers (PEs). A PE is an engineer (of any discipline) licensed by a state (in much the way that lawyers, CPAs, architects, and physicians are), but only about a fifth of all U.S. engineers are so licensed. The rest, who work in large organizations, do not need a license to practice because of what is known as “the industrial exemption.” Although they are not PEs, they 59  ABET was formerly the American Board for Engineering and Technology; the name change to ABET, Inc., reflects its expansion into new areas.

338 CONFLICT OF INTEREST are full members of the engineering profession.60 The five major societies (along with many of the smaller ones) frequently cooperate ad hoc as well as maintain many permanent bodies, of which ABET is among the oldest and the most important. This complexity reappears in the codes of ethics governing engineers. Except for a brief period a half century ago, each of the five major societies has had its own code (as have many of the smaller societies); and, as if this were not enough ethical complexity for one profession, ABET (or its predecessor) has had a separate code (which has not been amended since 1977),61 one that most engineering societies have endorsed. Although the relationship among these codes is complex, it is not muddled. The NSPE code62 is designed (like architecture’s NCARB code) primarily for adoption by state licensing boards. Its rules are supposed to be appropriate for use in a disciplinary hearing. In contrast, the ABET code is designed to guide individual engineers. ABET has no enforcement procedure whatsoever (and does not even have a committee to issue advisory opinions) and, apparently, no interest in having its code enforced through any formal procedure. The ABET code thus functions much as the AIA’s Ethical Stan- dards do and should therefore be more demanding that the NSPE code. In fact, today it is as often less demanding than more demanding. Until the 1980s, ABET’s code was clearly the most important in engi- neering. Most engineering societies, including two of the major ones (ASME and ASCE), had adopted it as their own (either the 1977 version or one of its predecessors). In the last decade, however, its importance has declined dramatically. Some societies have amended their codes now and then, to the point that there are now important differences between those codes and ABET’s code. Some of the differences arise from the adoption of provisions that the NSPE adopted; some arise from local innovations (which other societies may or may not have followed). In addition, some societies (most notably, the IEEE) have abandoned the ABET code altogether. The NSPE code now seems destined to become the de facto standard of the profession (in part because the ABET code has gone so long without 60  Because most engineers are unlicensed, most continuing education depends on employers or on individual engineers. Large employers generally have their own internal technical courses (which the employer funds). Some continuing education goes on in universities as degree programs, certificate programs, or specific technical courses. Most large employers pay for an engineer’s continued technical education. Engineers may also be trained by a supplier, once the employer has contracted for some new product (such as software). Most states require PEs to take accredited continuing education courses. Accreditation of such courses is handled much as it is in accounting, architecture, and law. 61  See http://ethics.iit.edu/codes/coe/accreditation.board.engineering.tech.a.html (Code); http:// ethics.iit.edu/codes/coe/accreditation.board.engineering.tech.b.html (Guidelines). 62  See www.nspe.org/ethics/.

APPENDIX C 339 amendment and the IEEE code lacks sufficient detail to provide much guid- ance). The NSPE code is now much more often reprinted at the back of a text in engineering ethics than any other code. Although it is distinct from ABET’s code, NSPE’s code resembles it in layout and language because both derive from the “unity code” of a half century ago. Only the IEEE has an independent code (2000)63—which some other engineering societies, including the AIChE (2002), have followed. The IEEE code is quite short (260 words) and applies to IEEE members (not to engineers), an important distinction because many IEEE members are not engineers but are computer scientists, physicists, mathematicians, or the like. This survey confines its review of engineering’s methods of dealing with conflict of interest to three codes, those of the IEEE, the NSPE, and ABET, the most important (and distinctive) in U.S. engineering. For all the small differences among these three codes, there is a fundamental agreement about how to deal with conflict of interest. IEEE Code The one sentence on conflict of interest in the IEEE code expresses that fundamental agreement succinctly. IEEE members are to “avoid real or perceived conflicts of interest whenever possible, and to disclose them to affected parties when they do exist.” The IEEE strategy for dealing with conflict of interest (avoidance whenever possible and disclosure whenever avoidance is not possible or has failed) is similar to that identified for ar- chitects but nonetheless differs in two important respects. First, the require- ment of avoidance applies not only to “real” conflicts of interest but also to “perceived” ones. Perception—that is, the appearance—of a conflict of interest is treated as being just as bad as the reality. The underlying idea seems to be that an engineer’s professional judgment (or, rather, an IEEE member’s professional judgment) should be above suspicion. Even per- ceived conflicts should therefore be avoided (whatever the underlying reality about the interests in question). The underlying reality does not matter to those who would like to rely on an engineer—until it is disclosed and the false appearance is dispelled. The second important respect in which the IEEE strategy differs from that identified for architects is that there is no indication of what is to be done after disclosure (for example, there is no requirement of consent before continuing). The other engineering codes do provide guidance con- cerning this question, although the particulars vary a good deal, depending on the circumstances in question. The IEEE has a committee to prepare guidelines to supplement its code. 63  See www.ieee.org/portal/pages/iportals/aboutus/ethics/code.html.

