There is growing pressure all around the world on business leaders to manage sustainably and responsibly and some international business schools have responded by trying to provide their students with the required skills [1
]. How to do this is not an easy task, but it should be based on good knowledge of the students’ attitudes towards the topic to be taught.
Students have shown a willingness to upgrade ethics and sustainability when they move from the description of present practice to expectations of what business should be doing [3
]. It has also been detected a certain evolution in favour of a greater expectation of sustainability among the younger demographic [4
]. In this sense, recent research on students belonging to the millennial generation have evidenced their sensitiveness to sustainability concerns and ethical issues [5
], which may be reflected in their career decisions [8
Nevertheless, it is important not only knowing the value students give to sustainable and responsible business practices but also finding out why they believe that they are valuable in the first place. Stakeholders management is a basic component of student understanding of responsible management, mainly due to its contribution to a holistic consciousness of the decision-making consequences (economic, environmental, and social) [9
]. Ditlev-Simonsen and Midttun [3
] found that students consider that stakeholders should be a key motivator for managers pursuing corporate social responsibility (CSR) but they did not go further to analyse if there was an instrumental or a normative reason for that. Additionally, little is known about how individual characteristics of the students may be related to their perception of responsible management. On the basis of previous work [10
], student social and moral values and their ethical orientations in terms of “self” versus “others” have been linked to attitudes supportive of sustainable practices [11
]. In this work we turned to behavioural economics in order to know some psychological traits of the students connected to their moral principles, which could explain to a certain degree why they believe stakeholder interests should be attended and address in this way the aforementioned research gap.
Proposing a management model that takes into account how making decisions affects others brings stakeholder theory closer to the social utility model [12
]. The ultimatum and dictator games were applied on a sample of undergraduate students in a Spanish university to find evidence for a psychological explanation of the orientation an individual adopts with regard to stakeholders. Thus, the effect of potential reciprocity and social preference related to inequality aversion were tested to see if they have any influence on the orientation towards stakeholders from the instrumental and normative points of view. To our knowledge, there have been no previous empirical studies on this. The use of a sample of university students can contribute to the academic debate on the ethical development of students [13
] and continues the research trend of experimental or quasi-experimental studies [15
This research may contribute to the existing literature on stakeholder theory because it adopts an individual level of analysis, emphasising the importance of behavioural motives [19
] and of talking about human beings [20
], as they are the cornerstone of ethical decision-making [21
]. In this sense, by focusing on micro-foundations, the complexity of human psychology is taken into account in order to build stakeholder theory on a more solid foundation [22
]. We are thus in line with the proposal of Bridoux and Stoelthorst [24
], which “replaces the traditional assumption of economic theory that all human behaviour is exclusively motivated by self-interest with less pessimistic assumptions derived from empirical research on human motivation”. Finally, discovering the psychological triggers behind management decisions with relation to company stakeholders may help firms to adopt control and management mechanisms guaranteeing that such relations are maintained within the guidelines set at corporate level.
The rest of the paper is structured as follows. Section 2
draws up the research hypotheses based on a review of stakeholder theory and social preferences. The data, method and results are described in Section 3
. Finally, Section 4
3. Analysis and Results
The data were collected by means of questionnaires and the experiments described in the following sub-section. A total of 212 second-year undergraduate students from three different faculties of a Spanish university participated. Only the individuals who took the role of first player in the ultimatum and dictator games were taken into account, that is, those who proposed a distribution of the received amount of money. Therefore, the final number of observations was 106. The students were distributed as follows: 56 were studying in the Faculty of Economic and Business Science, 30 in the Faculty of Education and 20 in the Faculty of Philosophy and Arts. Average age of the students was 20.8 years, 75.5% of them were female and 25.5% had previous working experience.
3.2. Design and Implementation of the Experiments
Our research proposal was coordinated with the Vice-Rectorate for Research, where issues related to research involving students were addressed, and previous authorisation by the Faculties involved was granted. In order to avoid the selection bias that might arise when the volunteers used come to a laboratory with economic motivations in response to a call for volunteers [65
], the within-subject experiments were performed in classrooms without any advance warning being given to the students nor any prior indication that there would be payment. Nevertheless, informed consent was obtained from all individual participants included in the study so that all students participated in the experiments voluntarily.
