Skip to main content
Log in

Multidimensional social identity and redistributive preferences: an experimental study

  • Published:
Theory and Decision Aims and scope Submit manuscript

Abstract

Social identity is embedded in social structures, generated by various social processes, and has multiple dimensions. We report findings from a laboratory experiment eliciting two-dimensional social identities: a horizontal identity determined either randomly or by preferences and a vertical identity defined by income status and determined either by luck or performance. We also vary income gaps between vertical identity groups. Participants make redistributive allocation decisions between two beneficiaries differing in identity attributes. We find robust evidence of in-group favoritism and that both the identity distance between the allocator and the in-group recipient and income gaps influence the degree of in-group favoritism.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Institutional subscriptions

Fig. 1
Fig. 2
Fig. 3
Fig. 4
Fig. 5

Similar content being viewed by others

Notes

  1. Section 2 also reviews the existing studies with multiple identities and discusses the relation of our paper to these studies.

  2. It is worth making a caveat that in reality, most group identities carry more or less hierarchical meanings. There perhaps does not exist a group identity from which any notion of hierarchy or social stratification is purged. However, some group identities, such as membership in hobby or sports clubs, carry little and less hierarchical meanings than others. Even if there is such a notion associated with these identities, it often relies on a subjective and biased interpretation or perception of the beholders. For example, in our experiment, we cannot stop a subject who prefers Klee’s paintings from thinking the Klee team is superior to the Kandinsky team. On the other hand, some group identities carry obvious hierarchical meanings even from a neutral and objective perspective. For example, the vertical identity in our setup, which is based on income level, unequivocally carries a social stratification. The high-income group is always considered having a higher social stratum. All in all, while no social identity in reality is completely exempt from any hierarchical meaning, our classification of horizontal and vertical identities is based on a comparative/objective sense. We call those group identities that carry little hierarchical meaning or do not carry any notion of hierarchy from an objective perspective horizontal identities, and those group identities with obvious hierarchical meanings vertical identities.

  3. Essentially each subject must make a distributive allocation decision just like what a social planner does. Although, on paper, a social planner should always make impartial allocations, in practice, it is not always guaranteed. Social planners often exhibit preferences or bias towards particular groups at the expense of other groups. Moreover, the redistributive policy is often shaped by public opinions. The resource allocation task in our experiment can be interpreted as an elicitation of individual taxpayers’ redistributive preferences. In our experiment, the allocation to a group depends on the aggregated allocation decisions made by the others.

  4. On a separate note, in a large replication study done by Camerer et al. (2016), the results shown by Chen & Chen (2011) failed to replicate. However, in a very recent paper, Chen et al. (2020) showed that the original results could be replicated as long as the same identity inducement procedures and experimental protocols are followed. These contrasting results highlight an interesting point. A (minor) deviation from the original identity inducement protocol is often enough to produce results that deviate from the original ones.

  5. It is worth noting that in our paper, we simply define higher allocation towards in-groups than towards out-groups as in-group favoritism. That is, in-group favoritismis is defined in a relative/comparative sense. A caveat of this definition is that we are not able to distinguish in-group favoritism and out-group negativity. These two discriminatory attitudes, although are similar, have some subtle differences. To be able to tease them out, we would need to run a neutral baseline, with which we can compare both in-group and out-group allocation decisions.

  6. For the grouping based on the horizontal dimension we use “team” rather than “group” to label different horizontal identity attributes. We will use “group” rather than “team” in labeling vertical identity attributes to avoid confusion.

  7. Note that the grouping is based on relative preferences. Subjects were also informed of this piece of information before they made decisions. Please refer to Online Appendix A.5 for more details. We find that subjects in general prefer Kandinsky to Klee. In total 39 out of the 48 subjects from treatment Choice_Random_KK and Choice_Random_XY chose at least 3 Kandinsky paintings in the five pairs. Among these 39 subjects, 15 were assigned to Team Klee. We include a dummy variable indicating this “misassignment” in our regression analysis, and do not find any evidence showing these “misassigned” subjects behave differently from the rest. Note that subjects were not informed of how many Klee/Kandinsky paintings they chose.

