The importance of outcome and effort in negative reciprocity.

(1)

1

The Importance of Outcome and Effort in

Negative Reciprocity

Micro-Foundation in Strategic Management

Master’s Thesis

MSc Business Studies

Strategy track

Geert-Jan Morskate (6081657)

Supervisor

Anouar el Haji

August 15

th

, 2014

Amsterdam

FACULTY OF ECONOMICS AND BUSINESS

(2)

2 Abstract

All societies have social norms on what kind of behavior needs to be rewarded and what needs to be punished. The current literature focusses more on positive reciprocity. This research looks at the how people reciprocate when faced with an unkind action. Previous experiments such as the moonlighting and ultimatum game find that in the case of a benevolent action, individuals reward less consistently than that they punish unkind behavior. Firstly, this research looks at the importance of the outcome of the unkind action. Secondly, the effort another person puts into an unkind action is examined. This research shows that the outcome of an unkind behavior is crucial when deciding to reciprocate. When the outcome is not negative then individuals do not reciprocate even when the other person put high effort in the unkind action. When the outcome is negative individuals punish the other person, irrespectively of their social value orientation. However, prosocial individuals also take into account the effort exerted and engage in more negative reciprocity when the other exerted more effort in the unkind action.

Acknowledgement

I want to thank Anouar el Haji who supervised me during my thesis. I valued the direct, insightful, and critical comments.

(3)

3

1. Introduction

All societies have social norms on what kind of behavior needs to be rewarded and what needs to be punished. These basic principles o rewarding and punishing naturally create the tendency to reciprocate (Perugini et al., 2003). Reciprocating can either be positive by returning a favor or negative by retaliating, for which the sayings such as “An eye for an eye and a tooth for a tooth”, “to give a dose of its own medicine” are illustrative. De Cremer & Van Lange, (2001) find that in the case of a benevolent action, individuals reward less consistently than that they punish unkind behavior. Experiments like the moonlighting game, where another person offers a one-time deal and the individual can accept or reject it, and the moonlighting game, where another person can give or take money from the next person who can either reward or punish the other person, show that negative actions are more consistently punished than kind actions are rewarded. In the moonlighting game players consistently punishes unkind behavior but players do not consistently rewards friendly actions (Abbink et al., 2000). Retribution seems to be a more compelling motivator than benevolent actions.

The natural tendency to reciprocate questions the validity of standard economic theory that individuals behave solely rationally and self-interested. An individual for example considers as well the level of fairness when making a decision, which is “irrational” according to standard economic theory (Kahneman et al., 1998). Standard economic theory does not allow for other factors than monetary payoffs such as the intention behind the action, while individuals value intentions (Falk & Fischenbacher, 2006). The exerted level of effort increases the value we give to otherwise identical objects (Kruger et al., 2004). The level of effort shows the severity of the intention and in the case of either good or bad intentions how much the other person wants to help or hurt you. Therefore, this is important in determining the counteraction. Furthermore, in contrast to economic theory individuals are heterogeneous and have different motivations (e.g. Gottschalg & Zollo, 2007; Bridoux et al., 2011). Social

(5)

5 scholars distinguish between four different kind of social value orientations (e.g. Murphy et al., 2011). One’s social value orientation helps explain the individual differences of reciprocity (Fehr et al., 2002; Fehr & Gintis, 2007).

This research focusses on the negative reciprocal response and the determinants of the level of negative reciprocity. This is hypothesized to be both influenced by the social value orientation of the person and the effort the other person exerted. Previous experiments show that individuals retaliate more consistently than they would return a favor. Therefore, it is expected that the social value orientation only matters with positive reciprocity and that negative reciprocity is more universal. This is researched by using a vignette study where the outcome and level of effort are manipulated. The research consists of 259 respondents. This research increases the understanding of human motivation, contributing to the strategic organization literature by expanding the micro-foundations. The core of any organization exists out of individuals. However, in the recent strategic organization research the individual was lost in collective conceptualizations (Felin and Foss, 2005). Individuals within an organization matter and are the micro-foundation in strategic organization (Felin and Foss, 2005). For example in collective value creation and cooperation, individual motivation and behavior matters (Flore et al., 2011).

The structure of this research is as following. Section 2 is the literature review, giving an overview of the relevant literature from sociology, economics, and strategic organization. Section 3 establishes the conceptual framework, incorporating the social value orientations, effort and negative reciprocity concept. Section 4 elaborates the used methodology and data. Section 5 consists of the results and section 6 is the discussion. Finally, section 7 concludes this thesis.

(6)

6

2. Literature Review

The literature review starts with an overview of the economics of reciprocity and the behavior of self-interest individuals. Secondly, the influence of effort in evaluating unkind actions and the heterogeneous motivations regarding negative reciprocity are examined. Thirdly, the different social value orientations are discussed and finally the gap in the micro-foundation of strategic organization is addressed.

2.1 Reciprocity and Self-interested Behavior

Individuals are reciprocators when they return a favor and retaliate when treated unfairly, also when this does not increase their personal payoff (Falk and Fischenbacher, 2006). Reciprocity distinguishes two kinds of reciprocity, negative and positive (Falk and Fischenbacher, 2006; Perugini et al., 2003; Fehr & Gächter, 1998). When another person shows kindness, a positive reciprocator will also respond with kindness. On the other hand when a person is unfair, a negative reciprocator will respond with hostility. The existence of these reciprocators explains why there is cooperation in certain circumstances when there is no cooperation expected by standard economic theory. In standard economic theory individuals are self-interested and do not cooperate if there is not a direct personal payoff. Hence, in groups self-interest individuals might ride. According to Fehr and Gärtner (2000), reciprocators will punish these free-riders and try to enforce social norms (negative reciprocity). The reciprocator is simple responding to a negative action (i.e. free-riding) and expects no individual material gain from punishing. Hence, the reciprocator contradicts the standard economic theory of self-interested individuals by not solely looking at personal gains.

When reciprocators punish self-interested (non-contributing) individuals, this might generate a negative counter reaction from the interested individual. A rational self-interest individual has two options, either to contribute just enough to not get punished, or in a sequential equilibrium the self-interested individual retaliates to deter future punishments. The

(7)

7 second option results in self-interested individuals engage in punishing the people who behave prosocially (the reciprocators), which is called the “antisocial” punishment phenomenon (Herrmann et al., 2008). The initial altruistic punishment by reciprocators can result in a negative reaction towards themselves, the self-interested individuals’ retaliate. The retaliation can be explained as self-interested behavior but this is questionable. The cost of punishing could have been used to contribute just enough and furthermore anti-social punishment most likely will result in lower cooperation (Hermann et al., 2008), which diminishes the payoff of the self-interested individual and therefore not rational. Another, behavioral, explanation is that self-interested individuals engage in negative reciprocity as well. This suggests that not only the initial reciprocators engage in negative reciprocity but also the self-interested individuals.

