Costly Information in Experimental Assets Markets

How to cite

Klapp, Francisco; Lunawat, Radhika. 2023. "Costly Information in Experimental Assets Markets." AEA RCT Registry. AEARCTR-0011425. https://doi.org/10.1257/rct.11425-1.1

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

Francisco Klapp* , Radhika Lunawat

* Primary investigator

Registry ID
AEARCTR-0011425
DOI
https://doi.org/10.1257/rct.11425-1.1
First registered
Last updated
Status
On going
Trial dates
2023-03-16 – 2023-12-31
Countries
Private
JEL codes
C91, C92, G13, G14
Registry record
http://www.socialscienceregistry.org/trials/11425

Abstract

In this study we add to the long tradition of assets market experiments going back to the seminal work of Plott and Sunder in the late 80’s and early 90’ by closely looking at the interaction between costly information acquisition and trading purpose. We conduct a laboratory assets market experiment wit an Overlapping Generations framework where there are speculating investors (who do not collect dividends and trade only for capital gains since they won't be "alive" when dividends realize) and dividend-collecting investors (who will be "alive" when state-contingent dividends are paid).

Experimental design

The experimental implementation consists of randomly assigning participants to different generations or cohorts in each of the eight 3-period sequences that make up a session. Members of each cohort face different trading motivations (speculation or dividend collection) in each period's double auction asset market given by the overlapping generations structure and state contingent dividend design.

In terms of the treatments we use both a between and within design. In half of the sessions information about the "state of the world" that determines dividend payments is freely given to half the market participants, in the other half this information is costly and can be acquire by up to half the participants. In each session, regardless of the information structure, all participants will eventually encounter both trading motivations depending on on the cohort for the sequence they been assigned to.

In terms of the actual interface, we use Otree and run sessions online with members of the UCI ESSL subject pool. Instructions, a comprehension test and 2 practice sequence are directed by a researcher trough Zoom and then participants are left for the remaining 8 sequence on their own.

Design details

Randomization method
Recruitment of volunteers using SONA software: Students enroll in one of the offered dates using our website and they are paired with 23 other students who decided to enroll in the same date. If more than 24 volunteers enroll for a single date we randomly select (using a random number generator) which students will participate. Those who are not selected to participate can enroll again for an upcoming session.
Randomization unit
Group (session) randomization for cost information availability treatment, and individual
Clusters
8 sessions of 24 students each
Observations
192 students
Arms
4 sessions for public information control and 4 sessions for costly information
Primary outcomes
- Information dissemination -but NOT information aggregation- for both investor types under both information structures, as measured by deviations from fundamental asset price. - Likelihood of information acquisition (% of predicted requesters by type) and whose information is more likely to be incorporated into prices? - Price informativeness (about the state of the world) and efficiency (deviations from fundamental price) of the interaction b/w trading purpose and information acquisition? - Trading intensity (trading volume) and its interaction with trading purpose and information acquisition?

Cited by

No citation data available for this registration yet.

Keywords