Boards of Directors' Sanction Judgments: The Effect of Situational Factors.
Governance theory provides foundations for corporate oversight by various groups, with the board of directors a key component. While aggressive monitoring by the audit committee tends to decrease the need for disciplinary action by the board, complexity and lack of transparency increase the need for such actions (Ndofor et al., 2015). To increase management accountability, the NYSE (2014: 75) requires all main board committees (audit, human resources and compensation, nominating and governance) be comprised of independent members. This appears to have led to increases in monitoring of CEOs and disciplinary actions imposed on management (Guo and Masulis, 2015). However, directors are often sympathetic to management (Cohen et al., 2012). Research suggests empathy can play an important role in determining board members' actions (Huse, 1998); thus, empathy can further impact judgment quality.
This paper reports results of an experiment that examined the effect of situational factors (history of similar incidents, compensation structure, and reason for violation) on board of director members' judgments regarding sanctions for a violation of regulations. Drawn from punishment guidelines provided by FINRA (2017), history of similar incidents captured whether the individual was a first-time or repeat offender; compensation structure varied whether the offender received financial benefit from the crime. Reason for violation was included based on literature suggesting empathy plays a role in judgments (Huse, 1998) and varied whether the reason for the violation was based on altruistic or opportunistic motivation. Members of boards of directors made sanction recommendations after reading a short scenario that varied these factors, based on experimental condition. Predictions were that board of directors' members would recommend more severe sanctions when there was a history of suspected (versus no suspicion of) previous violations, an incentive-based (versus fixed) compensation structure, and an opportunistic (versus altruistic) reason for the violation.
This study makes an important contribution to the literature and to practice. To the authors' knowledge, prior studies have not considered sanction judgments from the board of director's perspective. A firm's board is a critical component of internal control and governance. Directors are expected to fulfill fiduciary responsibilities to shareholders and provide effective organizational oversight. A strong functioning board can be viewed as one composed of "mediating hierarchs," where directors are charged with balancing the needs and desires of multiple stakeholders to achieve optimal outcomes with respect to the firm's assets and outputs (Blair and Stout, 2001). Their judgments and actions help set an appropriate tone at the top and define the organization's expectations relating to integrity and ethical conduct (COSO, 2004; 2012). However, directors are susceptible to common decision-making biases and influences, which can affect judgment quality (COSO, 2012). Therefore, directors' responses to situational factors are important to understand in order to set appropriate expectations for board members.
RELEVANT LITERATURE AND HYPOTHESES DEVELOPMENT
Evolutionary Governance Theory: Board of Directors
Governance theory establishes premises for oversight and control of organizations. Further, corporate governance is charged with meeting external regulatory and accountability expectations (Rhodes, 2000). External governance often begins with an outside regulator creating rules and auditing to check compliance (Bratton, 2007), then adjudicating violations with sanctions at the individual, organization, or executive board level (FINRA 2017). Similarly, internal governance typically begins with the board of directors establishing controls and auditors ensuring ongoing compliance (Gillan, 2006). Violations, depending on the source and type, may be handled by an external adjudicator or internally by the board. In the trading community, despite recurring news of board actions in response to controls violations (Cheng, 2017; Scheer and Griffin, 2016), there is a lack of research investigating board actions. Thus, this research is important to expand the understanding of governance mechanisms and the role of directors.
As auditors discover findings or the business environment changes, the external and internal governance model evolves (Van Assche et al., 2013). This process of evolutionary governance is further informed by punishment theory. As punishments are meted out, their severity and celerity establish grounds for deterrence (Apel, 2013). Boards and external regulators then adapt and modify controls and sanctions, and the governance model continues to evolve (Beunen et al., 2015). Thus, the decision-making of boards with respect to designing controls and sanctions provides an important bridge between governance and punishment, as well as the evolution of both concepts. This process and the relations are illustrated in Figure I.
Punishment Theory: Deterrence and Retribution
There are two prevalent justifications in punishment theory--deterrence and retribution. Threat of punishment is generally seen as a deterrent, while implementation of punishment is considered retribution (Byrd, 1989). Punishment theory provides insight as to how punishment discourages future crimes through the lens of perception in terms of perceived probability of detection and enforcement penalties (Apel, 2013). Threat of punishment--and retribution mechanisms within it--induces a modicum of deterrence, but implementation of punishment is key in setting expectations (Skatova and Ferguson, 2013). Further, even when criminals make restoration, peoples' sense of justice demands the possibility of retribution as well (Gromet and Darley, 2006). From a criminal's perspective, likelihood of detection correlates inversely with the net marginal utility of benefits associated with the act (Becker, 2013). Marginal costs rise as the perceived probability of detection increases, which reduces the likelihood of the individual committing the criminal act. This seems to be true regardless of punishment seventy, possibly because as an individual contemplates committing a criminal act, his or her internal belief that the behavior is wrong carries more weight than any consideration of the severity of a sanction (Akers, 2013).
Despite mixed findings regarding the effectiveness of varying punishment type and size (Gachter et al., 2008), research has suggested punishment was beneficial even when individual cost was high and the act was committed for altruistic reasons (Fehr and Gachter, 2002). While accounting regulatory bodies tend to focus on deterrence (e.g., COSO, 2016; SEC, 2016), retribution often accompanies guidelines for punishment. Within the financial marketplace, this often includes monetary fines; i.e., the SEC has a history of preferring such penalties as a part of its enforcement process (Correia, 2014). Similarly, FINRA publishes sanction guidelines for disciplinary decisions as part of its regulatory services to equities and options exchanges such as NYSE, NASDAQ, and the CBOE. FINRA's (2017) guidelines specifically call for consistency with an emphasis on deterrence. However, the guidelines indicate that recommended sanctions are not fixed, allowing for flexibility and judgment in assigning penalties. For instance, FINRA (2017: 2) General Principle 2 states "disciplinary sanctions should be more severe for recidivists." First-time offenders may commit a violation naively or unintentionally, while repeat offenders more likely act intentionally. Thus, repeat offenders require increased penalties for punishments to work correctly (Mungan, 2010).
Indeed, punishment theory suggests escalating penalties for previous offenders has a deterrent effect (Miceli and Bucci, 2005). As a result, adjudicators tend to treat repeat offenders more harshly than first-time offenders (Miceli, 2013). Likewise, FINRA's (2017) guidelines call for leniency for first offenses. Further, research has indicated that the mere suggestion of past criminality influences punishment decision-making, irrelevant of actual history of violations or convictions (Holzer et al., 2006). Therefore, when case information suggested suspicion of past criminal behavior, this was predicted to elicit a more negative reaction from adjudicators and stronger sanction recommendations. Conversely, when a violator was presented as a first-time offender, this was expected to yield more leniency. Thus, the first hypothesis was:
H1: Directors recommend more severe sanctions when there is a history of suspected violations than when there is no suspicion of previous violations.
In addition to a history of violations, FINRA (2017) advises stronger sanctions when violators benefit financially from the criminal act. General Principle 6 states "to remediate misconduct, adjudicators should consider a respondent's ill-gotten gain when determining an appropriate remedy" (FINRA 2017: 5). The guidelines further express that restitution orders are not limited by the amount of the ill-gotten gain (FINRA, 2017: 4). This guidance acknowledges that society has responded more negatively when an individual, particularly an agent acting on another's behalf (Tett, 2009), directly profited from committing a criminal financial act (Dyck et al., 2010). Therefore, when an agent profited from a violation due to increased compensation, it was expected that the recommended punishment would be greater than if the agent had no personal profit from the act. Formally, the second hypothesis was:
H2: Directors recommend more severe sanctions when there is an incentive-based compensation structure than when there is a fixed compensation structure.
Judgment and Empathy
Punishment is often driven by a desire for retribution, even when such retribution may not be logical or conventionally moral (Aharoni and Fridlund, 2012). Therefore, compassion and fairness are crucial in reaching a punishment decision (Singer and Steinbeis, 2009; Dong, 2014). While compassion can lead to perceptions of preferential treatment, the judgment process is driven by an underlying concern for justice (Blader and Rothman, 2014). When individuals encounter a situation requiring judgment, they generally strive for a virtuous decision--a process that involves empathie reasoning (Jensen, 2010). Research has suggested empathetic reasoning generates some level of distress as the individual imagines how it feels to be the person in the situation; this process creates empathy, which then creates distress, in a circular fashion (Hoffman, 1993). As a result, empathy may create unconscious bias in the decision-making process (Gibbs, 2013) and impact judgments.
