Uncertainty, economic expectations, and financial decisions

How to cite

Zafar, Basit; Bos, Marieke; Olafsson, Arna; Ravina, Enrichetta. 2023. "Uncertainty, economic expectations, and financial decisions." AEA RCT Registry. AEARCTR-0011707. https://doi.org/10.1257/rct.11707-1.0

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

Basit Zafar* , Marieke Bos , Arna Olafsson , Enrichetta Ravina

* Primary investigator

Registry ID
AEARCTR-0011707
DOI
https://doi.org/10.1257/rct.11707-1.0
First registered
Last updated
Status
On going
Trial dates
2023-06-01 – 2025-12-31
Countries
Private
JEL codes
D8, E3, E4, E5
Registry record
http://www.socialscienceregistry.org/trials/11707

Abstract

This project aims to shed light on the relationship between households’ expectations, their understanding of the relationship between inflation and other economic variables, and their high-frequency data on financial and labor market decisions. Specifically, in the first wave we survey clients of an Icelandic Bank and link these data to high-frequency data on their financial decisions. In the survey, at the baseline, we elicit respondents’ expectations regarding inflation, income, spending, consumption, choice of mortgage type, etc., as well as how they perceive the link between inflation and other various economic variables. A randomized subset of respondents is then informed either about (1) the past and current rate of inflation; (2) the past and current rate of inflation, and the real income effects of inflation (for example, erosion of income as well as debt in the case of high inflation; implications for inflation-indexed versus non-indexed mortgages); (3) the past and current rate of inflation, and the intertemporal substitution channel in the presence of inflation; (4) the past and current rate of inflation, and both the real income effect of inflation and the intertemporal substitution effect. At the endline, respondents are re-asked about their expectations and intended plans. We then also propose to investigate the impact on actual subsequent financial decisions, using the bank data. In the second wave, to be fielded a few months later, we will ask respondents again about their intermittent spending, financial decisions, and labor market outcomes, re-test their understanding of the mechanisms linking inflation and other macro and micro-economic variables, and study the degree and the persistence of learning about them.

Intervention

Treatments: Respondents are assigned to one of 5 possible groups:
(1) Control
(2) T1: Information about past and current inflation
(3) T2: T1 + information about the real income effects of inflation
(4) T3: T1 + information about the substitution effect (shifting consumption intertemporally in response to higher/lower inflation)
(5) T4: combining T2 and T3

Experimental design

Initial Survey:
Baseline: elicitation of past/current inflation and spending/consumption; expectations about future inflation and intended financial behaviors (for example, spending, debt, choice of mortgage type etc.). Perceptions about how inflation affects consumers (qualitative/open-ended questions about mental models)
Treatments: Respondents are assigned to one of 5 possible groups:
(1) Control
(2) T1: Information about past and current inflation
(3) T2: T1 + information about the real income effects of inflation
(4) T3: T1 + information about the substitution effect (shifting consumption intertemporally in response to higher/lower inflation)
(5) T4: combining T2 and T3
Endline: Re-elicitation of expectations and future intended behavior.

Follow-up survey: will be fielded to a subset of the same respondents a few months later. We will (1) ask respondents again about their financial and labor market outcomes, (2) re-elicit their expectations, (3) re-test their understanding of the mechanisms linking inflation and other macro and micro-economic variables, and study the degree and the persistence of learning about them, and (4) investigate the role of cues and memories and have a debrief session on the reasons why they made the choices they report and/or we see in the data.

Design details

Randomization method
Randomization done by a computer within the survey.
Randomization unit
Individual
Clusters
1000 respondents for each of the 5 arms. A total sample size of 5,000 respondents. Note that we will use the within-subject analysis to investigate the impact of the treatments on expectations and intentions. For the impact on actual behavior, we will conduct between-subject design, using randomized assignment to treatments.
Observations
A total sample size of 5,000 respondents.
Arms
1000 respondents for each of the 5 arms.
Primary outcomes
Our main outcomes are as follows. From the survey: (1) the main outcomes are the impacts of the different treatments on expectations and intended consumption plans etc. (2) how consumers perceive the link between inflation and the various variables (i.e., the mental models that consumers have in mind). From the bank data: the main outcomes are the impacts on actual financial behavior (for example, refinancing, spending, debt).
Secondary outcomes
We are also interested in heterogeneity in impacts.

Cited by

No citation data available for this registration yet.

Keywords