Supply vs. Demand Constraints for E-Commerce Adoption

How to cite

Osman, Adam; Coville, Aidan; Piza, Caio. 2022. "Supply vs. Demand Constraints for E-Commerce Adoption." AEA RCT Registry. AEARCTR-0009680. https://doi.org/10.1257/rct.9680-1.1

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

Adam Osman* , Aidan Coville , Caio Piza

* Primary investigator

Registry ID
AEARCTR-0009680
DOI
https://doi.org/10.1257/rct.9680-1.1
First registered
Last updated
Status
On going
Trial dates
2018-11-01 – 2022-12-31
Countries
Private
JEL codes
L18; O12; O14
Secondary IDs
Illinois IRB18668
Registry record
http://www.socialscienceregistry.org/trials/9680

Abstract

E-commerce continues to grow in importance, but many MSME’s have not invested in their firm’s ability to sell online. This may be due to a lack of know-how (a “supply-side” constraint), or simply due to a lack of perceived benefit for selling online (a “demand-side” constraint). We test the relative importance of these two types of constraints using two randomized evaluations in the country of Georgia. In a first experiment we randomize which firms are offered a training program on how to get their firms online. In a second experiment we offer some firms a
promise that we would make a large purchase from them if we were able to make that purchase online. We will compare the likelihood that firms sell online in the months after our treatment.

Intervention

Training intervention: A 3-day face-to-face training on e-commerce basics, and covered: (Day 1) How to use Google, Face-
book, Instagram and Trip Advisor to increase the visibility of your business; (Day 2) How to understand customersˆa profiles, and how to register in e-commerce platforms to increase sales (like Bookings.com, Airbnb, hotels.com and other local platforms) and; (Day 3) How to develop a business model, access financial opportunities and grants, and participate in public procurement opportunities.

Demand Shock: Eligible firms were contacted by phone and told that an implementing partner would like to order their goods or services, with the condition that it would have to be done online. Firms were given two weeks in which to ensure they had an online business profile that included, at a minimum, their business name, logo, description of their goods and services (including photos and/or videos), contact in-
formation for how to make an order, and an option for review feedback on the site. Once the site was in place, they would place an order. Payments were needed to be made by bank transfer or directly on the website (no cash option), and the firm had to be able to deliver the product to the partner offices in Tbilisi. In the case of services, this would come in the form of a voucher. Firms were either provided the ”high” or the ”low” demand shock offer. The ”low” demand shock offer was priced to be equivalent to the cost of the e-commerce training ($ 130) and the ”high” demand shock was set at $ 780 - five times the size of the ”low” offer.

Experimental design

Experiment 1: Interested businesses were randomized into being provided the e-commerce training or to serve as a control group.

Experiment 2: Businesses were randomized into 3 groups: small demand shock, large demand shock, control.

For experiment 2 we stratified by a firms "readiness" to sell online. We split firms into 3 groups, a "likely ready" group, a "more effort group", and an "unlikely to be ready" group. As of the writing of this registration we have not yet collected data on the "unlikely to be ready group" but we expect them to respond least to the demand shock.

Design details

Randomization method
Randomization don in office by a computer.
Randomization unit
Individual (firm owner) level
Clusters
Experiment 1: 858 firms Experiment 2: 283 firms
Observations
Experiment 1: 858 firms Experiment 2: 283 firms
Arms
Experiment 1: 220 control, 638 treatment Experiment 2: 76 small shock, 76 large shock, 131 control
Primary outcomes
Ability to sell good online
Secondary outcomes
Business performance (revenue, number of customers, innovation)

Cited by

No citation data available for this registration yet.

Keywords