WHEN CONSTRUAL LEVEL AND FINANCIAL INCENTIVE COLLIDES: AN EXPERIMENTAL STUDY ON COOPERATIVE BEHAVIOR AMONG JAVANESE AND BALINESE

This study attempted to answer the question of how the financial incentives can influence cooperative behavior in situations involving taboo tradeoffs, and the extent to which construal levels (abstract or concrete) also influences the willingness of participants to perform said tradeoffs. The study was conducted by taking three different contexts, namely a voluntary provision of tutorials from senior students to the junior (Depok study), mosque renovation activities (Mojokerto study) and renovation of temples (Bali study). The study used laboratory experimental methods for Depok study, and field experiment for Mojokerto and Bali studies. In Depok, we found that students were willing to do taboo tradeoffs and perceived them as not taboo. However, even though Mojokerto and Bali participants perceived the exchange as taboo, they were willing to do it. In general, we also found that participants tend to contribute higher taboo tradeoffs in abstract situation.


Introduction
Most of the studies in cooperative behavior consider trust and reciprocity as basic features (Putnam, 1995;Knack & Zak 2003, Knack & Keefer 1996Lazzarrini, Madalozo, & Siguera, 2005; Algan & Cahuc, 2010) of social capital, as well as network and norms. It can be inferred, therefore, that a society practicing trust, reciprocity, developing a networks system, and cultivating norms should have high social capital.
In Indonesia, there is a widely practiced value called gotong royong (lit. working together). In gotong royong, every member of a society is expected to help each other to achieve a common goal. Subejo (2004) listed the three "rules" of gotong royong: reciprocity, which means everyone provide equivalent contribution in a given time and expects compensation from it; trust, or having faith that other members follow the rules; and tightly knit, meaning that this form of cooperation operates in a specific relationship, such as family, neighborhood, or friendship, and meant to strengthen social bond among them rather than individual goals. It can be seen that Putnam's concept of social capital is embedded in gotong royong. For this reason, therefore, we postulate that as a society, Indonesians should have high social capital.
Interestingly, there are very few studies about cooperative behavior in Indonesia. Relevant studies include Scott (1976) who claimed that Javanese people were the strongest communitarian in the world, and Subejo (2004) who sees the decaying process of gotong royong among Central Java villagers due to industrialization. On the other hand, Balinese is still considered as holding onto their banjar tradition, even though there are also testimonials about how the practice is becoming unpopular among youth (Sumarta, 2009, personal communication).
The study aims to find the extent to which monetary incentive, in the form of money, influences cooperative behavior. We also seek to better understand the mechanism of cooperative behavior, especially among Javanese and Balinese. Our findings may play a part to the strengthening of Indonesia's social capital, as well as a theoretical contribution to economic psychology.
Social Dilemma and Cooperation. Some of the earliest psychological studies on social dilemma and cooperation were Dawes (1991) and De Cramer and Van Vugt (1999), and still being done until recent times (for review see Biel et al., 2008). The underlying question of these studies is why people still decided to cooperate, while not cooperating would make them better off. Initially, the conclusion was that free-riding, as the practice was called, was done not only out of selfish motivations, but also because the lack of trust. Greed and fear of not getting equal response (reciprocity) was found as factors as well. Furthermore, Fehr and Schmidt (1999) developed the inequality aversion theory, which assume that people dislike inequality, and the feeling escalate when they incur the worst loss.
Beside inequality aversion, other studies also found power as a factor influencing reciprocity (Narhetali, 2009), and the link between hierarchical power and group culture (Kopelman, 2009). In our understanding, in the context of banjar and gotong royong, the interaction between hierarchical factor and trust and reciprocity is not studied yet. There is, however, an Indonesian study that found gender and the understanding that one's wellbeing is self-sustained affect reciprocity but not trust (Narhetali, Nisa, & Syaebani, 2010).
The majority of studies on social dilemma focus on punishment to increase cooperation (Fehr & Gachter, 2000). However, it is not relevant to banjar or gotong royong, since both are regulated by tradition, and there is little room for change in tradition; in banjar, for example, a verdict is given through voting. This reduces the emergence of non-strategic factors such as anger (Rabin, 1993;Falk & Fischbacher, 2006) or inequality aversion (Fehr & Schmidt, 1999) as the basis of punishment.
There is very few research in cooperative behavior in Indonesia. Among them are Subejo (2004), which studied about gotong royong in farming communities, Warren (1999) about the differences between adat and dinas in banjar; Lietaer and De Meulenaere (2003), which analyzed banjar from an economic perspective, and a study on public goods in banjar (Veszteg & Narhetali, 2010).
Both gotong royong and banjar, as traditional cooperation mechanisms, are driven by values of non-monetary nature such as trust, reciprocity, and putting shared interests above personal ones. Therefore, the decline of cooperative behavior can be explained as a result of a shift in value, from non-monetary to monetary values.
According to the resource theory (Foa, 1971;Foa & Foa (in Foa, Converse, Tornblom & Foa, 1993), resource is defined as anything that can be transmitted in interpersonal relationship. Furthermore, "anything" in the definition refers to items, abstract or concrete, that are the object of the transaction. In addition, according to said theory, there are 6 types of reward and punishment transmitted through the exchange: love, status, information, money, goods, and service.
Given the consequences, the exchange between the six types of resources that have the effect of different and often not simple. This phenomenon is confirmed by the Fiske and Tetlock (1997) which states that taboo trade-offs is the comparative form of mental, social or any transaction which violates or values and beliefs that have long held about integrity, and even sacredness, in relations between individuals or between communities and individuals. Furthermore, people in general believe that things like love, respect or justice, or things that are pushing the primacy of values of life is sacred (sacred), which therefore would be taboo if exchanged with things who are secular or monetary, such as money (Tetlock, 2003;McGraw & Tetlock, 2005).
Exchanges between the six types of resources are differed, and often complicated. Fiske and Tetlock (1997) stated that taboo tradeoffs violate deeplyheld normative intuition about the integrity and sanctity of relationships between individuals or between a community and an individual. In general, people believe that concepts like love, respect, and justice, or sentimental objects to be taboo to be exchanged with secular or monetary objects such as money (Tetlock, 2003;McGraw & Tetlock, 2005).
On abstract and concrete objects, Liberman and Trope (2008) and Liberman and Wakslak (2007) propose construal level theory, which describes abstract and concrete concept as a function of psychological distance. The more abstract the object, the lower its construal level; accordingly, the more concrete the object, the higher its construal level. The distance in this concept is not only limited to geographic distance, but also space, time, and many other social dimension.

