Technical ReportPDF Available

Improving Knowledge, Inputs and Markets for Legume Expansion: A Contribution Analysis of N2Africa in Ghana and Ethiopia

Authors:

Abstract and Figures

This report documents the findings of an evaluation study of the impacts of N2Africa, a ten-year ‘research-in-development’ project that aimed to harness nitrogen fixation for the benefit of small-scale farmers of legume crops. The study analysed N2Africa’s contribution to development outcomes using the methodology of process tracing as a structured way of critically reviewing the change process. The evaluation focused on N2Africa’s activities in Ghana and Ethiopia. In Ghana, the evaluators examined the evidence base underlying the contribution claim, ‘N2Africa has been a relevant contributory factor in the expansion of soybean production in northern Ghana’. In Ethiopia, they verified the claim that ‘N2Africa has contributed to the increase in production, distribution, uptake and expansion of market demand for legume inoculants’. The evaluation identified the critical causal assumptions underlying the project’s Theory of Change in each case. After reviewing available project documentation and other literature, additional data were gathered through stakeholder interviews. The study sought to verify whether the expected changes had taken place, and assessed the size and importance of N2Africa’s contribution to each of the observed outcomes. The study found convincing evidence that N2Africa contributed substantially to a process of technological upgrading of soybean production in northern Ghana, yet the project played a relatively small part in the overall increase in soybean production in that region. In Ethiopia, there was clear evidence that N2Africa had made a decisive contribution to expanding the production and supply of legume inoculants, and had stimulated awareness of and demand for inoculants among small numbers of legume farmers. However, the project has only helped the market for legume inoculants to reach a small fraction of its potential.
Content may be subject to copyright.
IDS PRACTICE PAPER 10
1
PRACTICE PAPER
Volume 2019 Number 10
Improving Knowledge, Inputs and
Markets for Legume Expansion:
A Contribution Analysis of
N2Africa in Ghana and Ethiopia
Giel Ton and Dominic Glover
December 2019
IDS PRACTICE PAPER 10
2
About IDS
The Institute of Development Studies (IDS) is a global research and learning organisation for
equitable and sustainable change. In partnership with the University of Sussex, IDS is ranked
rst in the world for development studies by the QS University Rankings.
We believe passionately that cutting-edge research, knowledge and mutual learning are
crucial in shaping the transformations needed to reduce inequalities, accelerate sustainability,
and build more inclusive and secure societies. In line with our engaged excellence approach,
we are committed to collaborating closely with local civil society, governments, international
non-government organisations (NGOs), citizens, donors and many others to realise this positive
change.
IDS has been working with partners to tackle complex development challenges in this way since
1966. Our partnership with the University of Sussex supports and accredits our Master’s degrees
and PhD research programmes. We also offer a range of thematically and approach-focused
professional development courses, learning journeys, and online and blended learning. Through
our learning offer we aim to strengthen the skills and capabilities of those already working within
the development sector, as well as nurture a new generation of development thought, policy and
practice leaders.
IDS publishes a wide range of high-quality and high-impact outputs, including our agship
publication, the IDS Bulletin – an open access, peer-reviewed journal exploring emerging
international development challenges.
The Institute is home to more than 200 staff and 200 students at any one time. But the IDS
community extends far beyond, encompassing an extensive global network of more than
360 partners, 3,000 alumni and hundreds of former staff.
For more information visit: www.ids.ac.uk.
For further information on IDS publications, contact:
IDS Communications and Engagement
Institute of Development Studies
Brighton BN1 9RE, UK
Tel: +44 (0) 1273 915637
Web: https://www.ids.ac.uk/research/publications/
IDS is a charitable company, limited by guarantee and registered in England (No. 877338).