Sanctions may reflect a measure of conformity to established legal guidelines (Kim et al., 2016). However, research has recommended penalties should fit the circumstances of the crime, with whatever discretion is allowed (Andreoni, 1991; Carlsmith et al., 2002). FINRA (2017: 3-4) guidelines acknowledge this in General Principle 3, which states "adjudicators should tailor sanctions to respond to the misconduct at issue." The guidelines are clear that sanctions should vary with the severity of the circumstances, above or below the recommendations given, and that "adjudicators must always exercise judgment and discretion and consider appropriate aggravating and mitigating factors in determining remedial sanctions in each case" (FINRA, 2017: 4). Mitigating circumstances (Chin, 2012) and situational context (Ugazio et al., 2014) can influence empathy. As such, consideration of how empathy factors in enhances the understanding of punishment judgments. Based on this, the last hypothesis considered whether recommended punishments varied when presented with a situation that was likely to invoke a stronger empathetic reaction (when an act was committed for an altruistic reason) than a less empathetic situation (when the act was based on an opportunistic reason). Thus, the third hypothesis was:
H3: Directors recommend more severe sanctions when there is an opportunistic reason for the violation than when there is an altruistic reason for the violation.
Auditors within governmental (SEC, 2018) and nongovernmental (FINRA, 2018) regulators conduct field examinations, collect and investigate evidence, and suggest penalties (where appropriate). When a firm disagrees with a regulatory auditor's findings, an arbitrator adjudicates the process and determines the outcome. Much of this adjudication process can focus on the firm and its quality of internal controls. Since internal controls quality is a function of the control environment, this encompasses the board of directors (Krishnan, 2005). Further, COSO (2009) and the SEC (2014) have increased boards' responsibilities for risk oversight. With this, boards are allowed to set controls and penalization pathways, consistent with the environment and scenarios investigated in this experiment. Thus, examining the responses of board directors allows for a deeper understanding of how mitigatory circumstances influence those in charge of overseeing organizations' control environments and provides additional insight to the evolutionary governance model.
Participants were 76 individuals who have served on a board of directors. (2) Forty-seven participants were male and 29 were female. The average (median) age was 54 (56) years, ranging from 28 to 80 years. (3) Seventy-one participants had completed four or more years of college; eight participants were Certified Public Accountants (CPA). Fifty-six participants had twenty or more years of experience in their respective industry; nine had between ten and twenty years of experience; and ten had between three and ten years of experience. One participant had between one and three years of experience in her industry. This experience translates into opportunities to design controls and become familiar with rules violation proceedings. Indeed, thirty-eight participants had been associated with rules violation examinations, audits, or arbitration proceedings with roles as internal audit (one), external audit (one), arbitrator (four), legal counsel (fourteen), board of directors (eleven), or in other capacities (seven).
The study was approved by the Institutional Review Board at the university where the study was conducted. Participants were recruited through an on-line platform where board members agreed to take part in studies for a fee. An intermediary matched these participants with researchers, offering guarantees to both parties. (4) Participants were pre-qualified, rated, and tracked by the intermediary. Measures and filters were built into the survey instrument to ensure participants paid attention and spent reasonable time considering their responses.
The instrument provided a background scenario and asked participants to make judgments regarding possible sanctions. All participants received the same background information indicating that a brokerage firm's loyal client of 15 years called the firm to sell 100,000 shares of a fund just as the trading deadline passed. Taylor, the client's longtime broker and a 20-year firm employee, took the call and agreed to enter the order. Taylor recorded the order as having been placed just before the deadline, thus violating trading rules and the firm's code of conduct. This adversely affects the net asset value of one share by a fraction of a penny for all shareholders. In addition, participants received information regarding the reason the client needed the money immediately, as well as Taylor's compensation structure and history of similar incidents. These details varied depending on the experimental condition that participants were randomly assigned to (see the Independent Variables section).
All participants were also told Taylor was caught and a field examiner collected the evidence, and the firm managing the fund estimated net assets were adversely impacted by 190,000. Participants were asked to recommend sanctions, with the following guidelines given: (5)
The Financial Industry Regulatory Authority (FINRA) Sanctions Guidelines offers suggestions designed to deter future misconduct and improve overall business standards. Sanctions should be sufficient to achieve deterrence without being punitive. It was suggested that National Association of Securities Dealers (NASD) Rule 2110, which governs fair trading, and FINRA Rule 2020, which covers deceptive practices would be the rules that may be applicable. In general, 2110 suggests fines of $5,000 to $10,000 for the first action and $10,000 to $100,000 for subsequent actions. 2020 suggests fines of $2,500 to $50,000 for negligence and $10,000 to $100,000 for reckless misconduct. You have a range of non-monetary penalties available. In egregious cases, the guidelines allow for barring the individual and/or expelling the firm (2020). The penalties (both monetary and non-monetary) recommended should be tailored to the situation and, thus, can fall outside the recommended ranges.
There were three independent variables that varied situational factors (history of similar incidents, compensation structure, and reason for violation). History of similar incidents stated either that Taylor had a clean record with no prior suspicion of any wrongdoing, or that Taylor had a clean record but there was suspicion of several similar incidents in the past with other clients (though none were proven). The compensation structure for Taylor was either an $80,000 annual salary with no commissions (fixed compensation structure) or commission-based, at $0.80 per share for this fund, with no salary (incentive-based compensation structure). The reason for the violation was either the client needed money immediately for a medical procedure that the insurance company denied in payment pre-check (altruistic reason) or the client needed money immediately to buy an ultra-high-end vacation home in a live foreclosure auction (opportunistic reason). The 2x2x2 design resulted in eight experimental conditions, which are hereafter referenced by a three-letter Condition ID as follows: N=No prior incidents, P=Prior incidents suspected, S=Salary, C=Commissions, M=Medical procedure, and V-Vacation home. For example, NSM refers to the condition with no prior similar incidents, a salary compensation structure, and a medical procedure reason for violation. PCV refers to the condition with suspicion of prior similar incidents, a commission compensation structure, and a vacation home reason for violation.
The dependent variables measured two aspects of sanction severity--type of sanction and amount of monetary sanction. Based on FINRA (2017) guidelines for possible punishments, the instrument offered eight types of sanctions and asked participants to indicate the probability (zero to 100 percent) they would recommend each, without consideration of the other sanctions. The types of sanctions, ranked by the researchers' subjective assessment of increasing severity, were: (1) no sanction, (2) censure, (3) warning, (4) fine, (5) license suspension, (6) fire from job, (7) permanent bar, and (8) jail time. (The sanctions were not presented to participants in this order.) This offered a range of possible sanctions with varying levels of severity. Four (no sanction, censure, warning, fire from job) of the eight sanctions available in FINRA's guidelines align with internal penalties available to boards. The other four (fine, license suspension, permanent bar, jail time) represent external pathways board members are familiar with due to their interactions with regulators and the frequency with which these punishments are administered in practice. In addition, if participants indicated they recommended a fine, they were asked to specify the recommended fine amount and how much of the fine was due to unfair trading, deceptive practices, or something else (on a scale of zero to 100 percent).
Participants provided demographic data regarding age, gender, education, and CPA licensure. Participants also indicated whether they had been part of any examinations, audits, or arbitration proceedings (and their role, if so). Age, gender, education, and CPA licensure were not significant covariates for any of the dependent variables. (6) However, whether respondents had been part of examinations, audits, or arbitration proceedings (hereafter, "Experience") was a significant covariate (p=0.033) for the recommendation of fine. (7) As such, Experience was included in the multivariate analysis as an indicator variable with a value of 1 (-1) if the individual had (had not) been part of any examinations, audits, or arbitrations.
Participants also responded to items measuring Machiavellianism, adapted as a five-item scale from the Machiavellianism Personality Scale (Dahling et al., 2009). Many boards are run by individuals with dominant personalities (Ho and Wong, 2001), as well as former and current CEOs (Hillman et al., 2008) who tend to have Type-A personalities (Ahmad, 2010). Type-A personalities often exhibit higher levels of Machiavellianism (Thornton et al., 2011). Thus, Machiavellianism was measured as a control variable. Similarly, research has suggested high levels of empathy are present in various types of strong leaders (Bolton et al., 2010). Empathy affects legal decision-making and the severity of recommended punishment (Tsoudis, 2002). Therefore, empathy was measured as a control as well. A scale measuring the capacity one has to feel empathy (Jolliffe and Farrington, 2006) was adapted for this study.
