Please fill out this short user survey of only 3 questions in order to help us improve the site. We appreciate your feedback!
Social identity, taste bias and under provisioning of public goods
Last registered on May 01, 2020


Trial Information
General Information
Social identity, taste bias and under provisioning of public goods
Initial registration date
January 13, 2020
Last updated
May 01, 2020 3:01 AM EDT

This section is unavailable to the public. Use the button below to request access to this information.

Request Information
Primary Investigator
Indian Institute of Management Bangalore
Other Primary Investigator(s)
PI Affiliation
University of California, Davis
PI Affiliation
Shiv Nadar University
PI Affiliation
ISI Delhi
Additional Trial Information
In development
Start date
End date
Secondary IDs
Data shows that public good provision in India decreases with an increase in caste based fractionalization. While past research documents this association, the causal interpretation and the deeper mechanisms driving the result is an open question. This project aims to plug this gap by identifying the precise causal connection between caste fractionalization in India and under provisioning of public goods using an artefactual field experiment. Further, it aims to examine if higher fractionalization leads to lower co-consumption of public good due to distaste for the social other. To give an example, suppose people from different castes need to coordinate to build a well or lobby with local authorities to build that well. In a more caste wise fractured village, people may be less likely to coordinate and build the well because they have a distaste for sharing the public good with outgroup members. This form of associative distaste is markedly different from free riding, the dominant theme examined in the context of public good provision. Participants from Upper Castes and Scheduled Castes play a public good game under three treatment and two compositional conditions. In Treatment 1 they play it with money, while in Treatment 2 and 3 they play the game with biriyani (a delightful self contained meal consumed in many parts of India). In Treatment 2 participants can carry the food back home while in Treatment 3 they co-consume the food at the venue. Each treatment is played with only Upper Caste, only Scheduled Caste and mixed participants. Our experimental design allows us to separate free riding and unwillingness to share owing to associative distaste as two plausible reasons behind underprovision of public goods, through lab-in-the-field experiments conducted in rural areas in the Indian state of Bihar.
External Link(s)
Registration Citation
BANERJEE, RITWIK et al. 2020. "Social identity, taste bias and under provisioning of public goods." AEA RCT Registry. May 01. https://doi.org/10.1257/rct.4047-1.1.
Sponsors & Partners
Experimental Details
Intervention Start Date
Intervention End Date
Primary Outcomes
Primary Outcomes (end points)
Contribution in the public good game.
Beliefs about how much others will contribute.
Primary Outcomes (explanation)
Secondary Outcomes
Secondary Outcomes (end points)
Secondary Outcomes (explanation)
Experimental Design
Experimental Design
The experiment will involve male participants in the age group of 18-35 years, selected from the sampled villages of Patna-Rural district of Bihar, India. Participants or subjects will be selected from Upper Caste and Scheduled Caste categories. Each session will consist of three to four groups of four individuals playing a simple public good game: every player will be given a fixed amount of money/goods and will be asked to contribute towards a common pool. The treatment variations are along two basic dimensions: the composition of the group and the nature of the good in terms of which the endowment and pay-offs are defined. We now discuss the two differentiating criteria in more details.
In homogenous sessions, all four subjects participating in the session will be from the same caste group, either Upper Caste (UC, henceforth) or Scheduled Caste (SC, henceforth). In heterogeneous sessions, two of the players will belong to the SC category, while the remaining two will be from UC category. The caste composition in each group will be common knowledge, though the game itself will be anonymously played.
A standard money based public good game cannot possibly invoke associative distaste since the act of sharing itself is absent. As a result, we introduce a variant of the public good game where subjects play with biriyani (a delightful self-contained meal with substantial aspiration value in India) instead of money. a substantial aspiration value in India. Thus, in our experiment there will be three kinds of goods in terms of which sessions will be differentiated.
Experimental Design Details
Not available
Randomization Method
Randomization will be done at the site of experiment by a computer.
Randomization Unit
Randomization will be done at the session level. Each session will be randomly assigned to one of the treatments.
Was the treatment clustered?
Experiment Characteristics
Sample size: planned number of clusters
110 sessions, each with 16 subjects.
In our case, a session will assume the role of a cluster.
Sample size: planned number of observations
1800 subjects
Sample size (or number of clusters) by treatment arms
7 sessions per treatment arm, with 16 subjects in each.
Minimum detectable effect size for main outcomes (accounting for sample design and clustering)
IRB Name
Research Ethics Committee, London School of Economics
IRB Approval Date
IRB Approval Number