Adoption Pathways project discussion paper 7 February 2015 Maize and legume technology adoption in Malawi: Gender, social networks and SIMLESA effects Sam Katengeza, Henry Kankwamba, and Julius H. Mangisoni Lilongwe University of Agriculture and Natural Resources, Malawi Abstract This article investigates whether gender, social networks and being a SIMLESA beneficiary plays an important role in determining the level of maize and legume technology adoption. In order to do so, we exploit variation in random cross section data from 731 households in 2014. We use a multivariate probit regression model to analyze adoption of multiple technologies. Our approach allows sequential and simultaneous technology adoption and unobserved factors to be freely correlated across different technology practices. Our results unambiguously show that gender, social networks and being a SIMLESA beneficiary play a significant role. JEL classifications: Keywords: adoption; gender; social networks; technology; SIMLESA
15
Embed
Maize and legume technology adoption in Malawi: Gender ......Sustainable Intensification of Maize based Legume Systems (SIMLESA) program in order to enhance evidence based decision
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Adoption Pathways project discussion paper 7
February 2015
Maize and legume technology adoption in Malawi: Gender, social
networks and SIMLESA effects
Sam Katengeza, Henry Kankwamba, and Julius H. Mangisoni
Lilongwe University of Agriculture and Natural Resources, Malawi
Abstract
This article investigates whether gender, social networks and being a SIMLESA beneficiary
plays an important role in determining the level of maize and legume technology adoption. In
order to do so, we exploit variation in random cross section data from 731 households in 2014.
We use a multivariate probit regression model to analyze adoption of multiple technologies. Our
approach allows sequential and simultaneous technology adoption and unobserved factors to be
freely correlated across different technology practices. Our results unambiguously show that
gender, social networks and being a SIMLESA beneficiary play a significant role.
JEL classifications:
Keywords: adoption; gender; social networks; technology; SIMLESA
1. Introduction
Many Sub-Saharan countries see adoption of modern agricultural technology as an opportunity
to promote agricultural development. Malawi for example, has put technology adoption as one of
the key drivers of its agricultural sector wide approach. It is believed that technology adoption
will increase farm output, improve food security and eventually result in rising household
incomes through increased tradable surplus (Government of Malawi, 2011). Evidence suggests
that technology adoption can indeed accelerate agricultural growth. In Tanzania for example,
Amare et al. (2012) find that maize/pigeon pea adoption has a positive and significant effect on
incomes. This supports the widely held view that technology adoption increases household
incomes.
Technology adoption is often countered by the uncertainty of the timing when positive impacts
start to be realized. Giller et al. (2009) reported that some farmers who later adopted
conservation agriculture, ended up with negative returns in the first years. As such most
technologies are disseminated to farmers in bundles i.e. a technology package containing a
number of interventions aimed to increase productivity. This approach gives farmers the ability
to adapt the technologies to suit their own circumstances (citation). For example, female headed
households which are labor constrained tend to adopt technologies that demand less labor.
Women tend to play an important role in facilitating adoption of knowledge intensive
technologies (citation). Furthermore, farmers that are beneficiaries of particular projects tend to
adopt new technologies faster than non-beneficiaries (citation).
Adoption packages results in adopting packages that have possible complementarities and
tradeoffs. Farmers may adopt all the technologies in the package or may partially adopt. In
evaluating adoption of these technologies, it is necessary to take into account the simultaneous or
sequential decision making process and the possible trade-offs associated with these
technologies. Moreover, agriculture is area specific and farmer adoption behaviour changes
according to circumstances. As such, adoption of technology should be analyzed taking into
account area specific characteristics and farmer specific circumstances. To illustrate the merits of
such analysis, we develop a multivariate probit regression model of farmers who adopted
maize/legume intensification technologies taking into account the structure of social networks,
gender dynamics and the effect of being a beneficiary in the Sustainable Intensification of Maize
based Legume Systems (SIMLESA) project. The model takes into account how various
technologies relate with each other by providing a covariance matrix with correlation
coefficients. Section 2 outlines data sources and econometric approaches used; section 3 presents
results and section 4 outlines conclusions.
2. Methodology
2.1. The model
Agricultural technologies for sustainable intensification usually come in packages. Usually
farmers adopt part or complete packages. Adoption of mix of strategies makes dealing with
multiple production constraints a lot easier. Of note technologies might be adopted
simultaneously and/or sequentially as complements, substitutes or supplements and hence are
Of note, significant trade-offs occur in the fertility enhancing technologies such as manure-CR
and CF-CR. This might be an indication that these technologies are substitutes. This is not a
surprising result as Khakbazan et al. (2006) indicated that chemical fertilizers might have a
substituting relationship with crop rotation strategies. Teklewold et al. (2013) also find that
manure and crop rotation are used as substitutes. Marenya and Barrett (2007) found manure and
chemical fertiliser to be complementary, but a supplementary use (Teklewold et al. 2013) would
also make sense. It probably depends on the availability of fertilizer and household capital to
purchase chemical fertilizer. Crop rotation strategies usually use legumes to follow grass family
crops so that the legumes might provide nitrogen for the grass families that follow. However, if
chemical fertilizers are readily available it becomes easy to substitute them.
Further, RC-IC, manure-IC, CF-IC, manure-CF, IV-IC show positive results indicating that there
are complementarities between the technologies. Hoorman (2009) indicated that residue cover
can have some benefits on crops whether intercropped or standalone. It may provide much
needed moisture and nutrients to crops. Furthermore, since intercropping usually varies different
crops with different root depths together, manure and intercropping can have a complementary
relationship since crops have different demands (FAO, 2009).
3.2.Adoption of CA technologies
Table 3 presents results of the marginal effects of the multivariate probit model. Results indicate
that a number of socioeconomic, plot specific characteristics and location factors had
considerable explanatory power on adoption of CA technologies.
3.2.1. Socioeconomic factors
The number of individuals actively participating in labour provision positively influenced
intercropping adoption decisions. This is because most small holder farmers are labour
constrained and the number of workers in the field significantly conditions adoption of
intercropping (Citation). A similar explanation can be provided for the adoption of soil and
stone bunds.
Further, results also indicate that farmers who have had experience in growing improved legume
varieties were more likely to adopt residue cover and soil and stone bunds and a combination of
the two technologies. Nevertheless, farmers that had experience in growing improved maize did
not adopt residue cover and stone bunds. This might indicate that some technologies require
some experience before farmers adopt.
Membership to farmer associations influences adoption of residue cover positively. Farmers
influence each other when they are in groups. As such adoption can be accelerated when farmers
are in groups. Farmer groups in the study area were encouraged to utilize residue
Membership to credit associations positively influenced adoption of legume intercropping and
residue cover. When farmers have access to credit, they are able to adopt modern technologies
which might be resource intensive.
Noteworthy, if the EPA from which the household was sampled is a SIMLESA designated EPA,
individuals were more likely to adopt legume intercropping and soil and stone bunds,
respectively.
Land holding size negatively influenced adoption of legume intercropping and residue cover.
The main reason farmers intercrop is because they face land constraints. However, if farmers
have more land the incentive to intercrop becomes less of a problem. However, there might also
be an interaction between land and labour constraints. Further, residue cover is labour intensive
and bigger land sizes imply more labour to finish laying residue cover. Because of the extra cost
of labour associated with more land, farmers do not adopt residue cover. Table 3 Coefficient estimates of multivariate probit model (standard errors in parentheses) Explanatory variables