Randomization for Causality, Ethnography for Mechanisms: Illiquid Savings for Liquor in an Autarkic Society

Working Paper: CEPR ID: DP16810

Authors: Ricardo Godoy; Dean Karlan; Jonathan Zinman

Abstract: What should researchers do when confronted with surprising results? Financial accessinnovations usually leave “temptation” spending unaffected or reduced. However, wefound that promotion of savings lockboxes in a largely autarkic society increased alcoholconsumption and blood pressure, despite no one reporting intentions to save for alcohol.To probe mechanisms that could explain this pattern, we then used ethnographic methods,including direct observations of drinking (“scans”) and debriefing interviews to discuss theearlier trial results. We learn that sponsoring drinks confers prestige, but the stigmaattached to drinking by outsiders likely discouraged reporting intentions to save for it.

Keywords: mixed methods; randomized evaluation; native amazonians; tsimane; temptation; commitment; savings

JEL Codes: D12; I12; O15


Causal Claims Network Graph

Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.


Causal Claims

CauseEffect
alcohol consumption (L66)blood pressure (G21)
savings lockboxes (D14)alcohol consumption (L66)
savings lockboxes (D14)blood pressure (G21)

Back to index