In the ultimatum game of Camerer and Thaler (1995) they show that everybody engages in negative reciprocity. In this game person A can propose a one-time deal to person B. When person B rejects the proposed terms of the deal, both get a payoff of zero. This experiment shows that when A proposes an unfair deal to B, then B will reject the deal. This is not rational and contradicts the self-interest theory because by rejecting the deal the person will have a payoff of zero, while the deal might have been unfair but the payoff was most likely not below zero. This shows that by proposing an unfair deal and showing a greedy and unfair intention that all people will reciprocate negatively. The experiment shows that negative and positive reciprocity is quite different; everyone exhibits negative reciprocity, while only some exhibit positive reciprocity. The basic mechanism of reciprocity means that both negative and positive sanctions are expected, but people might develop a preference for one of the two over time (Perugini et al., 2003). This might mean that self-interested individuals develop a preference for negative reciprocity as well and that reciprocators engage in both negative and positive reciprocity, with the preference for positive reciprocity.

(8)

8 Furthermore, in the Moonlighting experiment (Abbink et al., 2000) where Player A can take money from or pass money on to player B, who can either return some money, which is multiplied, or punish player A for not passing on money, both negative and positive reciprocity are observed. In contrast with the ultimatum game, there is incremental punishment and higher punishments are increasingly costly. Regarding negative reciprocity Abbink et al. (2000) find that when A takes an amount from person B then person B will punish A, even though it costs more to punish than that it benefits her. Abbink et al. (2000) find that the player B consistently punishes unkind behavior but player B does not consistently rewards friendly actions. Retribution is a more compelling motivator than benevolent actions.

2.2 The Influence of Effort and the Motives Behind Negative Reciprocity

A key aspect of understanding reciprocity depends on how and why individuals reward and punish others. In evaluating an unkind action an individual will not solely look at the outcome of that action but she will also look at the underlying intention of the action (Falk & Fischbacher, 2006). The underlying intention most likely impacts the amount of negative reciprocity a person exhibits. Hence, an unkind action, which was not successful but had a bad intention, might be despised more than an unkind action, which did hurt, but had a less harsh intention. Kruger et al. (2004) and Morales (2006) determined that individuals use the effort heuristic when valuing items. This suggests that people take into account the amount of effort a person put into a project when determining the value of the outcome. Therefore, the observed level of effort a person exhibits when being unkind affects the amount of negative reciprocity. The effort shows the severity of the initial intention. Concluding, the amount of effort exercised by the person that initially engaged in an unkind action matters in the decision making process for the person exhibiting negative reciprocity.

(9)

9 Both reciprocators and self-interested individuals can engage in negative reciprocity. However, the motivation to retaliate might be heterogeneous. For example Wetzer et al. (2007) examine with what kind of intentions do customers engage in negative word-to-mouth communication, which does not benefits the customer. Customers’ motives vary from warning others, being a positive reciprocator, to retaliate on the company, negative reciprocity. More broadly, according to Murphy et al. (2011) there are four principle motives or goals, individual gain, competitive goal, inequality-averse, and maximizing collective gains. Firstly, the individual motive/goal to negative reciprocate could be to deter future unkind actions. Secondly, the competitive motive/goal is to become relatively better by increasing their lead or by decreasing the gap. Thirdly, when payoffs are unequal the goal is to decrease this. Therefore negative reciprocators might perform the action but with different motivations.

2.3 The Social Value Orientations of Individuals

In the same social situation, people can react differently. When facing the same decision people will react differently because they have different motivations and have different social value orientations (McClintock, 1972; Murphy et al., 2011). A social value orientation is the preference of resource distribution between oneself and others (McClinktock, 1972). According to Murphy et al. (2011) (similar to McClintock, 1972) a decision maker can either, maximize their payoff (individualistic), maximize (competitive) or minimize (inequality averse) the difference between her own and the other person’s payoff, or maximize the joint payoffs (prosocial). On the basis of these classifications most individuals can be considered either a self-regarding individual (individualistic and competitive) or as a reciprocator (inequality averse and prosocial). Reciprocity cannot be explained solely as a strategy to benefit personally in the long run, experiments where the repetitiveness was eliminated individuals showed reciprocity. Hence, personal norms are active where personal benefits are

(10)

10 absent (Perugini et al., 2003). Furthermore, Bogaert et al. (2008); Dehue et al. (1993) find that these motives are stable over time.

2.4 Micro-foundations in Strategy

The collective value creation depends on the micro-foundation of individuals and their motivation to cooperate. Hence, it is of key importance to understand individuals’ motives. Within management research the individual within the collective has been partially neglected. Therefore, understanding the motives to cooperate will improve the design of motivational systems. Flore et al. (2011) theorizes that different human motivations require different motivational systems. This paper contributes to this by improving the predictability of both self-regarding individuals as well as reciprocators to design better motivational systems.

Camerer and Thaler (1995), and Herrmann et al. (2008) show that individuals’ exhibit negative reciprocity. Herrmann et al. (2008) find that not only the reciprocators exhibited negative reciprocity, but also self-interested individuals, which is puzzling because this is not rational and in their monetary self-interest. Furthermore, the ultimatum game of Camerer and Thaler (1995) showed that all persons retaliate on unkind actions. These bargaining games research reciprocity irrespectively of to the person’s social value orientation and with what kind of intentions they reciprocate. This research examines if both self-interest and reciprocators exhibit negative reciprocity and why, i.e. is revenge a stronger motivator for a self-interested individual than rewarding benevolent behavior? Furthermore, the current literature (e.g. Fehr and Gärtner, 2000) focusses on the outcome, looking if an individual experiences a negative outcome, without looking at the intention of the action. This is not realistic concerning the importance and value individuals give to effort (Morales, 2006; Kruger et al., 2004). This research takes into account the amount of effort exerted by the counter party to punish the individual, and if this affects the amount of negative reciprocity.

(11)

11 To what extent do outcome and effort determine negative reciprocity?

3. Conceptual Framework

This section contains two parts. In the first part this research proposes a simple model to explain how people reciprocate when confronted with an action that affects them adversely. The second part formulates the hypothesis regarding the conceptual model and the theory.