340 CONFLICT OF INTEREST NSPE Code The NSPE code is divided into three main parts: a brief, four-sentence preamble; the body of the code, which consists of Part I. Fundamental Canons, Part II. Rules of Practice, and Part III. Professional Obligations; and an addendum (which may be ignored here) that quotes a 1978 federal court decision on competitive bidding and the response of the NSPE Ex- ecutive Committee. The fundamental canons (about 3 percent of the code) contains three sentences relevant to conflict of interest (in language dating from one of the first engineering codes): “Engineers, in the fulfillment of their professional duties, shall: . . . 3) Issue public statements only in an objective and truthful manner. 4) Act for each employer or client as faithful agents or trustees. 5) Avoid deceptive acts.” The Rules of Practice (which accounts for a quarter of the code’s 2,400 words) has six main rules, each of which corresponds to one of the Funda- mental Canons. The specific rules under a rule (designated with lowercase letters) are applications or elaborations of the prefacing rule. There are 3 rules under Rule II.3, 5 under Rule II.4, and 2 under Rule II.5, for 10 rules in all. All but two of these (Rules II.3a and II.3b) concern conflict of interest (more or less). In addition, Professional Obligations (about 40 percent of the code) contains six more rules related to conflict of interest. In all, about a fifth of the entire code is concerned with conflict of interest. Apparently, the NSPE takes conflict of interest very seriously. A detailed review of the provisions shows that they cover a surprisingly large number of specific issues. Rules of Practice Rule II.3c forbids engineers from issuing “statements, criticisms, or arguments on technical matters that are inspired or paid for by interested parties, unless they [the engineers] have prefaced their comments by explic- itly identifying the interested parties on whose behalf they are speaking, and by revealing the existence of any interest the engineers may have in the matters.” Although Rule II.3c is similar to NCARB’s Rule 3.1 (and AIA Rule 2.301), Rule II.3c differs in one striking respect. The engineer must not only reveal payment for a statement, criticism, or argument but even inspiration, presumably something more than NCARB’s “financial inter- est.” Although a financial interest might “inspire” a statement, so might friendship, the urging of a relative, or some other connection unrelated to compensation or financial interest. Although the language is vague, it is obviously meant to sweep wide (something that we might not expect in a code designed for discipline rather than for personal guidance). Why is there such a demanding rule?

APPENDIX C 341 For an engineer, the rationale for Rule II.3c is pretty straightforward. Because the rule is under Section 3, it concerns public statements. Engineers view the public much as physicians view patients. Engineers are—as Fun- damental Canon 1 puts it—to “[h]old paramount the safety, health, and welfare of the public.” There is no official definition of “public”; there is even some debate about the exact boundaries of the public, for example, whether the public includes employees of one’s client or employer. The most popular view, though, seems to be that the public includes all those who, owing to a lack of knowledge, power, or opportunity, are unable to protect themselves fully from what engineers do. What engineers call “the public” is (more or less) as dependent on engineering judgment as the physician’s patient is dependent on the physician’s judgment. Insofar as what engineers say in public may be influenced by an in- terested party other than the public, the public needs to know about that influence if it is to decide the appropriate weight to give the statement. The engineer, of course, tries to speak “in an objective and truthful manner” (as Canon 3 requires). If the engineer is not trying to do that, he or she should not speak at all (or, at least, not claim to speak as an engineer). However, if an engineer is aware of an influence that might (but also might not) undermine his or her ability to speak in an objective and truthful man- ner, he or she must warn the public of that danger to objectivity. There is nothing wrong with issuing public statements in the service of a client or an employer (as long as the statements are objective and truthful). There is, however, something wrong with an engineer giving the impression that he or she is doing something else, that is, expressing independent professional judgment (one independent of an employer, client, or other interested party) when it is not. An engineer must not give a false impression if he or she can reasonably avoid it. Experience and common sense suggest that an engineer may easily avoid giving that false impression by explicitly stating what he or she is doing as he or she begins the public statement in question. All the rules under the next heading (Rule II.4) are concerned with protecting the client or the employer, not the public. The strategy for deal- ing with conflict of interest is much the same as that expressed in Rule II.3c. Rule II.4a explicitly uses the term “conflict of interest.” Engineers are supposed to “disclose all known or potential conflicts of interest that could influence or appear to influence their judgment or the quality of their services.” There is no requirement to avoid those that can be avoided. The rule seems to be concerned with those conflicts that cannot be (or that perhaps just have not been) avoided. Rule II.4a makes something like the distinction between the IEEE’s “real” and “perceived” conflicts of interest, that is, between conflicts of interest that “could” influence and those that merely “could appear” to influence a judgment (or the quality of service). There is, however, a new distinction, that between “known” conflicts of in-

342 CONFLICT OF INTEREST terest and those that are merely “potential.” Although the distinction seems confused (should not the contrast be between “actual” and “potential” or between “known” and “unknown”?), the intent of the language seems to be clear enough: again (as in Rule II.3c), to sweep as widely as possible. Rule II.4a bars such excuses as, “I didn’t know it was a conflict of interest” and “It was only a potential conflict of interest.” Lastly, there is no attempt to distinguish financial interests from other kinds of interests. The rule applies to any conflict of interest whatsoever. Rules II.4b and II.4c are similar to NCARB’s Rules 2.1 and 2.2. They make avoidance the standard response to a conflict of interest. Rule II.4b forbids engineers from accepting “compensation, financial or otherwise, from more than one party for services on the same project, or for services pertaining to the same project, unless the circumstances are fully disclosed and agreed to by all interested parties.” Rule II.4c forbids engineers from soliciting or accepting “financial or other valuable consideration, directly or indirectly, from outside agents in connection with the work for which they are responsible.” The only significant difference between these two rules and the corresponding NCARB rules is (again) a wider sweep. Rule II.4b concerns compensation “financial or otherwise”; Rule II.4c concerns “other valuable consideration” as well as just “financial” (“directly or indirectly” solicited or accepted). So, a dinner, help finding another job, an all-expenses-paid trip to Jamaica, and a free course in some engineering subject, although they are not strictly financial compensation for work on a project (or “pertaining to the same project”), would clearly be close enough to require disclosure under Rule II.4b if they were, for example, offered in the way of offering thanks for favors done on a project. Similarly, although such things might not count as financial considerations, they still seem to count as consideration enough to be forbidden under Rule II.4c (coming from “outside agents”). The last two rules under Rule II.4 concern possible clashes between the engineer’s obligations to the public and obligations to a client or an employer. Rule II.4d forbids engineers in public service as members, ad- visors, or employees of a governmental or a quasigovernmental body or department to “participate in decisions with respect to services solicited or provided by them or their organizations in private or public engineering practice.” Engineers in public service are to recuse themselves whenever they, their employer, or their client has an interest in a decision. The client or employer may be a private firm, but the engineer should recuse himself or herself even if the client or the employer with an interest in the decision is another governmental (or quasigovernmental) agency. The rule makes no exception in the case of “full disclosure.” The idea seems to be that engineers serving in government (in whatever capacity) or in a quasigovern- mental agency (such as AMTRAK or the U.S. Postal Service) are supposed