Research Methods
Depok Study. The practice of seniors giving academic guidance to juniors is an organized tradition among generations of psychology students in Universitas Indonesia (UI). It is seen as a voluntary act, without the need of juniors to give anything back. It is, therefore, not usual to introduce monetary transaction into the mix. This study aims to seek the effect of said transaction, in different construal levels, to cooperative behavior among psychology students in UI. We also use findings from this study as a basis for the Mojokerto and Bali studies.
Participants and Design. Participants in this study were undergraduate psychology students in UI. The study design is 2 (type of exchange: taboo v. routine, between subject) x 2 (construal level: abstract v. concrete, within subject) mixed design. They were informed about the study through posters, short messages, and Facebook. Sixty students, aged 17-22 (M = 19.175, SD = 1.054), participated in the study. Upon arrival, participants were given Rp10,000 for coming to the study ("show-up" fee), and then another Rp10,000 endowments while participating. At random, all 60 participants were assigned to two groups, those with taboo treatment and routine treatment, and placed on two different rooms to work parallelly. We had to exclude one participant from further analysis for failure of following instructions.
Materials and Procedures. The experiment was held in a 6x6 meter classroom. Participants were asked to declare their consent, and then randomly assigned to numbers which will be used as their identification throughout the study. They were also told that 1) the study is about decision making, 2) the decision that they will make is confidential, and 3) they should not discuss about their decision with fellow participants. The apparatuses were then given one by one: data control sheet, a scenario sheet, endowment fee (in an envelope), and a questionnaire. Participants were then asked to complete the sheets as instructed. Once finished, participants were debriefed. The whole process took about 30 minutes to complete.
In the taboo group, participants were given the choice of paying tutors with money. The practice is considered taboo because the tutorial program is traditionally a voluntary movement-as an expression of compassion to maintain good relation and closeness between students. Involving money would violate the sacredness of this value, as suggested by Fiske and Tetlock (1997). We also asked how participants rate how taboo the practice described in the passage is, using a Likert scale of 1 (not taboo at all) to 6 (strongly taboo). The following is the English translation of the scenario to be read by the participants in the taboo group: You are a member of the student board, and also the one responsible for the academic tutor program. This program is where seniors can provide valuable academic tutorial to their juniors, especially around exam season, out of compassion. Right now, you are holding a meeting with juniors, or tutees, to hear about how the program went for them. Generally, the tutees are satisfied with the program, since they reported good score in tests. Upon seeing this good result, the board's chairman proposed the idea of paying some money to tutors. This practice has never been done before, so there is no budget allocation available. The solution is to ask tutees to pay for the program. The chairman also urges board members, including you, to contribute as well.
Next, participants were given an envelope containing Rp20.000, a combination of their show-up and endowment fee. Participants were then asked to state, by writing in a paper, how much money they will contribute if they were either a tutee (abstract situation) or a board member (concrete situation). They can contribute none, all, or only a part of their money.
The same procedure was applied to the routine group. The following is the English translation of the scenario to be read by the participants in the routine group: You are a member of the student board, and also the one responsible for the academic tutor program. Mojokerto Study. In 2006, there was a school-building project in Mojokerto, East Java, which end up costing Rp125 million instead of the projected Rp50 million. The drastic increase was possible because the people of Padang Asri District, where the school was built, decided to contribute not only manpower, but also materials. This is not a common form of gotong royong, but still embodies its principles. This is the rationale of choosing Padang Asri District as the site for this study.
Participants and Design. The study design is 2 (type of exchange: taboo v. routine, between subject) x 2 (construal level: abstract v. concrete, within subject) mixed design. Participants of this study were 78 locals from Padang Asri District, with ages range 15-52 years (M = 24,961, SD = 9,636). They were recruited during a pengajian (Islamic religious gathering) in the village hall. The recruitment was done in two days, with the help from local community leaders as well as a school principal, which proven effective since people in Padang Asri understand Javanese language better than Indonesian. Group assignment was done by placing participants who came in the first day to taboo group, and those who came in the second to routine group. Thirty-six participants were in the taboo group. Like participants in the Depok study, they also received Rp10,000 showup fee and another Rp10,000 as endowment.
Materials and Procedures. The experiment was held in the village hall.
Participants were asked to declare their consent, and then randomly assigned to numbers which will be used as their identification throughout the study. They were also told that 1) the study is about decision making, 2) the decision that they will make is confidential, and 3) they should not discuss about their decision with fellow participants. The apparatuses were then given one by one: data control sheet, a scenario sheet, endowment fee (in an envelope), and a questionnaire. Participants were then asked to complete the sheets as instructed. Once finished, participants were debriefed. The whole process took about 45 minutes to complete.
Like in the Depok study, participants in both groups were asked to read a passage and to write a certain amount of monetary contribution. The following is the English translation of the scenario to be read by the participants in the taboo group in Mojokerto: After reading the scenario, participants were also asked to rate how taboo the activity was. Then, they were given Rp20.000 to be used in the study. They were then asked to write down how much money they would contribute if they were the head of committee (abstract situation) and villager (concrete situation).
Bali Study. In Bali, there is an additional layer of official government called Banjar. Administratively, Banjar is positioned under administrative village (Kelurahan). However, unlike formal government, a Banjar is more socially motivated, with things organized included religious events, land sales, and, most relevant to this study, temple renovation. Details about the subject can be seen in Veszteg & Narhetali (2010) Participants and Design. The study design is 2 (type of exchange: taboo v. routine, between subject) x 2 (construal level: abstract v. concrete, within subject) mixed design. Participants in this study were 74 residents (aged 21-60, M = 38.067, SD = 9,213) from three different villages in Bali: Nyuhtebel, Karang Asem; Padang Bai, Klungkung; and Bedugul, Tabanan. Much like in Mojokerto, we received help from a local, who also happen to be in the research team. Participants received Rp10,000 for coming to the study and Rp10,000 endowment fee.
Materials and Procedures. Similar to the Mojokerto study, this study was held in the village hall. However, in Padang Bai, we had to do the experiment near the locals' workplace for they are at work at the time of data collection. Participants were asked to declare their consent, and then randomly assigned to numbers which will be used as their identification throughout the study. They were also told that 1) the study is about decision making, 2) the decision that they will make is confidential, and 3) they should not discuss about their decision with fellow participants. The apparatuses were then given one by one: data control sheet, a scenario sheet, endowment fee (in an envelope), and a questionnaire. Participants were then asked to complete the sheets as instructed. Once finished, participants were debriefed. The whole process took about 45 minutes to complete.
The procedure was the same as Depok and Mojokerto studies, where participants in both groups were asked to read the scenario and write a certain amount of monetary contribution. The following is the English translation of the scenario to be read by the participants in the taboo group in Bali: After reading the scenario, participants were also asked to rate how taboo the activity was. Then, they were given Rp20.000 to be used in the study. They were then asked to write down how much money they would contribute if they were the head of committee (abstract situation) and villager (concrete situation).

