Uncertain Choices in Environmental Contexts
How to cite
Roux, Sara le; Krishnamurthy, Chandra. 2025. "Uncertain Choices in Environmental Contexts." AEA RCT Registry. AEARCTR-0015791. https://doi.org/10.1257/rct.15791-1.0
Download: BibTeX · RIS — for reference managers without a browser translator. Zotero and most others pick up this page's citation metadata directly.
* Primary investigator
Abstract
While experiments related to the effects of ambiguity on many aspects of decision making are increasingly being conducted, there are only a handful of studies related to insurance and ambiguity, and none at all for forests, for Sweden or elsewhere. Helping fill this gap is the main goal of this project. More specifically, this project uses the most commonly accepted methodology today, behavioral experiments, to answer the following two questions: (i) how do individuals with some expertise in forestry respond to risk and ambiguity? And (ii) how are forest-related insurance decisions made under risk and under ambiguity? We are thus able to quantify the degree of risk and ambiguity aversion and to understand how different sets of risks affect insurance uptake, along with the degree of insurance chosen (full versus partial)
Experimental design
The experiment is divided into the following two phases:
Phase 1: Here, each participant is asked to provide a few basic details about themselves (with no sensitive data being required).
Phase 2: This phase consists of two parts, A and B. Part A pertains to elicitation of the participants’ attitude to risk and ambiguity. This will be carried out by asking the participants to make decisions in various hypothetical situations with the aim to elicit -
1. their general risk attitude
2. their general attitude to uncertain situations
3. their preferences related to time (that delaying or speeding up potential rewards)
We propose using the matching probabilities method proposed by Baillon et al. (2021) to elicit the model-free ambiguity parameters.
In Part B, the participants’ preference related to insuring their forests against low-probability high-impact events (e.g. large fires, storms) will be elicited. This again will be split into three treatments (a base and two ambiguity treatments), each specifying a given level of risk involved and certain insurance choices.
We note that the three experimental treatments referred to above (base and two ambiguity settings) for insurance will be randomised, meaning that the "treatment effect" is the effect of the two ambiguity settings relative to the base (which is a control group).
Design details
Cited by
No citation data available for this registration yet.