A lab-in-the-field experiment on Auction Design

Last registered on July 29, 2024

View Trial History

Pre-Trial

Trial Information

General Information

Title

A lab-in-the-field experiment on Auction Design

RCT ID

AEARCTR-0013641

Initial registration date

July 11, 2024

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

July 16, 2024, 3:30 PM EDT

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

Last updated

July 29, 2024, 2:19 PM EDT

Last updated is the most recent time when changes to the trial's registration were published.

Locations

Country

Burkina Faso

Region

Primary Investigator

Name

Andrew Dillon

Affiliation

Northwestern University

Contact Primary Investigator

Other Primary Investigator(s)

PI Name

Nicolo Tomaselli

PI Affiliation

University of Florence

Contact Investigator

Additional Trial Information

Status

In development

Start date

2024-05-01

End date

2024-09-30

Keywords

Agriculture, Behavior, Lab

Additional Keywords

JEL code(s)

D44 - Auctions, D82 - Asymmetric and Private Information, Mechanism Design, Q12 Micro Analysis of Farm Firms, Farm Input Markets

Secondary IDs

Prior work

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

Abstract

We study auction design in a low-income country where licenses to participate in a novel marketing system are auctioned among agricultural input dealers. We use auctions as a price discovery mechanism, and we measure bidders’ willingness to pay (WTP) to purchase licenses to take advantage of new business opportunities.

External Link(s)

Registration Citation

Citation

Dillon, Andrew and Nicolo Tomaselli. 2024. "A lab-in-the-field experiment on Auction Design." AEA RCT Registry. July 29. https://doi.org/10.1257/rct.13641-1.2

Sponsors & Partners

Experimental Details

Interventions

Intervention(s)

We have designed a field lab experiment with agricultural input dealers in Burkina Faso. In this country, adoption of agricultural technologies and innovations is low due to both supply and demand constraints. We use auction design to study the case of private-sector agricultural input dealers and how they price a market-based innovation, the Village Input Fair, that can increase their market share and, consequently, the uptake of agricultural inputs.

Auctions are used to determine prices (Berry et al., 2020), and our experiment consists of 5 rounds of auctions in which 60 input dealers bid to obtain licenses and sell agricultural inputs in rural villages during a Village Input Fair, or VIF — an event that allows input dealers to meet potential customers (Dillon and Tomaselli, 2023). We are particularly interested in understanding bidder responses to variations in reserve price rules and license grouping during the auction.

Intervention (Hidden)

Intervention Start Date

2024-07-15

Intervention End Date

2024-07-31

Primary Outcomes

Primary Outcomes (end points)

The main outcome of interest is the Willingness to Pay (WTP) of the bidders, which is proxied by the price they bid for the licenses for sale.

Primary Outcomes (explanation)

Secondary Outcomes

Secondary Outcomes (end points)

Secondary Outcomes (explanation)

Experimental Design

75 ag-input dealers will be invited to participate in the experiment and will be split into 3 groups. On each day, 25 ag-input dealers will participate in the experiment. Each ag-input dealer will participate in five rounds of bidding. Of the five rounds, one round is the control arm for the experiment, and the other four rounds are treatments. The arms as as follows:

In the control arm, no reserve price is set. In Treatment 1, the reserve price for each of the two licenses is set to half the marginal cost of organizing the fair. In Treatment 2, the reserve price is set to λ times the expected license revenues, where λ is the cost/revenue ratio of the whole license portfolio. In Treatment 3, the reserve price for each of the two licenses is set to half the marginal cost of organizing the fairs, and three fairs are auctioned together based on their geographical position. In Treatment 4, no reserve price is set, and three fairs are auctioned together based on their geographical position.

The lab in the field will generate data with the four treatment conditions described above plus the control. Since bidders will bid for licenses that, by construction, are statistically similar across treatment conditions, we expect to be able to neatly compare the moments of the bids with the different reserve prices and the with/without grouping.

Experimental Design Details

Randomization Method

Randomization done in the office on a computer using Stata.

Randomization Unit

Individual ag-input dealers randomized by day and randomized by round; order of the rounds randomized by day; order of the villages randomized within rounds.

Was the treatment clustered?

Yes

Experiment Characteristics

Sample size: planned number of clusters

There are 5 rounds, each one composed of 4 auctions. Each round will take one treatment.

Sample size: planned number of observations

1,200 bids (60 ag-dealers * 5 rounds * 4 auctions = 1,200 bids)

Sample size (or number of clusters) by treatment arms

240 bids per treatment arm

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

In our previous work in Mali similar sample sizes have been sufficient to detect statistically significant differences among treatment arms, ranging between 0.08 and 0.20.

Supporting Documents and Materials

IRB