Participants indicated the extent of their agreement with each empathy and Machiavellianism statement item on a scale of zero to 100 percent. Table 1 lists the statement items with means. For the empathy statements, items 4 and 7 were significant covariates for the recommendation of censure; item 5 was significant for the recommendation of fine; and item 4 was significant for the recommended amount o/fine. For the Machiavellianism statements, items 4 and 5 were significant for the recommendation of license suspension; item 5 was significant for the recommendation of jail time; and items 2, 4, and 5 were significant for the recommended amount of fine. Therefore, the average of the significant empathy items (4, 5, and 7) and the average of the significant Machiavellianism items (2, 4, and 5) for each participant was used to create the control variables "Empathy" and "Machiavellianism," respectively, for the multivariate and univariate analyses. (8) Table 1 also presents the means for Empathy and Machiavellianism variables.
Recommended Types of Sanctions
Figure II Panel A illustrates the mean likelihood that participants recommended each type of sanction, with each line graphing history of similar incidents (no prior incidents or prior incidents suspected), compensation structure (salary or commissions), or a reason (medical procedure or vacation home). The graphs follow a similar path for each independent variable for recommendations of no sanction, censure, license suspension, fire from job, permanent bar, and jail time. However, the paths diverge at warning and fine. Specifically, participants more likely recommended warning for the more altruistic reason (medical procedure) or when Taylor had no history of similar incidents or less incentive (salary compensation). Conversely, participants more likely recommended fine for the more opportunistic reason (vacation home) or when Taylor had prior incidents suspected or more incentive (commissions-based compensation). The same pattern for warning and fine recommendations emerges in the combined "extreme" experimental conditions (NSM and PCV). Figure II Panel B illustrates the mean likelihood that participants recommended each type of sanction in the NSM and PCV conditions.
Mean Comparisons. Hl, H2, and H3 state that directors recommend more severe sanctions when there is a history of suspected violations (versus no suspicion of previous violations), an incentive-based (versus fixed) compensation structure, and an opportunistic (versus altruistic) reason for the violation, respectively. As previously noted, types of sanctions were ranked by the researchers' subjective assessment of increasing severity as: (1) no sanction, (2) censure, (3) warning, (4) fine, (5) license suspension, (6) fire from job, (7) permanent bar, and (8) jail time. Further, the researchers placed the binary cutoff of more/less severity between (3) warning and (4) fine; thus, no sanction, censure, and warning were less severe, while fine, license suspension, fire from job, permanent bar, and jail time were more severe.
Table 2 Panel A presents the means corresponding to the Figure II Panel A graph, along with results of t-tests comparing means within each independent variable. Mean comparisons provided support for H1. Participants recommended more severe types of sanctions (fine p=0.031, license suspension p=0.017, fire from job p=0.021, permanent bar p=0.015, and jail time p=0.015) if there was suspicion of prior incidents and less severe (warning p=0.014) if there were no prior incidents. There was only partial support for H2 relating to compensation structure. Participants recommended more severe types of sanctions (license suspension p=0.052, fire from job p=0.004, and jail time p=0.029) with an incentive-based compensation structure and less severe (no sanction p=0.033) with a fixed compensation structure. However, there was no difference in recommendations for censure, warning, fine, and permanent bar. Mean comparisons provided little support for H3, as participants were moderately more likely to recommend less severe (no sanction p=0.060 and warning p=0.053) if the reason was altruistic and more severe (fire from job p=0.092) if the reason was opportunistic.
In addition, Table 2 Panel B presents the means by condition, including those corresponding to the Figure II Panel B graph of the "extreme" conditions (NSM and PCV), along with results of t-tests comparing the NSM and PCV means. Participants recommended more severe types of sanctions (fine p=0.045, license suspension p=0.053, and fire from job p=0.004) in PCV, and less severe types of sanctions (no sanction p=0.089 and warning p=0.008) in NSM. (9)
Multivariate Analysis of Variance with Covariates (MANCOVA). Table 2 Panel C presents results of a MANCOVA with the eight types of sanction recommendations as dependent variables; the three independent variables and their interactions; and the previously discussed covariates Experience, Empathy, and Machiavellianism. The MANCOVA results were consistent with findings from the mean comparisons. There was strong support for HI that directors recommend more severe sanctions when there is a history of suspected violations (versus no suspicion of previous violations). History of similar incidents was significant for warning (p=0.041), fine (p=0.027), license suspension (p=0.049), fire from job (p = 0.038), permanent bar (p=0.027), and jail time (p=0.038); and moderately significant for recommendation of no sanction (p=0.089).
Compensation structure was significant for recommendations of no sanction (p=0.028) and fire from job (p = 0.013) and moderately significant for recommendations for license suspension (p=0.059) and jail time (p=0.065), providing partial support for H2 that directors recommend more severe sanctions when there is an incentive-based (versus fixed) compensation structure. Reason for violation was not significant for any type of sanction recommendations; thus, there was no support for H3 that directors recommend more severe sanctions when there is an opportunistic (versus altruistic) reason for the violation. There was essentially no support for meaningful interactions of independent variables, with only the interaction of history of similar incidents and compensation structure having moderate significance for fine recommendations (p=0.051).
Taken together, the mean comparisons and MANCOVA results suggest history of similar incidents is an important factor that influences directors' recommendations in terms of types of sanctions, with fine recommended if there is suspicion of prior incidents and warning recommended if there are no prior incidents suspected. However, participants indicated the probability they recommend each sanction independent of the other sanctions; as such, they could recommend both warning and fine. Therefore, it was also necessary to consider differences in individuals' likelihoods of recommending warning versus fine.
Recommendations of Warning versus Fine. Table 3 Panel A presents the frequencies of when participants' recommendation of warning exceeded, equaled, and was less than recommendation of warning by independent variable. Recommendations of warning were more frequent than recommendations of fine when there was no history of prior incidents (N = 21, versus N = 15 when there was suspicion of prior incidents), there was a salary compensation structure (N = 20, versus N = 16 for commissions), and the reason for violation was medical procedure (N = 24, versus N = 12 for vacation home). Conversely, recommendations of fine were more frequent than recommendations of warning when there was suspicion of prior incidents (N = 21, versus N = 11 when there was no history of prior incidents), there was a commissions compensation structure (N = 22, versus N = 10 for salary), and the reason for violation was vacation home (N = 21, versus N = 11 for medical procedure). Table 3 Panel A also includes the "emphasis on warning versus fine," computed at the participant level by taking likelihood of recommendation for warning minus likelihood of recommendation for fine; thus, positive (negative) indicates greater mean likelihood of recommendation for warning (fine). There was an overall emphasis on warning for no prior incidents (mean = 14.44%), salary (mean=4.94%), and medical procedure (mean = 5.53%); and an overall emphasis on fine for suspicion of prior incidents (mean = -22.43%), commissions (mean = -13.84%), and vacation home (mean = -18.67%). These differences in emphasis were significant for history of similar incidents (p=0.005) and reason for violation (p = 0.047).
Table 3 Panel B presents the means by condition, along with results of t-tests comparing the extreme conditions, NSM and PCV. There was an overall emphasis on warning in NSM (mean = 28.55%) and on fine in PCV (mean = -40.93%); this difference was significant (p = 0.004). Table 3 Panel C reports results of analysis of variance with covariates (ANCOVA) for emphasis on warning versus fine, which supported HI that history of similar incidents is significant (p = 0.012). Taken together, the warning versus fine analyses confirm that history of similar incidents is an important determinant of directors' recommendations of warning versus fine, with a stronger emphasis on fine if there is suspicion of prior incidents and on warning if there are no prior incidents suspected.
Recommended Fine Amount
If participants recommended a fine, they also were asked to suggest a fine amount. Analyses of recommended fine amount were filtered for those participants who indicated a greater than 20 percent likelihood of recommending a fine and a fine amount greater than $1. In terms of amount of monetary sanction, mean comparisons did not support H2 and H3, that directors recommend more severe sanctions when there is an incentive-based (versus fixed) compensation structure and opportunistic (versus altruistic) reason for the violation, respectively. Table 4 Panel A presents median and mean recommended fine amounts, as well as mean comparison tests by independent variable that show differences for compensation structure and reason for violation were non-significant (p = 0.832 and p = 0.859, respectively). However, results again supported H1, that directors recommend more severe sanctions when there is a history of suspected violations versus no suspicion of previous violations. The mean recommended fine amount of $9,786 was significantly lower if there were no prior incidents than the mean recommendation of $21,324 if there were prior incidents suspected (p = 0.026).