3.1 The Model

Negative reciprocity is the behavioral response to perceived unkindness (Falk and Fischenbacher, 2006). Experiments, conducted by e.g. Camerer and Thaler (1995) and Abbink et al. (2000), show that individuals reciprocate based on what they perceive as an unfair outcome. Hence, when deciding to reciprocate the outcome is an important factor. Furthermore, individuals will evaluate the unkindness not solely on the outcome but also what the underlying intention is (Falk & Fischbacher, 2006). Kruger et al. (2004) and Morales (2006) show that people use the effort heuristic when valuing items. Therefore, in this model effort shows the underlying intention and affects how people value the unkind action, when valued as more unkind people will reciprocate stronger. A negative action which is perceived as being more unjust will results in a stronger retaliation. Hence, in the model people evaluate the other person’s action that adversely impacted them by the outcome, how much are they affected, and the level of effort the other person put into the unkind behavior.

Furthermore, personal norms are active where personal benefits are absent (Perugini et al., 2003). In the scenario personal benefits are absent and reciprocating is solely on the basis of personal norms. People are not the same and individuals respond differently in the same situation due to their heterogeneous beliefs (e.g. McClintock, 1972; Murphy et al., 2011). In this research people are categorized in two groups, interested and prosocial. The self-interested individual is more focused on its own payoff and the prosocial looks more at other

(12)

12 person’s payoff as well. The SVO slider measure of Murphy et al. (2011) is used to identify to which group a person belongs. The SVO measure identifies four kind of people: individualistic, competitive, prosocials and altruistic. The individualistic and competitive are regarded as self-interested while the prosocials and altruistic individuals are deemed prosocials. Because people are different and react differently on the same situation the SVO will have a moderating effect. Prosocials will be more focused on fairness and the underlying intention than solely the outcome.

All this is summarized in the conceptual model (Figure I). First another person is unkind. Subsequently, the outcome can be either negative or is unsuccessful and the effort put into it can be either high or low. Then the person will need to make to decide whether she wants to reciprocate. This is moderated by their personal social value orientation.

Figuur I: Conceptual model

3.2 The Hypotheses

In the conceptual model individuals reciprocate depending on the outcome of the unkind actions and the effort the other person put into the action. Firstly, when another person is unkind this can have a negative effect or no effect. In standard economic theory self-regarding individuals are only expected to reciprocate when it generates personal benefits. In this research the setting is designed so that there are no personal gains form reciprocating and that there are costs. Hence, reciprocating is not rational. When personal benefits are absent then the decision to reciprocate will be made based on personal norms (Perugini et al., 2003). All

Outcome of the action Effort of the other person Negative Reciprocity Social Value Orientation

(13)

13 societies have social norms on what kind of behavior needs to be rewarded and what needs to be punished. These basic principles on rewarding and punishing naturally create the tendency to reciprocate (Perugini et al., 2003). De Cremer & Van Lange (2001) find that in the case of a benevolent action, individuals reward less consistently than that they punish unkind behavior. Furthermore, in experiments from Camerer and Thaler (1995) and Abbink (2000) they also find that people more consequently reciprocate negatively than that they reciprocate positively. In both experiments the individuals negatively reciprocate because the outcome of the other person’s actions adversely impacted them. Hence, it seems that in cultures punishing unkind behavior is more universal. The current theory determines that reciprocators return the favor and punish unkind behavior and rational self-interested individuals will not engage in either one of them. However, retaliating for unkind actions seems to be more universal and a stronger motivator to reciprocate. Hence, in the case of negative reciprocity both self-interested and prosocials will engage in similar reciprocating when the unkind action adversely impacts them.

Hypothesis 1A: Negative reciprocity is positively related with the negative impact of the unkind behavior.

Hypothesis 1B: The level of negative reciprocity between self-interested and prosocial individuals is the same.

Secondly, individuals take into account the intention of the other person. This is supported by Falk & Fischbacher (2006) who show that individuals also take into account the underlying intention when reciprocating. Hence, when the other person’s effort increases the action is deemed more negative. Kruger et al. (2004) and Morales (2006) determined that individuals use the effort heuristic when valuing items. This suggests that people take into account the amount of effort a person put into a project when determining the value of the

(14)

14 outcome. Hence, when another person puts more effort in the unkind action individuals will value it as more negative while the outcome can remain the same.

Hypothesis 2: The amount of effort put into an unkind action is positively related to negative reciprocity.

People are not the same and individuals respond differently in the same situation due to their heterogeneous beliefs (e.g. McClintock, 1972; Murphy et al., 2011). In this research individuals are classified as either self-interested or prosocial. As mentioned before, when the unkind action results in negative outcome then people are expected to reciprocate more consistently. Furthermore, Kruger et al. (2004) and Morales (2006) do not distinguish between individuals and their SVO and state that all people in general value based on effort. Hence, when the outcome is negative then there is no difference between self-interested and prosocial individuals when looking at effort. Both self-interested and prosocial individuals will evaluate the unkind action as more unkind and will reciprocate more strongly. However, based on the nature of self-regarding individuals and prosocials this seems unlikely. Self-regarding individuals are more focused on their own payoff while prosocials more look at the joint payoff and intentions. Hence, based on these differences in SVO prosocials will be more inclined to reciprocate more strongly when they deem the unkind behavior as mean.

Hypothesis 3: Negative reciprocity is positively related to the level of effort when the outcome is negative instead of neutral.

(15)

15

4. The Methodology

4.1 Method

The research design is mixed, consisting of a vignette study and a questionnaire. Firstly, in a survey the SVO is determined by using the SVO slider measure of Murphy et al. (2011). Murphy’s et al. (2011) SVO measure has six primary items with nine secondary (and optional) items, where each item is a resource allocation choice. Murphy et al. (2011) classifies four different SVOs, individualistic, competitive, inequality averse, and prosocials. In this research both individualistic as well as competitive individuals are regarded as self-interested and the inequality averse and prosocials as reciprocators.

Secondly, a vignette study is conducted. In a vignette study a situation is described shortly and subsequently the person gives his judgment about the situation. The vignette study consists of the core element and specific characteristics (Atzmüler and Steiner, 2010). The benefit of vignettes is the ability to include both explanatory as well as contextual factors, making the scenarios more realistic (Atzmüler and Steiner, 2010). Furthermore, the vignettes are strong in finding causal relations because by adjusting one part the outcome can change and because only one part is altered this is the cause.

This vignette’s scenario is based on cooperation in a work environment with a bonus system. In the scenario the person works in a team and is rewarded depending on the evaluation of the client. This research wants to identify if the level of effort and the outcome causes individuals to engage in negative reciprocity differently. Hence, with a vignette study the level of effort exerted can be easily manipulated and potentially show a causal relationship between effort and negative reciprocity. Furthermore, the outcome is manipulated to see if the level of effort matters while the respondent is not affected adversely.