  8. On average a subject completed 13.55 questions within the 90 seconds. The best subject completed 23 questions correctly.

  9. Throughout our experiment, income is expressed in terms of experimental tokens. At the conclusion of the experiment, the total experimental tokens earned are converted into the Singapore Dollars equivalent based on our conversion rate.

  10. Experimental instructions, which contain details of the procedure of identity inducement, were read aloud to participants to ensure that all participants have common knowledge on the experiment, including the way their identity attributes were to be determined.

  11. It is important to note that the allocation decision does not affect the allocator’s payoff. The task is essentially a disinterested dictator game (DGG).

  12. Note that the comparisons can also be made for settings where the two beneficiaries differ in both identity attributes rather than only one like the one in our setup. There are four of such settings. Incorporating these settings would require us to run many more experimental sessions without any guarantee of new insights beyond those presented in this paper. We, therefore, opted to exclude such comparisons in this paper.

  13. The within-subject design may give subjects certain cues about the experimental objectives, and thus there might be an experimenter demand effect (EDE). Nevertheless, the subjects did not receive any pressure from peers nor from the experimenters. The experimental instructions did not provide any hint about how to play the game. Thus, according to the classification of Zizzo (2010), the potential EDE in our experiment is purely cognitive, and not social. Zizzo (2010) argues that purely cognitive EDE are typically weak. The within-subject design also has an advantage in that it gives us a richer dataset in a cost-effective way, especially in view of the fact that we have adopted the between-subject design for identity inducement and enhancement. If we opted for a between-subject design for various income-gap scenarios as well, we would need many more subjects in order to get a similar data size. Meanwhile, purely cognitive EDE also has its bright side: it helps subjects in task construal (Zizzo, 2010). For the above reasons, we opted for the within-subject design to elicit allocation decisions under various income-gap scenarios.

  14. In most of the treatments, the ratio of male subjects is close or equal to \(\frac{1}{2}\). One exception is Treatment Choice_Random_XY, where we have more female subjects (17) than male (7). However, our estimation of any treatment variable will not depend on any single treatment. For instance, to investigate the effect of identity enhancement at the horizontal dimension versus that done at the vertical dimension, we can compare Treatment Choice_Random_KK with Choice_Random_XY, Treatment Random_Performance_KK with Random_Performance_XY, or Treatment Random_Random_KK with Treatment Random_Random_XY. Thus, the high ratio of female subjects in Treatment Choice_Random_XY is unlikely to confound our estimation of treatment effects. Moreover, our regression analysis will control for gender.

  15. The psychological differences between the randomly determined horizontal identity and the painting-preference-induced horizontal identity might be weak in our experimen̄t and in Chen & Li (2009).

  16. A related study on peer effects by van Veldhuizen et al. (2018) shows that, in an environment where individuals can observe the productivity of co-workers, peer effects exist, but the direction varies across individuals. Therefore, an earlier finding that low-productivity workers perform better when they are observed by high-productivity co-workers was not replicated. The sensitivity of results to the underlying environment is also present in the studies of group identity, where some results are not replicated. See also Footnote 4 for a discussion. In a recent paper, Guala and Filippin (2017) contribute to the debate on this issue. In their experiment, the effect of group identity varies unsystematically across different experimental games and complexity of the decision tasks. This suggests that it is difficult to find a common standardized methodology for identity inducement that works best relative to other methodologies.

  17. It would look messy if the right panel plots the lines by treatment, as in the left panel of Fig. 5, because there would be eighteen lines displayed.

  18. A likelihood-ratio test on this extension shows the effect indeed varies between allocation decisions (p value\(< 0.01\)).