Results and Discussion
Depok Study: Manipulation Check. The assumption in this study is that the Indonesian people are now experiencing a shift from the values of collectivity, cooperation, and mutual aid to more materialistic values. Therefore, it is no longer taboo for participants in Depok to give or to receive money in activities that require mutual cooperation (i.e volunteering to be tutors). Manipulation check results supported this assumption by showing that giving money to the tutors in order to increase the desire to continue tutoring was considered "slightly taboo" (M = 2.423, SD = 1.06).
Depok Study: Types of exchanges. Our analysis showed that there was no significant difference between taboo and routine condition (F = .218, p <.643, ηp2 = .004). This is most likely due to participants' perceptions that the exchange was not taboo. However, given the very low number of partial eta squared acquired (= .004), the small sample size can also be a factor causing these differences are not significant.
On the other hand, there is a significant difference in average contributions made in both abstract and concrete conditions (F = 4783, p <.05, ηp2 = .08). Interestingly, participants in abstract-taboo condition contributed significantly more than participants in abstract-routine condition. While on the contrary, when presented with concrete situation, participants in routine condition contributed significantly more than participants in taboo condition.
This results indicate the presence of emotional strain in participants when making decision containing elements of taboo in concrete condition. This emotional strain seems to actually encourage the opposite response in the abstract condition, so that the average contribution of the participants for making decisions that contain elements of taboo in the abstract condition even greater than on purely routine decisions.

