Adoption and impacts of sustainable intensification practices in Ghana

ABSTRACT

Sustainable agricultural intensification requires the use of multiple agricultural technologies in an integrated manner to enhance productivity while conserving the natural resource base. This study analyses the adoption and impacts of sustainable intensification practices (SIPs) using a dataset from Ghana. A multivariate probit (MVP) model was estimated to assess the adoption of multiple SIPs. Moreover, we used a multivalued semi-parametric treatment effect (MVTE) model to estimate the effects of adopting multiple SIPs on maize productivity. The MVP model results show, among others, that access to market, capital, and information/knowledge would enhance the adoption of SIPs. The MVTE model results show that a higher number of SIPs is associated with higher productivity which is more visible when commercial inputs are used in combination with cultural practices. These results have the following policy implications. First, they imply that good rural infrastructure and agricultural services such as rural road network, village-level input delivery system, input credit, and multiple information/knowledge sharing approach (instead of the conventional singular formal information/knowledge sharing approach) can enhance adoption. Second, the results suggest that promoting an integrated use of technologies, instead of a single technology, would have a positive impact on farm productivity and farm household income.

KEYWORDS:

• Sustainable intensification practices
• adoption
• impacts
• Ghana

Acknowledgements

This research was conducted as part of the Africa RISING programme. The authors thank the project team members at IFPRI, specifically Carlo Azzarri, for sharing the dataset.

Disclosure statement

No potential conflict of interest was reported by the authors.

ORCID

Bekele Hundie Kotu http://orcid.org/0000-0002-5488-8426

Irmgard Hoeschle-Zeledon http://orcid.org/0000-0002-2530-6554

Notes

1 Data from the Ministry of Food and Agriculture, http://mofa.gov.gh/site (last accessed on 16 July 2016).

2 The time frame is five years before the survey time.

3 The default number of random draw is 5 and the adjustment was made as suggested by Cappellari and Jenkins (2003).

4 We used a conventional count model (Poisson) for the pre-estimation analysis related to the number of SIPs. For the other option (i.e. category of SIPs), we used the multinomial logit model.

5 We could not report the other alternatives of accessing improved seeds because of lack of detailed data on this.

6 Decomposability of organic fertilizers varies from 10% to 60% during the first year which shows that they can serve as a reservoir of minerals for multiple seasons (van Opheusden, van der Burgt, & Rietberg, 2012).

7 This is based on the analysis of stochastic efficiency with respect to a function (SERF) (Hardaker, Richardson, Lien, & Schumann, 2004). The risk aversion coefficients we considered varies from 0 (corresponding to a risk-neutral person) to 0.0001 (corresponding to a risk-averse person).

8 We considered costs of seeds, chemical fertilizers, pesticides, and other costs such as payments for tractor services.

9 Farmers mostly pay higher than official prices. For instance, during the season under consideration, farmers actually paid on average GhC110 for 100 kg of NPK fertilizer although the official price was pegged at GhC71.5. This might be because of inefficient input markets characterized by high transaction and transport costs.

Additional information

Funding

The authors are grateful to the USAID/FtF Initiative for its financial support.