Impacts of socio-psychological factors on smallholder farmers’ risk attitudes: empirical evidence and implications

ABSTRACT

In the literature, there are limited studies on how socio-psychological factors affect farmers' attitudes towards risks. Therefore, this paper looks into this issue in Ethiopia for better understanding of how socio-psychological issues influence farmers' risk attitudes. The study uses data from a cross-sectional survey and a generalised ordered logistic regression is used to analyse the data. The findings show that farmers perceive natural hazards, input and output price volatility, technological risks, financial shocks and human security as the main sources of risks for agricultural production and livelihoods. Farmers’ risk attitudes were assessed dividing them into three categories from more to less risk averse. Education, relational capital, attitudes, group membership, technical training and household sise were found to be the factors influencing farmers’ risk attitudes. Farmers who can read and write, who have strong social capital and who have received training are less risk averse. Finally, this study confirms the importance of positive attitudes, strong social capital (group membership and relational capital) and satisfactory competence to motivate farmers to take risks related to technological innovations and others in the agricultural production. Therefore, attention should be given to specific initiatives to enhance their awareness, build their adaptive capacity to shocks and hazards, provide timely and up-to-date information, and improve their skills and knowledge. These issues would help to avert the risk aversion of smallholder farmers, who are usually thought to be risk-averse in low-income economies and to stimulate them to adopt sustainable agriculture and technological innovations, which are expected to improve agricultural production and enhance the tolerance of local systems for risks and shocks.

KEYWORDS:

• attitude
• information
• social capital
• risk attitudes
• (generalised) ordered logit

JEL CLASSIFICATION:

• D14
• D21
• D71
• D91
• M53

Acknowledgments

We would like to extend our deepest gratitude to the organisations which covered the research cost of this study, especially Ghent University (BOF) and Mekelle University (e.g., NORAD-NORHED). We have also forwarded our special appreciation to the editor and anonymous reviewers for their constructive comment and valuable suggestions. Finally, we are grateful for the farmers who shared their time and provided valuable information.

ORCID

Woldegebrial Zeweld http://orcid.org/0000-0003-4017-5932

Notes

1. We would like to thank participants in the International Conference on Economics and Administration, 3–4 November 2016, Bucharest, Romania, who advised us to use a heterogeneous choice model to cross-check the validity of the results of the (generalised) ordered logit model. This heteroscedastic choice model verifies and accounts for scale or variance difference from some variables-hid difference in residual variance across the response levels. Therefore, it corrects for heteroscedasticity by simultaneously estimating two equations: determinants of risk attitudes (choice equation) and determinants of the residual variances (variance equation). Thus, null hypothesis is risk attitude effects do not differ across transitions by scale or residual variance factors.