Will Farmers Trade in Fire for Fertilizer? Evaluation of Regenerative No-burn Rice Stubble Management Technology in Myanmar

How to cite

Tauseef, Salauddin; Maredia, Mywish; Goeb, Joseph; Boughton, Duncan; Win, Myat. 2024. "Will Farmers Trade in Fire for Fertilizer? Evaluation of Regenerative No-burn Rice Stubble Management Technology in Myanmar." AEA RCT Registry. AEARCTR-0014587. https://doi.org/10.1257/rct.14587-2.0

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

Salauddin Tauseef* , Mywish Maredia , Joseph Goeb , Duncan Boughton , Myat Win

* Primary investigator

Registry ID
AEARCTR-0014587
DOI
https://doi.org/10.1257/rct.14587-2.0
First registered
Last updated
Status
In development
Trial dates
2024-03-01 – 2025-12-31
Countries
Private
Registry record
http://www.socialscienceregistry.org/trials/14587

Abstract

Rice cultivation is essential to Myanmar’s rural economy, but large-scale production results in substantial stubble, often burned to quickly prepare fields for the next planting. Farmers prefer burning as it is fast, cheap, and helps control pests. Consequently, this practice is widespread in Myanmar and other countries (Kanabkaew & Kim Oanh, 2011; Abdurrahman et al., 2020).
Yet, burning rice stubble carries high social and environmental costs. It can degrade soil structure, reduce organic matter (Sanchez et al., 2005), waste valuable biomass (Reicosky & Wilts, 2005), and pollute the air with greenhouse gases (Oanh et al., 2018). Alternatives that manage crop residues instead of burning can mitigate these issues and provide economic benefits.
Various options have been proposed, including reusing residues for economic purposes (e.g., mushroom cultivation, animal feed) and conservation agriculture practices (e.g., zero tillage, mulching). Though potentially profitable (Shyamsundar et al., 2019), these methods have not fully replaced burning due to practical challenges and lack of incentives for farmers (Oanh et al., 2013).
Understanding farmers' perceptions and responses to alternatives can aid in designing incentives to shift away from burning. This study utilizes efforts by Proximity Design, a social business working to promote climate resilient agricultural technologies in Myanmar, to promote a no-burn method in Myanmar, employing an Effective Microbial solution to decompose stubble. A randomized evaluation, in collaboration with Proximity Design, will assess the impact of this no-burn technology on various outcomes such as yield and fertilizer usage.

Intervention

The intervention being evaluated is the promotion of a no-burn rice farming practice by Proximity Designs (PD). This practice is implemented by PD’s frontline team, following these steps:

1. Village Outreach: A team of 50 sales representatives contacts village chiefs, either by phone or in person, to arrange appointments.

2. Village Presentations: On the scheduled day, the team visits the village and conducts presentations to groups of about 20 attendees, with an average of three meetings per village tract. These presentations target both double-cropping rice farmers (those planting in both the monsoon and post-monsoon seasons) and farmers who grow rice only during the monsoon season. The team promotes no-burn rice stubble management by emphasizing soil health benefits, cost savings from reduced fertilizer use, and potential yield improvements.

3. Sales Orders: During the visit, the team collects orders for EM (Effective Microorganism) bottles, which are integral to the no-burn practice. They expect about 35% of attendees to place orders, resulting in approximately 20 orders per village tract.

4. Order Fulfillment: After 3-6 weeks, the team returns to the village to deliver the EM bottles, priced at 7,500 MMK per bottle. Farmers pay in cash upon delivery and can modify or cancel their orders before the fulfillment date.

5. Product Quality Assurance: Farmers may also purchase EM from input stores, but PD guarantees the quality of its product, which ensures sufficient microorganism culture count for effectiveness.

Experimental design

The evaluation will employ a randomized-controlled trial (RCT) field experiment to explore whether adoption of the no burn stubble management technology have any economic impact as well as examine the household and farmer characteristics that influence the adoption of this technology among rice farmers in Myanmar.

Village tracts serve as both the units of intervention and randomization. Our observations will focus on farmers who participate in the intervention and purchase the product as described in step 4 above.

The intervention for this RCT will take place over three seasons with two cohorts. Village tracts are randomly assigned to treatment and control groups, and the intervention proceeds as follows:

• Cohort 1:
o Treatment Group: Receives the intervention at the start of the Monsoon 2024 season. Farmers who purchase the EM product are classified as cohort 1 treatment farmers (N=55 village tracts).
o Control Group: Receives the intervention at the start of the Post-Monsoon 2024-25 season (N=30 village tracts) or at the start of Monsoor 2025 (N=23 village tracts). Farmers who purchase the EM product during the intervention in their respective seasons are classified as cohort 1 control farmers.

• Cohort 2:
o Treatment Group: Receives the intervention at the start of the Post-Monsoon 2024-25 season. Farmers who purchase the EM product are cohort 2 treatment farmers (N=30 village tracts).
o Control Group: Receives the intervention at the start of the Monsoon 2025 season. Farmers who purchase the EM product are cohort 2 control farmers (N=32 village tracts).

Outcome Measurement:
• Surveys will be conducted at the end of Monsoon 2024 and at the end of Post-Monsoon 2024-25, For cohort 1, the outcomes will correspond to monsoon 2024 (season 1) and for cohort 2, they will correspond to post-monsoon 2024-25 (season 2).

The total sample size (i.e., units of analysis) for the surveys is targeted to be around 2442 households across 170 village tracts (the maximum number of village tracts feasible across the two seasons) with 14 to 15 farmers per village tract. We plan to develop survey instruments in October-November and obtain IRB approval by the end of November 2024.

Design details

Randomization method
Randomization of villages done in office by a computer
Randomization unit
Village tracts
Clusters
170 villages (85 villages each for treatment and control across the two cohorts)
Observations
The total sample size (i.e., units of analysis) for the impact evaluation survey is targeted to be around 2442 households across 170 village tracts (the maximum number of village tracts feasible across the two seasons) with 14 to 15 farmers per village tract (the most affordable within the budget constraint).
Arms
85 villages each for treatment and control
Primary outcomes
Key outcomes include: yield (expected to increase); inorganic fertilizer application (expected to decrease)

Cited by

No citation data available for this registration yet.

Keywords