The value of forecasts: Experimental evidence from developing-country agriculture

How to cite

Burlig, Fiona; Jina, Amir; Kelley, Erin; Lane, Gregory; Sahai, Harshil. 2022. "The value of forecasts: Experimental evidence from developing-country agriculture." AEA RCT Registry. AEARCTR-0008846. https://doi.org/10.1257/rct.8846-1.1

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

Fiona Burlig* , Amir Jina , Erin Kelley , Gregory Lane , Harshil Sahai

* Primary investigator

Registry ID
AEARCTR-0008846
DOI
https://doi.org/10.1257/rct.8846-1.1
First registered
Last updated
Status
Completed
Trial dates
2022-04-01 – 2023-12-31
Countries
India (Telangana)
JEL codes
D81, D25, O12, O13, Q12, Q54
Registry record
http://www.socialscienceregistry.org/trials/8846

Abstract

Climate risk is a key driver of low agricultural productivity in poor countries. We use a cluster-randomized trial to evaluate a novel risk-mitigation approach: long-range forecasts that provide information about the onset of the Indian summer monsoon well in advance of its arrival. In contrast to traditional approaches that allow farmers to cope with risk ex post, this new ex ante technology provides accurate information at least one month in advance of the monsoon's arrival, enabling farmers to alter cropping choices and other up front input decisions. Moreover, forecasts have the potential to be disseminated cheaply, even at scale. We assign 250 villages to one of three groups: a control group; a group that is given an opportunity to purchase the forecast; and a group that is offered insurance. This design allows us to investigate farmers' willingness-to-pay for forecasts; measure how forecasts affect farmer beliefs, up-front investments, and welfare; and study how these effects compare to the canonical ex post loss mitigation tool: weather-based index insurance.

Intervention

We implement a field experiment with a control group and two treatment arms. In the first treatment arm, farmers will be offered the opportunity to purchase a long-range monsoon forecast. In the second treatment arm, farmers will be offered the opportunity to purchase an index-based insurance product.

Experimental design

We conduct a cluster-randomized trial that assigns each of 250 villages to one of three groups. We stratify on village characteristics. We sample 5-10 households per village to survey and, if applicable, treat.

Design details

Randomization method
Stratified randomization done in office by a computer
Randomization unit
Village
Clusters
250 villages
Observations
1,250 -- 2,500 households
Arms
100 villages control, 100 villages forecast offer, 50 villages insurance offer
Primary outcomes
Willingness-to-pay; farmer beliefs; "ex ante" outcomes including area planted, labor use per area, fertilizer use per area, and crop choice; "ex post" outcomes, including total harvest, yield, value of crop sales, off-farm work/migration, household finances, and consumption expenditure.
Secondary outcomes
Please see our PDF PAP for detailed information on outcomes.

Cited by

No citation data available for this registration yet.

Keywords