The Currency Co-Holding Puzzle

How to cite

Koch, Melanie. 2024. "The Currency Co-Holding Puzzle." AEA RCT Registry. AEARCTR-0015029. https://doi.org/10.1257/rct.15029-1.0

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

Melanie Koch*

* Primary investigator

Registry ID
AEARCTR-0015029
DOI
https://doi.org/10.1257/rct.15029-1.0
First registered
Last updated
Status
Completed
Trial dates
2022-09-27 – 2024-12-16
Countries
Czechia (), Poland (), Serbia ()
JEL codes
D14, D91, G11, G51
Registry record
http://www.socialscienceregistry.org/trials/15029

Abstract

Foreign currency usage is very common in some economies in Central, Eastern and Southeastern Europe, where especially the euro is used next to local currencies. In this project, I conceptualize the joint decision in which currency to save and in which to borrow: in local or foreign currency. I coin this decision currency co-holding. Using an international survey, I look at which co-holding combinations are common among which socio-demographic groups, i.e. to which individual characteristics they are correlated to. I am especially interested in the role of trust, risk tolerance and financial literacy, in particular the knowledge about potential risks stemming from foreign currency loans (FX-literacy). First data for the project were already collected in the 2022 wave of the survey. To uncover a potential causal channel from FX-literacy to currency co-holding preferences, I additionally conduct an information experiment in the current wave of the survey, which is in field in fall 2024.

The experiment intends to vary the knowledge about exchange rates and the potential risks arising from an exchange rate depreciation. After answering several financial literacy questions, including one on exchange rate literacy, survey participants are randomized into either getting further information on exchange rates or not. Subsequently, they are asked whether they would prefer to deposit money in local or foreign currency and in which currency they would prefer to borrow. In addition, they are asked about potential motives to borrow in a certain currency, depending on the choice they have made before.

Intervention

I design an information treatment, which is supposed to vary knowledge about the concept of exchange rates. I want to test if that knowledge impacts the preferences in which currency to save and in which to borrow: in local or foreign currency. After answering a question on exchange rate literacy, respondents are randomized into two groups and depending on the group, receive a different follow-up question and with this different information.

Experimental design

The control group is only asked how they would assess their knowledge on exchange rates. In contrast, the treatment group first gets a more detailed explanation what an exchange rate is and then, is asked how they would assess their knowledge too.The explanation itself is designed in such a way that it conveys the concept of exchange rates without priming people into either the savings or the borrowing frame but rather keeps a neutral frame. Moreover, it is relatively short to avoid interviewer fatigue (i.e. interviewers not reading out the full information) and respondent fatigue (i.e. respondents not listening to the full information).

Design details

Randomization method
1. For Czechia and the CAPI-interviews in Poland: randomization is done by computer automatically 2. For Serbia: randomization is done by the interviewer typing in a running number based on which the used tablet randomizes the questionnaire 3. For PAPI-interviews in Poland (ca. 20% of the total interviews): paper questionnaires are stacked randomly In every case, the randomization is stratified by interviewer
Randomization unit
See pre-analysis plan attached
Clusters
There are usually more than 50 interviewers per country
Observations
1,000 respondents will receive the intervention, given results from previous years around 2%-15% might refuse to answer the primary outcome questions
Arms
In each country, the plan is to have 50% of the observations in the control and 50% in the treatment
Primary outcomes
"Currency co-holding", which takes values from 1 to 4 and then the share of respondents for distinct values
Secondary outcomes
1. Reason to prefer borrowing in local currency, if local currency was chosen in the previous question. Possible answer options: 1) The [LOCAL CURRENCY] is the official currency of [MY COUNTRY] 2) I trust the [LOCAL CURRENCY] more 3) Better loan conditions (better interest rate/ lower fees etc) 4) Foreign currency loans are too risky 5) It is not possible to borrow in foreign currency 6) Other reason 2. Reason to prefer borrowing in foreign currency, if local currency was chosen in the previous question. Possible answer options: 1) I believe that the [LOCAL CURRENCY] will be replaced by the euro soon 2) I do not trust the [LOCAL CURRENCY] enough 3) Better loan conditions (better interest rate/good exchange rate/lower fees etc) 4) Loans in [LOCAL CURRENCY] are too risky 5) Most/all of my income is denominated in a foreign currency 6) Other reason

Cited by

No citation data available for this registration yet.

Keywords