APPENDIX C 343 to put their independent judgment at the public’s service. Disclosure of a conflict of interest to the relevant agency does not, as such, protect the public. Even the agency’s informed consent does nothing to ensure protec- tion of the public interest. The disclosure would have to be made to the public directly in a way that allows the public to take appropriate action. This can seldom happen when the engineer is advising a public agency (as it can when an engineer is speaking to the public directly). Disclosure to the relevant agency does not necessarily reach the public, and even when it does, the agency and not the public would ordinarily make the decision. So, the only way to protect the public from an engineer’s conflict of inter- est when the engineer’s judgment, though exercised in behalf of the public, works through a public agency, is to have the engineer avoid participation in the decision. Rule II.4e adopts the same strategy with respect to soliciting or accept- ing “a contract from a governmental body on which a principal or officer of their organization serves as a member.” (For some reason, this rule is silent concerning quasigovernmental bodies.) Again, neither mere disclosure of the conflict nor disclosure with consent is enough. The engineer must never solicit or accept such a contract. Although Rule II.4e concerns con- flict of interest, it is not designed to protect the engineer’s judgment (as the others are) but is designed to protect the judgment of the principal or the officer of the organization that the engineer serves. Indeed, it seems to be designed to protect the principal or the officer in question from the appear- ance of conflict as well as from actual conflict. There is no requirement that the principal or officer know of the contract or have anything to do with obtaining it. Protecting the principal or officer in question from the appear- ance of conflict of interest is part of being a faithful agent or trustee. Rule II.5b consists of three long sentences mostly concerned with brib- ery, but a part of the first sentence seems designed to avoid both certain conflicts of interest and the mere appearance of them (as well as actual bribes): “Engineers shall not . . . receive, either directly or indirectly, any contribution to influence the award of a contract by public authority, or which may be reasonably construed by the public as having the effect or in- tent of influencing the awarding of a contract.” The expression “reasonably construed” is, of course, a somewhat lower standard than “perceived,” as used in Rule II.4a. “Perceived” may be interpreted to include unreasonable as well as reasonable construal. The reason for the change in terms is not obvious (or known). One explanation is that what the public might reason- ably construe as taking a bribe or as being a threat to judgment is too close to dishonesty (the concern of Rule II.5) to be good for engineering’s repu- tation. However, what might unreasonably be so construed is not. There are other ways to deal with unreasonable construal; for example, pointing

344 CONFLICT OF INTEREST out how unreasonable it is. In this context, an engineer should be above “reasonable suspicion” but cannot avoid all suspicion. One problem with the use of “the appearance [or perception] of con- flict of interest” is its subjectivity. What appears to be is in part a matter of the psychology of the person doing the perceiving. The mad, the overly suspicious, or the profoundly cynical might see a conflict of interest where no one else would. In contrast, what might reasonably be construed as a conflict of interest, given the evidence available to the person doing the construing, is an objective matter. Even if we know that there is no conflict of interest (for example, because we know that the investments in question are in a blind trust), we can see that the public would be right to draw the opposite conclusion if all it knew was, say, that the engineer in question held the compromising investment. The public’s construal of the situation is, on the basis of the evidence, reasonable. Although the distinction between reasonable construal and unreason- able construal is important, it may not be as important to interpreting the NSPE code as it seems. All codes of ethics must be applied by using rea- sonable interpretive principles. One principle of reasonable interpretation is that, unless it is unavoidable, an interpretation should not lead to logical impossibility or practical absurdity. Because the avoidance of all perception or appearance of conflict of interest is probably impossible or at least un- reasonable, it seems likely that even the code provisions that do not specify the “reasonableness” of the perception or appearance in fact assume it—or, at least, should be interpreted as so doing. Professional Obligations So far we have been examining the part of the NSPE Code of Ethics called Part II. Rules of Practice. That part explicitly provides interpretations of Fundamental Canons 1 to 5, rules designed to protect the public, client, and employer. We now turn to the next part, Part III. Professional Obliga- tions, rules that seem to offer interpretations of the remaining Fundamental Canon, which requires engineers to “Conduct themselves honorably, re- sponsibly, ethically, and lawfully so as to enhance the honor, reputation, and usefulness of the profession.” This section has nine main rules, numbered like the Rules of Practice, but not obviously derived from the wording of either the preamble or the Fundamental Canons. Except for not overlapping much with the Rules of Practice, there is no obvious unity in the subject matter of Part III. It is, in effect, a code within a code concerned (primar- ily) with enhancing the honor, reputation, and usefulness of the profession rather than with protecting the public, the client, or the employer (though following its rules would often have that effect too). Part III has three rules concerning conflict of interest: Rules III.4, III.5, and III.6.