  19. We also include various control variables, including the allocator’s gender, the number of the allocator’s correct answers in Stage 2, the allocator’s team membership (Klee or Kandinsky), whether the allocator was “misassigned” to this team as discussed in Footnote 7, and whether the allocation decision is the first one in Table 3. In all of our regressions, the coefficients of these variables are not significant at conventional levels.

  20. Since scenario is decomposed to \(scenario\times overkk\) (allocations over the horizontal identity), \(scenario\times D_{ingroup>outgroup}\) and \(scenario\times D_{ingroup<outgroup}\) (allocations over the vertical identity), the variable overkk is dropped in the regression.

References

  • Afridi, F., Li, S. X., & Ren, Y. (2015). Social identity and inequality: the impact of Chinas hukou system. Journal of Public Economics, 123, 17–29.

    Article  Google Scholar 

  • Akerlof, G. A., & Kranton, R. E. (2000). Economics and identity. Quarterly Journal of Economics, 115(3), 715–753.

    Article  Google Scholar 

  • Akerlof, G. A., & Kranton, R. E. (2005). Identity and the economics of organizations. Journal of Economic Perspectives, 19(1), 9–32.

    Article  Google Scholar 

  • Akerlof, G. A., & Kranton, R. E. (2008). Identity, supervision, and work groups. American Economic Review, 98(2), 212–217.

    Article  Google Scholar 

  • Alesina, A., & La Ferrara, E. (2005). Preferences for redistribution in the land of opportunities. Journal of Public Economics, 89(5), 897–931.

    Article  Google Scholar 

  • Alesina, A., & Perotti, R. (1996). Income distribution, political instability, and investment. European Economic Review, 40(6), 1203–1228.

    Article  Google Scholar 

  • Angerer, S., Dutcher, E. G., Glätzle-Rützler, D., Lergetporer, P., & Sutter, M. (2017). Gender differences in discrimination emerge early in life: evidence from primary school children in a bilingual city. Economics Letters, 152, 15–18.

    Article  Google Scholar 

  • Benjamin, D. J., Choi, J. J., & Strickland, A. J. (2010). Social identity and preferences. American Economic Review, 100(4), 1913–1928.

    Article  Google Scholar 

  • Cadsby, C. B., Servátka, M., & Song, F. (2013). How competitive are female professionals? A tale of identity conflict. Journal of Economic Behavior and Organization, 92, 284–303.

    Article  Google Scholar 

  • Camerer, C. F., Dreber, A., Forsell, E., Ho, T.-H., Huber, J., Johannesson, M., et al. (2016). Evaluating replicability of laboratory experiments in economics. Science, 351(6280), 1433–1436.

    Article  Google Scholar 

  • Charness, G., Cobo-Reyes, R., & Jiménez, N. (2014). Identities, selection, and contributions in a public-goods game. Games and Economic Behavior, 87, 322–338.

    Article  Google Scholar 

  • Charness, G., Rigotti, L., & Rustichini, A. (2007). Individual behavior and group membership. American Economic Review, 97(4), 1340–1352.

    Article  Google Scholar 

  • Chen, R., & Chen, Y. (2011). The potential of social identity for equilibrium selection. American Economic Review, 101(6), 2562–2589.

    Article  Google Scholar 

  • Chen, R., Chen, Y., & Riyanto, Y. E. (2020). Best practices in replication: a case study of common information in coordination games. Experimental Economics, 24, 1–29.

    Google Scholar 

  • Chen, Y., & Li, S. (2009). Group identity and social preferences. American Economic Review, 99(1), 431–57.

    Article  Google Scholar 

  • Chen, Y., Li, S. X., Liu, T. X., & Shih, M. (2014). Which hat to wear? Impact of natural identities on coordination and cooperation. Games and Economic Behavior, 84, 58–86.

