First-Price Auctions: Single- vs. Multi-Stage Mechanisms

Last registered on May 03, 2018

Pre-Trial

Trial Information

General Information

Title
First-Price Auctions: Single- vs. Multi-Stage Mechanisms
RCT ID
AEARCTR-0002937
Initial registration date
May 01, 2018

Initial registration date is when the trial was registered.

It corresponds to when the registration was submitted to the Registry to be reviewed for publication.

First published
May 03, 2018, 11:20 AM EDT

First published corresponds to when the trial was first made public on the Registry after being reviewed.

Locations

Region

Primary Investigator

Affiliation
ZEW Mannheim

Other Primary Investigator(s)

PI Affiliation
ZEW Mannheim
PI Affiliation
ZEW Mannheim

Additional Trial Information

Status
In development
Start date
2018-05-03
End date
2018-09-30
Secondary IDs
Abstract
We experimentally test our hypothesis, that participants bid more aggressively in multi-stage first-price auctions than in single-stage first-price auctions. This hypothesis is based on the assumption bidders are loss-averse narrow-bracketers and is driven by an increasing attachment effect. Standard theory predicts revenue equivalence.
The multi-stage format consists of a semi-final and a final. In the semi-final the set of bidders is reduced to two based on highest bids, and in the final the good is allocated to the bidder with the highest final bid.
External Link(s)

Registration Citation

Citation
Fugger, Nicolas, Philippe Gillen and Tobias Riehm. 2018. "First-Price Auctions: Single- vs. Multi-Stage Mechanisms." AEA RCT Registry. May 03. https://doi.org/10.1257/rct.2937-1.0
Former Citation
Fugger, Nicolas, Philippe Gillen and Tobias Riehm. 2018. "First-Price Auctions: Single- vs. Multi-Stage Mechanisms." AEA RCT Registry. May 03. https://www.socialscienceregistry.org/trials/2937/history/29051
Experimental Details

Interventions

Intervention(s)
Intervention Start Date
2018-05-03
Intervention End Date
2018-09-30

Primary Outcomes

Primary Outcomes (end points)
Auction Revenue
Bidding Behavior
Primary Outcomes (explanation)
Loss aversion implies stronger attachement effect in two-stage mechanism and thereby implies more aggressive bidding behavior.

Secondary Outcomes

Secondary Outcomes (end points)
Secondary Outcomes (explanation)

Experimental Design

Experimental Design
See in hidden section
Experimental Design Details
The experiment consists of 10 rounds, out of which exactly one is relevant for player´s payoff. In each round, participants who receive an initial endowment of 60 ECU, are asked to bid on a lot, on which they write their personal winning numbers prior to the bidding process.
If they then win a certain round, they are allowed to keep the lot, and if not an experimenter takes it away from them.
The amount of winning numbers is randomly drawn and independent between bidders and rounds. It can either be 1, 2, 3, 4 or 5 winning numbers, with the probability for each amount being 20%.
In each round, each participants bids against 3 competitors that are randomly assigned to him. After the experiment, the relevant round is determined by rolling a 10-sided dice. The winner of the relevant round then again rolls the 10-sided dice, and if the number matches one of his winning numbers, the payoff of the lot is 100 ECU. If it does not match, the payoff of the lot is 0 ECU. Total payoff of participants is then 60 ECU if they did not win the relevant round, and 60 ECU minus their bid plus the lot's payoff if they won the relevant round. 1 ECU equals 20 Cents.
The bidding process is the only variable differing between treatments: In one treatment it is a standard first-price auction, while in the other it is a two-stage first-price auction. In the latter there are two semi-finals and a final, where only the two winners (and thus the ones submitting the highest semi-final-bid) qualify for the final, that is again a standard first-price auction. Final-bids cannot fall below semi-final-bids.
Randomization Method
Groups and amount of winning numbers are randomized by a computer; Relevant round and lottery outcome are randomized by a 10-sided dice
Randomization Unit
Experimental session
Was the treatment clustered?
No

Experiment Characteristics

Sample size: planned number of clusters
12-20 cohorts
Sample size: planned number of observations
48 - 80 subjects per treatment, 96 - 160 subjects total
Sample size (or number of clusters) by treatment arms
6-10 cohorts per treatment
Minimum detectable effect size for main outcomes (accounting for sample design and clustering)
IRB

Institutional Review Boards (IRBs)

IRB Name
IRB Approval Date
IRB Approval Number

Post-Trial

Post Trial Information

Study Withdrawal

There is information in this trial unavailable to the public. Use the button below to request access.

Request Information

Intervention

Is the intervention completed?
No
Data Collection Complete
Data Publication

Data Publication

Is public data available?
No

Program Files

Program Files
Reports, Papers & Other Materials

Relevant Paper(s)

Reports & Other Materials