An Experimental Investigation of Trust and Cooperation in Strategic and Non-Strategic Settings

Ing. Diya Elizabeth Abraham, Ph.D.

Doctoral thesis

Abstract:

This thesis examines the drivers of trust, reciprocity and cooperative behavior in different strategic and non-strategic settings. In three chapters, I explore how information about an an agent's institutional environment or interaction partners affects their willingness to cooperate with one another. Across all three chapters, I employ a common methodological paradigm, namely, an economic experiment, to address my main research questions. The first chapter studies the behavior of buyers and sellers in an online market. Using a laboratory experiment, my coauthors and I investigate whether reducing the social distance between buyers and sellers by having sellers represented by their online avatar increases both buyers' trust and sellers' trustworthiness. We find that while buyers do not trust more when they see the seller's avatar, sellers behave in a more trustworthy manner when they know that buyers view their avatar. This positive effect on the sellers' trustworthiness is undone when they can freely change their avatars at any time and hence select them strategically. We find that female avatars tend to be more trusted by buyers regardless of whether or not the seller had the possibility to change their avatar in each trading round. Correspondingly, in the treatment in which seller avatars can be freely changed, male sellers are more likely than female sellers to select an avatar of the opposite gender. The second chapter explores two important factors that affect the perceived trustworthiness of charitable intermediaries, namely the extent of their discretion over donors' voluntary contributions, and the presence and size of unavoidable overhead costs. We find strong evidence that both of these factors play an important role in donors' willingness to use such intermediaries to coordinate their donation efforts. In particular, donors are only likely to use the intermediary when the latter is heavily restricted in terms of the proportion of donations she can keep for herself, but more strikingly, the positive effect of imposing stronger restrictions on the intermediary is undone when donors know that the intermediary incurs an unavoidable and sunk overhead cost. Our analyses suggests that the ex-ante inequality created between the donors and the intermediary as a result of this sunk cost reduces the intermediary's trustworthiness in the donors' eyes, and ultimately leads to a less efficient outcome. The third chapter explores trust-based cooperation in what is perhaps the most important social dilemma we face today, namely, climate change mitigation. We study how setting obligatory minimum standards for voluntary contributions to help mitigate climate change affects the levels of such contributions. We find that when the minimum is set too low, it has a net negative effect on overall contributions. This is true regardless of whether the minimum is imposed by the experimenters or determined endogenously through a majority vote among the participants. We find suggestive evidence that the net negative effect of setting a low minimum is driven by participants' pessimistic beliefs about the behavior of others once the minimum comes into effect.

Abstract:

This thesis examines the drivers of trust, reciprocity and cooperative behavior in different strategic and non-strategic settings. In three chapters, I explore how information about an an agent's institutional environment or interaction partners affects their willingness to cooperate with one another. Across all three chapters, I employ a common methodological paradigm, namely, an economic experiment, to address my main research questions. The first chapter studies the behavior of buyers and sellers in an online market. Using a laboratory experiment, my coauthors and I investigate whether reducing the social distance between buyers and sellers by having sellers represented by their online avatar increases both buyers' trust and sellers' trustworthiness. We find that while buyers do not trust more when they see the seller's avatar, sellers behave in a more trustworthy manner when they know that buyers view their avatar. This positive effect on the sellers' trustworthiness is undone when they can freely change their avatars at any time and hence select them strategically. We find that female avatars tend to be more trusted by buyers regardless of whether or not the seller had the possibility to change their avatar in each trading round. Correspondingly, in the treatment in which seller avatars can be freely changed, male sellers are more likely than female sellers to select an avatar of the opposite gender. The second chapter explores two important factors that affect the perceived trustworthiness of charitable intermediaries, namely the extent of their discretion over donors' voluntary contributions, and the presence and size of unavoidable overhead costs. We find strong evidence that both of these factors play an important role in donors' willingness to use such intermediaries to coordinate their donation efforts. In particular, donors are only likely to use the intermediary when the latter is heavily restricted in terms of the proportion of donations she can keep for herself, but more strikingly, the positive effect of imposing stronger restrictions on the intermediary is undone when donors know that the intermediary incurs an unavoidable and sunk overhead cost. Our analyses suggests that the ex-ante inequality created between the donors and the intermediary as a result of this sunk cost reduces the intermediary's trustworthiness in the donors' eyes, and ultimately leads to a less efficient outcome. The third chapter explores trust-based cooperation in what is perhaps the most important social dilemma we face today, namely, climate change mitigation. We study how setting obligatory minimum standards for voluntary contributions to help mitigate climate change affects the levels of such contributions. We find that when the minimum is set too low, it has a net negative effect on overall contributions. This is true regardless of whether the minimum is imposed by the experimenters or determined endogenously through a majority vote among the participants. We find suggestive evidence that the net negative effect of setting a low minimum is driven by participants' pessimistic beliefs about the behavior of others once the minimum comes into effect.

Keywords

cooperative behavior, experimental economics, public goods, reciprocity, trust

Language used: English

Date on which the thesis was submitted / produced: 3. 4. 2023

Thesis defence

• Date of defence: 22. 6. 2023
• Supervisor: prof. Mgr. Jiří Špalek, Ph.D.
• Reader: doc. Ing. Jana Péliová, PhD., Prof. Michail Drouvelis

Masaryk University

Faculty of Economics and Administration

Doctoral programme / field:
Public Economics / Public Economics