IDS PRACTICE PAPER 10
1
IDS PRACTICE PAPER 10
Improving Knowledge, Inputs and
Markets for Legume Expansion: A
Contribution Analysis of N2Africa in
Ghana and Ethiopia
Giel Ton and Dominic Glover
December 2019
Institute of Development Studies Brighton BN1 9RE UK
IDS PRACTICE PAPER 10
2
Improving Knowledge, Inputs and Markets for Legume Expansion: A Contribution Analysis of N2Africa in Ghana
and Ethiopia
Giel Ton and Dominic Glover
IDS Practice Paper 10
N2Africa Report 121
First published by the Institute of Development Studies in December 2019
© Wageningen University & Research 2019
ISSN: 2040-0209 ISBN: 978-1-78118-598-8
This is an Open Access paper distributed under the terms of the Creative Commons Attribution
Non Commercial 4.0 International licence (CC BY-NC), which permits use, distribution and
reproduction in any medium, provided the original authors and source are credited, any modifications or adaptations
are indicated, and the work is not used for commercial purposes.
http://creativecommons.org/licenses/by-nc/4.0/legalcode
This publication has been funded by the Bill & Melinda Gates Foundation through a grant to
Wageningen University entitled ‘Putting nitrogen fixation to work for smallholder farmers in Africa’.
Its content does not represent the official position of Bill & Melinda Gates Foundation, Wageningen
University & Research, Institute of Development Studies, or any of the other partner organisations
within the project and is entirely the responsibility of the authors.
The information in this document is provided as it is and no guarantee or warranty is given that the information is fit
for any particular purpose. The user thereof uses the information at their own sole risk and liability.
Giel Ton: ORCID – https://orcid.org/0000-0003-1730-9517;
Dominic Glover: ORCID – https://orcid.org/0000-0003-0956-0103;
Institute of Development Studies
Brighton BN1 9RE, UK
Tel: +44 (0) 1273 915637
Web: https://www.ids.ac.uk/research/publications/
IDS is a charitable company limited by guarantee and registered in England
Charity Registration Number 306371
Charitable Company Number 877338
IDS PRACTICE PAPER 10
3
Foreword
As N2Africa moved into its tenth and nal year, my project colleagues and I were
busy with impact studies. These comprised a number of formal, large-scale,
questionnaire-based evaluations, which specically addressed the impacts of our
efforts at farm household level. Talking to my colleagues, I became increasingly
concerned that the household surveys would not capture important learnings
from the project. I was thinking about how best we could capture other kinds of
outcomes and learnings.
Impact studies based on survey questionnaires are the standard tool for looking
at changes at household level, but they have some shortcomings. They provide
snapshots of the situation at a specic moment in time, although we know that
‘adoption’ cannot be understood simply as an on–off switch. At the same time, in
N2Africa we tried to get messages out as widely as possible – in effect, we were
intentionally ‘contaminating’ any potential control group of households. Further, we
had coordinated interventions at so many levels across the value chain that I was
concerned they could not be captured by household surveys.
I discussed these concerns with my colleague and collaborator, IDS Fellow Jim
Sumberg, when he was visiting my group in Wageningen. I suggested that one
way to try and understand the broader impacts of N2Africa could be to test each of
the assumptions made between the steps of the project’s Theory of Change. Jim
responded: ‘But that sounds like Contribution Analysis – an established method
that is used by my colleague Giel Ton.’
That is how the N2Africa team came to work with Giel and Dominic Glover on the
two case studies presented in this report. There is no doubt that with Contribution
Analysis we have learned more about the outcomes and impacts of actions taken
by N2Africa over the past years. We benet from having our work held up to the
light for critical examination, and this report provides a most useful complement to
the other impact studies that are currently underway.
I commend the Contribution Analysis approach to you – and will certainly try and
use it in a more concerted way in other ongoing and future projects.
Ken Giller
N2Africa Project Lead
Professor of Plant Production Systems
Wageningen University
IDS PRACTICE PAPER 10
4
Improving Knowledge, Inputs and Markets for Legume Expansion:
A Contribution Analysis of N2Africa in Ghana and Ethiopia
Giel Ton and Dominic Glover
Summary
This report documents the ndings of an evaluation study of the impacts of
N2Africa, a ten-year ‘research-in-development’ project that aimed to harness
nitrogen xation for the benet of small-scale farmers of legume crops. The study
analysed N2Africa’s contribution to development outcomes using the methodology
of process tracing as a structured way of critically reviewing the change process.
