Individual determinants of self protection among homeowners with(out) disaster risk insurance
Last registered on July 18, 2018

Pre-Trial

Trial Information
General Information
Title
Individual determinants of self protection among homeowners with(out) disaster risk insurance
RCT ID
AEARCTR-0002966
Initial registration date
May 07, 2018
Last updated
July 18, 2018 6:50 AM EDT
Location(s)
Region
Primary Investigator
Affiliation
Vrije Universiteit Amsterdam
Other Primary Investigator(s)
PI Affiliation
Vrije Universiteit Amsterdam
PI Affiliation
Vrije Universiteit Amsterdam
Additional Trial Information
Status
In development
Start date
2018-05-08
End date
2018-07-19
Secondary IDs
Abstract
The focus of this paper is to examine how the financial incentives emanating from insurance compare with the influence of other behavioral motivations for investing in flood risk reduction. The online lab-in-the-field experiment among homeowners in floodplains (N=2000) will demonstrate the effects of individual characteristics on risk mitigation decisions of relevant decision-makers. The expected results can be used to inform policy makers and insurance companies about how to stimulate people who voluntarily have flood insurance coverage to better prepare for flood disasters.
External Link(s)
Registration Citation
Citation
Blasch, Julia, Wouter Botzen and Jantsje Mol. 2018. "Individual determinants of self protection among homeowners with(out) disaster risk insurance." AEA RCT Registry. July 18. https://www.socialscienceregistry.org/trials/2966/history/31887
Experimental Details
Interventions
Intervention(s)
We will examine the effects of a premium discount and different types of insurance (mandatory, voluntary, no insurance) on damage reducing investments in an online experiment.
Intervention Start Date
2018-05-13
Intervention End Date
2018-06-17
Primary Outcomes
Primary Outcomes (end points)
damage reducing investments
Primary Outcomes (explanation)
Secondary Outcomes
Secondary Outcomes (end points)
willingness to pay for flood insurance
Secondary Outcomes (explanation)
Experimental Design
Experimental Design
We use a short version of the investment game which was programmed for an earlier experiment. This investment game is followed by a final survey to gather data on risk preferences, time preferences and other behavioral factors that could be important characteristics related to flood risk.
Experimental Design Details
In the investment game, respondents are asked to imagine owning a house in a floodplain for the next 25 years and a savings balance of 65,000 ECU. All payments in the game are subtracted from this balance. A scenario starts with instructions and the introduction of the parameters: yearly flood probability (1%), maximum damage (50,000 ECU), savings balance (65,000 ECU) and insurance (No/Yes, with 5% deductible). The yearly objective probability of flooding is given at 1%. The sequence of pages in a scenario is Invest, Pay premium, Flood risk result, Overview of results. The scenario concerns 25 years, but decisions are made only once to facilitate a short version of the investment game. The instructions are supported by graphics and are always available as a pop-up screen throughout the experiment. All respondents are shown the Invest page with five discrete investment levels and accompanying benefits. The Pay premium page is shown to individuals in the Insurance treatments: here the fair premium was paid from their savings balance for all 25 years at once. The Flood risk result page shows a grid with 100 houses, where the house of the participant isindicated with a square and all houses flooded (according to the yearly 1% flood probability) at least once in the 25 years of the scenario are blue. In case a participant's house is one of these, the deductible (or damage in the No Insurance treatment) is paid from the savings balance. The Overview of results page shows the history of the savings balance (65,000 ECU - premiums - deductible/damage - investment). The investment game starts with a test scenario to make participants familiar with the decision screens. To ensure understanding of the game and the savings balance, the test scenario is followed by comprehension questions, conditional on treatment. The answers are available in the pop-up instructions. After answering all comprehension questions correctly, subjects can start with the main scenario. Respondents in the Voluntary treatment will be moved to Insurance and No Insurance based on their WTP. If our parameters are such that a large majority self selects into either of those, we will assign more respondents to the Voluntary treatment (or change the parameters slightly) to ensure enough observations (at least 300) in both.
Randomization Method
randomization done by oTree software
Randomization Unit
individual
Was the treatment clustered?
No
Experiment Characteristics
Sample size: planned number of clusters
5 treatments
Sample size: planned number of observations
The sample targets 2000 homeowners who are located in the river delta areas of the Netherlands with flood probability standard of 1 in 1250.
Sample size (or number of clusters) by treatment arms
At least 300 homeowners per treatment. Exact sample size per treatment will be determined after a pilot with 100 observations in the Voluntary treatment.
---
The pilot results showed that about 25% of homeowners in the Voluntary treatments self selects into Insurance. No major changes were made in the design after the pilot results. We aim for at least 150 homeowners self selected into Voluntary Insurance and Voluntary Insurance Discount. For the premium discount versus Mandatory Insurance we expect a larger effect size. Therefore we will sample:

300 No Insurance
300 Baseline
200 Premium Discount
600 Voluntary
600 Voluntary Discount
-----------------------------
2000 in total (or more if budget allows)
Minimum detectable effect size for main outcomes (accounting for sample design and clustering)
A sample size analysis assuming an alpha of 0.05 and a power of 80% indicated that we need a sample size of at least 252 participants per treatment in order to detect the smallest effect size* found in the scenario of a previous lab experiment, closest to our current parameters, with a Wilcoxon Mann Whitney test. * Mean investments in ECU in previous experiment No Insurance versus Mandatory Insurance: 2711.9 (sd 4102.4) versus 1793.4 (sd 3976.8): effect size 27% -> sample size needed 252 * Mean investments in ECU in previous experiment Mandatory Insurance versus Premium Discount: 1793.4 (sd 3976.8) versus 3550 (sd4560.1): effect size 41% -> sample size needed 78
IRB
INSTITUTIONAL REVIEW BOARDS (IRBs)
IRB Name
IRB Approval Date
IRB Approval Number
Post-Trial
Post Trial Information
Study Withdrawal
Intervention
Is the intervention completed?
No
Is data collection complete?
Data Publication
Data Publication
Is public data available?
No
Program Files
Program Files
Reports and Papers
Preliminary Reports
Relevant Papers