APPENDIX C 345 Rule III.4 protects the confidentiality of business and technical informa- tion that engineers learn while they are acting in a professional capacity. Rule III.4a is concerned with the unreliability in judgment that arises from trying to judge as one would if one did not know what one in fact knows. That rule forbids engineers “without the consent of all interested parties, [to] promote or arrange for new employment or practice in connection with a specific project for which the engineer has gained particular and special- ized knowledge.” The engineer must have the consent of “all interested parties,” generally, the old employer and the new one (as well as clients, if any), because she or he would be in an ethically untenable position other- wise. The engineer has an obligation to maintain the confidentiality of all specific business and technical information learned at the present employer (apart from what has become the engineer’s skill, experience, or general knowledge). At the new job, the engineer will have an obligation to act as a faithful agent, using her or his best engineering judgment on behalf of the new employer (or client), just as she or he did at the old employer. The engineer cannot use her or his best engineering judgment while trying to ignore some of what she or he knows (say, the specifics of a new product under development). If the engineer “bends over backward” to be fair to the previous employer, she or he will not treat her or his new employer as she or he should. The engineer will give the new employer less than her or his best. If, however, the engineer does not bend over backward to be fair, she or he cannot know that she or he has treated the old employer as she or he should. (To avoid revealing too much, the engineer needs a margin of safety, which means revealing too little.) Without guidance from the fully informed “interested parties,” the engineer is likely to fail the past employer, the new employer, or both. Because much technical and business information consists of trade secrets, the engineer may even provoke a lawsuit between the past and the present employer. The only way to avoid all of these troubles, apart from never changing jobs or never seeking new employment closely related to projects that one has worked on before, is to have the parties work out an arrangement in advance of the move from one company to another. The arrangement may be as simple as the new employer agreeing to buy a right to use the technol- ogy in question or as complicated as an agreement stating what kinds of projects the engineer can work on for a specified period (say, 2 years). Why does Rule III.4a not simply forbid engineers to seek employment too closely related to previous work? Why does it allow the consent of interested parties to resolve the conflict of interest problem (even if it does not resolve the underlying threat to judgment)? The (primary) moral wrong that conflict of interest threatens is a betrayal of justified reliance (rather than actual biased judgment). Engineers undertake to provide a certain level of service, that is, to be a reliable source of independent professional judg-

346 CONFLICT OF INTEREST ment concerning engineering. Conflict of interest means that an engineer can no longer safely be relied on for such judgment within a certain range of activities. If an interested party, that is, someone justified in relying on that judgment, is alerted to the problem by its disclosure and, by its (informed) consent, accepts the risk, the possibility of betrayal (in that respect) is eliminated. The profession’s honor and reputation for honor are preserved. What remains is only a practical problem of protecting the various interests at stake from biased judgment (that is, protecting engineering’s usefulness and its reputation for usefulness). One of those interests is the public’s inter- est in the productive use of what the engineer knows. Forbidding engineers to move from one job to a closely related one would waste some of what the engineer has learned (a social resource as well as a personal one) and make it harder for employers to find the engineers that they need. The rationale for Rule III.4b is similar. The rule forbids engineers, “without the consent of all interested parties, [to] participate in or represent an adversary interest in connection with a specific project or proceeding in which the engineer has gained particular specialized knowledge on behalf of a former client or employer.” Engineers frequently testify in court, arbitra- tion hearings, and similar proceedings on behalf of one side or the other. The United States does not have a system of official experts for tribunals to rely on. Representation or even participating in an adversary representation (for example, by testifying as an expert witness for one side or the other) may seem like a betrayal of trust or reliance (whether or not it is), if the engineer’s expertise derives even in part from specialized knowledge gained in the course of working for the adverse party (“biting the hand that once fed him or her,” so to speak). The engineer should appear as an expert in such a proceeding (or otherwise participate in it) only if all the interested parties welcome the engineer as an independent expert or at least as some- one who is not going to betray their justified trust or reliance. Again, the honor and the reputation of engineering are preserved. (The number of engineers makes it unlikely that the adverse party will fail to find a qualified witness even if one party rejects the first engineer for conflict of interest.) Rule III.5 forbids engineers to “be influenced in their professional duties by conflicting interests.” The rule should not be interpreted as for- bidding engineers to be influenced by conflicts of interest because, so in- terpreted, it would be inconsistent with all of the rules discussed so far, which permit conflicts of interest when there are full disclosure and consent (however, the interests, in fact, influence the decision). The only way to avoid being influenced by a conflict of interest is to avoid the conflict of interest or, having failed to avoid it, to recuse oneself. There is no other way to ensure that the decision in question is not influenced. (Disclosure and consent protect against betrayal of trust, not the loss of independent judgment itself.) So, Rule III.5 must instead be understood as forbidding