The evaluation focused on N2Africa’s activities in Ghana and Ethiopia. In Ghana,
the evaluators examined the evidence base underlying the contribution claim,
‘N2Africa has been a relevant contributory factor in the expansion of soybean
production in northern Ghana’. In Ethiopia, they veried the claim that ‘N2Africa
has contributed to the increase in production, distribution, uptake and expansion
of market demand for legume inoculants’. The evaluation identied the critical
causal assumptions underlying the project’s Theory of Change in each case. After
reviewing available project documentation and other literature, additional data
were gathered through stakeholder interviews. The study sought to verify whether
the expected changes had taken place, and assessed the size and importance of
N2Africa’s contribution to each of the observed outcomes.
The study found convincing evidence that N2Africa contributed substantially to
a process of technological upgrading of soybean production in northern Ghana,
yet the project played a relatively small part in the overall increase in soybean
production in that region. In Ethiopia, there was clear evidence that N2Africa had
made a decisive contribution to expanding the production and supply of legume
inoculants, and had stimulated awareness of and demand for inoculants among
small numbers of legume farmers. However, the project has only helped the
market for legume inoculants to reach a small fraction of its potential.
Keywords: impact evaluation; technological change; legumes; rhizobium;
input markets.
Giel Ton is a social scientist at the Institute of Development Studies and Director
of the Centre for Development Impact. He specialises in the design of mixed-
methods impact evaluations in agricultural value chains and private sector
development, and applies Contribution Analysis as an overarching approach.
His previous research has focused on contract farming, collective marketing,
innovation grants, and certication.
Dominic Glover is an interdisciplinary social science researcher specialising in
technology, agriculture and agrarian change. His previous research has focused on
the emergence and spread of new technologies, cultivation methods and farming
practices through small-scale agricultural systems, including transgenic crops and
alternative methods of rice cultivation.
IDS PRACTICE PAPER 10
5
Executive summary
The N2Africa project was a ten-year ‘research-in-development’ project, which
aimed to harness nitrogen xation for the benet of African small-scale farmers of
legume crops. The key productivity-enhancing technologies promoted by N2Africa
were improved legume varieties in conjunction with fertilisers and rhizobium
inoculants. Rhizobia are living organisms that have a symbiotic relationship with
leguminous plants and x nitrogen from the atmosphere. The project also worked
to strengthen value chains and improve legume farmers’ access to input and
output markets.
N2Africa gathered and monitored information about the impact of its work.
However, the diversity of N2Africa’s interventions, their dynamism, and the widely
different contexts where these have been implemented, make it challenging
to derive strong inferences about the project’s impacts from survey-based
impact evaluations. Therefore, N2Africa also used institutional information and
community-level data collected using qualitative methods.
N2Africa’s objective was to construct a coherent and evidence-based contribution
narrative – in other words, a narrative that explains and demonstrates how the
project’s activities contributed to the impacts indicated in the Theory of Change.
The aim of this type of analysis is to support those involved in a project to develop
a fuller understanding of what they had achieved, and to construct more rened
Theories of Change that may be used to scale and replicate interventions in the
future.
To assist N2Africa with this effort, a team of external evaluators from the Institute
of Development Studies (IDS, Brighton, UK) gathered and critically analysed
evidence underlying N2Africa’s claims about its impact. The evaluation team used
an approach known as Contribution Analysis. This approach is informed by the
methodology of process tracing, which is a structured way of critically reviewing a
process and sequence of change.
The evaluation focused on two case studies, in Ghana and Ethiopia respectively,
which were considered by N2Africa stakeholders to be promising for replication or
upscaling. After reviewing available project documentation and other literature, the
evaluation proceeded by rst identifying critical assumptions underlying the Theory
of Change. The evaluation team then gathered additional data through stakeholder
interviews. The interviews were used alongside the documentary evidence to verify
whether the expected changes had indeed taken place, and assessed the size and
importance of N2Africa’s contribution to each of these outcomes.
In the Ghana case, the evaluators examined the evidence base underlying
the contribution claim, ‘N2Africa has been a relevant contributory factor in the
expansion of soybean production in northern Ghana’. In Ethiopia, they veried
the claim that ‘N2Africa has contributed to the increase in production, distribution,
uptake and expansion of market demand for legume inoculants’.
This report documents the evaluation process and summarises the ndings.
