Competition and Bribery in Auction Experiments

Last registered on November 17, 2022

View Trial History

Pre-Trial

Trial Information

General Information

Title

Competition and Bribery in Auction Experiments

RCT ID

AEARCTR-0010402

Initial registration date

November 12, 2022

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

November 17, 2022, 3:45 PM EST

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

Locations

Country

Norway

Region

Oslo

Primary Investigator

Name

Thomas McKay

Affiliation

BI Norwegian Business School

Contact Primary Investigator

Other Primary Investigator(s)

Additional Trial Information

Status

In development

Start date

2022-11-14

End date

2022-12-30

Keywords

Behavior, Governance

Additional Keywords

Economics, Auction

JEL code(s)

C92, D44, D73

Secondary IDs

Prior work

This trial does not extend or rely on any prior RCTs.

Abstract

Does competition affect the impact of bribery on auction allocation and revenue? Bribery auctions are a collection of auction models where the auctioneer extracts a share of the seller revenue by soliciting bribes from potential bidders. This study aims to experimentally test this effect of competition in reducing the ineffeciency of bribery. To do this I introduce a two stage auction game. In the first stage the players can bid for the right of first refusal (ROFR) followed by an entry decision. Stage two is a standard first price auction (FPA) auction for the good with ROFR assigned to the winner of the first stage. I then vary the degree of competition in the first stage. This model is designed with experimental testing in mind, and thus has unique symmetric equilibria for any level of competition.

External Link(s)

Registration Citation

Citation

McKay, Thomas. 2022. "Competition and Bribery in Auction Experiments." AEA RCT Registry. November 17. https://doi.org/10.1257/rct.10402-1.0

Sponsors & Partners

Experimental Details

Interventions

Intervention(s)

This study is economics lab experiment testing a multistage bribery auction game when varying the degree of competition in bribery.

Intervention Start Date

2022-11-14

Intervention End Date

2022-12-30

Primary Outcomes

Primary Outcomes (end points)

The level of entry in each treatment.

Primary Outcomes (explanation)

There are three players in each auction game, thus entry will be the mean number of entrants measured at the block level

Secondary Outcomes

Secondary Outcomes (end points)

Efficiency, Bribery Auction Price, Goods Auction Price

Secondary Outcomes (explanation)

Efficiency is constructed as 1 when the subject with the highest valuation is allocated the good, and zero otherwise. Bribery Auction Price is the mean price in the bribery auction, measured at the block level. Goods Auction Price is the mean price in the goods auction, measured at the block level.

Experimental Design

This is a 3x1 study with between subjects design. Each game consists of 3 subjects who participate as bidders, while the seller and auctioneer are automated. The difference between treatments is the number of players competing in the first stage (3, 2 or 1). In addition, when there would be only a single subject in the bribery auction, I randomly assign ROFR to a random subject, as having a single bidder in an auction complicates the implementation and may be confusing for subjects. My design thus has three treatments. In treatment PC all subjects compete in the bribery stage. In treatment IC, 2 of the 3 subjects are randomly selected to participate in the bribery stage. In treatment NC, there is no bribery stage and instead the ROFR is assigned to a random subject.

Experimental Design Details

see prestudy plan

Randomization Method

Random assignment to treatment, and blocked lab design in groups of 6 with random rematching within blocks.

Randomization Unit

Block of 6 subjects.

Was the treatment clustered?

Yes

Experiment Characteristics

Sample size: planned number of clusters

30 Blocks

Sample size: planned number of observations

180 participants, 1200 Auction games.

Sample size (or number of clusters) by treatment arms

10 blocks per treatment, 30 blocks total

Minimum detectable effect size for main outcomes (accounting for sample design and clustering)

Power analysis on two sided rank sum test based on pilot data indicated that 7 blocks were required in order to reach power of 90% for primary variable. Since I chose blocks of 10 power is thus 98.3%

Supporting Documents and Materials

IRB