Understanding of Compound Risk and Investment Decisions

How to cite

Haar, Lotte van der; Morsink, Karlijn; Shikuku, Kelvin. 2026. "Understanding of Compound Risk and Investment Decisions." AEA RCT Registry. AEARCTR-0018255. https://doi.org/10.1257/rct.18255-1.0

Download: BibTeX · RIS — for reference managers without a browser translator. Zotero and most others pick up this page's citation metadata directly.

Lotte van der Haar* , Karlijn Morsink , Kelvin Shikuku

* Primary investigator

Registry ID
AEARCTR-0018255
DOI
https://doi.org/10.1257/rct.18255-1.0
First registered
Last updated
Status
On going
Trial dates
2025-11-24 – 2027-04-01
Countries
Private
Registry record
http://www.socialscienceregistry.org/trials/18255

Abstract

This study aims to answer how individuals' understanding of compound risks, as opposed to preferences over compound risks, impacts investment behavior. We study this question using a lab-in-the-field experiment among a sample of 3143 pastoral household heads in Kenya and Ethiopia. We randomly vary the provision of a decision aid that helps individuals with processing compound risks and test the effects on investment decisions in an investment task.

Experimental design

The experimental design consists of a short lab-in-the-field experiment with two experimental arms, randomized at the household level:
1. Compound framing only (control)
2. Compound framing + Decision aid
In both arms, participants face an investment task involving compound risk. In the second arm, participants also receive a decision aid that supports the processing of compound risk. Because the way the compound risk is explained (the framing) may also influence understanding, we cross-randomize two alternative framings: an abstract framing and an intuitive framing. The abstract framing describes compound risks using tokens, whereas the intuitive framing uses livestock.

Design details

Randomization method
The first two randomizations were conducted in Stata 18 and stratified by districts and unit area of insurance. The third randomization was pre-programmed in the survey data-collection tool and revealed to the respondent and enumerator after the tasks were completed.
Randomization unit
The project includes three household-level randomizations: (1) Randomization of version and cross-randomization of the framing of the lab-in-the-field task, (2) Randomization of order of the two investment tasks to measure compound risk ability and simple risk ability, (3) Randomization of selection for payment and of the task to be incentivized.
Clusters
NA
Observations
3143 household heads
Arms
1571 compound framing only, 1572 compound framing + decision aid
Primary outcomes
- Investment behavior in investment task (see PAP) - Reaction time

Cited by

No citation data available for this registration yet.

Keywords