Search and Information Frictions on Global E-Commerce Platforms: Evidence from AliExpress

Last registered on November 12, 2020

Pre-Trial

Trial Information

General Information

Title
Search and Information Frictions on Global E-Commerce Platforms: Evidence from AliExpress
RCT ID
AEARCTR-0006728
Initial registration date
November 10, 2020

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 12, 2020, 8:19 AM EST

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

Locations

Region

Primary Investigator

Affiliation
Harvard

Other Primary Investigator(s)

Additional Trial Information

Status
Completed
Start date
2017-05-01
End date
2018-12-01
Secondary IDs
Abstract
Global e-commerce platforms present new export opportunities for small and medium-sized enterprises in developing countries by significantly lowering the entry barriers of exporting. However, the lack of market selection can lead to a large number of online firms competing for consumers' attention, resulting in severe congestion in consumers' search process. When firms' intrinsic quality is not perfectly observed, these search frictions can further slow down the resolution of the information problem and hinder market allocation towards better firms. In this paper, we investigate how search and information frictions shape firm dynamics and market evolution in global e-commerce. Using detailed data from AliEpxress as well as a rich set of self-collected objective quality measures, we provide stylized facts that are consistent with the presence of search and information frictions. Moreover, using a randomized experiment that offers exogenous demand and information shocks to small prospective exporters, we establish that firms with larger past sales have an advantage in overcoming the search friction and generating future orders. This indicates that initial demand shocks could confound firms' true quality in determining firm growth and the long-run market structure. We construct and estimate an empirical model of the online market that are consistent with our descriptive and experimental findings and use the model to quantify the extent of demand-side frictions. Counterfactual analyses show that alleviating information frictions and reducing the number of firms can help to improve allocative efficiency and raise consumer welfare.
External Link(s)

Registration Citation

Citation
Bai, Jie. 2020. "Search and Information Frictions on Global E-Commerce Platforms: Evidence from AliExpress." AEA RCT Registry. November 12. https://doi.org/10.1257/rct.6728-1.0
Experimental Details

Interventions

Intervention(s)
We conduct an experiment in which we generate exogenous demand shocks to a set of small exporters via randomly-placed online purchase orders. We further interact the order treatment with a review treatment about firms' product and shipping quality to examine the role of information provision.
Intervention Start Date
2018-05-20
Intervention End Date
2018-07-30

Primary Outcomes

Primary Outcomes (end points)
firms' online sales (orders)
Primary Outcomes (explanation)

Secondary Outcomes

Secondary Outcomes (end points)
Secondary Outcomes (explanation)

Experimental Design

Experimental Design
Prior to the start of the experiment, a baseline data collection was conducted in May 2018 that covers the majority of the product listings in the sector of children's t-shirts on AliExpress. We group the product listings into 4,640 distinct variety groups. We focus on the ``popular'' varieties with greater than 100 orders, aggregated across all stores, and sold by at least two ``small'' stores with fewer than 5 cumulative orders. This screening procedure enables us to create a treatment and control group of ``identical'' product listings. In total, 133 varieties satisfy the above criteria, containing 1,265 product listings from 638 stores.

Of the 1,265 product listings, 790 are small listings with fewer than 5 orders. We randomize the 790 small listings into three groups of different order and review treatments: a control group C without any order and review treatment, T1 which receives 1 order randomly generated by the research team and a star rating, and T2, which, in addition to receiving an order and a star rating, further receives a detailed review on product and shipping quality.
Experimental Design Details
Randomization Method
randomization done in office by a computer
Randomization Unit
seller-product listing
Was the treatment clustered?
No

Experiment Characteristics

Sample size: planned number of clusters
NA
Sample size: planned number of observations
790 listings over 13 weeks
Sample size (or number of clusters) by treatment arms
We have 303 listings in C, 259 in T1, and 228 in T2
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