The study found convincing evidence that N2Africa contributed substantially to
IDS PRACTICE PAPER 10
6
a process of technological upgrading of soybean production in northern Ghana,
which, however, is still only weakly contributing to the expansion of soybean
production in that region. In Ethiopia, there was clear evidence that N2Africa had
made a decisive contribution to expanding the production and supply of legume
inoculants, and had stimulated awareness of and demand for inoculants among
small numbers of legume farmers. However, stakeholders believe that the market
for legume inoculants has only reached a small fraction of its potential.
IDS PRACTICE PAPER 10
7
Contents
Acknowledgements 8
Acronyms 8
1 Introduction 9
2 Case 1 – Verifying the contribution claims of N2Africa related to
soybean expansion in northern Ghana 11
2.1 Critical assumption 1: Soybean production in Northern Ghana is
expanding 12
2.2 Critical assumption 2: N2africa recommended technologies are
used by smallholders 13
2.3 Critical assumption 3: N2Africa recommended technologies are
available to smallholders 15
2.4 Critical assumption 4: N2Africa recommended technologies can
improve yields and income 17
2.5 Conclusion 19
3 Case 2 – Verifying the contribution claims of N2Africa related to
the adoption of legume inoculants by smallholders in Ethiopia 21
3.1 Critical assumption 1: N2Africa support increased the demand for
legume inoculants by smallholders 24
3.2 Critical assumption 2: N2Africa increased the quality and
availability of legume inoculants in Ethiopia 27
3.3 Critical assumption 3: N2Africa strengthened the potential of
inoculant technologies with legumes as a viable commercial
opportunity for small-scale producers and other stakeholders 30
3.4 Conclusion 32
4 Discussion: the value of Contribution Analysis 34
Annexe 36
References 37
Figures
Figure 1.1 Hoop tests to assess the strength of the contribution claim 10
Figure 2.1 The Theory of Change of N2Africa contribution to soybeans
expansion in northern Ghana 11
Figure 3.1 The Theory of Change of N2Africa’s contribution to the
expansion of inoculant technologies in Ethiopia 23
Tables
Table 3.1 The N2Africa Ethiopia programme 22
Boxes
Box 3.1 N2Africa in Ethiopia 21
IDS PRACTICE PAPER 10
8
Acknowledgements
We acknowledge the detailed comments on earlier drafts of this report from
Eva Thuijsman, Esther Ronner, Imogen Bellwood-Howard, Samuel Adjei-Nsiah
and Endalkachew Woldemeskel. Moreover, we want to thank all interviewees in
Ghana and Ethiopia, and participants at the workshop held on 26–27 February 2019
in Wageningen, the Netherlands.
Acronyms
AGRA Alliance for a Green Revolution in Africa
BMGF Bill & Melinda Gates Foundation
CRS Catholic Relief Services
EPDRA Evangelical Presbyterian Development and Relief Agency
HARC Hollota Agricultural Research Centre (Ethiopia)
ILRI International Livestock Research Institute
MBI Menagesha Biotech Industry (Ethiopia)
NARS National Agricultural Research System (Ethiopia)
NGO non-government organisation
NPK nitrogen, phosphorus, potassium
NSTC National Soil Testing Centre (Ethiopia)
P-fertiliser Fertiliser with a phosphorus component
PFJ Planting for Food and Jobs (Ghana)
PPP public–private partnership
R&D research and development
SSTP Scaling Seeds and Technologies Partnership (AGRA)
TSP triple superphosphate
USAID United States Agency for International Development
IDS PRACTICE PAPER 10
9
1 Introduction
The N2Africa project was a ten-year ‘research-in-development’ project, which
aimed to harness nitrogen (N2) xation for the benet of African small-scale
farmers of legume crops. The project carried out a number of interlinked
activities in several countries, including Ghana and Ethiopia, to improve the lives
of smallholders by increasing the productivity of legume cultivation. The key
productivity-enhancing technologies promoted by N2Africa were improved legume
varieties in conjunction with fertilisers and rhizobium inoculants. Rhizobia are
living organisms that have a symbiotic relationship with leguminous plants, and x
nitrogen from the atmosphere. The project also worked to strengthen value chains
and improve legume farmers’ access to input and output markets.1
N2Africa gathered and analysed data to monitor its progress and study the impact
of its activities. The project team’s principal approach to quantitative evaluation
was the matched difference-in-difference survey, a method that compares the
situations before and after N2Africa’s intervention, for the groups targeted by
the project and one or more control groups. The project team also gathered
institutional information and community-level data, using qualitative methods.
