Urban Amazonians use Fishing as a Strategy for Coping with Food Insecurity

Abstract

Fishing provides livelihoods and food for millions of people in the Global South yet inland fisheries are under-researched and neglected in food and nutrition policy. This paper goes beyond the rural focus of existing research and examines how urban households may use fishing as a livelihood strategy for coping with food insecurity. Our study in Brazilian Amazonia is based on a random sample of households (n = 798) in four remote riverine towns. We quantitatively examine the inter-connections between fishing and food insecurity, and find that fishing is a widespread coping strategy among disadvantaged, food insecure households. Fisher households tend to be highly dependent on eating fish, and for these households, consuming fish more often is associated with a modest reduction in food insecurity risks. Fishing provides monthly non-monetary income worth ≤ USD54 (equivalent to ∼12% of mean monetary income), potentially reducing food insecurity risks almost as much as the conditional cash transfer Bolsa Família. We estimate that nearly half a million inhabitants of the region’s remote, riverine urban centres are directly dependent on a household member catching fish, a nutritious and culturally preferred food. Consequently, small-scale urban fishers must be recognised in policy debates around food and nutrition security and management of natural resources.

Keywords:

• Towns
• capture fisheries
• safety net
• diet
• natural resources

Acknowledgements

We are grateful to G Davies, B Taylor, P Diggle, M Cunha, J Orellana and N Filizola for input into project design and planning. We thank data collectors including M Tavares, N Migon, G Correia, G Fink, M Freiré, L M L Silva and R F R Costa. Christina Hicks provided useful advice during the development of this paper.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Data availability statement

Code (in the R programming language; for data cleaning, processing, and analysis) and processed data can be provided upon reasonable request to the corresponding author.

Additional information

Funding

This work was supported by the UK’s ESRC under Grant ES/K010018/1 (Future Research Leaders Fellowship to L Parry); Newton Fund/FAPEAM under Grant ES/M011542/1; Brazil’s CNPq under Grant CsF PVE 313742/2013-8, Universal 455378/2014-2, CNPq/ANA/MCTI 446309/2015-0; CAPES-ProAmazonia under Grant Projeto 3322-2013; European Commission Horizon 2020 RISE programme under Grant Project 691053 (ODYSSEA).