The social experiments were designed and performed taking maximum care to define the process in order to guarantee maximum quality in the information obtained and homogeneous replication. For experiments performed in the classroom as in this case, the ways in which the materials are distributed and collected, how confidentiality is guaranteed and information leaks avoided are crucial [65
]. The experiments were therefore conducted following exactly the same patterns and at the same time to ensure that participants could not receive any information as to their content. In addition, since the fact that the experimenter was a teacher and therefore a figure of authority who might affect participants’ decisions if they feel judged [66
], the tests were performed with groups of students with whom the experimenters had no relation.
Before the experiments began, the students in the classrooms were divided into two equal groups and the roles of first and second player were assigned to the students in each group at random. Even though this amounted to a loss of data, it was decided that each participant should have a single role in the experiments, to prevent the first players from being influenced in their distribution decision by the fact that it would then be them who would receive the offer. The students were then informed that they were going to be paired with another person who was in another of the various rooms being used and that they would never know who that person was either during the experiment or after it, and that that person would never know who had been his or her partner. We thus avoided the effect that a lack of anonymity can have on decisions [67
]. At no time could students comment on their decisions with their peers.
The instructions handed out to the participants informed them that they were about to participate in two economic within-subject experiments (ultimatum and dictator) and that, as a result, they would receive the amount resulting from one of the two games chosen at random. The aim was to find out which decision was adopted by each individual that was to take the role of first player in both experiments, that is, to find out and compare their distribution offer in an ultimatum and dictator situation without being influenced by knowing which of the two would determine the final reward.
The starting amount used in both experiments was €10 and each participant also received a set payment of €3 for participating. A result that has been observed in previous studies is the small effect that the amount used for distribution has on the decisions of both players [68
]. However, Andersen et al. [69
] performed their experiment in poor towns in Northeast India with amounts between 20 and 20,000 rupees (equivalent to 1600 h of work), which amounted to a greater difference than those in previous experiments, and they noted that in the ultimatum game the amount offered increased as the amount available increased but the percentage of the offer decreased as did the probability of rejection of the offer received. Cherry et al. [70
] did not find relevant differences in the offers in the case of the dictator game when they raised the amount from $
10 to $
40, but Carpenter et al. [71
] did detect a reduction of 33–25% in the average offered when the amount rose from $
10 to $
3.3. Measurement of Variables
Dependent variable: Firstly, a brief explanation of the stakeholder concept was provided for those who were not familiar with it. In order to identify the instrumental and normative orientations to stakeholders (INSTRUMENTAL and NORMATIVE), as covered respectively by the two hypotheses posed, the following two items measured on a 7-point Likert scale (1 = Totally disagree; 7 = Totally agree) were used:
“Firms should take into account their stakeholders’ interests only when such groups may affect the firms’ activities and goals now or in the future”.
“Firms should always take into account their stakeholders’ interests because that is the right thing to do, even when such groups are not able to affect the firms’ activities and goals now or in the future”.
Independent variable: The independent variable used to test the first hypothesis (RECIPROCAL) aimed to find how much of the inequality offered by the first player in the ultimatum experiment corresponds to his interest in achieving a positive response to his offer from the other player and not to his aversion to unfair distribution or the search for fairness, that is, the reciprocal component in his decision. It was calculated, therefore, by subtracting from the amount offered by an individual in an ultimatum situation the amount proposed in the dictator case [64
The independent variable corresponding to the second hypothesis (NON-RECIPROCAL) aims to measure the intention of an individual in a dictator experiment to share with an unknown person, fairly and avoiding inequality, the money he has just received while knowing that the other person has no possibility of rejection however the money is distributed. The value of this variable corresponded to the amount that the first player actually handed over in the dictator experiment.
Control variables: The analyses included three control variables for individual factors.