APPENDIX C 347 certain kinds of interests, those that always (or at least too often) conflict with an engineer’s professional duties. The two rules under Rule III.5 confirm this inference. Rule III.5a bars the acceptance of “financial or other considerations, including free engi- neering designs, from material or equipment suppliers for specifying their product.” (Free engineering designs, sometimes including free software, are the engineering equivalent both of the free samples of drugs that phy- sicians receive and of drug company-sponsored courses.) Rule III.5b also bars the acceptance of “commissions or allowances, directly or indirectly, from contractors or other parties dealing with clients or employers of the engineer in connection with work for which the engineer is responsible.” There is no exception for disclosure and consent. The engineer can easily avoid such interests without failing to do anything that an engineer should do for the public, a client, or an employer. The engineer should not put his or her interests in financial gain ahead of the interests of the public, a client, or an employer in having the engineer’s independent judgment. Finally, Rule III.6 is concerned with obtaining employment. Although most of the rules under it have nothing to do with conflict of interest, one does. Rule III.6a forbids engineers to “request, propose, or accept a com- mission on a contingent basis under circumstances in which their judgment may be compromised.” At one time, most engineering codes simply forbade engineers from working on a “contingent basis” (that is, where payment, all or just part, depends on success). One consequence of a series of antitrust cases brought against professions in the 1970s was that the rule against con- tingent fees was declared an unreasonable restraint of trade. The NSPE then sought to restate the rule to make clear its intent, which was not to raise the fees that engineers could charge for failure but to protect engineering judgment. Engineers might, it was thought, take chances that they should not take if their income depended even in part on “success”—success not in the sense in which engineers understand it (which takes into account the public’s long-term interests) but in the sense in which a client or employer might understand it (for example, getting a product out the door by a cer- tain date). Hence, Rule III.6a is another absolute rule. The consent neither of the client nor of the employer would permit an engineer to enter a fee contingent arrangement that might compromise her or his judgment. That completes the survey of how the NSPE code of ethics regu- lates conflict of interest. That does not, however, complete the survey of NSPE’s regulation of conflict of interest. In addition to the code, the NSPE maintains the Board of Ethical Review (BER), which receives inquiries from NSPE members concerning questions of ethics and issues opinions in response.64 BER publishes between 6 and 13 opinions each year. About 64  Many of these, including most since 1990, are available at www.niee.org/cases/.

348 CONFLICT OF INTEREST a quarter of these are indexed under “conflict of interest” (among other categories). Space does not allow for an examination of these, but such an examination would only confirm the guiding principles sketched so far: avoid all conflicts of interest that can reasonably be avoided, whether they are actual, potential, or merely apparent. Tolerate the remainder only if full disclosure to interested parties and their informed consent give them the tools that they need to protect against less reliable judgment. Neither consent nor disclosure is enough when an affected party cannot protect itself once it is fully informed. ABET Code The ABET code consists of two major parts: the Code of Ethics proper, which is a short document (210 words), and the much longer Suggested Guidelines for Use with the Fundamental Canons of Ethics (2,667 words). The Code of Ethics is divided into Fundamental Principles (which have much the same content as the NSPE preamble) and Fundamental Canons (which have much the same content as NSPE’s Fundamental Canons). The ABET guidelines correspond to the rest of the NSPE code (Parts II and III). For the purposes of this paper, the only significant difference between the two codes (apart from the guidelines) is that ABET’s Fundamental Canon 4 has been amended to append to the language of the NSPE code a comma and the words “and shall avoid conflicts of interest.” Most engi- neering codes of ethics now include that amendment, the result of a scandal in the middle 1970s that ended in a $7.5 million judgment against ASME.65 Some volunteers in one of ASME’s standard-setting bodies, although faith- ful agents and trustees of their employer (as Fundamental Canon 4 then required), had a conflict of interest when acting as members of the commit- tee. Because the engineers involved were members of ASME, as well as vol- unteers, ASME had not been their client or employer (in the ordinary sense of these terms). They had not (it seemed) violated Fundamental Canon 4 (or any other rule in effect at the time). Yet, most engineers thought that they had clearly done something that an engineer should not do. Since Fundamental Canon 4 did not seem to cover the case, although it should have, ABET revised Fundamental Canon 4 (adding the reference to conflict of interest), and most other engineering societies followed (with the notable exception of NSPE—which dealt with the problem by adding or amending rules under its equivalent of Fundamental Canon 4). 65  See ASME v. Hydrolevel Corp., 456 U.S. 556 (1982).

APPENDIX C 349 ABET Guidelines Although the ABET Guidelines were explicitly designed for use with the Fundamental Canons, they are an independent code in structure (and they were in fact the body of the code itself until 1977). The guidelines consist of seven main divisions, each of which is identical to one of the code’s Fundamental Canons (and carries the same number). Under each of these are lettered sections and sometimes numbered subsections interpreting or applying the canon. Many of the sections are identical in language to provisions of the NSPE code. Some differ in ways not important here. For example, ABET Rule 3d differs from NSPE Rule III.3c in using “engineering matter” rather than “technical matter,” by requiring engineers to identify themselves (as well as the party for whom they are speaking), and requiring them to describe any “pecuniary interest” that they may have in what they are about to say (rather than just any interest). In what follows, we ignore such small differences, focusing on rules that add something important to what we found in the NSPE code. There are only two such rules. They are, not surprisingly, both under Canon 4 (the canon explicitly concerned with avoiding conflict of interest). Rule 4a of ABET’s code differs from its NSPE counterpart (Rule II.4a) in requiring engineers to “avoid all known conflicts of interest” (rather than simply to disclose them) and to disclose promptly to their clients or employers the rest, what the NSPE code identified as “potential” conflicts of interest: “any business association, interests, or circumstances which could influence their judgment or the quality of their services” (emphasis added). This guideline makes explicit what we had found implicit in NSPE’s Rule II.4a. The ABET rule may, nonetheless, be less demanding than its NSPE counterpart. If the adjective “business” applies to “interest” and “circum- stance” as well as to “association” (a natural reading), then Rule 4a does not require disclosure of all conflicts of interest but only those arising from business associations, business interests, or business circumstances. The theme of avoidance is carried through the rest of the conflict of interest provisions under Rules 4b to 4g—all of which, except for Rule 4b, are (more or less) identical to the rules in the NSPE code (and, in fact, date from some of the earliest codes of engineering ethics). The exception, Rule 4b, forbids engineers to “knowingly undertake any assignments which would knowingly create a potential conflict of interest between themselves and their clients or their employers.” The two uses of “knowingly” sug- gests not only sloppy editing but also a great concern that engineers not be blamed for undertaking such assignments inadvertently. The require- ment of knowledge is a break with the ABET code’s general policy, which is to require avoidance or disclosure without providing for the excuse “I did not know.” (In other words, engineering codes generally treat ethical conduct as a question of competence for which “I did not know” is an

