Nudging to Save: An Online Experiment on Savings Behaviour

How to cite

Saulitis, Andris; Kroeger, Lea. 2022. "Nudging to Save: An Online Experiment on Savings Behaviour." AEA RCT Registry. AEARCTR-0009666. https://doi.org/10.1257/rct.9666-1.0

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

Andris Saulitis* , Lea Kroeger

* Primary investigator

Registry ID
AEARCTR-0009666
DOI
https://doi.org/10.1257/rct.9666-1.0
First registered
Last updated
Status
In development
Trial dates
2022-06-27 – 2022-07-04
Countries
Private
JEL codes
D14; G41; G51;
Registry record
http://www.socialscienceregistry.org/trials/9666

Abstract

This experimental study will provide causal evidence on the potential of nudges and financial incentives to effectively address present-bias preferences and lack of self-control that decrease savings and lead to immediate consumption and excessive borrowing.

Intervention

This study will offer a payment for taking part in the survey equivalent to €10. The payment for taking part in a survey such as this is usually transferred to the respondent’s account immediately after finishing the survey. This study will offer a choice to deposit all or part of the money in virtual savings account for six months to increase the payoff to a maximum of €12.

Experimental design

The goal of this experimental study is to investigate the savings behaviour of individuals and potential ways to increase savings levels. The experiment uses a natural setting where individuals are being paid by a professional survey company. Individuals are asked to choose between immediate and delayed consumption of the money earned from taking part in a study. The treatments introduce a prize-linked savings account option and a nudge based on social norms.

Design details

Randomization method
Randomization is done by the survey company. Randomization is carried out in blocks by gender, age, country and region.
Randomization unit
Randomization is done at the individual level
Clusters
450 individuals who have taken part in the survey (non-experimental) carried out in Fall 2021
Observations
450 individuals
Arms
150 respondents in each experimental group: 75 from each country included in the study (Latvia & Estonia)
Primary outcomes
The variable of interest is the amount of earnings invested in the savings account (€0-€10). We will use OLS linear regression to estimate the causal effect of treatments on savings behaviour. The reference group will be the baseline offer. We will use additional explanatory variables in the second model to gather more precise coefficients. These include socio-demographics and questions on savings behaviour from the current survey and the one carried out with the respondents in Fall 2021. We will use robust standard errors clustered at the country level. We will consider the treatment effect as a statistically significant effect with p<0.05 threshold. We will examine two hypotheses: • Social norm nudges lead to saving more • PLS account is the most effective way to increase savings
Secondary outcomes
In addition to the primary analysis, we will examine the country differences. We predict that the level of savings observed in this experiment will be higher in Estonia, as savings behaviour, in general, is more common in Estonia than in Latvia. We also expect that Estonians are more likely going to opt for a PLS account than Latvians, as financial incentives for increasing savings are effective among risky-behaviour individuals. The results from the survey of Fall 2021 show that Estonians are more willing to invest their earnings and prizes in investments and stocks, therefore showing a higher risk preference than Latvians.

Cited by

No citation data available for this registration yet.

Keywords