Firstly, the training received by the respondent was identified by a dummy variable (ECON_EDUCATION) to know if the person had previously studied or was currently studying economics and business management, in which case the variable took value 1, otherwise it took 0. This variable was included because previous works emphasize the potential effect of economics or business education on the students’ moral reasoning and decision-making [72
]. For instance, economics students have shown a greater non-cooperative behaviour, such as more freeriding or less charitable giving [74
], and it has also been found that they are significantly more corrupt than others [76
] and that graduate business students cheat more than their non-business peers in order to achieve their goals [77
]. Specifically, these findings can be explained by the fact that most business programmes reinforce a “norm of self-interest” [78
], instilling the belief that everyone intends to maximise personal material outputs [79
Secondly, gender (GENDER) was also measured with a dummy variable, taking 1 for men and 0 for women. Many articles focused on student attitudes and perceptions on CSR and/or stakeholders have considered gender as a relevant explanatory variable (e.g., [80
]). Although different effects have been found, most studies conclude that women tend to exhibit greater sensitivity to CSR [83
], have stronger ethical orientation [80
], and attach more importance to environmental issues [81
]. These results are consistent with the ethics of care [84
], which proposes that women speak and act in the language of caring and responsibilities. This means that, in the context of a relationship, women are likely to believe that all parties in the relationship have responsibilities [85
Finally, the models also included a dummy variable (EXPERIENCE) indicating if the respondent had work experience, in which case it took value 1 and 0 otherwise. This variable has been included as it can affect a person’s ethical judgement [86
], being expected in this case that the more experienced students were more ethically oriented [88
]. A possible explanation for this could be that, due to their own work experience, they see themselves more as a stakeholder in the business context and may therefore value the consideration of stakeholders’ claims beyond instrumental motives [89
and Table 2
give the main descriptive statistics and the correlation coefficients for the variables used in the study. Although some of them show a statistically significant correlation, analysis of the variance inflation factors (VIF) showed that there was no evidence of multicollinearity because in no case were the VIF above 10 [90
]. It is worth mentioning that the amounts offered by the first participant in the experiments were not the lowest amounts possible but on average were €4.5 in the ultimatum and €3.2 in the dictator case, with the respective medians being €5 and €4. It can also be observed that the instrumental approach to stakeholders obtained on average (5.4) a higher value than the normative approach (4.5).
shows the results of the regression analyses applied to test the hypotheses and relations, controlling for the respondents’ training in business management, gender and work experience. It is worth mentioning that we repeated the analysis taking into account the possibility of a heteroskedasticity problem and the results did not vary.
As the table shows, in the regression analysis of the extra amount offered in the ultimatum experiment over the money shared in the dictator game, which represented the reciprocal component of the decision, on the instrumental approach towards stakeholders the coefficient was positive and statistically significant (β = 0.166; p < 0.1), thus confirming the first hypothesis. In the second regression, it was found that the amount delivered in the dictator experiment, which reflected the decision-makers non-reciprocal component, was positively and significantly related to the normative approach to stakeholders (β = 0.160; p < 0.1), as proposed in the second hypothesis.
Regarding the control variables, gender was marginally significant to the instrumental approach (β = 0.195; p < 0.1), so that being a man seems to have a positive influence on this approach towards stakeholders. In addition, having received training in economics and business management had a negative impact on the normative approach (β = −0.285; p < 0.01).
Awareness about the social and environmental impact of business organisations beyond their economic role has increased in the last two decades. In particular, young people have shown a special sensitiveness to sustainability concerns and ethical issues in relation to business practices. Previous works on university students have evidenced the significance they give to stakeholder management within this context and this study contributes by including social preferences in the research on their stakeholder approach. With that purpose in mind, social experiments were carried out to be subsequently related to the participants’ orientation towards stakeholders.
The ultimatum and dictator experiments corroborated the presence of decisions that are not optimal for the monetary incentives given. The respective proposals about how to split the sum of money given to the first players were 45% and 32%. Although this context can be characterised by a low social distance and it could have positively conditioned the offers made by participants [91
], our results are in line with the average values obtained by other studies applying within-subject social experiments in different contexts [68
]. On average, players in this standard version of the ultimatum game offer about 40–50% of the total assigned, and these amounts are almost always accepted, while the rate of rejection increases for offers below 20% [68
]. The average percentage offered in the dictator game is about 30%, and almost 65% of players shared the money with the other player [92
]. The fact that there were no minimal offers in the dictator case assumes there is a concern for fair distribution, and the fact that more was offered in the ultimatum adds the reciprocal component that stems from considering the possibility of a reprisal or rejection by the other player if the amount proposed is perceived as inappropriate.
The reciprocal and non-reciprocal components in decision-making may be present when the need and legitimacy of meeting company stakeholders’ interests are considered. Our results indicate that the greater weight of the concern for the other party’s response to the payment granted may to some extent justify a greater instrumental approach towards stakeholders. The greater importance of the non-reciprocal component corresponds to a more normative approach with a greater moral implication regarding stakeholders.
