What guidance is provided by ethical guidelines for professional practice in economics and statistics and data science?

Dear Amstat News,

The article by professor Rochelle Tractenberg on “ethical guidelines” in the May 2017 Amstat News was interesting, but the information about guidelines for economists was not up-to-date.

Several years ago, the National Association for Business Economics (NABE), a professional association with more than 2,500 members, introduced a program to become a “Certified Business Economist” (CBE). Successful candidates are required to be familiar with NABE’s “Professional Ethics Guidelines,” which are spelled out on the NABE website. In addition, this year, the CBE program is being offered in the graduate economics programs at Boston College, Brandeis International Business School, The George Washington University, and the University of Cincinnati, as well as in the undergraduate program at John Carroll University.

Also, in 2012, the larger American Economics Association issued guidelines for articles appearing in their journals. However, these guidelines appear to cover only conflicts of interest, and do not cover other professional activities of its members.

I think ASA members should be informed about recent activities on ethics by professional organizations of economists, which whom they frequently collaborate.

Sincerely,
Robert Parker
Consultant and ASA Fellow

RESPONSE

Rochelle E. Tractenberg and George F. DeMartino

Rochelle E. Tractenberg is a professor in the departments of neurology, biostatistics, bioinformatics, biomathematics, and rehabilitation medicine at Georgetown University in Washington, DC. She is a PStat®, Fellow of the ASA, and chair of the ASA Committee on Professional Ethics.

George F. DeMartino is a professor of international economics and co-director of the program in global finance, trade, and economic integration in the Josef Korbel School of International Studies at the University of Denver.

In a May 2017 column (“Advocating for Ethical Guidelines Strengthens Statisticians, Data Analysts”), one of us (Tractenberg) described a symposium at the 2017 AAAS annual meeting that was organized specifically “to highlight the crucial ways ethical professional practice is essential to promoting ethical policy and decision-making.” The May 2017 column included a summary of themes that emerged during that AAAS symposium, namely pointing out that the other of us (DeMartino) argued the field of economics has neither “guidance” nor “guidelines” when it comes to professional practice.

We appreciate the informative response provided by an ASA Fellow who is also an economist. This ASA member pointed out that the National Association of Business Economics (NABE) adopted professional guidelines in 2010. Individuals who seek to become “Certified Business Economists” (CBE), an accreditation that became available in 2012, are required to sign and adhere to the CBE Professional Ethics Guidelines. To date, of NABE’s 2,500 members, about 130 individuals have been awarded the CBE designation.

The NABE initiative is commendable, and we are happy to correct the record here. In adopting its guidelines, NABE joins several smaller economic associations in taking up the challenge of thinking through their professional responsibilities. The leader in this regard is the 550-member National Association of Forensic Economists (NAFE). Its Statement of Ethical Principles and Principles of Professional Practice resulted from careful discussions among its members. Equally important, NAFE views its statement as central to its mission and it promotes examination of and debate over these principles in its journal and other activities. In these ways, the NAFE helps to ensure their statement is not a dead letter.

Other associations around the world are now just beginning to move in similar directions. The hope is that their initiatives will begin to normalize the idea that economists face nontrivial, daunting ethical duties that the profession must engage if it is to fulfill its obligations to its members and those whose lives are affected by economists’ practice.

One Step Forward, Two Steps Back: The American Economic Association and the Situation Across Economics

But, in our view, economics has a long, long way to go to achieve that goal. Although difficult to count, there are between 21,500 (Bureau of Labor Statistics estimate from 2014) and 27,000 (estimated in The Economist’s Oath) economists working in the United States. Of these, about 12,500 work in academia, with the rest spread across government and private, multilateral, and nonprofit sectors.

To date (July 2017), of NABE’s 2,500 members, about 130 individuals have been awarded the CBE designation, meaning 5.2% of NABE membership is known to have read and agreed to their guidelines. Considering NABE CBE holders and NAFE membership (n=630) within the population of economists in the US (n=27,000), 2.3% of active economists would definitely be aware of the NAFE and NABE initiatives.

