Shareholder value ideology, reciprocity and decision making in moral dilemmas.

Second, the vignettes presented the detailed information about the decision in the given business situation. In vignette #1, participants were asked whether they would continue the business relationship with a long-serving supplier, which was in a major financial crisis and needed the purchasing deal to resolve the crisis, or switch to a new supplier, which offered comparable supplies at a slightly lower price. For vignette #2, participants were asked whether they would keep all of the cost savings their company recently achieved through efficiency improvements, or pass on some cost savings by way of discounts/promotions to customers who had already been satisfied with the quality and the current price of the company's products. In vignette #3, participants were asked whether they would lay off long-serving employees to fully increase the company's operational efficiency leading to greater profits, or keep the employees and forgo this opportunity to instantly increase profits (see the Appendix for the full description).

Third, the options available to the participants in each vignette would not lead to the company's illegal conduct and financial difficulty. While legal and moral issues could be intertwined in various scenarios, in this study we focused only on the moral side and not the legal side of managerial decision making. Thus, we carefully set the three hypothetical situations in a way that both options available to participants were perfectly legal. We were also aware that participants might choose the option that has less financial risk involved. To control for the risk factor, which is not the focus of our study, we framed all three situations to make sure that both options would not bring any financial difficulty to the company (i.e., profitable vs. more profitable deals in vignette #1; already high customer satisfaction vs. higher customer satisfaction in vignette #2; already high company profitability vs. higher profitability in vignette #3).

The vignettes presented to the participants in both the experimental and control groups had the same business situations. However, the additional information, providing experimental stimuli or manipulations, was bundled into the vignettes that were presented only to the participants in the experimental groups. This manipulation technique involving differing information inputs given to participants in control and experimental groups has been adopted by previous researchers (e.g., Joshi and Arnold, 1998; van Dijk and Zeelenberg, 2003). The additional information for the participants in experimental group #1 was the duty of the top managers to maximize profits and shareholders' wealth (i.e., shareholder value manipulation). The additional information for those in experimental group #2 was the reciprocation between the company and the stakeholders (i.e., reciprocity manipulation), and the additional information of both manipulations was presented to those in experimental group #3 (see the Appendix for more detail).

The sequence of vignettes, presented to participants enrolled in fall and spring semester courses, was also reordered to neutralize the potential effect of the vignette sequence on the decision outcomes. In addition, the manipulation check was successfully performed with 87% of the participants in the experimental groups, indicating that they took the additional information (experimental stimuli) into consideration when they made the decisions. Participants that did not respond to our manipulations were dropped from the study. The results of chi-square and t tests indicated that there were no significant difference in terms of academic status, gender, ethnicity, age group, work experience, and managerial experience between the participants in the study and those who were dropped out.

Variables, Data Coding and Statistical Model

The decision outcomes favorable and unfavorable to stakeholders (i.e., suppliers, customers, and employees) in the three vignettes were the dependent variables and were coded as 0 and 1, respectively. Shareholder value and reciprocity manipulations were the independent variables and were coded as 0 and 1 for their absence from and presence in the vignettes, respectively. Control variables included gender, age group, years of managerial experience, years of work experience, levels of courses (from which participants were drawn), and ethnicity. Male and female were coded as 1 and 0, respectively. Age group, categorized into six groups (below 20, 21-24, 25-29, 30-39, 40-49, and 50 years and above), was coded as 1, 2, 3, 4, 5, and 6, respectively, while course levels (i.e., introductory junior, senior, and graduate) were coded as 1, 2, and 3, respectively. Ethnicity, simply categorized into White and non-White, was coded as 1 and 0, respectively, whereas years of managerial work and years of work experience were kept as continuous variables. In addition, since we had binary dependent variables, we used logistic regression analyses and the following statistical model to test our hypotheses:

Decision Outcome = constant + [b.sub.1]Shareholder

Value + [b.sub.2]Reciprocity + [b.sub.3](Shareholder Value x

Reciprocity) + [b.sub.4]Course Level + [b.sub.5]Gender + [b.sub.6]Age

Group + [b.sub.7]Managerial Experience + [b.sub.8]Work Experience

+ [b.sub.9]Ethnicity + errors

RESULTS

Table 2 shows the descriptive statistics results, including the percentage of favorable and unfavorable decision outcomes in vignette #1 (supplier-related), vignette #2 (customer-related), and vignette #3 (employee-related) in the experimental and control groups. The descriptive data indicated that participants in the control group had the tendency to maximize profits and shareholders' wealth even at the expense of employees (71.29% of them decided to lay off employees for more profits). Conversely, the participants in the control group tended to make decisions in favor of customers and suppliers, although compromising some profits (79.21% decided to pass on cost saving to customers through promotions/discounts, and 61.39% decided to purchase from their current supplier at a slightly higher price to rescue the supplier from potential bankruptcy). The descriptive data also revealed that the shareholder value manipulation seemed to increase the likelihood that the participants would make a decision unfavorable to suppliers, customers and employees in order to increase profits (control group vs. experimental group #1), whereas the reciprocity manipulation seemed to lessen such likelihood (control group vs. experimental group #2). The shareholder value and reciprocity manipulations also appeared to nullify each other as the percentage of favorable/unfavorable decision outcomes in experimental group #3 were almost the same as in the control group for all three dilemmas.

Table 3 shows the results of logistic regression analyses. The overall chi-square test of model #1 and the improvement of the goodness-of-fit (measured by the difference of-2 log likelihood in the full and control models) were both statistically significant (p < 0.01). The model #1 results indicated that in the supplier-related dilemma, the ideology of shareholder value was positively related to the likelihood that participants would make a decision unfavorable to suppliers in order to pursue more profits (p < 0.05), while the norm of reciprocity was negatively related to such likelihood (p < 0.05), providing support for Hypotheses la and 2a. The overall chi-square test of model #2 was not significant, providing no support for Hypotheses lb and 2b. This indicates that in the customer-related dilemma, there was no significant difference in decision outcomes of participants in the control group and of those in the experimental groups, despite the shareholder value and reciprocity manipulations. For model #3, the overall chi-square test and the improvement of the goodness-of-fit were both statistically significant (p < 0.01 and p < 0.05, respectively). Model #3 results also indicated that the norm of reciprocity was negatively related to the likelihood that participants would make the decision to lay off employees in order to increase profits (p < 0.05), while the ideology of shareholder value was not significantly related to such likelihood, yielding support for Hypothesis 2c but not for Hypothesis 1c. The interaction between the ideology of shareholder value and the norm of reciprocity was also not significant across all three dilemmas. Given that Hypotheses la and 2a were supported, coupled with no significant interaction effect, Hypothesis 3a (the opposite effect of the ideology of shareholder value and the norm of reciprocity in the supplier-related dilemma) was supported while Hypotheses 3b and 3c were not. In addition, the results of model #3 analysis indicated that participants in higher-level courses were more likely to lay off-employees to increase profits and shareholders' wealth (p < 0.01).

DISCUSSION

General Implications

The results of this study suggest that the ideology of shareholder value and the norm of reciprocity affect managerial decision making in stakeholder moral dilemmas differently when the decision outcome affects different stakeholders. From the experiment, the ideology of shareholder value significantly increased the likelihood of participants' decision outcome to increase profits at the expense of suppliers (not customers and employees), while the norm of reciprocity significantly decreased the likelihood of participants' decision outcome to increase profits at the expense of suppliers and employees (not customers). This provides mixed support for our ideology of shareholder value and norm of reciprocity hypotheses.