(16)

16 Finally, another questionnaire is conducted. Here, the intentions to engage in negative reciprocity or to abstain from it are questioned. Additionally, there are questions on whether the respondent interpreted the scenario as intended. Finally, there are general background questions.

4.2 The Survey Design

In this research the effect of both the outcome and the effort is researched. Both effort and outcome are the independent variables and are manipulated in the vignettes. The scenario describes a situation where the respondent needs to cooperate with another person. Subsequently, the other person makes the respondent look bad and this can affect the bonus the respondent get. In total there are four scenarios (See Table I). First, the other person puts a high effort in making the respondent look bad and the outcome is also negative. Secondly, the other person puts a high effort in making the respondent look bad but it has no effect. Thirdly, the other person exerts low effort in making the respondent look bad but the outcome is negative anyway. Finally, the other person exerts low effort and the action does not hurt the individual. After the scenario the respondent is asked if they want to reciprocate (the dependent variable). In this design punishing is also not free and it is explicitly stated that the respondent will not work with the person in the future. Therefore, it would not be rational to punish because you make yourself even worse of and the respondent will not work with the other person again so there are no future gains. The possible redistribution options are shows in Appendix A. The respondent could choose to reduce the other person’s bonus, between 0 percent and 80 percent. However, each 10 percent reduction in the other person’s bonus reduced the respondents own bonus with 2.5 percent.

(17)

17 The second part of the survey determines the social value orientation by using the SVO measure of Murphy et al. (2011). Within the SVO framework people differ in their preferences in resource allocations between themselves and others. People are not all ‘rational’ and choose the self-maximizing option. The SVO measure consists of six primary items. All items have the same form and ask the respondent to choose a resource allocation. To make sure that the respondent understands the concept they are asked to confirm the allocation by writing it down. For a more detailed explanation on the validity see Murphy et al. (2011). Finally, the respondents are also asked general questions about their background.

4.3 The Sample

The sample is consists of 260 respondents. Each respondent randomly received one of the four vignettes. The scenarios are: The other person exerts high effort and the outcome for the

Rules of the game

You do not like that Bob made you look bad. Regardless of the outcome you can set this straight. How much effort do you put into doing this depends on you. The effort will cost you something.

Context

Other than the final presentation the management board cannot assess how much each one has contributed. The management is satisfied with the report and decides to give a team bonus of EUR 10,000.

Outcome

Negative. However, they reward you with only 40% of the total bonus because they believe that you contributed less.

Neutral. Furthermore, they reward both of you with 50% of the total bonus because they believe that you contributed as much.

Level of effort

High. Bob presents his part first but on purpose he covers also your main contributions and now it seems you contributed less.

Low. Bob presents his part first but in his enthusiasm he covers also some of your parts and now it seems you contributed less.

You are working with Bob on a consulting case for an international firm. This unique case requires both of your skills and in the future you will not work with Bob again. You have put both the same amount and quality of work into the final report. The last step is to present the report to the management board who also determines your bonus if applicable.

Context

Table I. The vignette

This table shows the four scenarios. The context is the same in each scenario while the level of effort (high or low) and outcome (negative or neutral) are manipulated.

(18)

18 respondent is negative, the other person exerts high effort but does not affect the respondent, the other person exerts low effort but the outcome is negative, and finally the other person exerts low effort and the outcome is not negative for the respondent.

The respondents are geographically dispersed, from 22 different countries. Nevertheless the largest group is from the United States (87%). Other respondents come for example from China (2%), or Mexico (2%). Furthermore, from the respondents 57% is male. The age of the respondents is between 18 and 71 but roughly 50 percent is between 18 and 30 (See Table I). The education level is as well divers (Table I). The largest group, 45 percent, has a bachelor’s degree and 10 percent has a Master’s degree. Furthermore, 25 percent has finished primary school. Most have a general education as background (33%), followed by physics or engineering (23%).

The most respondents are familiar with teamwork. Only 5 percent is limitedly familiar with it, while 90 percent states to be familiar or very familiar with teamwork. Furthermore, 20 percent of the respondents play a team sport. Finally, the social value orientation of the respondents (Table II): The self-interested group consists of 154 respondents (59%) and is mainly individualistic according to the Murphy et al. (2011) SVO measure. The prosocial group consists mainly of pure prosocials according to Murphy et al. (2011) and some altruistic respondents.

Age 18-30 31-40 41-50 51-60 60>

53% 26% 12% 7% 3%

Work experience <10 10-15 16-20 20>

48% 23% 10% 18%

Education Primary Secondary Bachelor Master

25% 20% 45% 10%

Field of study General Physics, etc. Humanities Business Other

33% 23% 15% 15% 14%

(19)

19 Individualistic Competitive Altruistic Prosocials

N 58.8% 0.4% 1.9% 38.8%

Self-interested Prosocial

(20)

20

5. The Results

This section analyzes the sample. First, the vignettes are discussed. Secondly, the sample is analyzed for correlations between variables. Finally, the hypotheses are tested.

5.1 The Vignettes

The respondents got a scenario where they worked together with another person. The other person subsequently with either high or low effort tries to make the respondent look bad. The result of this can be either negative for the respondent or the other person did not succeed and the respondent is not affected. After this the respondent is asked how much effort and money they are willing to invest in retaliating.

Table III and Figure II show the average amount that respondents reciprocate in each vignette. The first vignette the effort of the other person is high and subsequently the unkind action affects the respondent negatively. In this vignette on average respondents are willing to reduce their own payoff with 6.1 percent to reduce the other person’s payoff with 24.3 percent. However, when in second vignette the effort remains high but the respondent is not negatively affected the willingness to punish decreases sharply. On average the respondent is now willingly to accept 1.8 percent of personal costs to reduce the other’s payoff with 7.3 percent. In the third vignette the effort of the other is low but it does negatively affect the respondent. In this scenario the respondent is willingly to reduce its own payoff with 3.8 percent to reduce the other person’s payoff with 15.1 percent. When the effort remains low and the respondent is not affected then willingness to negatively reciprocate drops. The respondent is on average willing to sacrifice 1.6 percent of its own payoff to reduce the other person’s with 6.3 percent.

(21)

21 Figure II shows a pattern. When the respondent is not affect the effort does not seem to matter. The average amount of reciprocity is low and similar for both high and low effort scenarios. Furthermore, Table III shows that in both scenarios only 21 and 24 percent reciprocate at all. However, when the respondent is negatively affected effort does seem to matter. More respondents engage in negative reciprocity, the percentage that reciprocates increases from 49 percent to 62 percent. The respondents reciprocate on average 0.92 more meaning that they are willing to sacrifice 2.3 percent of their own payoff to reduce the other person’s payoff with 9.2 percent.