In contrast to NABE and NAFE, the largest and most important professional association for economists, the American Economic Association (AEA), has 18,000 members. What the AEA does reverberates throughout the profession. Its journals are top ranked, its officers and board members are recipients of the most prestigious prizes and hold the most coveted academic positions, and—in various ways—its practices and norms diffuse throughout the profession.

It is for this reason it is such a tragedy that the AEA has distanced itself from any serious engagement with the matter of professional economic ethics. There’s no need to hedge here: The AEA does not provide ethical guidance of any sort for its members.

The AEA does have a requirement that economists publishing in one of the AEA journals disclose potential conflicts of interest (COI) that might arise from funding sources for their research, coupled with the recommendation that economists disclose COI in their other work.

The AEA disclosure guidelines have been around for some time. They were strengthened in 2012 in response to stinging criticism from the business press following the release of the documentary Inside Job and other research (here and here) demonstrating that leading economists who wrote reports and testified before the US Congress on financial policy matters in the lead up to and following the crisis of 2008 routinely failed to disclose that they were paid by those with financial interests in their results.

Just as the film was released, DeMartino’s book, The Economist’s Oath, was published. It called out the profession for its ethical lapses and advocated the creation of the new field of professional economic ethics. In response to the brouhaha, the AEA Executive Council established an ad hoc ethics committee to investigate its responsibilities and potential courses of action. The result of the committee’s efforts was the strengthened disclosure rules. The committee was then disbanded.

Beyond that, the AEA has never in its 130-year history (and up to today) provided an ethical code or statement of ethical principles or issued suggestions, instructions, or ethical guidance of any sort. It has no journal—or even a bulletin—that engages the matter of professional economic ethics, no professional ethicists (as we find in professions that take the matter seriously), and no regular ethics committee. Also, it sponsors no conferences on economists’ professional responsibilities.

Moreover, the leading economic departments at universities across the US have no curriculum in ethical practice, and they provide no ethical training whatsoever to new initiates to the profession who will quickly exert substantial influence over thinking about economic policy. Indeed, the AEA leadership has dismissed out of hand calls to advance professional economic ethics, as DeMartino chronicles in The Economist’s Oath.

The AEA’s view, in short, is that having clarified and strengthened its COI disclosure rules for its journals in 2012, its ethical work is done. Importantly, whether this is or is not AEA’s view, its activities are perceived as such. Many leading economists concur with this view. Possibly, they would also hold to the view so firmly articulated recently by former World Bank Chief Economist Anne Krueger in her scathing review of the new Oxford Handbook of Professional Economic Ethics (which DeMartino co-edited with Deirdre McCloskey). She argues that attending to ethical issues would stall economic projects that badly need doing—as if thinking through the ethical entailments of economic practice was simply a foot-dragging strategy that interferes with economists’ professional practice.

The AEA treatment of professional ethics reflects a naïve sensibility that has, by now, thankfully been displaced in those professions that take ethics seriously. Its endurance in economics is a consequence of the immature state of professional ethical thinking and even willful ignorance in the profession. Unfortunately, non-economists unwittingly bear the consequences of that naiveté.

How Does This Compare with the Situation in Statistics?

The 2014 Bureau of Labor Statistics estimated there were 30,000 statisticians in the United States. Like the NABE, the ASA also requires that Accredited Professional Statisticians (PStat®) and graduate-level (GStat) statisticians are familiar with—and agree to uphold—the ASA Ethical Guidelines for Statistical Practice.

Assuming that only those who are required to actually have read and follow the guidelines, and given that there are roughly 400 current holders of PStat and GStat, about 2.2% of an estimated 19,000 ASA members are specifically asked to follow the guidelines. This number must be the lower bound on the estimate, but amounts to about 1.3% of the statisticians in the US.