Figure 1. Average Contribution in Depok Study
Analysis of Motivation. Analysis of participants' motivation to contribute conducted qualitatively by two people with a method of inter-raters. In abstracttaboo condition, participants showed varied motivations, namely: 1) being the more benefited party, 2) being the indirectly benefited, 3) rational reasons, that when the person considers market price before contributing, and the fact that other people will contribute as well.
The motivation of participants in concrete-taboo conditon were identified as two major answers:1) rational reason, and, and 2) reward, this motivation has not yet appeared in the previous category, where the participants felt the urge to compensate justly for the tutor and as a symbol of gratitude. Rational motivations as participants in concretetaboo condition: 1) rational and 2) reward.

Figure 2c. Motivation in Abstract-Taboo Condition
In abstract-taboo condition, participants' motivations were categorized in 5 major answers: 1) they considered themselves as the benefited party, or 2) as indirectlybenefited part, 3) rational reasons, 4) fairness (mutual benefit tradeoff between tutor and tutee), and 5) reward (i. e. by contributing, the person feels like saying thank you).
Mojokerto Study: Manipulation Check. Manipulation check results showed that giving money to villagers in order to increase the desire to volunteer in mosque renovation was considered "somewhat taboo" (M = 3.22, SD = 1.551).
Mojokerto Study: Types of Exchanges. The average contribution from participants in routine group was significantly lower than their taboo counterpart (F = 4.249, p <.05, ηp2 = .053). Such result might occur because, as indicated in the manipulation check, participants regard the exchange as taboo. However, given the very low number of partial eta squared acquired (= .004), the small sample size can also be a factor causing these differences are not significant On the other hand, unlike the Depok study, there was no difference between the contributions made in abstract and concrete situations (F = 3529, p> .05, ηp2 = .044). However, like the Depok study, average contribution to taboo exchange was higher when done in abstract situation-the opposite dynamics was in play in concrete situation. This finding aligns with our tension theory from before, that it happened when participants faced taboo exchange in concrete situation. The tension encouraged an opposite response, that is to contribute higher.
However, the same trends, namely that in an abstract context, contributing to the condition taboo higher than routine conditions. While on the contrary, in the context of concrete, contributing to a higher condition than routine conditions taboo.

