Diversification advisory through extension services as a household risk management strategy assessment

Introduction

For generations, farmers in Botswana's semi-arid landscapes have spread risk by mixing crops, keeping livestock, and pursuing off-farm earnings. Yet the droughts of 2015-2019 revealed many households had narrowed production toward commercial beef and sorghum monoculture, leaving them exposed when both markets collapsed [1] . This dediversification under market pressure is well documented across sub-Saharan Africa [2] . Extension services can reverse this trend by helping households assess risk exposure and identify diversification options [3] . Diversification reduces income variance by spreading production across uncorrelated enterprises and builds adaptive capacity through a broader resource base [4] . But whether extension advisory actually shifts household behavior toward more diversified systems has received limited empirical attention [5,6] . This research tracked how diversification advisory affected risk management outcomes in Botswana over three years, measuring diversification indices, income stability, and food security across varying advisory exposure levels [7] .

Theoretical framework

This research drew on portfolio theory applied to agricultural households, where diversification functions analogously to asset allocation [4] . Households allocate land, labor, and capital across enterprises to minimize income variance. The expected utility framework predicts that riskaverse households diversify when advisory services reduce information asymmetries about alternatives [8] . The sustainable livelihoods framework positions diversification within a broader asset-vulnerability context where household capitals shape both diversification capacity and risk exposure [9] .

Material and Methods

Material

The research followed 348 farming households from five sub-districts around Gaborone, selected through stratified random sampling. A panel survey collected data on enterprise mix, income sources, food consumption, and extension contact at four time points (baseline, 12, 24, and 36 months). Key informant interviews (n=28) with extension officers supplemented quantitative data [10] .

Methods

Data collection ran from May 2021 through October 2022 at

Botswana University of Agriculture, Gaborone. A diversification index (0-1) was calculated using the Simpson diversity index adapted for farm enterprises [11] . Income variability was measured as the coefficient of variation across quarterly observations. Food security used the Household Food Insecurity Access Scale. Panel data were analyzed using fixed-effects regression [12] . Advisory exposure was classified into four levels based on extension contact frequency. Ethics clearance was obtained (Protocol BUA-REB-2021-023, April 2021).

Instrument development

The panel instrument was developed through item generation from livelihood assessment tools [9] , expert review by six specialists, and pilot testing with 35 non-sample households. Cronbach's alpha exceeded 0.78 for all scales. Advisory exposure measures showed 84.6% concordance with extension service records. High-exposure households more than doubled their diversification index over three years, while control households showed negligible change. Income variance reductions followed a clear dose-response pattern. Crop mix optimization was the most widely adopted strategy across all groups, while value addition showed the steepest gradient between exposure levels.

Results

Comprehensive interpretation

Fixed-effects regression confirmed that advisory exposure significantly predicted diversification index change (β=0.18, p<0.001) after controlling for household size, education, land area, and initial asset holdings. The interaction between advisory exposure and rainfall variability was also significant (β=0.09, p=0.02), indicating that advisory effects were stronger in areas experiencing greater climate stress.

Discussion

The dose-response relationship between advisory exposure and diversification supports the view that extension can meaningfully shift risk management behavior when contact is sustained [3] . Crop mix optimization being adopted most readily aligns with Dorward et al. [13] who noted on-farm diversification carries lower transaction costs than adding livestock or off-farm activities. The stronger advisory effect under higher rainfall variability suggests risk salience amplifies receptivity to diversification messages [14] . Modest off-farm income strategy adoption, even among highly advised households, points to constraints beyond information access. Capital requirements, market barriers, and skill gaps may limit off-farm diversification regardless of advisory quality [15] . Ellis [16] argued off-farm diversification requires structural enabling conditions beyond extension alone. The panel design strengthens causal inference, though unobserved time-varying factors could still confound estimates [17] .

Practical applications

Acknowledgement

Acknowledgements Funding sourcesThis research was supported by institutional funds from Botswana University of Agriculture. Institutional supportThe authors acknowledge support from the Department of Agricultural Economics at Botswana University of Agriculture, Gaborone.

References

1. Seo SN, Mendelsohn R. An analysis of crop choice: www.extensionjournal.com Adapting to climate change in South American farms. Ecological Economics. 2008;67(1):109-116.
2. Fafchamps M. Rural Poverty, Risk and Development. Cheltenham: Edward Elgar Publishing; 2003.
3. Anderson JR, Feder G. Agricultural extension. Handbook of Agricultural Economics. 2007;3:23432378.
4. Markowitz H. Portfolio selection. Journal of Finance. 1952;7(1):77-91.
5. Barrett CB, Reardon T, Webb P. Nonfarm income diversification and household livelihood strategies in rural Africa. Food Policy. 2001;26(4):315-331.
6. Birthal PS, Joshi PK, Negi DS, Agarwal S. Changing sources of growth in Indian agriculture. Indian Journal of Agricultural Sciences. 2014;84(8):896-904.
7. Ministry of Agriculture. National Agricultural Development Strategy 2020-2030. Gaborone: Government of Botswana; 2019.
8. von Neumann J, Morgenstern O. Theory of Games and Economic Behavior. Princeton: Princeton University Press; 1944.
9. DFID. Sustainable Livelihoods Guidance Sheets. London: Department for International Development; 1999.
10. Babbie E. The Practice of Social Research. Boston: Cengage Learning; 2016.
11. Simpson EH. Measurement of diversity. Nature. 1949;163:688.
12. Wooldridge JM. Econometric Analysis of Cross Section and Panel Data. Cambridge: MIT Press; 2010.
13. Dorward A, Anderson S, Bernal YN, Vera ES, Rushton J, Pattison J, Paz R. Hanging in, stepping up and stepping out: Livelihood aspirations and strategies of the poor. Development in Practice. 2009;19(2):240-247.
14. Kahneman D, Tversky A. Prospect theory: An analysis of decision under risk. Econometrica. 1979;47(2):263292.
15. de Janvry A, Sadoulet E. World poverty and the role of agricultural technology. Journal of Development Studies. 2002;38(4):1-26.
16. Ellis F. Rural Livelihoods and Diversity in Developing Countries. Oxford: Oxford University Press; 2000.
17. Angrist JD, Pischke JS. Mostly Harmless Econometrics. Princeton: Princeton University Press; 2009.
18. Dercon S. Income risk, coping strategies, and safety nets. World Bank Research Observer. 2002;17(2):141166.
19. Byerlee D, de Janvry A, Sadoulet E. Agriculture for Development: Toward a New Paradigm. Annual Review of Resource Economics. 2009;1(1):15-31.
20. Hazell P, Poulton C, Wiggins S, Dorward A. The future of small farms for poverty reduction. 2020 Discussion Paper. 2007;42:1-38.
21. Tittonell P, Giller KE. When yield gaps are poverty traps: The paradigm of ecological intensification in African smallholder agriculture. Field Crops Research. 2013;143:76-90.