    Article  Google Scholar 

  • Chuah, S.-H., Hoffmann, R., Ramasamy, B., & Tan, J. H. W. (2014). Religion, ethnicity and cooperation: an experimental study. Journal Of Economic Psychology, 45, 33–43.

    Article  Google Scholar 

  • Collier, P. (2000). Economic causes of civil conflict and their implications for policy. Washington: World Bank.

    Google Scholar 

  • Cramer, C. (2003). Does inequality cause conflict? Journal of International Development, 15(4), 397–412.

    Article  Google Scholar 

  • Cramer, C. (2005). Inequality and conflict: a review of an age-old concern. Geneva: United Nations Research Institute for Social Development.

    Google Scholar 

  • Deffains, B., Espinosa, R., & Thöni, C. (2016). Political self-serving bias and redistribution. Journal of Public Economics, 134, 67–74.

    Article  Google Scholar 

  • Eckel, C. C., & Grossman, P. J. (2005). Managing diversity by creating team identity. Journal of Economic Behavior and Organization, 58(3), 371–392.

    Article  Google Scholar 

  • Engelmann, D., & Strobel, M. (2004). Inequality aversion, efficiency, and maximin preferences in simple distribution experiments. American Economic Review, 94(4), 857–869.

    Article  Google Scholar 

  • Fischbacher, U. (2007). Z-tree: Zurich toolbox for ready-made economic experiments. Experimental Economics, 10(2), 171–178.

    Article  Google Scholar 

  • Fong, C. (2001). Social preferences, self-interest, and the demand for redistribution. Journal of Public Economics, 82(2), 225–246.

    Article  Google Scholar 

  • Gioia, F. (2017). Peer effects on risk behaviour: the importance of group identity. Experimental Economics, 20, 100–129.

    Article  Google Scholar 

  • Grosskopf, B., & Pearce, G. (2017). Discrimination in a deprived neighbourhood: an artefactual field experiment. Journal of Economic Behavior and Organization, 141, 29–42.

    Article  Google Scholar 

  • Grosskopf, B., & Pearce, G. (2020). Do you mind me paying less? Measuring other-regarding preferences in the market for taxis. Management Science, 66(11), 5059–74.

    Article  Google Scholar 

  • Guala, F., & Filippin, A. (2017). The effect of group identity on distributive choice: social preference or heuristic? The Economic Journal, 127(602), 1047–1068.

    Article  Google Scholar 

  • Guala, F., Mittone, L., & Ploner, M. (2013). Group membership, team preferences, and expectations. Journal of Economic Behavior and Organization, 86, 183–190.

    Article  Google Scholar 

  • Hargreaves Heap, S. P., & Zizzo, D. J. (2009). The value of groups. American Economic Review, 99(1), 295–323.

    Article  Google Scholar 

  • Hewstone, M., Rabiul Islam, M., & Judd, C. M. (1993). Models of crossed categorization and intergroup relations. Journal of Personality and Social Psychology, 64(5), 779–93.

    Article  Google Scholar 

  • Holm, S. (1979). A simple sequentially rejective multiple test procedure. Scandinavian Journal of Statistics, 6(2), 65–70.

    Google Scholar 

  • Klor, E. F., & Shayo, M. (2010). Social identity and preferences over redistribution. Journal of Public Economics, 94(3–4), 269–278.

    Article  Google Scholar 

  • Kranton, R., Pease, M., Sanders, S., & Huettel, S. (2020). Deconstructing bias in social preferences reveals groupy and not-groupy behavior. 117 35, Proceedings of the National Academy of Sciences, pp. 21185–21193.

  • Kranton, R., & Sanders, S. (2017). Groupy versus non-groupy social preferences: personality, region, and political party. American Economic Review, 107, 65–69.

    Article  Google Scholar 

  • Ku, H., & Salmon, T. C. (2013). Procedural fairness and the tolerance for income inequality. European Economic Review, 64, 111–128.

