Risk Disclosures and Investor Behavior

How to cite

Gathergood, John; Stewart, Neil; Quispe-Torreblanca, Edika. 2024. "Risk Disclosures and Investor Behavior." AEA RCT Registry. AEARCTR-0013150. https://doi.org/10.1257/rct.13150-1.0

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

John Gathergood* , Neil Stewart , Edika Quispe-Torreblanca

* Primary investigator

Registry ID
AEARCTR-0013150
DOI
https://doi.org/10.1257/rct.13150-1.0
First registered
Last updated
Status
On going
Trial dates
2024-02-12 – 2024-04-15
Countries
Private
JEL codes
G1, G4, G5
Registry record
http://www.socialscienceregistry.org/trials/13150

Abstract

This study uses a randomised control trial to evaluate the effects of contextualised risk warnings on investor behavior

Intervention

In this randomised control trial we test the effects of experimental variation in the risk disclosures shown on screen on an online trading website. A control group is shown a standard risk warning emphasising that the value of investments can fall as well as rise. A first treatment group is shown a varied risk warning that adds a further disclosure stating that in the medium run the returns from investing in stocks and shares usually exceed those from cash savings. A second treatment group is shown the a varied risk warning that further adds cited empirical evidence on the higher returns from investing compared with cash savings.

This is a live trial involving individuals on the website of a stockbrokers. Previously we undertook survey-based trials to test the risk disclosures. Details of these are available at https://aspredicted.org/ZW3_6VT and https://aspredicted.org/MNV_CKX.

Experimental design

The experimental design is a randomised control trial. Subjects will be individuals who navigate to the stockbroker's website and view one of the pages in the trial. For each page in the trial, the subject will be allocated at random to control, treatment one or treatment two. Using cookie data, the subject will remain in the treatment arm for the duration of the trial (unless they wipe their cookies, in which case they will be re-assigned at random to an arm the next time they navigate to the website).

Design details

Randomization method
Randomisation via Adobe Target
Randomization unit
Individual
Clusters
NA
Observations
The trial will be live for 5 weeks and the number of observations will be determined by web traffic volume
Arms
Three trial arms of equal size, with the actual number of observations being inherently dependent on the volume of web traffic during the trial's timeframe.
Primary outcomes
There are two main outcomes. First, whether an individual in the trial opens an investing account with the stockbroker. Second, conditional upon opening an account, the level of investment in pounds sterling and the split of the investment between stocks/mutual funds and bonds. For existing customers, the study will examine changes in the investment portfolio, specifically focusing on the adjustment in amounts allocated to stocks/mutual funds and bonds following exposure to the modified risk warnings.

Cited by

No citation data available for this registration yet.

Keywords