Table 4 Panel B reports median and mean recommended fine amounts by condition. In NSM, the mean amount was $4,571, which was significantly lower than the mean amount $18,269 in PCV (p = 0.039). Table 4 Panel C presents ANCOVA results, showing history of similar incidents was moderately significant (p=0.080) for recommended fine amount. Taken together, these results show some support for HI, but not H2 or H3, in terms of severity of sanctions using monetary amount of fine.
The primary contribution of this research comes from a bridging of evolutionary governance theory and punishment theory to investigate how boards of directors see deterrence and retribution within their organizations. This is important, as there is a lack of previous research regarding the sanctions decisions of board members. Governance continues to evolve, in part, based on the punishments received in the marketplace. Thus, understanding the decisions of board members from the perspective of punishment helps complete a missing but nantral link between governance and punishment. This is vital because research suggests boards continually adapt and modify control processes based on changing business environments. While much of the literature focuses on the operating environment, there is a paucity of knowledge of how board members decide to punish and how internal and external punishment decisions and sanctions guidance might inform their decisions.
These findings increase understanding of how suspicion of earlier wrongdoing alone is enough to increase penalization by boards. Further, the nuances of the punishment choices available combine with these findings to extend retribution and deterrence theory. Results of this study suggest board members focused on the offender's history of violations (repeat or first-time offender) and were not influenced by whether the individual received financial benefit (in terms of compensation structure) or by empathetic situations (altruistic versus opportunistic motivation for the act) when making sanction recommendations. The findings hold for both sanction type and amount of monetary sanction. History of similar incidents influenced recommendations in terms of sanction type, with a stronger emphasis on warning if there were no prior incidents suspected and on fine (and a greater fine amount) if there was suspicion of prior incidents. Thus, it appears board members place greater weight on whether the situation involves a first-time offense. However, consistent with prior research (Holzer et al, 2006), the mere suggestion of previous violations influenced participants' judgments, without concrete proof of prior offenses. The findings suggest, on average, directors will hold firm in sanction judgments and will not be swayed by emotion. While the punishment recommendations do not appear to reflect all the discretion allowed in punishment guidelines (e.g., FINRA, 2017), this position sets a stronger tone at the top, which is an important role of the board (COSO, 2004, 2012).
However, it is interesting to note that most participants' recommendations for a fine amount did not reach the firm's $90,000 estimate of the adverse impact on net assets. Only three participants recommended a fine amount of $90,000 or higher--one in NCM recommended $90,000; one each in CS M and CSV recommended $100,000. Comparison of the recommended fine amounts with the NASD Rule 2110 and FINRA Rule 2020 sanction guidelines that were provided to participants suggests participants focused on the fair-trading aspect when recommending fine amounts. Participants were told NASD Rule 2110 governs fair trading (recommendations of $5,000 to $10,000 for the first action and $10,000 to $100,000 for subsequent actions) and FINRA Rule 2020 covers deceptive practices (recommendations of $2,500 to $50,000 for negligence and $10,000 to $100,000 for reckless misconduct). As Table 4 Panel A shows, the median amount of $5,000 if there was no history of prior incidents equals the minimum recommendation for a first action under NASD Rule 2110, while the median amount of $10,000 if prior incidents were suspected equals the minimum recommendation for subsequent actions. In fact, this is consistent with participants' responses when asked what percentage of their fine was due to unfair trading (mean response was 60 percent) and to deceptive practices (mean response was 42 percent). Therefore, while the results of this study suggest a strong tone at the top set by the board in terms of recommending punishment action, the amount of monetary penalty assessed may not send as strong of a message.
The authors acknowledge this study is subject to limitations. Participants were recruited through an intermediary and the study was conducted in an uncontrolled setting, with participants completing the experiment remotely online. However, the intermediary's vetting process and guarantees, as well as the attention-checking and time measures built into the experiment, give confidence that relevant human subjects participated with the appropriate attention given to the task. Also, while individual simulated decision-making does not capture the depth and richness of the decision-making of a live board discussion, it does provide insight into the starting points for such a discussion. In addition, the authors acknowledge that the experimental setting is that of a brokerage firm, not a more general organization. However, there is little reason to believe that board members' judgments would differ for other types of organizational settings, given that boards across industries are regularly faced with choices to do nothing, censure, warn, or fire executives. Additionally, external regulators of many industries impose similar types (fine, suspend, bar from industry, jail time) of penalties to those examined in this study. Further, board independence requirements (NYSE 2014) appear to have resulted in many board members serving in a similar capacity on multiple boards (Larcker et al., 2016), even when such service may be detrimental (Cashman et al., 2012). Therefore, their judgments likely carry over and have similar effects on many organizations (Chiu et al., 2013).
For boards reviewing or implementing punishment guidance within their organizations, this study offers three central concepts for reflection. First, because the mere suggestion of previous violations was enough to influence participants' recommendations, boards should consider whether others in management and enforcement capacities might have similar reactions. Boards should be clear about the amount of discretion they allow key decision-makers in implementing punishment decisions. Second, boards should identify several alternative remedies for situations. In the scenario explored in this study, warning and fine were a primary punishment mechanism. However, participants' responses suggested desire for a range of available sanctions. Providing various alternatives may help responsible decision-makers more closely align punishments with guidance from the board. Lastly, because results suggested giving a range for monetary penalties was ineffective, boards should consider whether greater discretion in terms of monetary penalization might better serve deterrence and retribution, as opposed to setting ranges on which individuals might anchor.
Considerations for Future Research
Future research should further explore directors' monetary sanction recommendations and whether they set an effective tone at the top. Also, since boards often make sanction recommendations, but other parties in enforcement or arbitration roles often implement the penalties, future research should examine whether there is congruence between the recommendations and implementations. In addition, prior literature has expressed dissatisfaction with penalties imposed on wrongdoers (Verschoor, 2014; Wall and Fogarty, 2016). Future research should also consider the perspective of investors and whether board members' judgments are perceived as fulfilling a fiduciary role and representing shareholder interests. Lastly, while FINRA (2017) has updated Rule 2020 since the time of the study to reflect increased monetary penalization recommendations on the top end of the range (up to $73,000 for negligence and $146,000 for reckless misconduct), this study's findings suggest the top ends may be irrelevant. Future research should evaluate whether removing monetary guidelines entirely to avoid any anchoring effects (COSO, 2012) might better serve justice and achieve deterrence. Considering these factors in aggregate would constitute a novel line of literature and form the basis for a conceptual model of punishment within organizations and the related internal and external perceptions.
Aharoni, E., and A. J. Fridlund. 2012. "Punishment without Reason: Isolating Retribution in Lay Punishment of Criminal Offenders." Psychology, Public Policy, and Law 18(4): 599-625.
Ahmad, H. M. 2010. "Personality Traits among Entrepreneurial and Professional CEOs in SMEs." International Journal of Business and Management 5(9): 203-213.
Akers, R. L. 2013. Criminological Theories: Introduction and Evaluation, 2nd Edition. Hoboken, NJ: Taylor and Francis.
Andreoni, J. 1991. "Reasonable Doubt and the Optimal Magnitude of Fines: Should the Penalty Fit the Crime?" The RAND Journal of Economics 22(3): 385-395.
Apel, R. 2013. "Sanctions, Perceptions, and Crime: Implications for Criminal Deterrence." Journal of Quantitative Criminology 29(1): 67-101.
Becker, G. S. 2013. The Economic Approach to Human Behavior. Chicago, IL: University of Chicago Press.
Beunen, R., K. Van Assche, and M. Duineveld. 2015. Evolutionary Governance Theory. New York, NY: Springer Publishing.
Blader, S. L., and N. B. Rothman. 2014. "Paving the Road to Preferential Treatment with Good Intentions: Empathy, Accountability and Fairness." Journal of Experimental Social Psychology 50: 65-81.
Blair, M. M., and L. A. Stout. 2001. "Director Accountability and the Mediating Role of the Corporate Board." Washington University Law Quarterly 79(2): 403-447.
Bolton, P., M. K. Brunnermeier, and L. Veldkamp. 2010. "Economists' Perspectives on Leadership." In Handbook of Leadership Theory and Practice. Ed. N. Noria and R. Khurana. Boston, MA: Harvard Business Press.
Bratton, W. W. 2007. "Private Standards, Public Governance: A New Look at the Financial Accounting Standards Board." Boston College Law Review 48(1): 5-53.
