Many accountants do not receive formal training in psychology or judgment and decision-making, but foundational concepts in these fields have broad implications for everyone (even accountants).
Academic research in both psychology and accounting reveals numerous pitfalls in the human decision-making process that can negatively affect everything from audit quality to the appropriateness of tax positions advocated by accountants for their clients.
By understanding key concepts about judgment processes and weaknesses, we can avoid making predictable mistakes and professional blunders, maintain professional skepticism, and perhaps be more aware of mistakes made by others (see the sidebar, "How to Use Psychology to Your Advantage," below).
Directional goals are a form of motivated reasoning that cause people to evaluate information selectively and search for information that allows them to justify their desired conclusion, Ziva Kunda wrote in Psychological Bulletin in 1990. Although people believe they are being objective, directional goals motivate them to reach specific conclusions, but only after they can identify enough evidence to support the goal. The unconscious nature of directional goals makes them dangerous in a professional setting such as accounting where practitioners are expected to remain objective and not be inappropriately persuaded with insufficient information. Perhaps even more troublesome is the consistent finding that just being aware of another person's opinion may bias us toward that position. Studies have shown that:
- Auditors made going concern judgments that were more consistent with the judgments of the partner in charge of a hypothetical engagement when they knew the partner's opinion before they evaluated the evidence, according to a paper published in The Accounting Review by T. Jeffrey Wilks in 2002. Those who learned the partner's opinion after they evaluated the evidence made judgments that were less consistent with the partner's.
- Auditors cautioned about needlessly investigating audit details because of inefficiencies were significantly less skeptical than other groups of auditors, as long as client integrity was believed to be high, Mark Peecher wrote in the Journal of Accounting Research in 1996.
- Tax professionals completing a hypothetical analysis stressed the importance of tax research that supported their client's desired position but gave less consideration to equally relevant court cases that suggested a contrary position, C. Bryan Cloyd and Brian Spilker wrote in The Accounting Review in 1999. This persisted even when the client supported an inappropriate tax position.
These examples show that accountants can be susceptible to the effects of directional goals, including those that we unintentionally inherit from superiors or clients. These studies also suggest the need for caution when communicating our views to others, because even the most objective accountants can fall victim to the effects of motivated reasoning. All is not lost, however.
Being aware of the potential to be directionally motivated is a step in the right direction. Armed with this knowledge, CPAs can take careful steps to make sure they:
- Spend just as much effort seeking information that may contradict their desired conclusion as they spend seeking confirming information.
- Give equal, objective consideration to all information that may be relevant in a decision-making task.
When we like another person, we tend to like everything about that person, Nobel Prize-winning psychologist Daniel Kahneman notes in his 2011 book, Thinking, Fast and Slow. Likewise, when we dislike a person, we tend to dislike everything about that person.
Moreover, we tend to develop our impressions about one another by more heavily weighting early information we receive and largely ignoring later information, once our initial impression has been formed (first impressions matter). Psychologists call this phenomenon "exaggerated emotional coherence" or "the halo effect."
The halo effect is so strong that once we have developed an initial impression, we begin to assume things about another person's character that we have not yet observed, according to Kahneman. These positive and negative attributions further enhance our impressions.
Pitfalls of the halo effect are especially apparent when auditors evaluate information about clients (and members of client management) in a somewhat arbitrary order. In this setting, negative information about a client's control structure or operational features may precede other, positive information about client attributes, or vice versa.
For example, audit seniors who analyzed a client's business strategy before performing unrelated, account-level analytical procedures documented less risk associated with unexpected account fluctuations than seniors who did not complete strategic analyses before performing the analytical procedures, Ed O'Donnell and Joseph Schultz Jr. wrote in The Accounting Review in 2005. Evidently, merely developing an initial impression about the client's business environment caused auditors to discount the importance of unexpected account balance fluctuations.
Moreover, O'Donnell and Schultz found that auditors who thought business strategy risks were low (favorable halo) assessed inconsistent account fluctuations as lower-risk than auditors who thought business strategy risks were high (unfavorable halo).
According to Kahneman, the halo effect is common largely because this approach is easier and requires less mental effort than evaluating information in a consistent and objective fashion, regardless of when it is received. This threat to unbiased decision-making is why some teachers grade the responses from all students to a given exam question before moving to the next question, and then grade all responses to that question, instead of simply grading each student's exam, front-to-back, individually.
By adopting this approach, graders are less likely to let their impression of a student's ability on one response contaminate their assessment of the student's other work. Adopting similar strategies in our work that force us to impartially evaluate evidence may help reduce the influence of the halo effect on decision-making processes.
An equally frustrating psychological phenomenon known as "ego depletion" dictates that acts of volition (the use of a person's will), such as exercising self-restraint or exerting mental effort, deplete a limited but renewable self-control resource, according to the Journal of Personality and Social Psychology paper "Ego Depletion: Is the Active Self a Limited Resource?"
This resource can wane throughout the day. Everyday decision-making depletes finite self-restraint resources. In fact, researchers Maryam Kouchaki and Isaac Smith wrote in Psychological Science in 2013 that people are more likely to be dishonest or engage in cheating later in the day than early in the morning.
