Does Non-farm Income Affect Food Security? Evidence from India

Abstract

Livelihood diversification through greater non-farm activities has been considered as an important mechanism to propel growth, lower rural poverty and augment farm income across developing countries. Little, however, is known about its implications for nutritional outcomes such as dietary diversity. Using a nationally representative panel survey of rural households in India, and night-time light intensity as an instrumental variable (IV) for non-farm income, we show that engaging in non-agricultural livelihood has a positive effect on overall food expenditure, especially on non-cereal items, enabling greater dietary diversity. These findings have crucial policy implications for nutrition transition in India where agricultural incomes have been stagnant during the last decade. Our findings further contribute to the existing knowledge of agriculture-nutrition pathways.

Acknowledgements

We are grateful to Prabhu Pingali, Marc Bellemare, Jordan Chamberlin, Anaka Aiyar, Digvijay Negi, S. Chandrasekhar, Soham Sahoo, Amlan Das Gupta, Sripad Motiram, Goedele Van Den Broeck, Karthikeya Naraparaju and two anonymous referees for helpful discussions and comments on the draft. The paper has also benefited from participant comments at the Annual Meetings of the American Agricultural Economics Association (AAEA), International Association of Agricultural Economists (IAAE), Canadian Economic Association (CEA) and Population Association of America (PAA) in 2018, Global Food Symposium (University of Gottingen), PEGNet Conference (ETH Zurich), and IFMR seminar in 2017. Research assistance by Kiera Crowley is duly acknowledged. Research funding for this paper came from the initiative “System of Promoting Appropriate National Dynamism for Agriculture and Nutrition (SPANDAN)”, housed at the Indira Gandhi Institute of Development Research (IGIDR), Mumbai and supported by the Bill & Melinda Gates Foundation (BMGF). Any remaining errors are the authors’ own. Usual disclaimers apply.

Disclosure statement

No potential conflict of interest was reported by the authors.

Supplementary Materials

Supplementary Materials are available for this article which can be accessed via the online version of this journal available at https://doi.org/10.1080/00220388.2019.1640871

Notes

1. Multiple agriculture-nutrition pathways have been proposed in the literature over the last decade, such as (Bhagowalia, Headey, & Kadiyala, 2012; Kadiyala et al., 2014; Kanter, Walls, Tak, Roberts, & Waage, 2015).

2. Comparing India with other countries, income from non-farm sources account for about 35 per cent of total rural income in Africa and around 50 per cent in Asia and Latin America (Haggblade et al., 2010).

3. Since the effects of migration could be driven by the nature of migration – job quality, amount of income and so forth. – and household labour re-allocation, the set of studies reviewed by Thow et al. (2016) do not provide us with an idea of the true nature of this association. Most of the papers they reviewed include contexts where migration is a short-term coping phenomenon and hence plausibly little long-term productive impact.

4. For more details on the data, sampling strategy, attrition rates, and the design of consumption surveys, refer to Appendix A.

5. Refer to Appendix A for greater detail.

6. NREGS stands from the National Rural Employment Guarantee Scheme. This is a large-scale public works programs under which every rural household is guaranteed 100 days of employment. Implementation of NREGS began in India in 2006. For the same reason, the 2004-05 survey reports no income from NREGS.

7. Public Distribution System (PDS) is a targeted food-based assistance program in India.

8. For instance, Kilic, Carletto, Miluka, and Savastano (2009); Pfeiffer, López-Feldman, and Taylor (2009) employ the municipio (district) level share of non-agri employment as an instrument for off-farm income. Table O1 in the Supplementary Materials reports the list of instruments used in the existing literature on non-farm income.

9. Estimates from the set of panel regression model are presented in Appendix Table B1.

10. Village schedule of the survey instrument collects information on whether the village attracts inflow of workers or not.

11. For the sake of further robustness and comparability, we also use district-level share of non-agricultural workers as an instrument, which is most commonly used in the literature. Results are reported in Table O2 in the Supplementary Materials.

12. A detailed note on the method is presented in Appendix D.

13. Results and discussion are presented in Appendix E of the paper.

14. See http://indianexpress.com/article/india/india-news-india/farm-incomes-dreaming-to-double-2939405/.

Additional information

Funding

This work was supported by the Bill and Melinda Gates Foundation [OPP1029478].