    Article  Google Scholar 

  • Martinangeli, A., & Martinsson, P. (2020). We, the rich: inequality, identity and cooperation. Journal of Economic Behavior and Organization, 178, 249–266.

    Article  Google Scholar 

  • Moffatt, P. G. (2015). Experimetrics: econometrics for experimental economics. London: Macmillan International Higher Education.

    Google Scholar 

  • Morita, H., & Servátka, M. (2013). Group identity and relation-specific investment: an experimental investigation. European Economic Review, 58, 95–109.

    Article  Google Scholar 

  • Mujcic, R., & Frijters, P. (2013). Still not allowed on the bus: it matters if you’re black or white!. Discussion Paper 7300, IZA Discussion Paper.

  • Nafziger, E. W., & Auvinen, J. (2002). Economic development, inequality, war, and state violence. World Development, 30(2), 153–163.

    Article  Google Scholar 

  • Paetzel, F., & Sausgruber, R. (2018). Cognitive ability and in-group bias: an experimental study. Journal of Public Economics, 167(C), 280–292.

    Article  Google Scholar 

  • Piketty, T. (1995). Social mobility and redistributive politics. Quarterly Journal of Economics, 110(3), 551–584.

    Article  Google Scholar 

  • Shih, M., Pittinsky, T. L., & Ambady, N. (1999). Stereotype susceptibility: identity salience and shifts in quantitative performance. Psychological Science, 10(1), 80–83.

    Article  Google Scholar 

  • Skrondal, A., & Rabe-Hesketh, S. (2004). Generalized latent variable modeling: multilevel, longitudinal, and structural equation models. London: Chapman and Hall/CRC.

    Book  Google Scholar 

  • Stangor, C., Lynch, L., Duan, C., & Glas, B. (1992). Categorization of individuals on the basis of multiple social features. Journal of Personality and Social Psychology, 62(2), 207–218.

    Article  Google Scholar 

  • Tajfel, H., & Turner, J. C. (1979). An integrative theory of intergroup conflict. The social psychology of intergroup relations. (W. G. Austin & S. Worchel, Eds.) (p. 33). USA: Taylor & Francis Psychology Press.

  • Tremewan, J. (2010). “Group identity and coalition formation: Experiments in a three-player divide the dollar game,” CEPREMAP Working Papers (Docweb) 1020, CEPREMAP.

  • van Veldhuizen, R., Oosterbeek, H., & Sonnemans, J. (2018). Peers at work: evidence from the lab. PLOS One, 13(2), e0192038.

    Article  Google Scholar 

  • Weng, Q., & Carlsson, F. (2015). Cooperation in teams: the role of identity, punishment, and endowment distribution. Journal of Public Economics, 126, 25–38.

    Article  Google Scholar 

  • Zizzo, D. J. (2010). Experimenter demand effects in economic experiments. Experimental Economics, 13, 75–98.

    Article  Google Scholar 

Download references

Author information

Authors and Affiliations

Authors

Corresponding author

Correspondence to Fuhai Hong.

Additional information

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

We thank two anonymous referees and seminar and conference participants at Nanyang Technological University, NUS Behavioral and Experimental Economics Reading Group, the 5th Annual Xiamen University International Workshop on Experimental Economics, 2016 ESA European Meeting (Bergen), 2016 North-American ESA Conference (Tucson), 2018 ESA World Meeting (Berlin), and 2019 WEAI International Conference (Tokyo) for helpful comments. We acknowledge financial support from the Singapore MOE Academic Research Fund Tier 1 Grant.

Supplementary Information

Below is the link to the electronic supplementary material.

Supplementary file1 (PDF 103 KB)

Rights and permissions

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Hong, F., Riyanto, Y.E. & Zhang, R. Multidimensional social identity and redistributive preferences: an experimental study. Theory Decis 93, 151–184 (2022). https://doi.org/10.1007/s11238-021-09834-z

Download citation

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s11238-021-09834-z

Keywords

Navigation