# Recognizing Power Imbalances in Decision Making

Participants receive sealed envelopes holding zero to five beans that strictly limit how many times they may speak during a simulation. They adopt community stakeholder identities matching their perceived influence, debate a resource allocation dilemma, and examine how voice restrictions shaped the outcome.

**Time:** 45 minutes. 45 minutes total (5m scenario, 4m envelopes, 6m introductions, 20m debate, 10m debrief).
**People:** 4 to 24. Parallel small groups of 4 to 8 participants. For 9 participants, form groups of 5 and 4. For 17 participants, form three groups: 6, 6, and 5.

## Choose this exercise

**Purpose:** Demonstrate through structured discussion limits how speaking power and perceived stakeholder status alter group decisions.
**Use it when:** Use when a team needs to examine covert power dynamics, speaking habits, or who gets excluded during collaborative deliberations.
**Skip it when:** Skip when groups lack trust to handle unequal participation dynamics, or when working on real immediate project deadlines needing full contribution from all members.

## Materials and tools

- One opaque envelope per person
- Dried beans or counting tokens (0 to 5 per envelope, roughly 10 to 20 beans per group)
- One copy of the Scenario Handout per participant or displayed on screen
- One small bowl or tray per table to collect spent beans

## Before participants arrive

1. Count and seal beans into envelopes beforehand so each table receives an assortment ranging from 0 to 5 beans. For 4 people, pack 0, 1, 2, and 3 beans. For 5 people, pack 0, 1, 2, 3, and 4 beans. For 6 people, pack 0, 1, 2, 3, 4, and 5 beans. For 7 people, pack 0, 1, 2, 2, 3, 4, and 5 beans. For 8 people, pack 0, 1, 1, 2, 2, 3, 4, and 5 beans.
2. Set one token collection dish on each table.
3. Distribute or project the scenario description.

## Say this to open

Today we are exploring how power and assumptions shape decision-making. You will each receive an envelope that contains a specific number of tokens. Do not open it until asked, and do not show anyone your contents. Each token in your envelope represents one opportunity to speak during our community meeting. When your tokens are spent, you can no longer speak. You will select a persona based on your own perception of power in this issue, and our group must reach a final decision before all speaking chances are gone.

## How to run it

### 1. Introduce scenario and constraints (5 minutes)

Read the community funding scenario aloud to the entire room. Check for comprehension. Explain that each group represents a local council choosing between repairing a faulty train line or building a youth skill center. Answer basic clarifying questions about the dilemma.

### 2. Distribute envelopes and adopt personas (4 minutes)

Shuffle and hand out one sealed envelope to every participant. Instruct them to peek inside without revealing their count to neighbors. Direct each person to privately decide on a stakeholder role based on their own perception of influence. Leave the persona choice open rather than assigning roles. For example, a participant might choose an unemployed parent, a transit engineer, a town mayor, or an isolated youth.

### 3. Introduce stakeholder identities (6 minutes)

In small groups, participants take turns introducing their character name, background, and perspective on the dilemma in 30 seconds or less. Remind participants not to reveal their bean counts during introductions. Introductions are cost-free and do not require spending a bean.

### 4. Deliberate and decide (20 minutes)

Groups debate the scenario to reach a concrete decision. Each time a participant speaks in the debate, they drop one bean into the center dish. When a person runs out of beans, they must remain silent for the rest of the discussion. If a person started with 0 beans, they observe without speaking. Stop deliberation at 20 minutes or when all beans are spent. Have each group record their final chosen option and vote tally.

### 5. Debrief whole room (10 minutes)

Gather the full room. Allow participants to disclose their bean counts. Call on one spokesperson from each small group to share their outcome, then facilitate discussion using the debrief prompts. Limit debrief answers to two or three speakers per question.


## Debrief questions

- What did your group decide, and how did your bean count influence what happened?
- What did you notice yourself doing or holding back that you would not normally do in a team meeting?
- Which stakeholder identities ended up commanding the floor, and how did that reflect or distort real-world influence?

**Finished output:** A recorded group decision on the scenario, followed by individual participant observations on speaking allocation and power dynamics.

## Remote

Assign breakout rooms of 4 to 8 people. Direct message each participant privately with their assigned speaking number (0 to 5). When speaking in the breakout room, each person tracks remaining turns by posting their updated count in the chat immediately after speaking (for example: '3 left', '2 left', '0 left'). Unmute is permitted only while a person has remaining speaking credits.

## Hybrid

Keep in-person and remote participants in separate small groups if possible. If mixed, pair each remote participant with an in-person partner who drops a physical bean into the bowl whenever the remote partner speaks, while video feeds ensure in-person members see the chat tracking.

## Access and participation choices

- Participants assigned 0 beans can communicate non-verbally via thumbs-up or down during voting, or write down private observations to share during the whole-room debrief.
- Anyone uncomfortable role-playing may opt to act as an objective timekeeper and bean auditor for their small table without penalty.

## If the session gets stuck

**A participant with zero or one bean feels alienated or frustrated by enforced silence.**

Validate their reaction during debrief as a direct insight into institutional silencing. Emphasize that their silence exposes how procedural barriers suppress valuable perspectives.

**Groups finish all their beans within 8 minutes and stop talking.**

Instruct them to conduct their final vote immediately based only on what was said, then spend remaining small group time silently writing individual reflections on who dominated.


## Terms used here

- Stakeholder: A person, group, or organization with a direct interest or concern in the outcome of a decision.

## Participant material: Community Decision Scenario

You are a community group living in a place where jobs are scarce, poverty is rife, and there are limited opportunities for making a decent living. The community is cut off from neighboring towns because the train link stopped operating due to faulty tracks. Your locality has received a one-time grant, but it only funds one project.

Your group must decide between two options:
Option A: Build a community resource center focused on youth skill-building and training.
Option B: Repair the train line to restore transit and physical access to regional jobs.

A community meeting is now underway to select which single option the locality will submit for funding.

## Sources and adaptation

Observed in the Collaborative Discussion Project curriculum as activity 4.7, 'Recognizing Power Imbalances in Decision Making'.

Retained the envelope and bean mechanic, participant-chosen stakeholder personas, and the core train line versus youth center scenario from the Collaborative Discussion Project source. UWT standardized the 45-minute timeline, specified clear envelope bean assortments for groups of 4 to 8 participants, clarified odd-number splits including 9 participants, and aligned persona selection with source guidance so participants decide stakeholder identities freely based on their own perceptions of power.

[Collaborative Discussion Project](https://www.collaborativediscussionproject.com/items-1/47recognizingpowerimbalancesindecisionmaking)

Current run sheet: https://unitedwetransform.com/exercises/7-recognizing-power-imbalances-in-decision-making/