5.2 Correlations

In this section the main variables of this research are examined for correlations (Table IV). The variables included are the dependent variables, effort, outcome of the other person’s

Reciprocate

Scenario N Percentage Mean S.D. Personal Other's

1. High effort, negative outcome 63 62% 2.43 (2.31) -6.1% -24.3%

2. High effort, neutral outcome 62 24% 0.73 (1.72) -1.8% -7.3%

3. Low effort, negative outcome 67 49% 1.51 (2.03) -3.8% -15.1%

4. Low effort, neutral outcome 68 21% 0.63 (1.45) -1.6% -6.3%

Table III. Negative reciprocity

Effort in reciprocating Reciprocating cost

0% 5% 10% 15% 20% 25% 30%

High effort, negative outcome

Low effort, negative outcome

High effort, neutral outcome

Low effort, neutral outcome

Figure II: The Amount of Reciprocity

Personal costs Other's costs

(22)

22 action and the SVO. Furthermore, the control variables age, education, gender, work experience, and familiarity with teamwork are included. Finally, the independent variable, negative reciprocity, is also included.

First of all the dependent variable seems to correlate with the outcome of the other person’s action, the perceived effort, work experience, and age. When the outcome is negative respondents engage in more negative reciprocity (ρ = 0.32, p < 0.01). Furthermore, when the perceived effort increases then respondents also engage in more negative reciprocity (ρ = 0.35, p < 0.01). However, when the person has more work experience they reciprocate less (ρ = -0.15, p < 0.05). Furthermore, when the respondent is older they reciprocate less as well (ρ = -0.14, p < 0.05). The dependent variable, the outcome is negative, correlates with the perceived effort, if someone practices a team sport, and work experience. It seems that respondent evaluate actions more negatively when the outcome is also negative. Hence, the outcome drives the perception of how much effort the other exercised (ρ = 0.16, p < 0.05). The other two correlations, practicing a team sport and work experience, are a coincidence. Respondents are randomly assigned to each vignette. Hence, their personal characteristics should be insignificant. Additionally, prosocials and respondents with a university background perceive the effort of the other person as less negative (ρ = 0.14, p < 0.05; ρ = -0.19, p < 0.01). The moderator variable, the social value orientation (“SVO”), correlates with respondents studying physics, engineering, etc., which are more often selfregarding (ρ = -0.17, p < 0.01).

(23)

23

T

ab

le

I

V

. M

e

an

s, S

tan

d

ar

d

D

e

vi

at

ion

s, C

or

re

lat

ion

s

V

ar

ia

bl

es

M

SD

1

2

3

4

5

6

7

8

9

10

11

12 1. N

eg

at

iv

e r

ec

ipr

oc

ity

1.32 (2.02

)

1 2. N

eg

at

iv

e ou

tc

om

e

0.50 (0.50

)

.32**

1 3. E

ffor

t pe

rc

ept

ion

3.64 (1.54

)

.35**

.16*

1 4. S

V

O

(

P

ros

oc

ia

l =

1)

0.59 (0.49

)

-.01

.06

-.14*

1 5. F

am

ilia

rit

y w

ith

t

ea

m

w

or

k

4.38 (0.76

)

-.09

-.10

-.07

.05

1 6. P

ra

ct

ic

e a

t

ea

m

s

por

t

1.79 (0.41

)

-.06

.19**

-.01

.06

-.03

1 7. W

or

k e

xpe

rie

nc

e

12.45 (10.5

6)

-.15*

.03

.02

.28**

.15*

1 8. G

en

de

r (

F

em

ale

=

2)

1.42 (0.49

)

-.09

.02

-.09

.10

.00

.12*

.15*

1 9. U

niv

er

sit

y e

du

ca

te

d

0.55 (0.50

)

-.06

.06

-.19*

*

-.12

.04

.00

-.02

.07

1

10. S

tu

dy

: B

us

in

es

s or

e

con

om

ic

s

0.15 (0.36

)

-.03

.12

-.03

.11

-.04

-.05

.07

-.07

.07

1

11. S

tu

dy

: P

hy

sic

s, e

ng

in

ee

rin

g, e

tc

.

0.23 (0.42

)

.08

.05

-.02

-.17*

*

-.01

-.06

-.12

-.23*

*

.19**

-.23*

*

1

12. A

ge

33.09 (11.5

1)

-.14*

-.11

-.01

.01

.15*

.17**

.91**

.20**

.06

-.12

1 * C

or

re

la

tion

is

s

ig

nif

ic

an

t a

t t

he

0.0 5 l

ev

el (

2-ta

ile

d)

.

** C

or

re

la

tion

is

s

ig

nif

ic

an

t a

t t

he

0.0 1 l

ev

el (

2-ta

ile

d)

.

(24)

24 5.3 Hypotheses

The first regression in Table V shows how individuals negatively reciprocate when the other person’s actions negatively affects them. When the other person’s action had a negative impact then the respondent put significantly more effort in reciprocating (β = 1.15, P < 0.01). People were willing to forgo 2.9 percent of their own payoff in order to reduce the other person’s payoff by 11.5 percent. This confirms hypothesis 1A that expects that negative reciprocity is positively correlated with the outcome. This is in line with previous research that finds that individuals reciprocate even when it is costly (e.g. Camerer and Thaler, 1995; Abbink, 2000). According to standard economic theory a true self-regarding individual would not negatively reciprocate when it is costly and does not benefit them. Other scholars distinguish also reciprocators in general that do are willing to incur costs in order to enforce social values (e.g. Perugini et al., 2003). Hence, in reciprocating there should be a significant difference: Self-interested people will not reciprocate and prosocials will. However, as expected in hypothesis 1B, the first regression also shows that there is no significant difference between individuals with different social value orientations in how they reciprocate based on a negative outcome. This seems to confirm the idea that societies have social norms on rewarding and punishing which creates a natural tendency to reciprocate irrespectively of a person’s SVO. When personal benefits are absent then the decision to reciprocate will be made based on personal norms (Perugini et al., 2003).