Byrd, B. S. 1989. "Kant's Theory of Punishment: Deterrence in Its Threat, Retribution in Its Execution." Law and Philosophy 8(2): 151-200.
Carlsmith, K. M., J. M. Darley, and P. H. Robinson. 2002. "Why Do We Punish? Deterrence and Just Deserts as Motives for Punishment." Journal of Personality and Social Psychology 83(2): 284-299.
Cashman, g' D., S. L. Gillan, and C. Jun. 2012. "Going Overboard? On Busy Directors and Firm Value." Journal of Banking and Finance 36: 3248-3259.
Cheng, E. 2017. "Wells Fargo Reportedly Fires Foreign Exchange Execs Amid Investigations." CNBC. October 20, 2017. Retrieved from https://www.cnbc.com/2017/10/20/wells-fargo-fires-fx-trading-execu tives-amid-report-of-probe.html.
Chin, D. 2012. "Sentencing: A Role for Empathy." University of Pennsylvania Law Review 160(6): 1561-1584.
Chiu, P., S. H. Teoh, and F. Tian. 2013. "Board Interlocks and Earnings Management Contagion." The Accounting Review 88(3): 915-944.
Cohen, L., A. Frazzini, and C. J. Malloy. 2012. "Hiring Cheerleaders: Board Appointments of "Independent" Directors." Management Science 58(6): 1039-1058.
Committee of Sponsoring Organizations of the Treadway Commission (COSO). 2016. Fraud Risk Management Guide. New York, NY: COSO.
Committee of Sponsoring Organizations of the Treadway Commission (COSO). 2012. Enhancing Board Oversight--Avoiding Judgment Traps and Biases. New York NYCOSO.
Committee of Sponsoring Organizations of the Treadway Commission (COSO). 2009. Effective Enterprise Risk Oversight: The Role of the Board of Directors New York NYCOSO.
Committee of Sponsoring Organizations of the Treadway Commission (COSO). 2004. Enterprise Risk Management--Integrated Framework. New York, NY: AICPA.
Correia, M. M. 2014. "Political Connections and SEC Enforcement." Journal of Accounting and Economics 57(2): 241-262.
Dahling, J. J., B. G. Whitaker, and P. E. Levy. 2009. "The Development and Validation of a New Machiavellianism Scale." Journal of Management 35(2): 219-257.
Dong, J. 2014. "Prosecutorial Discretion at the International Criminal Court: A Comparative Study." Journal of Politics and Law 2(2): 109.
Dyck, A., A. Morse, and L. Zingales. 2010. "Who Blows the Whistle on Corporate Fraud?" The Journal of Finance 65(6): 2213-2253.
Fehr, E., and S. Gachter. 2002. "Altruistic Punishment in Humans." Nature 415(6868)137-140.
Financial Industry Regulatory Authority (FINRA). 2018. Rules and Regulations. Retrieved from http://finra.compIinet.com/en/display/displayviewall.html?rbid=2403&element_id=3878
Financial Industry Regulatory Authority (FINRA). 2017. Sanction Guidelines. Retrieved from http://www.finra.org/sites/default/files/Sanctions_Guidelines.pdf.
Gachter, S., E. Renner, and M. Sefton. 2008. "The Long-Run Benefits of Punishment" Science 322(5907): 1510.
Gibbs, J. C. 2013. Moral Development and Reality: Beyond the Theories of Kohlberg, Hoffman, and Haidt, 3rd Edition. New York, NY: Oxford University Press.
Gillan, S. L. 2006. "Recent Developments in Corporate Governance: An Overview." Journal of Corporate Finance 12(3): 381-402.
Gromet, D. M., and J. M. Darley. 2006. "Restoration and Retribution: How Including Retributive Components Affects the Acceptability of Restorative Justice Procedures." Social Justice Research 19(4): 395-432.
Guo, L., and R. W. Masulis. 2015. "Board Structure and Monitoring: New Evidence from CEO Turnovers." The Review of Financial Studies 28(10): 2770-2811.
Hillman, A. J., G. Nicholson, and C. Shropshire. 2008. "Directors' Multiple Identities, Identification, and Board Monitoring and Resource Provision." Organization Science 19(3): 441-456.
Ho, S. S. M., and K. S. Wong. 2001. "A Study of the Relationship between Corporate Governance Structures and the Extent of Voluntary Disclosure." Journal of International Accounting, Auditing and Taxation 10(2): 139-156.
Hoffman, M. L. 1993. "The Contribution of Empathy to Justice and Moral Judgment." In Reading in Philosophy and. Cognitive Science. Ed. A. I. Goldman. Cambridge, MA: MIT Press.
Holzer, H. J., S. Raphael, and M. A. Stoll. 2006. "Perceived Criminality, Criminal Background Checks, and the Racial Hiring Practices of Employers." The Journal of Law and Economics 49(2): 451-480.
Huse, M. 1998. "Researching the Dynamics of Board-Stakeholder Relations." Long Range Planning 31 (2): 218-226.
Jensen, M. C. 2010. Moral Markets: The Critical Role of Values in the Economy. Ed. P. J. Zak. Princeton, NJ: Princeton University Press.
Jolliffe, D., and D. Farrington. 2006. "Development and Validation of the Basic Empathy Scale." Journal of Adolescence 29(4): 589-611.
Kim, B., M. V. Cano, K. Kim, and C. Spohn. 2016. "The Impact of United States v. Booker and Gall/Kimbrough v. United States on Sentence Severity: Assessing Social Context and Judicial Discretion." Crime & Delinquency 62(8): 1072-1094.
Krishnan, J. 2005. "Audit Committee Quality and Internal Control: An Empirical Analysis." The Accounting Review 80(2): 649-675.
Larcker, D. F., S. Miles, T. Griffin, and B. Tayan. 2016. Board of Director Evaluation and Effectiveness. CGRI Survey Series. Corporate Governance Research Initiative, Stanford Rock Center for Corporate Governance, and The Miles Group. Retrieved from https:/Avww.gsb.stanford.edu/sites/gsb/files/publication-pdf/cgri-survey-boarddirectors-evaluation-effectiveness-2016.pdf.
Miceli, T. J. 2013. "Escalating Penalties for Repeat Offenders: Why are They So Hard to Explain?" Journal of Institutional and Theoretical Economics 169(4): 587-604.
--, and C. Bucci. 2005. "A Simple Theory of Increasing Penalties for Repeat Offenders." Review of Law & Economics 1(1): 71-80.
Morgan, B. L., and C. R. Wilson Van Voorhis. 2007. "Understanding Power and Rules of Thumb for Determining Sample Sizes." Tutorials in Quantitative Methods for Psychology 3(2): 43-50.
Mungan, M. C. 2010. "Repeat Offenders: If They Learn, We Punish Them More Severely." International Review of Law and Economics 30(2): 173-177.
Ndofor, H. A., C. Wesley, and R. L. Priem. 2015. "Providing CEOs with Opportunities to Cheat: The Effects of Complexity-Based Information Asymmetries on Financial Reporting Fraud." Journal of Management 41(6): 1774-1797.
New York Stock Exchange (NYSE). 2014. Corporate Governance Guide. Retrieved from https://www.nyse.com/publicdocs/nyse/listing/NYSE_Corporate_Governance_Guide.pdf.
Pallant, J. 2013. SPSS Survival Manual. New York, NY: McGraw-Hill.
Rhodes, R. A. 2000. "Governance and Public Administration." In Debating Governance: Authority, Steering, and. Democracy. Ed. J. Pierre. Oxford, England, UK: Oxford University Press.
Rose, A. M., J. M. Rose, and C. S. Norman. 2013. "Is the Objectivity of Internal Audit Compromised When the Internal Audit Function is a Management Training Ground?" Accounting & Finance 53(4): 1001-1019.
Scheer, D., and D. Griffin. 2016. "Deutsche Bank May Punish Staff for Personal Trade With Firm." Bloomberg. May 20, 2016. Retrieved from https://www.bloomberg.com/news/articles/2016-05-20/deutsche-bank-may-punish-ernployees-for-personaltrade-with-firm.
Securities and Exchange Commission (SEC). 2018. Invest in Your Career at the SEC. Retrieved from http://www.sec.gov/jobs/jobs_fulllist.shtml#accountants.
Securities and Exchange Commission (SEC). 2016. Enforcement Manual. Retrieved from https://www.sec.gov/divisions/enforce/enforcementmanual.pdf.