350 CONFLICT OF INTEREST admission of wrongdoing and not an excuse.) There is only one other use of “knowingly” in the entire code, one of which is unrelated to conflict of interest (Rule 6a, avoiding association with a disreputable business). Rule 4a’s knowledge requirement (like the use of “conflict of interest”) is new to the 1977 code. However, it is an innovation that some other codes have followed. For example, Rule 4b in ASME’s current code (2002) is simply a cleaner version of ABET’s: “Engineers shall not undertake any assignments which would knowingly create a potential conflict of interest between themselves and their clients or their employers.” Conclusions from Survey of Engineering Codes For engineering, then, conflict of interest is a threat to the profession’s usefulness (the reliability of its judgment) as well as to its honor and reputa- tion. For most purposes, the best response to an actual or potential conflict of interest is to avoid it as soon as one learns of it. In a few cases, recusal is allowed (or required); in others, those cases in which (1) disclosure allows the public, client, and employer an adequate response and (2) the engineer cannot be replaced or cannot be replaced at reasonable cost, tolerance of a conflict of interest is allowable. However, it is only allowable. Even then, there is a risk to all who rely on the engineer’s judgment that the engineer’s judgment will not be as good as it should be (and would be but for the conflict of interest). Disclosure is not a cure-all. COMPARATIVE OVERVIEW This survey has discussed both similarities and differences in the treat- ment of conflicts of interest by four important professions. What can be learned from this survey of lawyers, certified public accountants, architects, and engineers? The obvious point is that all four professions take conflict of interest in professional practice very seriously. The recent history of public accountancy shows that failure to take conflicts of interest seriously enough can result in federal regulation. Engineering had a similar experience three decades ago with civil liability.66 With the exception of engineering, these professions do not under- take the kind of scientific research carried out by some in the medical 66  American Society of Mechanical Engineering, Inc. v. Hydrolevel Corp., 456 U.S. 556 (1982), in which the U.S. Supreme Court held that ASME was strictly liable for the acts of its agent (the chairman of ASME’s Boiler and Pressure Vessel Codes Committee) if those acts are in breach of antitrust laws (the chairman, an officer in the competitor of Hydrolevel, had a financial interest in his committee, finding that Hydrolevel’s product did not meet ASME’s code).

APPENDIX C 351 profession.67 They have therefore not had to deal with controversies over conflicts of interest in research. Although each profession sets standards for providers of continuing professional education, they have, it seems, not faced any significant conflict of interest in education either.68 One reason for the relatively low level of attention given to conflicts of interest in research and education may be that lawyers and accountants do not act as gatekeepers for significant numbers of products and services as physicians do when they prescribe medications, use medical devices, order diagnostic tests, and the like. To the extent that architects and engineers are gatekeepers for supplies, their codes of ethics carefully regulate relation- ships with suppliers, gifts from suppliers, and other entanglements with suppliers that might threaten their professional judgment. All four professions treat conflict of interest situations as risk situa- tions; bias, breach of confidentiality, fraud, and malpractice are dealt with separately. Conflicts of interest are understood to threaten the quality of the individual professional’s judgment and, as a consequence, the well-being of the client or employer in question, the profession’s usefulness to the public (depending on the specific circumstance), and the reputation of the profes- sion as a whole. The four professions express concern about conflict of interest in somewhat different ways and justify their management measures by appealing to different core values. The three most prominent values are loyalty to the client (or the employer), professional judgment, and public service. Beyond the fact that the four professions share these three most prominent values, we can draw at least 13 other conclusions: 1. Each profession has, over time, developed at least one detailed na- tional code of professional ethics. Each of these codes is generally adopted (sometimes with amendments) by state-level professional organizations, licensing boards, or both. All these codes include general principles as well as more specific rules. A substantial part of each of these codes addresses conflicts of interest, describing what the profession understands conflict of interest to mean and how members of the profession should deal with 67  Publishable engineering research generally goes on in (1) universities, (2) government laboratories, or (3) private laboratories (such as IBM’s Watson Research Center). Most of this engineering research is scientific and is therefore subject to federal conflict of interest rules much as most medical research is. Relatively little engineering research is the equivalent of testing by the FDA. Some is, however, for example, the testing done by Underwriters Labo- ratories. So far, it seems, engineering’s strict rules concerning conflict of interest seem to have protected it from the sorts of scandals medical research has suffered. 68  Nevertheless, some guidance on conflict of interest in scholarship is available from the Association of American Law Schools, which requires that professors disclose any economic interest that they have in the subject matter of their scholarship. Insofar as professors are themselves members of their respective professions, they will be subject to the same conflict of interest rules and codes of conduct as their nonscholarly colleagues.

