Round-Ups, Consumers' Savings Rates and Stock Market Participation: Evidence from FinTech

How to cite

Nardini, Moritz; Renkl, Emanuel. 2024. "Round-Ups, Consumers' Savings Rates and Stock Market Participation: Evidence from FinTech." AEA RCT Registry. AEARCTR-0010512. https://doi.org/10.1257/rct.10512-1.0

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

Moritz Nardini* , Emanuel Renkl

* Primary investigator

Registry ID
AEARCTR-0010512
DOI
https://doi.org/10.1257/rct.10512-1.0
First registered
Last updated
Status
In development
Trial dates
2024-04-22 – 2025-03-31
Countries
Private
JEL codes
D14, G11, G40, G51
Registry record
http://www.socialscienceregistry.org/trials/10512

Abstract

We study the impact of an open banking powered round-up App on consumer spending, saving, and stock market participation using data from a FinTech App. In the first step, we run a panel regression with observational data of existing users. In the second step, we conduct a lab-in-the-field experiment with newly recruited users to establish causality. Study participants are invited to download the App and connect their main bank account, but are not required to use the round-up service. We estimate the effect of the App on consumer financial behavior with transaction and survey data. While round-ups are a standard bank service nowadays, their causal effect has not been investigated yet. We want to close this research gap and contribute to the discussion on how financial technology can be used to promote consumer savings and stock market participation.

Intervention

Study participants are required to download a savings App that provides a bank independent round-up service as its main feature and link their primary bank account to the App. However, participants are not incentivized to use the round-up service. Additionally, study participants have to answer two surveys. The first one needs to be answered at the start of the experiment and the second one at the end of the intervention, i.e., after two months.

Experimental design

This study will be conducted in collaboration with a German FinTech, which provides a savings App with a bank independent round-up service as its main feature. Specifically, users can link their bank accounts to the FinTech App and each transaction is rounded up by default to the next Euro. Additionally, users can link the clearing account of their brokerage account and transfer the round-ups directly via the App.

To estimate the causal effects of the App on the users' savings and spending behavior, we will run a lab-in-the-field experiment where participants are required to download the App and link their primary bank account. To this end, we will recruit study participants through the Munich Experimental Laboratory for Economic and Social Sciences at LMU Munich and possibly also through experimenTUM, which is the laboratory for experimental research in economics at Technical University Munich.

The experiment consists of two sessions, with two months between the sessions. Study participants will be randomly allocated to the treatment or the control group. In the first session, both groups answer a pre-survey that asks participants about their savings and spending behavior and collects information on demographics. Additionally, the treatment group is asked to download the App and link their primary bank account. By linking the bank account to the App, we receive a user's financial transactions for the past 90 days prior to the treatment due to the open banking regulation in Germany. Moreover, as long as the user does not delete her linked bank account in the App, all transactions can be monitored over the next 90 days even if the participant does not use the App.

Two months later, in the second session, both the treatment and the control group take part in a second survey to investigate whether the round-up amounts were actually invested into the stock market and to rule out substitution effects between savings within and outside the App (treatment group) or whether participants changed their savings behavior in the last two months (control group). Additionally, the control group is required to download the App during the second session, which gives us again access to all financial transactions over the past 90 days, building up an artificial control group.

Design details

Randomization method
The randomization is done in the office by a computer.
Randomization unit
The unit of randomization is at the individual level.
Clusters
There are no planned clusters, as the unit of randomization is at the individual level.
Observations
500 participants
Arms
Ratio between treatment and control group will be 50:50.
Primary outcomes
The key outcomes of interest are whether the App intervention changes a participant's overall savings rate, spending behavior, and stock market participation compared to the control group. Using anonymized transaction data, we will measure a user's direct savings rate (i.e., the amount saved in dedicated accounts, e.g., savings account, securities account, etc. ) but also her indirect savings rate by comparing income and spending. Moreover, we will measure a user's stock market participation (i.e., the amount transferred to a brokerage account; first transaction categorized as “Securities investment”) and spending behavior (e.g., spending frequency, average spending amount, discretionary and non-discretionary spending). These outcomes will be enriched with self-reported information from the surveys. For instance, we elicit study participants' self-reported savings rates and whether and how they invested the round-ups they transferred to their brokerage account. Additionally, we ask for changes in savings behavior outside of the App, and whether they opened a brokerage account and/or started investing during the intervention period.
Secondary outcomes
In general, secondary outcomes of interest are if and how users engage with the App. To investigate this, we observe in-app behavior, such as clicks on in-app sections and logins. Moreover, in the surveys, we elicit established factors determining (non-)stock market participation, following Choi and Robertson (2020), such as having too little wealth, missing trust, missing knowledge, etc. Additionally, among those who have not used the service, we ask about the reasons for not using it, and among those who used the service, we elicit the channels that drive adoption.

Cited by

No citation data available for this registration yet.

Keywords