Who Do People Cooperate For? Experimental Measures of Cooperation Behavior and Organizational Performance

How to cite

Nigus, Halefom Yigzaw; Abay, Kibrom; Meles, Tensay Hadush; Abate, Gashaw T. 2026. "Who Do People Cooperate For? Experimental Measures of Cooperation Behavior and Organizational Performance." AEA RCT Registry. AEARCTR-0018882. https://doi.org/10.1257/rct.18882-1.0

Download: BibTeX · RIS — for reference managers without a browser translator. Zotero and most others pick up this page's citation metadata directly.

Halefom Yigzaw Nigus* , Kibrom Abay , Tensay Hadush Meles , Gashaw T Abate

* Primary investigator

Registry ID
AEARCTR-0018882
DOI
https://doi.org/10.1257/rct.18882-1.0
First registered
Last updated
Status
In development
Trial dates
2026-06-08 – 2026-09-30
Countries
Private
JEL codes
C91; C92; C71; D71; H41
Registry record
http://www.socialscienceregistry.org/trials/18882

Abstract

Group-based organizations such as youth agribusiness associations, savings and credit groups, cooperatives, and forest user groups are central instruments of development policy. Their success depends on whether members and leaders are willing to forgo private resources for collective goals, but this form of cooperation is difficult to enforce or observe directly in real organizations. This study uses an incentivized public goods game (PGG) with Youth Agricultural Innovation Groups (YAIGs) in Ethiopia to elicit and compare different measures of cooperation among members and link these measures to real organizational outcomes. To assess relative importance of cooperative leaders and members, we elicit levels of cooperation among leaders and members. To evaluate and compare cooperation for various causes and motives, we randomly assign 183 youth groups to one of three experimental arms for the member experiment. Members are anonymously paired with three other members within their youth organization and decide how much of an endowment to contribute. In the first arm (Standard PGG arm), the total contribution of the four members is doubled and shared equally between them. In the second arm (YAIG contribution), the total contribution is doubled and transferred to the participant’s own youth organization. In the third arm (Forest contribution arm), the total contribution is doubled and transferred to a broader environmental public good, the Yayu Forest Coffee Biosphere Reserve. We first compare these experimentally elicited cooperations across these three arms, and then examine which measures predict pre-specified organizational outcomes, including economic performance, financial capacity, governance quality, organizational stability, and internal conflict. By varying the beneficiaries of cooperation and eliciting cooperation from both leaders and ordinary members, we provide a large-scale test of the external validity of experimental cooperation measures and identify which forms of cooperation best predict the success of real world organizations.

Experimental design

This section describes the experimental design of the three variants of the incentivized public goods game (PGG). In this study, YAIG members, including leaders, participate in the experiment. Each participant receives an initial endowment of 200 Birr and is randomly matched with three other anonymous participants from the same YAIG. At the time of the study, $1 approximately 155 Ethiopian Birr. Participants privately decide how much of the 200 Birr to keep for themselves and how much to contribute to a group account. Contributions can range from 0 to 200 Birr in increments of 20 Birr (0, 20, 40, 60, 80, 100, 120, 140, 160, 180, and 200 Birr). The amount kept directly benefits the participant. The amount contributed to the group account is doubled, but the beneficiary of the doubled contribution differs across the three treatment arms.

The experiment follows YAIG level clustered randomization. The 183 YAIGs are randomly assigned to one of three experimental arms, stratified by kebele, as there are multiple YAIGs in each kebele. All participating members and leaders within a YAIG are assigned to the same experimental arm. With each YAIG, participants are anonymously matched into four-person groups for the contribution decision.

Arm 1: Standard PGG arm
In this arm, participants are randomly paired with three other anonymous participants from the same YAIG. Each participant decides how much of the 200 Birr endowment to keep and how much to contribute to the group account. The experimenter doubles the total amount
contributed by the four matched participants and shares the doubled amount equally among them. A participant’s payoff in this arm equals the amount kept for oneself plus one-fourth of the doubled group account.

This arm is a standard, context-free PGG with a social dilemma: contributing is socially beneficial but privately costly. With four participants and a multiplier of two, each Birr contributed returns 0.5 Birr to the contributor but increases total group earnings by 2 Birr. Of the 183 YAIGs, 62 are randomly assigned to this arm.

Arm 2: YAIG contribution arm
In this arm, the contribution is framed as a contribution to the participant’s YAIG account. Participants are anonymously matched with three other participants from the same YAIG and decide how much of the 200 Birr endowment to keep and how much to contribute. The total contribution of the four matched participants is doubled and transferred to the participant’s own YAIG account. Participants directly receive only the amount they keep for themselves, while the contribution benefits the YAIG as an organization. This arm captures cooperation
directed toward an identifiable in-group institution, where benefits accrue to members of the participant’s own youth organization. Of the 183 YAIGs, 61 are randomly assigned to this arm.

In Arm 3: Forest contribution arm
In this arm, the contribution is framed as support for a local forest reserve, the Yayu Forest Coffee Biosphere Reserve. Participants are anonymously matched with three other participants from the same YAIG and decide how much of the 200 Birr endowment to keep and how much to contribute. The total contribution of the four matched participants is doubled and transferred to support conservation activities in the reserve. Participants directly receive only the amount they keep for themselves, while the contribution supports a broader environmental and social public good. This arm captures cooperation directed toward a public good that extends beyond the participant’s immediate youth organization. Of the 183 YAIGs, 60 are randomly assigned to this arm.

Design details

Randomization method
The randomization is conducted at the YAIG level. Each YAIG is assigned to one of the treatment arms. A total of 183 YAIGs are randomly assigned into three groups, with stratification at kebele level. Within each YAIG, participants are randomly paired with three other YAIG members for the game. Pairing is done using a pre-generated random list, and participants are not informed of their partner's identity before and after making their decision.
Randomization unit
Youth groups (YAIGs) in Ethiopia
Clusters
183 YAIGs, 732 YAIG leaders, 2500-2600 YAIG members
Observations
183 YAIGs, 732 YAIG leaders, 2500-2600 YAIG members
Arms
C (Control: Standard PGG): 62 YAIGs, T1 (YAIG contribution): 61 YAIGs, and T2 (Forest contribution): 60 YAIGs
Primary outcomes
1. Individual contribution to the common account (ETB) by leaders 2. Individual contribution to the common account (ETB) by members 3. Contribution rate by youth group leaders 4. Contribution rate by youth group members 5. Group-based performance of YAIGs (savings, income, revenue, assets, perceived performance, participation of members, frequency of meetings) 6. Group-level cohesion and trust among members and leaders

Cited by

No citation data available for this registration yet.

Keywords