In the second and third regression in Table V the effort of the other person and the control variables are included. Hypothesis 2 expects that the amount of effort is positively related to negative reciprocity. In the second and third regression the effect of a negative outcome is still significant and remains stable. In the second regression the effort on itself does not significantly determines how people reciprocate. However, the moderating variable is significant (β = 0.92, p < 0.05). When the individual is prosocial and the other person’s

(25)

25 effort is high it values the effort and negatively reciprocates stronger. When an individual is prosocial and the other exhibited a high amount of effort then the person reduces the other person’s payoff with 9.2 percent and this cost 2.3 percent of his/her own payoff. Hence, only prosocial individuals value the effort or are willing to act on it. This partially contradicts the second hypothesis. Not all individuals engage in more negative reciprocity when the other person’s effort increases, just the prosocials. In the research of Kruger et al. (2004) and Morales (2006) they do not make the distinction between people. Hence, this shows that the effort heuristic might not be applicable to every person in the same way. The third regression adds the control variables but only the variable practicing a team sport is significant (β = -0.52, P < 0.05). Individuals that practice a team sport reciprocate less strongly. Due to the insignificance of the control variables these are omitted in further regressions.

Finally, the fourth and fifth regression look at the difference between prosocials and self-regarded individuals by restricting the sample based on the outcome. The fourth regression in Table V looks solely at the vignettes where the other person’s action had a negative impact. Hypothesis 3 expects that when the outcome is negative the level of effort is positively related to negative reciprocity but only for prosocials. This regression shows a similar result as regressions two and three and supports the hypothesis. The level of effort only matters when the individual is prosocial and the effort was high then the negative reciprocity is stronger. When the individual is prosocial and the effort was high she puts more effort in punishing (β = 1.35, P < 0.05). Prosocial individuals punish the other person with 13.5 percent more than the self-regarding individuals when the effort put into the unkind action was high. This confirms the third hypothesis. Comparing this to Kruger et al. (2004) and Morales (2006) this shows that the effort heuristic is not the same for everyone. It does confirm that the SVO and that self-regarding individuals are more focused on their own payoff. It seems that both self-regarding and prosocials take into account the negative affect

(26)

26 on them when reciprocating but that the prosocials reciprocate additionally when the other person’s effort is high. The fifth regression in Table V looks at the vignettes where the other person’s action did not have a negative impact on the individual. In this regression nothing is significant. Hence, it seems that prosocials do not always take effort into account only when the outcome is also negative.

(27)

27 (1) (2) (3) (4) (5) Negative impact 1.152*** 1.152*** 1.298*** (0.375) (0.370) (0.381) High effort -0.044 -0.032 0.087 -0.156 (0.370) (0.375) (0.617) (0.421) SVO -0.232 -0.639 -0.503 -0.618 -0.425 (0.340) (0.411) (0.416) (0.555) (0.393)

Negative impact × SVO 0.223 0.207 0.063

(0.486) (0.481) (0.485)

High effort × SVO 0.921* 0.849* 1.347* 0.411

(0.481) (0.487) (0.783) (0.563)

Study: Business and Economics -0.25

(0.346)

Study: Physics, engineering, etc. 0.16

(0.304) University educated -0.343 (0.251) Age 0.01 (0.025) Work experience -0.025 (0.027)

Practice a team sport -0.517*

0.304

Constant 0.807*** 0.830*** 1.833** 1.913*** 0.889***

(0.255) (0.318) (0.766) (0.450) (0.305)

Observations 259 259 259 129 129

F-Statistic 9.71*** 7.59*** 4.28*** 2.97** 0.43

* significant at the 0.10 level (2-tailed). ** significant at the 0.05 level (2-tailed). *** significant at the 0.01 level (2-tailed).

Table V.

The Linear Regressions

This table shows the results from the linear regressions. The first regression shows the regression taking only the outcome into account and the different social value orientations ("SVO"). The second regression adds the effort of the other person. The third regression adds control variables. The fourth regression restricts the sample to only the vignettes that had a negative effect on the respondents. This regression looks at the effect of effort and the difference between SVOs. The fifth regression restricts the sample to vignettes where there was no negative effect and looks at the influence of effort and the differences between SVOs.

(28)

28 5.4 Additional Findings

Respondents were asked to also indicate why they reciprocated or abstained from it. Table VI shows the reasons why individuals retaliate. The most important reason for people to not retaliate is due to the fact that retaliating would not benefit them. Between the four vignettes there are no big differences. When the outcome is negative then respondents that not retaliate state 60 percent of the time that retaliation would not benefit them, against 52 percent when the unkind action had no effect. Furthermore, roughly 30 percent stated that retaliating was too costly.

Secondly looking solely at why people do retaliate there are some differences. When the effort was high more people retaliate to return the favor (20% of the 62% that retaliates). Furthermore, around 30 percent of the people are concerned about the inequality. 33 percent states that she retaliated to diminish the inequality. This is quite stable over the four vignettes. This however is strange due to the fact in two vignettes there was no inequality. A possible explanation is that they valued the unkind action and got “even” by retaliating.

Looking at table V there is a considerable difference between high and low effort when the outcome is negative, respectively 62 percent versus 49 percent decides to retaliate. The increased amount reciprocators state that they want to return the favor or to deter the person from future unkind actions. It seems that individuals pick up on the more unkind behavior and want to correct this. Furthermore, as for the results the SVO does not seem to matter for the person’s motivation. Bot self-interested and prosocial individuals have similar motivations to retaliate and both retaliate on average 40 percent of the time.

(29)

29

The vignette:

High effort Low effort High effort Low effort

N 63 67 62 68

Percentage that retaliate 62% 49% 24% 21%

Why retaliate?

To return the "favor" 11% 6% 5% 0%

Te deter from future unkind actions 25% 15% 10% 10%

To become relative better 3% 9% 2% 3%

To minimize the inequality 21% 19% 8% 6%

Other 2% 0% 0% 1%

Total 62% 49% 24% 21%

Why not retaliate?

It was costly 11% 13% 29% 21%

It would not benefit me 22% 31% 35% 46%

Other 5% 6% 11% 13%

Total 38% 51% 76% 79%

Why retaliate?

To return the "favor" 7 7% 7 5%

Te deter Bob from future unkind actions 15 14% 24 16%

To become relative better 7 7% 4 3%

To minimize the inequality 13 12% 22 14%

Other 0 0% 2 1%

Total 42 40% 59 38%

Why not retaliate?

It was costly 23 22% 25 16%

It would not benefit me 37 35% 51 33%

Other 4 4% 19 12%

Total 64 60% 95 62%

Self-interested Prosocials

Table VI. Why (not) retaliate?

