Activism and Fair Pricing: A Lab Experiment

We consider a game in which subjects interact as either an “activist/certifier” or a “retailer”. In this game, the monopoly retailer jointly exercises market power in the upstream product market and downstream consumer market. The producer is simulated in our game and offers a quantity that is proportional to the wholesale price paid by the retailer.

In the baseline game (retailer only), that we consider as the control group, the retailer determines his/her retail price and the wholesale price paid to the producer (Baseline).

In our experimental treatments, we introduce an “activist/certifier” into the game. The activist proposes a fairness standard to the retailer, defined as a specific share of the “food dollar” (the ratio of the wholesale price to the retail price). Additionally, the activist/certifier sets a monetary sanction to be imposed on the retailer if the proposed standard is not accepted.
The retailer then determines the retail price and decides the wholesale price paid to the producer. The retailer can either comply with the activist/certifier’s proposed standard or reject it, setting both prices independently while incurring the monetary sanction.
Finally, the simulated producer offers a quantity that is proportional to the wholesale price receives, and the market clears.

We consider four different treatments:
- One with a “low cost” of sanctioning and without information on the producer’s surplus displayed on the activist/certifier’s screen (Low-NoInf);
- One with a “high cost” of sanctioning and without information on the producer’s surplus displayed on the activist/certifier’s screen (High-NoInf);
- One with a “low cost” of sanctioning and with information on the producer’s surplus displayed on the activist/certifier’s screen (Low-Inf);
- One with a “high cost” of sanctioning and with information on the producer’s surplus displayed on the activist/certifier’s screen (High-Inf).

The game is repeated for a total of 12 rounds. We do not inform subjects about the total number of rounds to avoid end-game effects.

Subjects play in one experimental condition only (control group or one of the treatments).

2024-11-25

2025-03-31

We will primarily examine participants' individual decisions in each round, focusing on the fairness standard and sanction level set by the activist/certifier, as well as the retail and wholesale price determined by the retailer.

At each repetition of the game, subjects make decisions as previously emphasized:
- the activist/certifier proposes a relationship between the retail price and the wholesale prices to the retailer (fairness standard), and determines a sanction that reduces the retailer monetary payoff if he/she does not accept the activist’s standards;
- then the retailer determines his/her retail price and proposes a wholesale price to the producer.

We will also take into account subject’s socio-economic characteristics in our econometric analyses.

At the end of the experiment, after the main game, we ask subjects about their socio-economic characteristics (age, gender, major, etc.).

1) General setting:

Subjects play a retail game in which the retailer determines both the retail price and the wholesale price paid to the producer. In this game, the producer is simulated and produces a quantity of the good that depends on the wholesale price.

The retail price determines the market demand, while the wholesale price determines the offered quantity.

Participants will make their decisions using sliders. To eliminate any potential bias from the initial slider positions, they will first need to click anywhere on the screen to activate and display the sliders.

To assist participants in making their decisions, a simulator will be provided at the bottom of their screen. This tool will allow them to estimate their potential payoff based on different retail and wholesale price values.

To minimize subjects’ fatigue, a non-binding 90-second timer will be displayed. While the timer counts down, participants can still take action even after it expires.

The experiment consists of 12 rounds; however, participants will not be informed of the exact number to minimize potential end-game effects. At the end of each round, participants will receive feedback on their earnings for that round.

To make sure participants understood the rules of the game, they will first answer some understanding questions.


A) Baseline game:

In the baseline game, subjects play as a retailer only. In each round, they independently select a wholesale price and a retail price. There is no interaction between participants.

The primary objective of this baseline is to ensure that participants set prices in a way that aligns market demand with the quantity supplied.

B) Treatments:

Participants play the same retail game, but with the introduction of the activist/certifier. At the start of the experiment, participants are randomly assigned by the computer to one of two roles: activist/certifier (a third-party in the game) or retailer.

Each participant is placed in a group of six players, consisting of three activists/certifiers and three retailers, and is informed of their role. Participants remain in the same group and retain their assigned role throughout the entire session.

In each round, one activist/certifier is randomly paired with one retailer using an imperfect stranger matching protocol. Participants are informed that they will only interact with others within their group of six. While they may be paired with someone they have interacted with previously, all interactions are anonymous, ensuring decisions remain confidential.

Participants are informed that their role involves making sequential decisions:
i) First, the third-party proposes a fairness standard, defined as the ratio of the wholesale price to the retail price, and sets a sanction to be imposed on the retailer if the standard is not accepted.
ii) Next, the retailer decides whether to accept or reject the third-party’s proposal. If the retailer accepts, they only need to set the retail price, with the wholesale price adjusting automatically. If the retailer rejects the proposal, they independently set both the retail and wholesale prices but incur the monetary sanction.

At the end of each round, participants will receive feedback on their earnings for that round. Additionally, the third-party will be informed whether the retailer accepted or rejected their proposed fairness standard regarding the relationship between the retail price and the wholesale price.

Per session, we will consider between three and four groups of six subjects. There will be no interaction between these groups within a session.


2) Additional information on the treatments:

As previously explained, we consider four different treatments:
- One with a “low cost” of sanctioning and without information on the producer’s surplus displayed on the activist/certifier’s screen (Low-NoInf);
- One with a “high cost” of sanctioning and without information on the producer’s surplus displayed on the activist/certifier’s screen (High-NoInf);
- One with a “low cost” of sanctioning and with information on the producer’s surplus displayed on the activist/certifier’s screen (Low-Inf);
- One with a “high cost” of sanctioning and with information on the producer’s surplus displayed on the activist/certifier’s screen (High-Inf).

The distinction between "high" and "low" sanctioning costs is directly derived from our theoretical model, which demonstrates that the extent of excess supply is influenced by the activist/certifier’s ability to impose sanctions at a low cost.

The decision to provide information on the producer’s surplus aims to assess whether participants, particularly the activists/certifiers, adjust their choices in response to the potential negative impacts on producers' welfare.

Participants will take part in only one treatment.

3) Subjects’ payoff:

In the baseline version of the game (retailer only), subjects’ payoff will be determined by their own choices.

In the treatments (activist/certifier and retailer), subjects’ payoff will depend on their choice as well as the choice of the other subject they are matched with.

They are informed that their final payoff will be determined by their earnings from a single round, randomly selected by the computer at the end of the experiment.

Not available

Randomization made by computer

Individual (student)

No

Not relevant

We will recruit a minimum of 212 subjects. At least 20 subjects will participate in the control group (one subject corresponds to one independent observation since there are no interactions between subjects in the control). For each treatment, at least 48 subjects will be recruited (2 sessions of at least 24 subjects, in 4 closed groups of 6 subjects composed of 3 sellers and 3 activists/certifiers, for a minimum of 8 independent observations per treatment).

There will be a minimum of eight independent observations per treatment arm (i.e., a minimum of 48 subjects per treatment arm).