Uptake and Impact of Interlinked Index-based Insurance with Credit and Agricultural Inputs: Experimental Evidence from Ethiopia

How to cite

Marr, Ana; Belissa, Temesgen; Lensink, Robert. 2018. "Uptake and Impact of Interlinked Index-based Insurance with Credit and Agricultural Inputs: Experimental Evidence from Ethiopia." AEA RCT Registry. AEARCTR-0003504. https://doi.org/10.1257/rct.3504-1.0

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

Ana Marr* , Temesgen Belissa , Robert Lensink

* Primary investigator

Registry ID
AEARCTR-0003504
DOI
https://doi.org/10.1257/rct.3504-1.0
First registered
Last updated
Status
Completed
Trial dates
2017-05-01 – 2018-09-30
Countries
Ethiopia ()
Registry record
http://www.socialscienceregistry.org/trials/3504

Abstract

This paper examines the results of randomized experiments in Ethiopia that assess the relevance of bundling index-based insurance (IBI) with credit and inputs. We compare four IBI options and their impact on adoption of modern technologies, consumption and productivity: (1) standard IBI; (2) newly-developed IBI, i.e. promoted via farmer groups and featuring a delayed premium option; (3) new IBI bundled with credit; and (4) new IBI bundled with credit and inputs. We find that only when farmers adopt a package comprised of insurance, credit and inputs, do they significantly increase their investment in modern agricultural technologies and, consequently, productivity grows.

Intervention

We conducted our experiment with a local insurance company, Oromia Insurance Company (OIC), in the Rift Valley zone of Ethiopia. In the Rift Valley zone we randomly selected two kebeles , Desta Abjata and Qamo Garbi. Then, from the two kebeles, we randomly selected 47 farmer groups, called Garees in Ethiopia. To avoid ethical issues and to mitigate spillover effects, we used a cluster randomization to randomly assign the 47 garees into four groups: T1, T2, T3 and T4. All household heads from the 47 garees (in total 1661) are part of our experiment; all of them are farmers. Nobody is member of more than one garee.

Experimental design

Randomised Control Trial.

Design details

Randomization method
We randomly determined four groups, T1, T2, T3 and T4. T1 is the “control” group in the sense that this group did not get any specific treatment (or encouragement, or training). However, we were not allowed to impede T1 farmers from buying the standard index-based insurance from Oromia Insurance Company (OIC). In other words, T1 refers to the group who has access to the standard index-based insurance of OIC but nothing else. T2 group: We offer farmers a newly-designed index-based insurance product, whereby the product is promoted/delivered via trusting farmer groups and the premium payment is delayed after harvest. T3 group: We offer farmers the above-mentioned newly-developed IBI plus credit. T4 group: We offer farmers the above-mentioned newly-developed IBI plus credit and inputs.
Randomization unit
Individual
Clusters
47 clusters
Observations
1661 farmers
Arms
T1 group=420 farmers; T2 group=420 farmers; T3 group=401 farmers; T4 group=420 farmers. Total=1661 farmers.
Primary outcomes
Investment in modern agricultural technologies, e.g. better seeds, fertiliser. Agricultural production.
Secondary outcomes
Consumption

Cited by

No citation data available for this registration yet.

Keywords