352 CONFLICT OF INTEREST specific conflicts of interest (usually describing which conflicts will be pro- hibited, consentable, or allowable even without consent). 2. Compliance with each profession’s codes of ethics depends—as the AICPA code of ethics says—“primarily on members’ understanding and voluntary actions, secondarily on reinforcement by peers and public opin- ion, and ultimately on disciplinary proceedings, when necessary, against members who fail to comply with the Rules.” In other words, the codes of ethics of all four professions are enforced in much the same way that the AMA enforces its code of ethics. They are not designed for use by state licensing boards. 3. Protecting against conflict of interest occurs not only at the level of the professional society and state licensing board. Some conflicts of interest constitute malpractice or breach of criminal law or civil regulation (e.g., the Sarbanes-Oxley Act and its regulations). Some failures to deal properly with conflicts of interest can have serious consequences for the professionals involved. Statutes and case law, however, generally sets a standard for con- duct lower than that set by codes of ethics: law is designed to set minimum standards below which no member of the profession should fall, whereas codes of ethics are designed at least in part to set a higher standard (some- thing closer to the best that can reasonably be expected of members of the profession). For many professions, the minimum standard with respect to conflict of interest has risen substantially over the last four decades. There is no reason to expect that trend to change anytime soon. 4. There is general agreement that professionals will find themselves in some conflict of interest situations even when all reasonable precautions have been taken to avoid them. When avoidance cannot reasonably be expected or has failed, censure attaches not so much to having a conflict of interest (except for prohibited relationships) as to a professional’s failure to take proper steps to deal with it. 5. The conflicts of interest discussed in this paper can arise in at least three ways: • The interests of two or more of a professional’s current or former clients (or employers) can conflict and the professional can therefore be in a situation in which serving one client competently (for example, preserv- ing confidential information) would mean not serving another client com- petently (that is, the professional is not able to use all of the information that he or she knows). This is a major concern for lawyers as well as for engineers. • The financial, familial, or other interests or relationships of a professional can conflict with the interests of one or more clients (or em- ployers) and thereby compromise judgment (a major concern for all four professions).

APPENDIX C 353 • The interests of a client (or employer) can conflict with the public interest and thereby risk compromising the quality of the professional’s judgment (a major concern for CPAs, particularly when they conduct au- dits, but also a concern for architects working as adjudicators or making public statements and for engineers making public statements or working for or with government). 6. Each of the professions, as a general matter, understands that conflicts of interest can be created not only by financial considerations but also by other considerations, such as nonmonetary gifts, friendships, family relationships, and previous employment. The crucial question is always the known or suspected tendency of the fact in question to affect professional judgment adversely. 7. Each profession understands that conflict of interest is in part a threat to the trustworthiness (or reliability) of the profession as well as to judgments in specific cases. This is clear from the way in which the profes- sions, each in a somewhat different way, address appearances. In general, members of these professions are supposed to avoid giving clients, employ- ers, and the public even a plausible reason to suppose that they have an interest, relationship, or the like that might impair their objectivity (the reliability of their judgment). 8. Not all conflicts of interest are treated in the same way. We may distinguish three ways of treating them. The codes of ethics for each of the four professions begin with the instruction to avoid conflicts of interest. This general instruction is then modified or further refined by distinguishing between (1) conflicts of interest that must be avoided regardless of the spe- cific circumstances (i.e., conflict of interest situations that are prohibited), (2) those that are permitted under certain circumstances following disclo- sure and, generally, that are accompanied by the informed consent of the client or other parties directly affected or some other management strategy, and (3) those conflicts of interest that are permitted because of their relative insignificance. Because clients or employers are often sophisticated individu- als or businesses, they are capable of refusing consent or setting conditions for consent (once a conflict is disclosed). Modifiers such as “substantial” or “significant” as well as “direct” (in contrast to “indirect”) indicate that not all conflicts of interest are of equal concern. The professions understandably attempt to focus their rules on interests that seem likely to have more than a minor impact on professional judgment or on trust in the profession. 9. Because so many conflicts of interest are either prohibited outright, require disclosure and consent, or are hard to manage, avoidance is, all else being equal, the preferred technique for dealing with conflict of interest. Avoidance is facilitated by certain practices; for example, a lawyer runs a “conflicts check” inside the firm before a new file is accepted. In all four

354 CONFLICT OF INTEREST professions, the avoidance of a conflict of interest sometimes means forgo- ing personal gain or gain for a client or an employer, a fact that all four professions acknowledge. Avoiding conflict of interest certainly has costs (as well as benefits). 10. When the conflict of interest has not been avoided (for whatever reason and whether intentionally or unintentionally), various options to es- cape from or manage the conflict exist. Recusal is one option. For example, engineers who are members, advisors, or employees of a governmental department must withdraw from decisions in which they, their employers, or their clients have an interest. The engineer must comply with this ethical rule even if governmental regulation allows for disclosure and consent as an alternative way of managing the conflict. Despite the general requirement to avoid conflicts of interest, professionals can proceed despite a conflict of interest under specified circumstances. Generally, certain precautions must then be taken: (1) disclosure of the interest to the parties concerned (who can include current and former clients, current and former employers, and third parties), (2) the informed consent of these parties (although, occasion- ally, disclosure alone is sufficient), and (3) the implementation of additional management measures (for instance, the use of screens in law firms). The codes try to make clear when disclosure followed by consent (or disclosure alone) will be considered sufficient to preserve both the fact and the ap- pearance of proper judgment (independence, loyalty to client, reliability, or the like). When proper judgment cannot be ensured, the conflict must be avoided, despite the advantages (to the professional, the professional’s employer or client, or any other party) of accepting it. 11. Patterns of difference between (what lawyers call) “consentable” and “nonconsentable” conflicts of interest are sometimes difficult to discern (and, indeed, may be evolving). Overall, it seems that the more dependent that the client, employer, or public is on the professional and the less abil- ity that the client, employer, or public has to manage the conflict, the more likely that consent, even after full disclosure, will not override the general prohibition of conflict of interest. In legal practice, for example, a typical nonconsentable conflict of interest arises if a lawyer undertakes the drafting of a will granting him or her a substantial gift from a client. A typical con- sentable conflict of interest arises if, for example, a lawyer bought a share in a hotel owned by a client (what lawyers call an “arm’s-length” business transaction). 12. Instruction in understanding, identifying, and managing conflicts of interest is included in graduate education, licensing examinations, and (often) in mandated CPE for all of the professions evaluated here. 13. CPE in law, accounting, architecture, and engineering is provided by companies that are authorized by the relevant state-designated licensing boards or a national accreditation body to provide CPE. Individual pro-