By comparison, the Royal Statistical Society’s (RSS) Code of Conduct is “mandatory” for all professionally qualified Fellows of the RSS (equivalent to the ASA’s PStat/GStat designations). In spite of this mandate, like the ASA, only those who are accredited are known to have agreed to read and abide by the ethical guidelines; as of July 26, 2017, 27% of RSS Fellows are thus mandated to follow their code. Similarly, given their current accreditation numbers, 18% of members of the Statistical Society of Canada (SSC) are known to have read and agreed to follow their guidelines.

So, the issue that was pointed out in the original (incorrect) characterization of the AAAS session about ethical guidance and guidelines for the domain of economics was that there was none; the implicit message was that the domain of statistics and data science is better off because the ASA, RSS, and SSC do have them. Instead, while the NABE has more than twice the “acceptance rate” of their ethical code (5.2% of the 2,500 NABE members are known to have agreed to abide by them) than what can be confidently claimed for the ASA guidelines (2.2% of 19,000 ASA members), an astonishingly low level of guidance is actually being offered by both codes for practitioners to whom those codes should be familiar and by which practitioners should feel empowered.

Professional Ethics: Cultivating Ethical Professional Identities

Merriam-Webster defines “professionalism” as comprising “the skill, good judgment, and polite behavior that is expected from a person who is trained to do a job well” and “the conduct, aims, or qualities that characterize or mark a profession or a professional person.”

These characteristics may (or may not) resonate with readers, but researchers in the field of professional identity development point out that “… the sense of being a professional … the use of professional judgment and reasoning … critical self-evaluation and self-directed learning …” (Paterson et al. 2002. “Clinical Reasoning and Self‐Directed Learning: Key Dimensions in Professional Education and Professional Socialisation.” Focus on Health Professional Education) and “(p)rofessional identity formation means becoming aware of … what values and interests shape decision-making.” (Trede, F. 2012. “Role of Work-Integrated Learning in Developing Professionalism and Professional Identity.” Asia-Pacific Journal of Cooperative Education). We cannot assume that if a professional association creates a set of core professional conduct principles, even if these are reflected in a mission statement, that they have any hope of seeping into the consciousness and habits of mind of current or future practitioners. The American Statistical Association has revised its Ethical Guidelines for Statistical Practice, explicitly articulating that, “(t)he principles expressed here should guide both those whose primary occupation is statistics and those in all other disciplines who use statistical methods in their professional work … comprising statisticians at all levels of the profession and members of other professions who utilize and report statistical analyses and their implications.”

The original STATtr@k post may have inadvertently created the perception that, “while economics has neither guidance nor guidelines, statistics and data science has both.” As is hopefully much clearer now, while there are clear and highly relevant guidelines for professional practice for statistics and data science, the amount of actual guidance the ASA ethical guidelines can be inferred to be providing is low—too low for comfort.

While it may be true that economics, as a field, must do a better job of committing itself to “normalizing the idea that economists face nontrivial, daunting ethical duties that the profession must engage if it is to fulfill its obligations to its members and those whose lives are affected by economists’ practice,” the AEA “ethical guidelines” will not be enough. True normalization requires that new practitioners see this engagement, and must learn how to engage with those duties themselves.

The ASA has taken steps to achieve this normalization, having established professional practice guidelines going back to the mid-1970s. Articulating the “guidelines and guidance” as the ASA has done is necessary, but not sufficient. We need to engage all practitioners in statistics and data science, whose ethical professional conduct we hope to support, in contributing to the same normalization we hope to see one day for economics.

The ASA Committee on Professional Ethics is working (2017–2019) on a series of case studies that can be used to teach ethical statistical practice to all those the ASA ethical guidelines are intended to support. With the assistance of the ASA Board and the Professional Issues and Visibility Council, the committee is also exploring ways to build “normalization”—and guidance—that our Ethical Guidelines for Statistical Practice can and do represent.