Analysis of the control variables gave two main results. Firstly, it was detected that, although at a marginally significant level, men place greater value than women on the need to consider stakeholders according to their capacity for affecting the company results. That is, they valued more an instrumental stakeholder approach, although no gender effect on the normative approach was found. Prior empirical studies have shown that women tend to be more aware of their obligations towards the various stakeholders [93
] and to be more sensitive to others than men [94
]. Secondly, it was also found that training in economics and business management has a negative effect on the normative stakeholder approach [89
]. To the extent that the normative approach is ethics-based, this result is in line with some of the criticisms made of training in business management (e.g., [95
]) claiming that it is based on amoral theories that give maximum priority to profit, omitting to provide moral development to managers and leading to questionable ethics in their companies.
Some policy implications for management education can be drawn from the overall conclusions of this research about the relevance of individual characteristics when dealing with stakeholders and the differential effect of training in business and economics. We must keep in mind that, as a social science, what is taught in business studies is particularly relevant since theories can become self-fulfilling [97
], in that they might influence the phenomena studied, and even make them true because a management theory can change the behaviour of managers, who start acting in accordance with the theory [72
]. If educators wish their students contribute to building delightful organisations [98
], they should teach them how they can do so [99
Helping students to become more socially responsible and ethically sensitive is a significant part of the responsibility of business schools, who need to incorporate a more humanistic vision of the world [79
]. Moreover, admitting a broader perspective of human nature in business management would help to come up with new ways of facing complex and turbulent situations [99
In order to change business learning and education in this direction, academic programs should draw upon humanistic management postulates [101
] and move from an organisation-centered towards a human-centered worldview [79
]. Taking into account both notions, a people-oriented management education should emphasise the dignity of each person and the preeminence of people over material goods [102
] and teach students to explore the multifaceted (not simply economic) repercussions of their decisions, forcing them to look at what is really happening as a result of such decisions [79
]. As reflected by Trencher et al. [104
], the ability to anticipate and envision differing types of harmful or desirable consequences in the future is essential for empowering societal change agents and required for sustainable development.
In this sense, management education should focus more on value-oriented skills [79
] and aim at teaching students to think critically and act responsibly as managers, as employees, as customers, as citizens, and so on [101
]. To develop such a macro-competence, students should be equipped with multiple theories, supplementing neoclassical economics with other economic perspectives (such as ecological and feminist economics) and diverse views from other disciplines (such as sociology or positive psychology) on economic behaviour [105
]. Moreover, social experiments and resource-allocation contexts, as designed in the present research, may provide an opportunity to better align different courses or academic programs with societal needs, to the extent that they help simulate real-world contexts and stakeholder-thinking, beyond an instrumental approach.
Following the previous comments, to the extent that education is a relevant force in shaping students’ attitudes and perceptions, it could be interesting to include other forces, particularly at an institutional level, in further research. In this sense, studies with samples composed of students from different countries or regions might be relevant to analyse the potential influence of cultural factors. Previous empirical research has shown that these factors contribute to generate different attitudes and perceptions on stakeholders and CSR issues (e.g., [15
]). Specifically, the cultural dimension of individualism-collectivism [109
] may be relevant since more collectivist individuals have been found to be more concerned with society and the environment and to hold more socially responsible attitudes [110
]. In particular, with business students as their unit of analysis, Ng and Burke [11
] found that more collectivist individuals reported more positive attitudes toward sustainable business practices.
The line of research opened up here could be also broadened and enriched by taking into account the specific characteristics that may be present in the relation between a company and its stakeholders [112
]. For instance, access to valuable but limited resources by any of the parties involved could be included as this determines the degree of power available according to social exchange theory [113
] and the balance or imbalance of power in their relations. The ultimatum and dictator experiments are clearly different at this level of balance and power assignation, with greater compensation in the first case and absolute dependence for the second player in the dictator game. A specific study could therefore be carried out, comparing the approach towards specific stakeholders depending on the power attributed to them for influencing the company. Van Dijk and Vermunt [63
] observed that the lack of power may increase the fairness motivation, in such a way that perceiving that others are more dependent for obtaining rewards generates a feeling of responsibility towards them and encourages an effort to promote their interests, even if there is no mutual reciprocity.
Finally, our study analyses only the negative utility arising when there are differences between the reward received by a person and what others receive, but other issues such as the maximisation of joint efficiency or consideration of the party that receives least in a distribution could also be included [48
]. These issues may also be related to the decision on how the relation with the various stakeholders should be managed, which would be another direction for expanding this line of research.