(30)

30

6. Discussion

6.1 Discussion of the Results

Individuals are reciprocators when they return a favor and retaliate when treated unfairly, also when this does not increase their personal payoff (Falk and Fischenbacher, 2006). This research shows that self-regarding individuals, that normally that are less likely to engage in positive reciprocity, engage as much in negative reciprocity as prosocials. However, self-interested people are not seen as reciprocators in general but do reciprocate. Unkind actions seem to be a stronger motivator and also seduce self-interested people to reciprocate. When deciding to reciprocate this research looks at the outcome and the effort. The findings show the importance of the outcome when deciding to engage in negative reciprocity. All the regressions in Table V show that when the outcome is negative people reciprocate more strongly. This is the case even though punishing is costly, which contradicts the standard economic theory and their rational agents. Furthermore, the social value orientation does not impact how individuals reciprocate when the outcome is negative. Hence, it seems that individuals punish more consistently regardless of their social value orientation when solely looking at the outcome. This is consistent with previous experiments form Camerer and Thaler (1995) and Abbink et al. (2000) where people exhibit more consistently negative reciprocity when faced with an unkind action that hurt them. Those experiments furthermore show that retaliation is more universal and this research shows that the SVO indeed does not matter.

The second aspect in deciding to reciprocate is evaluating how much you dislike the action. In this paper the effort heuristic is used. Kruger et al. (2004) and Morales (2006) found that people evaluate based on effort. Hence, when the effort is high it is valued higher. This research looks at the importance in effort for negative reciprocity. The effort on itself is not a significant determinant when a person decides to negatively reciprocate. However, the second

(31)

31 and fourth regression shows that when the person is prosocial then the effort does matter. Prosocial individuals reciprocate more strongly when the effort of the other person is higher. Hence, individuals with different social value orientations do differ in their decision making when they consider reciprocating. The fact that people with different SVO reciprocate differently based on the effort potentially indicates that people use the effort heuristic of Kruger et al. (2004) and Morales (2006) heterogeneously; self-interested individuals might rely on it less. This was analyzed further by doing a regression on solely the vignettes where the unkind action had no effect. In this regression it did not matter if the person exerts high or low effort at all. It seems that the individual first looks at the outcome and subsequently at the effort. When the outcome is not negative then the effort also does not matter. This is supported by the fourth regression where only the vignettes with a negative outcome are taken into account. Now only the prosocials increase their negative reciprocity when the effort is high.

Therefore, the previous conceptual model might be overly simplistic. The decision making process might be twofold (Figure III). First the individual determines if the outcome was negative. If this was not the case then most individuals both prosocials and self-regarding individuals mostly will abstain from engaging in negative reciprocity. However, when the outcome adversely impacted the individual a second step is added. The self-regarding individuals only care about the negative impact and will engage in negative reciprocity. The prosocials however look also at the effort and if the effort is high will engage in punishing also the other person’s effort. The punishment for the effort is in addition to the punishment for the negative impact.

(32)

32 6.2 Implications

The main implications are theoretical. In evaluating an unkind action an individual will not solely look at the outcome of that action but she will also look at the underlying intention of the action (Falk & Fischbacher, 2006). Hence, an unkind action, which was not successful but had a bad intention, might be despised more than an unkind action, which did hurt, but had a less harsh intention. Kruger et al. (2004) and Morales (2006) determined that individuals use the effort heuristic when valuing items. However, this research shows that the intention measured by effort in negative situations is different than in positive situations. The first contribution is that both self-regarding and prosocial individuals negatively reciprocate when the outcome affects them adversely. Secondly, only prosocial individuals take into account the intention of the other person but only when the outcome was negative. Prosocials punish additionally when the effort of the other person was high. Thirdly, many individuals engage in negative reciprocity while this is costly. Hence, self-regarding individuals are not strictly rational and are tempted to retaliate when they deem the situation as unfair to themselves.

Figure III: Conceptual model

Outcome

Negative reciprocity

Effort

Social Value

Orientation

Negative

(33)

33 6.3 Limitations

The first limitation is the method. Vignettes try to stage a realistic scenario but that people will actually behave like they say they will is always questionable. For example in this vignette the rage a person feels when she worked on a project for several months and feels that she produced the same quality but does get rewarded less hurts more when it actually happens. Hence, respondents might not reciprocate (or less) because they do not feel the actual injustice. Hence, future research could try to capture realistic scenarios where individuals observe both the effort and the outcome. Furthermore, the current setup there is only a negative outcome and no effect, respectively a 40 percent - 60 percent distribution and a 50 percent – 50 percent distribution. Hence, there were limited scenarios. In future research more outcomes can be tested were the distribution is more unfair.

Another limitation is the effort measure. First of all the effort is used to quantify the intention of the other person. However, in the used vignettes the effort was operationalized by the extent and intent of the unkind action. This might not purely measure the effort but also other social factors. Furthermore, the perceived effort is affected by the outcome. Table VI in Appendix B shows that when the outcome is negative the perceived effort is higher. Hence, in the outcome variable affects the effort variable making it difficult to distinguish what causes the decision to reciprocate. Hence, future research can more clearly describe a situation that untangles the both and effort is better measurable.

Finally, this research looks at a one-shot game. This to make sure that negative reciprocity would be irrational. However, outcomes of one-shot games are different from sequential games where you know the other person. These sequential situations would be more realistic when looking at cooperation among employees.

(34)

34

7. Conclusion

All societies have social norms on what kind of behavior needs to be rewarded and what needs to be punished. These basic principles on rewarding and punishing naturally create the tendency to reciprocate (Perugini et al., 2003). De Cremer & Van Lange (2001) find that in the case of a benevolent action, individuals reward less consistently than that they punish unkind behavior. This research shows that with unkind behavior the result of the other person’s action is crucial and not the intention. When the result of the unkind action is negative then the individual will reciprocate otherwise not. The second step of the decision to reciprocate is looking at the effort. However, only prosocials will take into account the effort when reciprocating and punish more unkind behavior more harshly.

Other papers (e.g. Falk & Fischbacher, 2006) find that fairness and intentions are important when reciprocating. However, most of the research focusses on positive reciprocity. This research shows that negative reciprocity is different. This research shows that the result of unkind behavior is crucial. Only after the result was negative, and only prosocials, take into account the amount of effort put into the unkind behavior.

(35)

35

References

Abbink, K., Irlenbusch, B., & Renner, E. (2000). The moonlighting game: An experimental study on reciprocity and retribution. Journal of Economic Behavior & Organization, 42(2), 265-277.

Atzmüller, C., & Steiner, P. M. (2010). Experimental vignette studies in survey research. Methodology: European Journal of Research Methods for the Behavioral and Social Sciences, 6(3), 128-138.

Benedikt Herrmann, Christian Thöni, Simon Gächter,* (2008) Antisocial Punishment Across Societies. Science, Vol. 319 no. 5868 pp. 1362-1367

Bogaert, S., Boone, C., & Declerck, C. (2008). Social value orientation and cooperation in social dilemmas: a review and conceptual model. British Journal of Social

Psychology, 453-480.

Bridoux, F., Coeurderoy, R., & Durand, R. (2011). Heterogeneous motives and the collective creation of value. Academy of Management Review, 711-730.

