Biased income expectations and debt taking
How to cite
Grohmann, Antonia. 2017. "Biased income expectations and debt taking." AEA RCT Registry. AEARCTR-0002634. https://doi.org/10.1257/rct.2634-1.0
Download: BibTeX · RIS — for reference managers without a browser translator. Zotero and most others pick up this page's citation metadata directly.
* Primary investigator
Abstract
The aim of this experiment is to test whether biased income expectations due to overconfidence lead to higher levels of debt-taking. We exogenously vary subject’s income expectation by letting income depend on relative performance (rank) in hard and easy quiz tasks. The so called “hard easy gap” leads us to expect that subjects have higher income expectations in the easy task than in the hard task. Before subjects receive feedback on their income they can make consumption decisions using a loan. In a third treatment we assign income randomly in order to assess how much debt subjects take out when they have unbiased information over expected income. Additionally we measure subject’s risk preferences, self-control, and financial literacy.
Intervention
Three treatment groups:
i) hard quiz
ii) easy quiz
iii) random
We have a battery of 24 general knowledge questions for both the heard and the easy treatment. The questions correspond in domain between treatments. We have randomly drawn 4 sample questions that are presented to subjects as part of the instructions. In the random treatment, subjects see no sample questions and are informed that their income is determined randomly. The distribution of income is the same for all treatments.
Experimental design
At the beginning of the experiment, participants receive detailed instructions about the experiment. The instructions contain information on how money can be earned, how it can be spend, and how debt can be taken out. Subjects begin the experiment after answering a comprehension test.
Following this, four sample questions of the quiz task are shown which reflect the difficulty of the quiz (easy/hard) since they were randomly drawn from the pool of questions.
Income expectations
Income is determined by the relative performance in the quiz task in groups of eight subjects. After the instructions, income expectations are elicited. We ask for the income expectation in terms of a Euro value and in terms of probabilities to be in each rank.
Importantly, income expectations are not incentivized to avoid strategic behavior in the quiz tasks.
Markets
All markets (rounds 1, 2, 3) have the same structure. A total of ten goods are on display, which can be purchased. The goods are real goods (sweets, beverages, pens) that are handed out to subjects at the end of the experiment depending on their purchases. The current price for each good is shown as well as the future prices in the remaining rounds. Prices for goods increase substantially over time. Within the market screen, participants see their account balance and their current loan. If income does not suffice their account balance is negative. In market 2 it is again possible to take out debt. In market 3 no further debt can be taken, but all earned funds can be spent.
Quiz task
In each round of the quiz subjects are asked ten general knowledge questions. The questions are taken from a broad range of topics and are asked in a multiple choice format. The performance ranking builds on the number of correct responses with completion times as a tie-breaker.
Design details
Cited by
No citation data available for this registration yet.