Leadership selection modalities and leader characteristics

The interventions focus on three leadership selection modalities to select leaders of so-called forest user groups (FUGs).

The intervention include:
• Status quo leadership: The group chairperson and additional members of the executive committee (EC) are “elected” by group members during a public voting event through show of hand. Business-as-usual selection of leadership implies that leader selection committee is appointed followed by village elites “volunteer” themselves or suggested by group members for positions, and group members raise their hands to show their vote.

• Anonymized voting: The group chairperson and additional members of the executive committee (EC) are “elected” through anonymous ballot boxes. In anonymized voting, leader selection committee is appointed followed by anonymized selection candidates and subsequent anonymized voting using ballot boxes.

• Gender quota: This follows the status quo leadership selection but with a requirement that at least 3 members of the group executive committee should be women.

2024-09-20

2025-04-30

Leader capability: relevant experience, education, and managerial ability (ability to organize, delegate, mobilize support), ability to coordinate
Leader corruption: honesty and accountability
Leader connectivity: links to key actors within village and beyond
Leader trustworthiness: group members trust on leaders

Leader capability
1. Survey data on business experience and years of schooling
2. Survey data on managerial capacity, the outcome of which is a score between 0 and 117 based, which is the sum of infrastructure experience (0 to 26), community activity experience (0 to 14), project proposal skills (0 to 5), project costing skills (0 to 14), leadership experience (0 to 10), reading test scores (0 to 24) and numeracy test scores (0 to 24)
3. Lab-in-the-field experiment
We invite 8 members from each forest user group: 3 leaders (chairperson, vice chairperson and secretary) and 5 randomly selected members. At least one of the three leaders should be female. The 8 members are divided into two groups: (i) the first group consists of the chairperson and 3 randomly selected members, and (ii) the second group consists of the vice chairperson and 3 randomly selected members. All members play a linear public good game for 15 rounds, which consists of two stages and two treatments.
Stage 1: All players play a linear public good game for 5 rounds
Stage 2: This consists of two parts (5 rounds in each part):
a. Linear public good game in which a group leader suggests a contribution level before group members decide on their contribution to the public good, but the actual contribution of the leader is not shown to group members
b. Linear public good game in which a group a group leader suggests a contribution level before group members decide on their contribution to the public good, but the actual contribution of the leader is shown to group members
All players play both parts of stage 2, but we randomize the order of these parts.
Every participant gets a show-up fee of 100 ETB for coming to the experiment place. At the beginning of each round, everyone gets 20 ETB to decide on the allocation to private and group accounts. At the end of the experiment, participants will be paid: show-up fee + total earning from the experiment.
Leader ability to coordinate will be measured as the leader’s ability to raise group contributions, i.e., group contributions in stage 2 as a result of the leader’s suggested contribution.

Leader corruption (honesty and accountability)
1. Lab-in-the-field experiment
Honesty is measured as the difference in leader contribution between stage part 1 and part 2 games in stage 2 of the lab-experiment described above.
2. Structured activity
We inform the chairperson that the project has a gift for the group: a ‘thank you’ money amounting to USD 100, that the leader should use for the benefit of the group. This was done in private, so no other group member saw us handing the money to the leader. We visit the village 2-3 weeks later and ask (a) the leader on what she/he did with the USD 100 and ask for a proof, and (b) ask one other member of the executive committee and 2-3 ordinary members about whether they have been invited into a meeting or consulted by the leader about the group. But we do not inform them that we have given money to the leader and we do not specifically ask about the money. Note that we did not inform the chairperson that we will come back to ask about the money.
We will measure the share of the money spent by leaders to the benefit of the group and whether the leader informs and consults group members.

Leader connectivity
1. Survey data on number, frequency and strength of contacts inside and outside the village
2. Structure activity
We hand the chairperson a paper with a list of questions to make an incentivized guess on (a) number of hectares of forest in their district, (b) the number of FUGs in their district (or woreda), (c) the number of government projects in their district, and (d) allocated budget for schools in their district. We inform the chairperson that we will come back 2-3 weeks later to collect the answered paper and if the answers are the same or closer to the true answers, the leader will get 20USD as a reward.
We will measure the number of answers that are correct or close to the correct answer.

Leader trustworthiness
1. Survey measure on group members’ trust in leaders
2. Lab-in-the-field experiment
We will use group members’ contribution difference between part 1 and part 2 of stage 2 in the lab-in-the-field experiment described above.

We collaborate with Oromia Forest and Wildlife Enterprise (OFWE) in its expansion of the community-based forest management program. In this program, new FUGs are created with a right to manage a specific forest block and extract forest resources. One of the key steps in the establishment of these groups is group members are required to establish an executive committee, composed of a chairperson, vice chairperson, secretary and 4 additional Executive Members (ECs).

129 new FUGs in 129 villages (one FUG per village) are randomly and equally assigned to one of the three interventions.
1. Status quo leadership (43 FUGs)
2. Anonymized voting (43 FUGs)
3. Gender quota (43 FUGs)

Not available

Randomization done in office by a computer

Forest user groups (or villages)

Yes

129 groups

1032 group members

43 groups (344 members) per each treatment arm.