Together, the quantitative and qualitative data were intended to provide evidence
that would show whether and how N2Africa may have contributed to the impacts
envisaged in the project’s Theory of Change. Because N2Africa made a range of
different interventions, and implemented them exibly and dynamically across a
broad spectrum of African contexts, it proved challenging to derive generalised
inferences about the project’s impacts using monitoring information and survey-
based quasi-experimental designs.
As an additional impact evaluation method, N2Africa appointed us, the authors, as
external evaluators, to critically analyse the evidence behind claims that N2Africa
contributed to the impacts envisaged in the project’s Theory of Change, and reect
on the contextual conditions and causal mechanisms that led to the project’s
outcomes. We introduced and applied the methodology of Contribution Analysis
(Mayne 2001; Mayne 2011; Ton 2017) to test and reect on the project’s Theory of
Change, learn lessons, and develop more rened theories that could be used in
the future to replicate successful activities and to scale up and out.
We worked as external evaluators and had not been involved in N2Africa before;
however, we worked closely with N2Africa researchers and staff. We used
N2Africa’s documentation as major sources of information, and the project team
members as the key informants who were most knowledgeable about the project,
most keenly interested in it, and best placed not only to reect on its successes
and failures but to consider the emerging ndings of the Contribution Analysis, and
to learn lessons from the conclusions.
In two case studies, we used an analytical approach informed by process tracing
(Punton and Welle 2015; Beach and Pedersen 2013; Befani and Mayne 2014).
Process tracing offers a structured way to verify causal inferences, by critically
1 See www.n2africa.org.
IDS PRACTICE PAPER 10
10
reviewing the process and sequence of change. The analysis begins by inferring
the existence of a causal claim, by reviewing programme documents and eliciting
the (implied) Theory of Change. Then, we gathered additional data through
stakeholder interviews, to verify whether the claimed outcomes were indeed
delivered, and whether the project can be considered a necessary (non-redundant)
contributory factor in these change processes. Finally, we used the available
evidence to characterise the strength of the contribution claims, using ‘hoop tests’
(Figure 1.1). The gure illustrates the approach and indicates that a stronger
contribution claim is linked to the passing through a smaller (more stringent) hoop.
The hoop tests are ‘counterfactual thought experiments’, based on rigorous logical
reasoning (Belkin and Tetlock 1996).
Figure 1.1 Hoop tests to assess the strength of the contribution claim
Source: Authors’ own.
In each case study, stakeholder interviews were used to answer ve generic
questions, which were designed to assess the strength of the contribution claim
made by N2Africa:
Did the change occur?
Did it result from a process in which N2Africa support was used?
Can this support be considered as a necessary (non-redundant) causal
factor for that process to have taken place?
If not, was it a necessary causal factor in accelerating or scaling of the
outcomes?
Were there any other institutions or programmes that may have provided
similar support to the change process, if N2Africa had not been present?
IDS PRACTICE PAPER 10
11
In a workshop with all N2Africa country coordinators, we started with a reection on
the rationale behind N2Africa, leading to a rened Theory of Change with detailed
impact pathways for the two cases. The impact pathways imply sets of interlinked
assumptions about the immediate and intermediate outcomes that would need to
be in place in order to have a fair chance to generate impact at scale.
2 Case 1 – Verifying the contribution
claims of N2Africa related to soybean
expansion in northern Ghana
This section discusses the evidence behind the contribution story of N2Africa
regarding the expansion of soybean cultivation in northern Ghana. The contribution
claims are clearly documented in N2Africa Success Story: Putting Nitrogen
Fixation to Work for Smallholder Farmers in Northern Ghana (Dotse and Badu
2018). In the February 2019 workshop with all N2Africa staff, the work in Ghana
was considered to have had impact at scale and contributed to the expansion of
soybean production in the region.
The work in N2Africa’s second phase (2014–19) builds on the technology
development and validation work of an earlier phase (2009–13). It has a clear
objective to increase the availability and adoption of nitrogen xation techniques
by smallholder farmers. The intervention logic that provides the framework for this
contribution story is shown in Figure 2.1.