Figure 3. Average Contribution in Mojokerto Study
Bali Study: Manipulation Check. Manipulation check results showed that giving money to villagers in order to increase the desire to volunteer in temple renovation was considered "somewhat taboo" (M = 3.926, SD = 1.967).
Bali Study: Types of Exchange. There was no difference of average contribution between taboo and routine groups (F = 3.406, p> .05, ηp2 = .045). Such result might occur because, as indicated in the manipulation check, participants regard the exchange as taboo. However, given the very low number of partial eta squared acquired (= .045), the small sample size can also be a factor causing these differences are not significant. Larger group could potentially make a clearer distinction.

Figure 5a. Contribution Comparisons Between Mojokerto and Bali Studies
Results and Interactions: Mojokerto & Bali Studies. Participants in Bali study showed significantly higher contribution in both abstract and concrete condition than Mojokerto. (F = 6.903, p <.05, ηp2 = .045). This implies that Bali participants were more generous in terms of contributing for the benefit of the community. This was likely to happen because of the Balinese ngayah philosophy, meaning to expect nothing but God's blessing. It was proven by the higher manipulation check score than Mojokerto participants, which means that Bali participants consider the monetary exchange as more taboo.

Figure 5b. Contribution Comparisons Between Mojokerto and Bali Studies
The average contribution from Bali participants in both abstract and concrete situation was significantly higher than Mojokerto (F = 6.903, p <.05, ηp2 = .045). This implies that they were more generous Mojokerto participants in terms of contribution in the context of collaborative action for the benefit of the community. Both groups of participants also experienced the tension effect, which lead them to contribute more in taboo exchange when the situation is concrete.

Conclusions
In Depok, it was shown that participants did not consider volunteerism in tutorial activities as sacred and need to be maintained; the majority of students saw the act of exchanging tutorial and money was seen as not taboo. However, originally, the students agree that monetary exchange violates the value rooted in the tutorial program, but when the exchange is framed as a way to appreciate the time and effort given by the tutors, it becomes acceptable. This may show that financial incentive is no longer considered taboo, and more accepted as a means of exchange. This phenomenon can be explained by the fact that the students live in a city, where material needs are normally high. On the other hand, taken to a community context, the opposite occurred. In Mojokerto, regard financial incentive as slightly taboo, while the Balinese were no different.
From the amount of contribution given, there was no difference found between taboo and routine groups among UI students, both in abstract and concrete situations. This corresponds to our previous finding where the students regard financial incentive as not taboo. Likewise, Mojokerto and Bali participants showed similar amount of contributions between the two groups. There are two possible explanations: failed manipulation and a shift to a more materialistic lifestyle. However, we believe the latter to be more probable, since 1) manipulation check result revealed that both Mojokerto and Bali participants consider giving financial incentive to volunteers as taboo, although only slightly. In addition, our motivation analysis stated that participants viewed the incentive was reasonable because volunteers need it. On the other hand, those who reject the idea did so because they consider the renovation as a holy deed, hence they should do it only to seek God's blessing, not for financial reasons.
The amount of contribution in Depok and Mojokerto were significantly higher in abstract than concrete situation. However, it was not the case in Bali. A plausible explanation is because for Balinese, temple renovation is more of a form of worship than social activity. Therefore, they were indifferent towards whether they were the head of committee (abstract situation) or a villager (concrete situation).
Another interesting finding is that participants in Depok, Mojokerto, and Bali tend to contribute higher in abstract-taboo situation than abstract-routine. This could probably happen because in abstract situation, people see the problem (taboo tradeoffs) less clearly. On the other hand, in concrete situation, participants contribute more on routine than taboo exchange. These results implied that there are taboo beliefs that held participants from contributing higher-because when they see taboo tradeoffs clearly, they tend to opt for the "safer" decision that is routine tradeoff.