Securities and Exchange Commission (SEC). 2014. Boards of Directors, Corporate Governance and Cyber-Risks: Sharpening the Focus: June 10, 2014. Retrieved from http://www.sec.gov/News/Speech/Detail/Speech/1370542057946.
Singer, T., and N. Steinbeis. 2009. "Differential Roles of Fairness--and Compassion-based Motivations for Cooperation, Defection, and Punishment." Annals of the New York Academy of Sciences 1167(1): 41-50.
Skatova, A., and E. Ferguson. 2013. "Individual Differences in Behavioural Inhibition Explain Free Riding in Public Good Games When Punishment, is Expected but Not Implemented." Behavioral and Brain Functions 9(1): 3.
Spencer Stuart. 2017. "Spencer Stuart U.S. Board Index." Retrieved from https://www.spencerstuart.com/~/media/ssbi2017/ssbi_2017_final.pdf?la=en.
Tett, G. 2009. Fool's Gold: The Inside Story of J. P. Morgan and How Wall St. Greed Corrupted Its Bold Dream and Created a Financial Catastrophe. New York, NY: Free Press.
Thornton, B., R. M. Ryckman, and J. A. Gold. 2011. "Competitive Orientations and the Type A Behavior Pattern." Psychology 2(5): 411-415.
Tsoudis, O. 2002. "The Influence of Empathy in Mock Jury Criminal Cases: Adding to the Affect Control Model." Western Criminology Review 4(1): 55-67.
Ugazio, G., J. Majdandzic, and C. Lamm. 2014. "Are Empathy and Morality Linked? Insights from Moral Psychology, Social and Decision Neuroscience, and Philosophy." In Empathy and Morality. Ed. H. L. Maibom. New York, NY: Oxford University Press.
Van Assche, K., R. Beunen, and M. Duineveld. 2013. Evolutionary Governance Theory: An Introduction. New York, NY: Springer Publishing.
Verschoor, C. C. 2014. "Penalties for Fraud are Insufficient to Deter Wrongdoing." Strategic Finance 96(3): 15-61.
Wall, J., and T.J. Fogarty. 2016. "Foxes in the Henhouse: An Exploratory Inquiry into Financial Markets Fraud "Journal of Forensic and Investigative Accounting 8(1): 120-139.
Joseph M. Wall
Assistant Professor of Accounting
Jodi L. Gissel
Assistant Professor of Accounting
(1) The authors thank study participants for their time to respond to the survey and Michael Akers and Timothy Fogarty for insightful comments and suggestions. The authors also thank the editor and two anonymous reviewers for their valuable feedback. This paper is based on data collected as part of Professor Wall's dissertation at Case Western Reserve University.
(2) Sample size is in alignment with accounting research (Rose et al., 2013) and general principles within the Social Sciences (Morgan and Wilson Van Voorhis, 2007).
(3) Thirty-eight percent of the sample was female, which is slightly higher than national figures that report 22 percent of directors on S&P 500 boards are women (Spencer Stuart, 2017). The sample also skews slightly younger (56 years) than the average director age for S&P 500 boards (63 years) per Spencer Stuart's (2017) report. However, both sample demographics align with current board hiring practices (36% women, average age 57) in 2017 (Spencer Stuart, 2017).
(4) This study was personally funded by the researchers. The intermediary provides a nominal fee to participants it has previously located who desire to help with academic research and builds a blind panel of participants. The intermediary receives a fee for its services. Thus, no link (recruiting, monetary, or otherwise) exists between the researchers and the participants.
(5) The guidelines stated here and given to participants reflected FINRA Rule 2020's language at the time of the study. FINRA (2017) Rule 2020 has since been updated and now suggests fines of $2,500 to $73,000 for negligence and $10,000 to $146,000 for reckless misconduct. Given that participants' responses stay within suggested ranges, and these ranges have not changed substantially under the new guidelines, this update does not affect the authors' conclusions.
(6) Significance was judged as p [less than or equal to] 0.050 and moderate significance as 0.050 < p [less than or equal to] 0.100.
(7) Reported p-values in text and tables are one-tailed for hypothesized directional expectations in H1, H2, and H3 only (bold in tables), and two-tailed otherwise.
(8) An alternative was to use exploratory factor analysis to develop empathy and Machiavellianism control variables. However, while the ratio of participants to survey items was relatively adequate (10:1 for empathy and 15:1 for Machiavellianism), the sample size of 76 was below the recommended 150 observations (Pallant, 2013). Despite this, the model and conclusions were robust to other iterations of the model that included: (1) the individual empathy and Machiavellianism items (run with all of die items or just the three items from each type that are included in the averaging method employed), and (2) factors based on the empathy and Machiavellianism items.
(9) The authors conducted mean comparison tests for all pairs of conditions for the likelihood of each recommendation (Table 2), emphasis on warning versus fine (Table 3), and recommended fine amount (Table 4). For the non-extreme pairs of conditions, the authors did not find consistent significance that could be supported by theory. Therefore, these results are excluded from the presentation and discussion of results. However, this data is available from the authors upon request.
Caption: Figure 1 Evolutionary Governance Model through the Board of Directors' Lens
Table 1 Empathy and Machiavellianism Measures Participants indicated extent of agreement with each independent item on a scale of zero to 100 percent. Mean Item N=76 Empathy 1. When someone is upset, I get upset too. 28.64% 2. It is easy for me to get carried away 21.79% by other people's emotions. 3. My feelings are my own and do not reflect how others 58.80% feel. [Reverse-Coded] 4. I feel other people's pain. 51.57% 5. I get a warm feeling for someone if 74.07% I see them helping another person. 6. Being around happy people makes me feel happy too. 76.70% 7. I feel happy when I see people laughing and enjoying 76.81% themselves. Empathy variable (average of items 4, 5, and 7) 67.48% Machiavellianism 1. I am willing to be unethical if I believe 4.01% it will help me succeed. 2. I am willing to sabotage the efforts of other 6.96% people if they threaten my own goals. 3. I would cheat if there was a low chance 3.99% of getting caught. 4. I believe that lying is necessary to maintain a 3.57% competitive advantage over others. 5. The only good reason to talk to others is to get information that I can use to my benefit. 