Figure 2.1 The Theory of Change of N2Africa contribution to soybean
expansion in northern Ghana
Note: Revised version of the gure discussed in the February 2019 N2Africa workshop.
Source: Authors’ own.
IDS PRACTICE PAPER 10
12
Figure 2.1 depicts ve interlinked impact pathways with their respective immediate
outcomes in green. It shows: one pathway related to training; three related to
technologies (seeds, inoculants, fertiliser) that are used in a farmer-to-farmer
process of adoption and diffusion; and a fth pathway to create a favourable policy
environment of subsidies in the Planting for Food and Jobs Programme to facilitate
the adoption of the technologies by smallholder farmers. It is also clear that
N2Africa is well aware that the ve pathways are meant to work in combination,
making it a synergetic process of change.
We reviewed the evidence base behind the contribution claim: ‘N2Africa has
been a relevant contributory factor in the expansion of soybean production in
northern Ghana’. We differentiated four assumptions that underpin the contribution
claim, more or less aligned with the links between the activities, immediate
outcomes, intermediate outcomes and ultimate outcomes in the Theory of Change
(Figure 2.1), and veried whether these were in fact delivered, and what the
importance of N2Africa has been on each of these.
Critical assumption 1: Soybean production in northern Ghana is expanding.
Critical assumption 2: N2Africa-recommended technologies are used by
smallholders.
Critical assumption 3: N2Africa-recommended technologies are available
to smallholders.
Critical assumption 4: N2Africa-recommended technologies can improve
yields and incomes.
2.1 Critical assumption 1: Soybean production in northern Ghana is
expanding
Soybean production existed in northern Ghana well before N2Africa started, but
as a minor crop only, mixed-cropped with maize. Soybean was introduced in
smallholder agriculture in the 1980s, rst mainly as a fodder crop and gradually
incorporated as a minor crop for human consumption. In 2005, a study by the
Overseas Development Institute (ODI) mentioned that two companies buy soya
from farmers organised by non-government organisations (NGOs):
Oilseeds are promising both for the internal market and, if enough can be
produced, for export. Soybean, in particular, seems capable of expanding
as a commercial crop supplied to oil mills, where it is difcult for mills to
get enough groundnuts as these are in heavy local demand. There has
been signicant growth in soybean production, encouraged by some
NGOs. Both soybeans and groundnuts are both food and cash crops, and
therefore very attractive to farmers. Both contribute nitrogen to farming
systems. A rst step in expanding production for the market would be
the organisation of a producers’ association (which could be oilseeds or
crop-specic).
(Shepherd et al. 2005: 32)
The establishment of the processing industry (Ghana Nuts, Bosbel Oil Industries)
around 2004 was a decisive factor in positioning soybean production as a cash
IDS PRACTICE PAPER 10
13
crop in smallholder farming systems. In the farmer diet, soybeans were specially
used for the preparation of dawadawa (a fermented food condiment originally
made from the locust bean tree but nowadays increasingly made from soybeans).
The uptake of soybean as a cash crop is facilitated by the fact that it has some
advantages over other crops, like maize and groundnut. Soybeans are procured
year-round by agents that deliver to the oil processors, and can be stored without
much post-harvest loss during several months inside the farmhouse. Therefore,
farmers do not need to sell all their produce immediately after harvest. According
to the interview with Joshua Nyaaba of the Evangelical Presbyterian Development
and Relief Agency (EPDRA)-Yendi, farmers tend to sell around half of the harvest
at harvest time to pay for the outstanding loans used for production (and also for
school fees or other expenses), and sell the other half spread across the months
before planting. According to Yara Ghana, due to the possibility to save soybeans
in-house until planting time, soybean production may develop as an important
source of cash for smallholders to buy fertiliser for maize.
Data on regular soybean production in northern Ghana are notoriously unreliable
according to N2Africa staff, though all interviewees (see Annexe) believed that
there has been an expansion of soybean production. In the absence of ne-
grained ofcial data (e.g. Mohammed, Al-hassan and Jatoe 2018), there are
supportive indirect indications of this expansion, especially through the increased
domestic supply of soybean to processors, and growing exports of soybeans
(to Turkey). Based on interviews with the biggest oil processors, N2Africa and the
International Institute of Tropical Agriculture (IITA) (Baars pers. comm.) estimate
that the local production of soybeans in Ghana in 2018 was (at least) 170,000
metric tons, with ample potential for growth.2 The report estimates that in the same
period, a similar quantity of soy-derived products (grain equivalents) was imported,
suggesting ample room for growth.3
2.2 Critical assumption 2: N2Africa-recommended technologies are
used by smallholders
N2Africa aims to increase the availability of three types of technologies to farmers:
seeds, inoculants and fertiliser. All three technologies need to be bought from the
market, though seeds can last several years before they need to be replaced.