APPENDIX C 355 fessionals must regularly complete a set amount of CPE, often including training in conflict of interest, to maintain their professional licenses. They or their employers pay the cost of the CPE, although some CPE courses are offered for free by local or national professional organizations. Table C-1 summarizes the responses of the four professions discussed here to conflicts of interest.

356 TABLE C-1 Summary of the Responses of Four Professions to Conflicts of Interest Certified Public Conflict of Interest Lawyers Accountants Architects Engineers Gifts or rebates Lawyers cannot solicit or Gifts are prohibited unless Gifts cannot be accepted. Engineers cannot solicit prepare instruments to the value is “clearly Rebates are permitted or receive gifts or other receive substantial gifts. insignificant to the only with the informed valuable consideration. recipient.” consent of all relevant parties. Public speaking, speaking Not addressed. Public speaking is not Architects must disclose Engineers are forbidden about professional issues addressed, except to the any personal financial from making statements extent that audits and interests in public on technical matters “that attestations are public statements. are inspired or paid for by statements (conflict of interested parties,” unless interest management is the engineer prefaces the very strict in such cases). statement with disclosure of the interest. Financial or other Some fair financial Neither a CPA nor Disclosure and consent No contingent fee under relationships with client relationships (business members of a firm can are required if the client is conditions that could transactions, real estate, have direct or many not aware. affect professional etc.) with a client are indirect financial interests judgment is allowed. possible following or familial interests in Disclosure and consent disclosure and informed the client being audited. are required if the client is consent. Many personal Restrictions on nonaudit not aware. relationships are services are offered during nonconsentable. the time of audit.

Financial or other No representation of Few restrictions. Disclosure and consent of Such relationships are relationships with relevant opposing parties in all interested parties are generally prohibited. (opposing) nonclients litigation is allowed. required. Exceptions (such as Representation following testifying in court as disclosure and consent of an expert witness) both parties at other times are allowed with full is allowed. disclosure and consent. Sponsorship of CPE CPE is usually paid for CPE is paid for by the AIA or NCARB provides CPE is primarily required by the lawyer or his or accountant or his or her CPE. CPE is paid for by for a PE license, generally her firm. firm. CPE is not approved the architect or the firm state certified. Most if it is “devoted to the when it is not provided engineers are not required promotion of particular for free. to take CPE, but whether products or services.” it is required or not, it is generally paid for by the engineer or the employer. 357

Next: Appendix D: How Psychological Research Can Inform Policies for Dealing with Conflicts of Interest in Medicine »
Conflict of Interest in Medical Research, Education, and Practice Get This Book
×
Buy Hardback | $54.95 Buy Ebook | $43.99
MyNAP members save 10% online.
Login or Register to save!
Download Free PDF

Collaborations of physicians and researchers with industry can provide valuable benefits to society, particularly in the translation of basic scientific discoveries to new therapies and products. Recent reports and news stories have, however, documented disturbing examples of relationships and practices that put at risk the integrity of medical research, the objectivity of professional education, the quality of patient care, the soundness of clinical practice guidelines, and the public's trust in medicine.

Conflict of Interest in Medical Research, Education, and Practice provides a comprehensive look at conflict of interest in medicine. It offers principles to inform the design of policies to identify, limit, and manage conflicts of interest without damaging constructive collaboration with industry. It calls for both short-term actions and long-term commitments by institutions and individuals, including leaders of academic medical centers, professional societies, patient advocacy groups, government agencies, and drug, device, and pharmaceutical companies. Failure of the medical community to take convincing action on conflicts of interest invites additional legislative or regulatory measures that may be overly broad or unduly burdensome.

Conflict of Interest in Medical Research, Education, and Practice makes several recommendations for strengthening conflict of interest policies and curbing relationships that create risks with little benefit. The book will serve as an invaluable resource for individuals and organizations committed to high ethical standards in all realms of medicine.

  1. ×

    Welcome to OpenBook!

    You're looking at OpenBook, NAP.edu's online reading room since 1999. Based on feedback from you, our users, we've made some improvements that make it easier than ever to read thousands of publications on our website.

    Do you want to take a quick tour of the OpenBook's features?

    No Thanks Take a Tour »
  2. ×

    Show this book's table of contents, where you can jump to any chapter by name.

    « Back Next »
  3. ×

    ...or use these buttons to go back to the previous chapter or skip to the next one.

    « Back Next »
  4. ×

    Jump up to the previous page or down to the next one. Also, you can type in a page number and press Enter to go directly to that page in the book.

    « Back Next »
  5. ×

    To search the entire text of this book, type in your search term here and press Enter.

    « Back Next »
  6. ×

    Share a link to this book page on your preferred social network or via email.

    « Back Next »
  7. ×

    View our suggested citation for this chapter.

    « Back Next »
  8. ×

    Ready to take your reading offline? Click here to buy this book in print or download it as a free PDF, if available.

    « Back Next »
Stay Connected!