4.75% Machiavellianism variable (average of items 2, 4, and 5) 5.09% Table 2 Recommended Types of Sanctions Reported p-values are one-tailed for hypothesized directional expectations in H1, H2, and H3 only (bold in tables), and two-tailed otherwise. Panel A. Means by Independent Variable Likelihood of Recommendation History of Similar Incidents (H1) No Prior Prior Incidents Recommendation Incidents Suspected (Ranked Severity) N = 34 N=42 p-value No Sanction (1) 6.85% 2.90% 0.221 Censure (2) 41.15% 45.36% 0.603 Warning (3) 67.00% 45.43% 0.014 Fine (4) 52.56% 67.86% 0.031 License Suspension (5) 17.18% 31.69% 0.017 Fire from Job (6) 13.68% 27.00% 0.021 Permanent Bar (7) 0.85% 4.40% 0.015 Tail Time (8) 0.32% 2.62% 0.015 Panel A. Means by Independent Variable Likelihood of Recommendation Compensation Structure (H2) Recommendation Salary Commissions (Ranked Severity) N=32 N = 44 p-value No Sanction (1) 8.38% 1.98% 0.033 Censure (2) 38.31% 47.23% 0.273 Warning (3) 61.56% 50.36% 0.268 Fine (4) 56.63% 64.20% 0.361 License Suspension (5) 18.47% 30.09% 0.052 Fire from Job (6) 11.25% 28.16% 0.004 Permanent Bar (7) 2.91% 2.75% 0.931 Tail Time (8) 0.47% 2.41% 0.029 Panel A. Means by Independent Variable Likelihood of Recommendation Reason for Violation (H3) Medical Recommendation Procedure Vacation (Ranked Severity) N = 40 Home N = 36 p-value No Sanction (1) 6.85% 2.25% 0.060 Censure (2) 42.28% 44.81% 0.754 Warning (3) 62.70% 46.61% 0.053 Fine (4) 57.18% 65.28% 0.323 License Suspension (5) 25.98% 24.33% 0.815 Fire from Job (6) 16.80% 25.75% 0.092 Permanent Bar (7) 3.40% 2.17% 0.489 Tail Time (8) 2.08% 1.06% 0.379 Panel B. Means by Condition Condition ID: N = No prior incidents, P = Prior incidents suspected, S= Salary, C = Commissions, M = Medical procedure, V = Vacation home Likelihood of Recommendation Recommendation NSM PSM NCM PCM NSV (Ranked Severity) N = ll N = 9 N = 8 N = 12 N = 5 No Sanction (1) 13.00% 5.78% 3.25% 4.42% 12.60% Censure (2) 35.91% 48.67% 50.63% 37.75% 27.80% Warning (3) 72.09% 53.67% 71.13% 55.25% 64.20% Fine (4) 43.55% 66.22% 62.75% 59.17% 31.00% License Suspension (5) 10.18% 26.22% 18.63% 45.17% 10.20% Fire from fob (6) 6.73% 18.33% 3.63% 33.67% 8.60% Permanent Bar (7) 0.73% 5.67% 0.50% 6.08% 2.00% Jail Time (8) 0.36% 0.11% 0.38% 6.25% 0.00% Panel B. Means by Condition Condition ID: N = No prior incidents, P = Prior incidents suspected, S= Salary, C = Commissions, M = Medical procedure, V = Vacation home Likelihood of Recommendation Mean Comparison Recommendation PS V NCV PCV NSM vs PCV (Ranked Severity) N = 7 N = 10 N = 14 p-value No Sanction (1) 1.43% 0.10% 0.50% 0.089 Censure (2) 36.29% 46.00% 54.29% 0.161 Warning (3) 53.29% 59.50% 27.79% 0.008 Fine (4) 83.14% 65.10% 68.71% 0.045 License Suspension (5) 27.43% 27.20% 25.79% 0.053 Fire from fob (6) 11.14% 31.90% 34.79% 0.004 Permanent Bar (7) 3.43% 0.70% 2.64% 0.117 Jail Time (8) 1.43% 0.40% 1.71% 0.191 Panel C. Multivariate Analysis of Variance with Covariates (MANCOVA) Results Source dependent variable SS df MS History of No Sanction 311.519 1 311.519 Similar Censure 278.703 1 278.703 Incidents Warning 5882.378 1 5882.378 (H1) Fine 4374.919 1 4374.919 License Suspension 2437.682 1 2437.682 Fire from Job 2534.647 1 2534.647 Permanent Bar 235.022 1 235.022 Jail Time 75.603 1 75.603 Compensation No Sanction 631.803 1 631.803 Structure Censure 2327.493 1 2327.493 (H2) Warning 999.817 1 999.817 Fine 1007.830 1 1007.830 License Suspension 2179.326 1 2179.326 Fire from Job 4016.638 1 4016.638 Permanent Bar 0.864 1 0.864 Jail Time 54.548 1 54.548 Reason for No Sanction 181.178 1 181.178 Violation Censure 18.815 1 18.815 (H3) Warning 2065.081 1 2065.081 Fine 109.046 1 109.046 License Suspension 337.770 1 337.770 Fire from Job 622.183 1 622.183 Permanent Bar 11.329 1 11.329 Jail Time 14.319 1 14.319 History of No Sanction 355.398 1 355.398 Similar Censure 212.047 1 212.047 Incidents Warning 321.054 1 321.054 x Fine 4471.150 1 4471.150 Compensation License Suspension 2.377 1 2.377 Structure Fire from Job 270.425 1 270.425 Permanent Bar 1.427 1 1.427 Jail Time 34.053 1 34.053 History of No Sanction 21.889 1 21.889 Similar Censure 454.346 1 454.346 Incidents Warning 65.762 1 65.762 x Fine 829.724 1 829.724 Reason for License Suspension 1086.948 1 1086.948 Violation Fire from Job 1144.111 1 1144.111 Permanent Bar 34.108 1 34.108 Jail Time 6.704 1 6.704 Compensation No Sanction 1.084 1 1.084 Structure Censure 1304.556 1 1304.556 x Warning 1432.536 1 1432.536 Reason for Fine 4.819 1 4.819 Violation License Suspension 8.503 1 8.503 Fire from Job 1868.521 1 1868.521 Permanent Bar 0.277 1 0.277 Jail Time 18.083 1 18.083 History of No Sanction 1.150 1 1.150 Similar Censure 1802.809 1 1802.809 Incidents Warning 550.567 1 550.567 x Fine 6.510 1 6.510 Compensation License Suspension 194.559 1 194.559 Structure Fire from Job 541.258 1 541.258 x Permanent Bar 0.513 1 0.513 Reason for Jail Time 46.181 1 46.181 Violation Experience No Sanction 126.796 1 126.796 (covariate) Censure 156.721 1 156.721 Warning 243.426 1 243.426 Fine 4460.009 1 4460.009 License Suspension 98.875 1 98.875 Fire from Job 1583.838 1 1583.838 Permanent Bar 132.622 1 132.622 Jail Time 28.549 1 28.549 Empathy No Sanction 69.249 1 69.249 (covariate) Censure 6168.433 1 6168.433 Warning 14.793 1 14.793 Fine 1550.074 1 1550.074 License Suspension 1028.489 1 1028.489 Fire from Job 60.107 1 60.107 Permanent Bar 11.783 1 11.783 Jail Time 0.818 1 0.818 Machiavellianism No Sanction 114.574 1 114.574 (covariate) Censure 8.488 1 8.488 Warning 734.979 1 734.979 Fine 1016.673 1 1016.673 License Suspension 4597.168 1 4597.168 Fire from Job 130.891 1 130.891 Permanent Bar 0.635 1 0.635 Jail Time 7.407 1 7.407 Error No Sanction 10880.126 65 167.387 Censure 79467.931 65 1222.584 Warning 122811.288 65 1889.404 Fine 73800.967 65 1135.399 License Suspension 56254.424 65 865.453 Fire from Job 50292.351 65 773.728 Permanent Bar 3964.305 65 60.989 Jail Time 1505.550 65 23.162 Panel C. Multivariate Analysis of Variance with Covariates (MANCOVA) Results Source dependent variable F p-value History of No Sanction 1.861 0.089 Similar Censure 0.228 0.635 Incidents Warning 3.113 0.041 (H1) Fine 3.853 0.027 License Suspension 2.817 0.049 Fire from Job 3.276 0.038 Permanent Bar 3.853 0.027 Jail Time 3.264 0.038 Compensation No Sanction 3.775 0.028 Structure Censure 1.904 0.172 (H2) Warning 0.529 0.470 Fine 0.888 0.350 License Suspension 2.518 0.059 Fire from Job 5.191 0.013 Permanent Bar 0.014 0.906 Jail Time 2.355 0.065 Reason for No Sanction 1.082 0.302 Violation Censure 0.015 0.902 (H3) Warning 1.093 0.300 Fine 0.096 0.758 License Suspension 0.390 0.534 Fire from Job 0.804 0.373 Permanent Bar 0.186 0.668 Jail Time 0.618 0.435 History of No Sanction 2.123 0.150 Similar Censure 0.173 0.678 Incidents Warning 0.170 0.682 x Fine 3.938 0.051 Compensation License Suspension 0.003 0.958 Structure Fire from Job 0.350 0.556 Permanent Bar 0.023 0.879 Jail Time 1.470 0.230 History of No Sanction 0.131 0.719 Similar Censure 0.372 0.544 Incidents Warning 0.035 0.853 x Fine 0.731 0.396 Reason for License Suspension 1.256 0.267 Violation Fire from Job 1.479 0.228 Permanent Bar 0.559 0.457 Jail Time 0.289 0.592 Compensation No Sanction 0.006 0.936 Structure Censure 1.067 0.305 x Warning 0.758 0.387 Reason for Fine 0.004 0.948 Violation License Suspension 0.010 0.921 Fire from Job 2.415 0.125 Permanent Bar 0.005 0.947 Jail Time 0.781 0.380 History of No Sanction 0.007 0.934 Similar Censure 1.475 0.229 Incidents Warning 0.291 0.591 x Fine 0.006 0.940 Compensation License Suspension 0.225 0.637 Structure Fire from Job 0.700 0.406 x Permanent Bar 0.008 0.927 Reason for Jail Time 1.994 0.163 Violation Experience No Sanction 0.758 0.387 (covariate) Censure 0.128 0.721 Warning 0.129 0.721 Fine 3.928 0.052 License Suspension 0.114 0.736 Fire from Job 2.047 0.157 Permanent Bar 2.175 0.145 Jail Time 1.233 0.271 Empathy No Sanction 0.414 0.522 (covariate) Censure 5.045 0.028 Warning 0.008 0.930 Fine 1.365 0.247 License Suspension 1.188 0.280 Fire from Job 0.078 0.781 Permanent Bar 0.193 