Auditors who exhibit higher levels of professional skepticism were actually more susceptible to the effects of ego depletion in terms of their risk assessment performance in a case study setting than auditors who were less skeptical, accounting researchers Lori Bhaskar, Tracie Majors, and Adam Vitalis reported in their paper on the Social Science Research Network. This is thought to be the case because more skeptical auditors likely engage in more complex thinking during the audit process, which more quickly depletes their decision-making resources. Once these resources are depleted, these auditors are inhibited from exerting additional effort.
Fortunately, ego depletion is not a permanent state, and its causes are somewhat physiological in nature. Rest, sleep, and replenishing glucose levels in the blood also appear to renew our decision-making capacity, according to a paper published in the Journal of Personality and Social Psychology in 2007. So reaching for an afternoon snack or thinking about important decisions after a good night's sleep are more than just routine habits. They may actually improve decision-making processes and outcomes.
RANDOMNESS AND PATTERNS
People tend to be poor intuitive statisticians and are not very adept at distinguishing patterns from random events, according to Kahneman. Specifically, we tend to misunderstand what "randomness" actually looks like, and we tend to see patterns in data when none actually exist.
Truly random data frequently look like clusters instead of symmetrical distributions, at least until sample sizes get very large. Conversely, large sets of random numbers frequently behave in foreseeable ways where patterns of data distributions are predictable with relative accuracy.
CPAs can use awareness of this concept to detect fraud. Fraudsters frequently don't understand randomness, which can lead to the detection of attempts to cover their tracks. For example, a popular anti-fraud textbook, Fraud Examination, states that, "[a]ccording to Benford's Law, the first digit of random data sets will begin with a 1 more often than with a 2, a 2 more often than with a 3, and so on. ... People usually assume that the digits 1—9 have an equal chance of appearing in the first position, but reality is very different."
When fraud perpetrators attempt to cover their tracks by creating fictitious accounting information, they often attempt to fabricate entries or transactions that start with the digits 1—9 in equal proportions. Forensic accountants can use Benford's Law to search for numerical patterns that do not follow the statistical regularities of nature.
REGRESSION TO THE MEAN
Within any given population of data, the population average represents the most likely value we should expect for an observation drawn from the group, assuming we know nothing else about the observation in question.
This regularity holds for data related to human and business performance as well. Kahneman argues that success is a function of both luck and talent combined, and that extreme examples of success (or failure) are likely to regress to the average in subsequent periods as the portion attributable to luck is unlikely to persist.
Put differently, below-average performance (as far as it is attributable to luck) is likely to reverse and get better, while above-average performance (as far as it is attributable to luck) is likely to reverse and get worse.
In an accounting context, the tendency of performance-to-mean regression suggests that common performance metrics such as variance analyses or flexible budget comparisons should be used with caution when evaluating a manager's recent performance. Managers' and employees' performance (both good and bad) is probably somewhat attributable to luck and somewhat attributable to enduring talent and should probably be evaluated over a long time horizon. Thus, evaluating a body of work or an average of outcomes paints a more accurate picture of overall performance.
How to use psychology to your advantage
Understanding certain psychological concepts can help CPAs maintain professional skepticism, make better decisions, and improve their performance. Here are some tips on how to put psychology to work:
Consider your motivation. Be aware of the potential to seek and emphasize information that supports your initial ideas or your client's position. Taking care to search for and evaluate all relevant information can provide the objectivity necessary to avoid invalid conclusions that can be brought about by a form of motivated reasoning known as directional goals.
Don't rely on first impressions. According to the "halo effect," we rely so heavily on our first impressions of people that we ignore later information about them. Maintaining an awareness of this tendency and developing systems to eliminate bias can keep the halo effect from causing problems.
Replenish your resources. Your capacity for exerting mental effort may wane as the day wears on. Getting enough rest and eating properly—and taking a snack break to raise your glucose levels—may prevent the effects of "ego depletion," a sapping of resources that can lead to reductions in objectivity.
Get educated on randomness and patterns. CPAs can use knowledge of randomness and patterns to spot fraud. Fraudsters often don't understand how random patterns occur in nature, and understanding these patterns can help accountants discover cases of data falsification.
Evaluate on a long time horizon. When evaluating people and results, keep in mind that below-average performance is likely to get better, while above-average performance is likely to get worse, particularly when only small sample sizes have been considered. Understanding this concept of "regression to the mean" can help CPAs see the wisdom in evaluating people and results over a long time horizon.
About the author
John Lauck (firstname.lastname@example.org) is an assistant professor of accounting at Louisiana Tech University in Ruston, La.
To comment on this article or to suggest an idea for another article, contact Ken Tysiac, editorial director, at email@example.com or 919-402-2112.
- "Highlights of Fraud Research," Nov. 2015, page 40
- "I'm Not Biased, Am I?" Feb. 2015, page 26
- "Highlights of Ethics Research," June 2014, page 34
- "Highlights of Tax Research," July 2013, page 56
- Ethics: Nonattest Service, Integrity, and Objectivity (#159419, one-year online access)
- Professional Ethics: 2016 Update and Refresher (#159437, one-year online access)
- Professional Ethics: The AICPA's Comprehensive Course (#732319, text; #155902, one-year online access (for licensure); #155702, one-year online access)
- Professional Skepticism and Maintaining the Public Trust—Audit Staff Essentials, Level I (#150024, one-year online access)
For more information or to make a purchase, go to cpa2biz.com or call the Institute at 888-777-7077.
AICPA Enhancing Audit Quality webpage, aicpa.org/EAQ