Camerer, C., & Thaler, R. H. (1995). Anomalies: Ultimatums, dictators and manners. The Journal of Economic Perspectives, 9(2), 209-219.

De Cremer, D., & Van Lange, P. A. (2001). Why prosocials exhibit greater cooperation than proselfs: The roles of social responsibility and reciprocity. European Journal of Personality, 15(S1), S5-S18.

Dehue, F. M., McClintock, C. G., & Liebrand, W. B. (1993). Social value related response latencies: Unobtrusive evidence for individual differences in information processing. European Journal of Social Psychology, 23(3), 273-293.

Falk, A., & Fischbacher, U. (2006). A theory of reciprocity. Games and Economic Behavior, 54(2), 293-315.

Fehr, E., Fischbacher, U., & Gacher, S. (2002). Strong reciprocity, human cooperation, and the enforcement of social norms. Human Nature, 1-25.

Felin, T., & Foss, N. J. (2005). Strategic organization: A field in search of micro-foundations. Strategic Organization, 3(4), 441.

Fehr, E., & Gachter, S. (1998). Reciprocity and economics: The economic implications of Homo Reciprocans. Economic European Review, 845-859.

Fehr, E., & Gächter, S. (2000). Fairness and retaliation: The economics of reciprocity. The Journal of Economic Perspectives, 14(3), 159-181.

Bridoux, F., Coeurderoy, R., & Durand, R. (2011). Heterogeneous motives and the collective creation of value. Academy of Management Review, 36(4), 711-730.

Gächter , S., & Herrmann, B. (2009). Reciprocity, culture and human cooperation: previous insights and a new cross-cultural experiment. Philosophical Transactions of the Royal Society B, 791–806.

Gottschalg, O., & Zollo, M. (2007). Interest alignment and competitive advantage. Academy of Management Review, 32(2), 418-437.

Kruger , J., Wirtz, D., Van Boven, L., & Altermatt, T. W. (2004). The effort heuristic. Journal of Experimental Social Psychology, 91-98.

Murphy, R., Ackermann, K., & Handgraaf, M. (2011). Measuring social value orientation. Available at SSRN 1804189.

Kahneman, D., Knetsch, J. L., & Thaler, R. (1986). Fairness as a constraint on profit seeking: Entitlements in the market. The American economic review, 728-741.

McClintock, C. G. (1972). Social motivation: A set of propositions. Behavioral Science, 438-454.

Perugini, M., Gallucci, M., Presaghi, F., & Ercolani, A. P. (2003). The personal norm of reciprocity. European Journal of Personality, 17(4), 251-283.

(36)

36 Wetzer, I. M., Zeelenberg, M., & Pieters, R. (2007). “Never eat in that restaurant, I did!”:

Exploring why people engage in negative word‐of‐mouth communication. Psychology & Marketing, 24(8), 661-680.

(37)

37

Appendix A: The Survey

Part I: The Vignette

1. You need to decide how much effort you want to put into explaining the situation. (1) when the outcome was negative

- Reduction: -0%, -0% | Final reward (40%, 60%) - Reduction: -1%, -6% | Final reward (39%, 54%) - Reduction: -2%, -12% | Final reward (38%, 48%) - Reduction: -3%, -18% | Final reward (37%, 42%) - Reduction: -4%, -24% | Final reward (36%, 36%) - Reduction: -5%, -30% | Final reward (35%, 30%) - Reduction: -6%, -36% | Final reward (34%, 24%) - Reduction: -7%, -42% | Final reward (33%, 18%) (2) when the outcome was neutral

- Reduction: -0%, -0% | Final reward (50%, 50%) - Reduction: -1.25%, -5% | Final reward (48.75%, 45%) - Reduction: -2.50%, -10% | Final reward (47.50%, 40%) - Reduction: -3.75%, -15% | Final reward (46.25%, 35%) - Reduction: -5.00%, -20% | Final reward (45.00%, 30%)

Rules of the game

You do not like that Bob made you look bad. Regardless of the outcome you can set this straight. How much effort do you put into doing this depends on you. The effort will cost you something.

Context

Other than the final presentation the management board cannot assess how much each one has contributed. The management is satisfied with the report and decides to give a team bonus of EUR 10,000.

Outcome

Negative. However, they reward you with only 40% of the total bonus because they believe that you contributed less.

Neutral. Furthermore, they reward both of you with 50% of the total bonus because they believe that you contributed as much.

Level of effort

High. Bob presents his part first but on purpose he covers also your main contributions and now it seems you contributed less.

Low. Bob presents his part first but in his enthusiasm he covers also some of your parts and now it seems you contributed less.

You are working with Bob on a consulting case for an international firm. This unique case requires both of your skills and in the future you will not work with Bob again. You have put both the same amount and quality of work into the final report. The last step is to present the report to the management board who also determines your bonus if applicable.

Context

Table I. The vignette

This table shows the four scenarios. The context is the same in each scenario while the level of effort (high or low) and outcome (negative or neutral) are manipulated.

(38)

38 - Reduction: -6.25%, -25% | Final reward (43.75%, 25%)

- Reduction: -7.50%, -30% | Final reward (42.50%, 20%) - Reduction: -8.75%, -35% | Final reward (41.25%, 15%)

2. To what extent do you think Bob wanted to work against you?

0 | 1 | 2 | 3 | 4 | 5

3. What was your motivation to take revenge? - To return the "favor"

- To deter Bob from future unkind actions

- To become relatively better, I lose but he loses more - To minimize the inequality

- Other

3. What was your motivation to not retaliate? - It was costly

- It would not benefit me - Other

Part II: The SVO Slider Measure

In this task you have to imagine that you are randomly paired with another person, whom we will refer to as the other. This other person is someone you do not know and who will not know who you are. Your task is to allocate money between you and this other person. For each of the following questions, please indicate the allocation you prefer most by selecting the respective option at the bottom. You can only select one allocation for each question. After you have selected your option, enter the resulting allocation of money in the question below each allocation to confirm your choice.

In the example below, a person could chose to allocate money so that she receives 50 euros, while the anonymous other person receives 40 euros (Option 5).

There are no right or wrong answers, this is all about personal preferences. As you can see, your choices influence both the amount of money you receive as well as the amount of money the other receives.

Allocation 1

Note that the top row shows what you receive and the bottom row shows what the other receives.

(39)

39 How much do you and the other receive in the allocation you chose?

You ... Other … Allocation 2

How much do you and the other receive in the allocation you chose? You ...

Other … Allocation 3

(40)

40 Allocation 4

The importance of outcome and effort in negative reciprocity.