0.662 Jail Time 0.035 0.852 Machiavellianism No Sanction 0.684 0.411 (covariate) Censure 0.007 0.934 Warning 0.389 0.535 Fine 0.895 0.348 License Suspension 5.312 0.024 Fire from Job 0.169 0.682 Permanent Bar 0.010 0.919 Jail Time 0.320 0.574 Error No Sanction Censure Warning Fine License Suspension Fire from Job Permanent Bar Jail Time Table 3 Emphasis on Warning versus Fine Analyses of recommended fine amount were filtered for those participants who indicated a greater than percent likelihood of recommending a fine and a fine amount greater than $1. Emphasis on warning versus fine was computed at the participant level by taking likelihood of recommendation for warning minus likelihood of recommendation for fine; thus, positive (negative) indicates greater mean likelihood of recommendation for warning (fine). Reported p-values are one-tailed for hypothesized directional expectations in H1, H2, and H3 only (bold in tables), and two-tailed otherwise. Panel A. Means by Independent Variable History of Similar Incidents (HI) Prior No Prior Incidents Incidents Suspected Measure N=34 N=42 p-value Frequencies for recommendation of: Warning > Fine 21 15 Warning = Fine 2 6 Warning < Fine 11 21 Emphasis on Warning versus Fine (mean) 14.44% -22.43% 0.005# Panel A. Means by Independent Variable Compensation Structure (H2) Salary Commissions Measure N=32 N=44 p-value Frequencies for recommendation of: Warning > Fine 20 16 Warning = Fine 2 6 Warning < Fine 10 22 Emphasis on Warning versus Fine (mean) 4.94% -13.84% 0.201 Panel A. Means by Independent Variable Reason for Violation (H3) Medical Procedure Measure N=40 Home N=36 p-value Frequencies for recommendation of: Warning > Fine 24 12 Warning = Fine 5 3 Warning < Fine 11 21 Emphasis on Warning versus Fine (mean) 5.53% -18.67% 0.047# Panel B. Means by Condition Condition ID: N=No prior incidents, P=Prim incidents suspected, S=Salary, C=Commissions, M=Medical procedure, V=Vacation home NSM PSM NCM Measure N=ll N=9 N=8 Frequencies for recommendation of: Warning > Fine 9 6 4 Warning = Fine 0 0 2 Warning < Fine 2 3 2 Emphasis on Warning versus Fine (mean) 28.55% -12.56% 8.38% Panel B. Means by Condition Condition ID: N=No prior incidents, P=Prim incidents suspected, S=Salary, C=Commissions, M=Medical procedure, V=Vacation home PCM NSV PSV Measure N=12 N=5 N=7 Frequencies for recommendation of: Warning > Fine 5 4 1 Warning = Fine 3 0 2 Warning < Fine 4 1 4 Emphasis on Warning versus Fine (mean) -3.92% 33.20% -29.86% Panel B. Means by Condition Condition ID: N=No prior incidents, P=Prim incidents suspected, S=Salary, C=Commissions, M=Medical procedure, V=Vacation home Mean Comparison NCV PCV NSM vs PCV Measure N=10 N=14 p-value Frequencies for recommendation of: Warning > Fine 4 3 Warning = Fine 0 1 Warning < Fine 6 10 Emphasis on Warning versus Fine (mean) -5.60% -40.93% 0.004 Panel C. Analysis of Variance with Covariates (ANCOVA) Results Dependent Variable: Emphasis on Warning versus Fine Source SS df History of Similar Incidents (HI) 20403.217 1 Compensation Structure (H2) 4015.279 1 Reason for Violation (H3) 3123.206 1 History of Similar Incidents x 2395.974 1 Compensation Structure History of Similar Incidents x 1362.665 1 Reason for Violation Compensation Structure x 1603.521 1 Reason for Violation History of Similar Incidents x 437.342 1 Compensation Structure x Reason for Violation Experience (covariate) 2619.512 1 Empathy (covariate) 1262.012 1 Machiavellianism (covariate) 3480.504 1 Error 244359.453 65 Panel C. Analysis of Variance with Covariates (ANCOVA) Results Dependent Variable: Emphasis on Warning versus Fine Source MS F p-value History of Similar Incidents (HI) 20403.217 5.427 0.012 Compensation Structure (H2) 4015.279 1.068 0.305 Reason for Violation (H3) 3123.206 0.831 0.365 History of Similar Incidents x 2395.974 0.637 0.428 Compensation Structure History of Similar Incidents x 1362.665 0.362 0.549 Reason for Violation Compensation Structure x 1603.521 0.427 0.516 Reason for Violation History of Similar Incidents x 437.342 0.116 0.734 Compensation Structure x 0.407 Reason for Violation Experience (covariate) 2619.512 0.697 Empathy (covariate) 1262.012 0.336 0.564 Machiavellianism (covariate) 3480.504 0.926 0.340 Error 3759.376 Table 4 Recommended Fine Amount Reported p-values are one-tailed for hypothesized directional expectations in H1, H2, and H3 only (bold in tables), and tuio-tailed otheruiise. Panel A. Means by Independent Variable History of Similar Incidents (H1) Prior No Prior Incidents Incidents Suspected Measure N=22 N=34 p-value Median $5,000 $10,000 Mean $9,786 $21,324 0.026 Panel A. Means by Independent Variable Compensation Structure (H2) Salary Commissions Measure N--23 N=33 p-value Median $5,000 $10,000 mam Mean $15,957 $17,373 0.832 Panel A. Means by Independent Variable Reason for Violation (H3) Medical Vacation Procedure Home Measure N=27 N=29 p-value Median $5,000 $10.000 Mean $16,185 $17,355 0.859 Panel B. Means by Condition Condition ID: N=No prior incidents, P=Prior incidents suspected, S=Salan, C=Commissions, M=Medical procedure, V=Vacation home NSM PSM NCM PCM Measure N=7 N=8 N=5 N=7 Median 15,000 $5,000 $5,000 $10,000 Mean $4,571 $17,813 $19,000 $23,929 Panel B. Means by Condition Condition ID: N=No prior incidents, P=Prior incidents suspected, S=Salan, C=Commissions, M=Medical procedure, V=Vacation home Mean Comparison NSV PSV NCV PCV NSM vs PCV Measure N=2 N=6 N=8 N=13 p-value Median $7,500 $15,000 $10,000 $10,000 Mean $7,500 $29,583 $9,163 $18,269 0.039 Panel C. Analysis of Variance with Covariates (ANCOVA) Results Dependent Variable: Recommended Fine Amount Source SS df df History of Similar Incidents (H1) 863600978.127 1 Compensation Structure (H2) 92128177.375 1 Reason for Violation (H3) 58449225.457 1 History of Similar Incidents x 43821980.003 1 Compensation Structure History of Similar Incidents x 5176411.580 1 Reason for Violation Compensation Structure x Reason 132577018.970 1 for Violation History of Similar Incidents x Compensation Structure x Reason for Violation 204792403.282 1 Empathy (covariale) 837401017.577 1 Machiavellianism (covariale) 9020856907.907 1 Error 19371471787.889 46 Panel C. Analysis of Variance with Covariates (ANCOVA) Results Dependent Variable: Recommended Fine Amount Source MS F p-value History of Similar Incidents (H1) 863600978.127 2.051 0.080 Compensation Structure (H2) 92128177.375 0.219 0.642 Reason for Violation (H3) 58449225.457 0.139 0.711 History of Similar Incidents x 43821980.003 0.104 0.748 Compensation Structure History of Similar Incidents x 5176411.580 0.012 0.912 Reason for Violation Compensation Structure x Reason 132577018.970 0.315 0.577 for Violation History of Similar Incidents x Compensation Structure x Reason for Violation 204792403.282 0.486 0.489 Empathy (covariale) 837401017.577 1.989 0.165 Machiavellianism (covariale) 9020856907.907 21.421 0.000 Error 421118951.911
|Printer friendly Cite/link Email Feedback|
|Author:||Wall, Joseph M.; Gissel, Jodi L.|
|Publication:||Journal of Managerial Issues|
|Date:||Mar 22, 2019|
|Previous Article:||"Aren't Bosses Supposed to Manage You?" A Qualitative Study of Managing Your Boss.|
|Next Article:||Merger and Acquisition Targets: The Role of Moral Hazard and Acquiring Firm Executive Compensation.|