Overall, farmers do not use many external inputs in soybean production, except
2 The data do not permit distinguishing between smallholder or large farmer production. Nevertheless,we
can make some rough calculations of the upper bound of the number of smallholder farmers involved in
soybean production. Considering a total production of 170,000 tons, the average smallholder plot size
as 0.5 ha, with a yield 850 kg/ha,–1 this would give an upper limit of 400,000 smallholders in northern
Ghana that may be involved in soybean production. The number of farmers that received direct training
on the N2Africa ‘technology package’, estimated at around 2,000 in 2017, is relatively small compared
with the number of farmers involved in this soybean expansion.
3 An unintended negative outcome of the mechanised expansion of soybean production may be the
potential soil degradation of the Savannah area. Most interviewees do not mention this as being a
problem. The relatively low density of trees in the Savannah region might preclude seeing deforestation
as a serious effect. This has also been the case in the Chaco region in Latin America where forested
areas were converted for large-scale soybean production in Argentina, Paraguay and Bolivia
(Hecht 2005).
IDS PRACTICE PAPER 10
14
for land preparation. Amanor (2019) shows that tractor services and herbicides are
the norm in most smallholder farming systems in northern Ghana but fertilisers and
certied seeds are rarely used in crops other than maize and rice. The very low
use of external inputs in legumes (groundnuts, soybean) is partly because these
crops are less responsive to fertiliser than maize is. Farmers have the perception
that legumes x their own nitrogen and therefore do not need mineral fertiliser. This
is also partly a result of the fact that they are planted later than other crops on the
farm, at a moment when cash is (even) more constrained.
The agricultural cycle in northern Ghana is based on one crop per year, often with
a crop rotation of yam, groundnuts, maize and soybean,4 using multiple plots.
Planting is sequenced; rst, farmers plant the yam, groundnuts and maize, and
later in the season they plant soybeans. The decision to buy legume-specic inputs
is, therefore, made when (large) parts of the cash or credit available to a farmer
have already been used for inputs for other crops, especially maize. This drives
non- and partial adoption of the N2Africa proposed technologies. This higher
preference of input use for maize and the application of incomplete input packages
for soybean production is supported by Hoppenbrouwers (2018), who analysed the
risk preference of farmers and the use of mineral fertilisers. He concludes:
I nd that usage of high-cost inputs is low... Furthermore, I nd that usage
of complete technology packages is often unprotable under specic
circumstances – although this is highly context-dependent. When farmers
adapt technology packages, leaving out mineral fertiliser, inoculant or
both, the protability estimates are nancially more benecial in some
cases. Adding inoculant always increases prot or decreases loss.
(Hoppenbrouwers 2018: iii)
The improved soybean seed varieties are widely used in Ghana, though farmers
do not buy new seeds every year. Our own interviews with the farmers’ group
Taaganoba Tibigangso Farmers Union in Yendi indicated that they preferred
renewing their soybean seeds every four or ve years. They explained this as a
rational decision to lower the cash expenses compared to the (perceived) yield
effects of using new, certied seeds compared with self-saved seeds. The Gender
and the Legume Alliance (Musebe, Njuge and Silvestri 2018: 9) documents the
widespread adoption of improved soybean varieties:
Notably, farmers are not necessarily growing their preferred seeds – they
may have very limited choice. Due to the shortage of seed in the formal
system, they buy grain in the markets and store seed year on year... Most
farmers were using improved varieties promoted during the CABI-led
2017 and 2018 campaigns, describing them as less-shattering and high
yielding. Jenguma, a cream variety which is tolerant to Striga and bacterial
pustule was the most popular among all categories of farmers. All the male
youth reported growing Afayak, a yellow-coloured improved variety which
is also tolerant to cercospora leaf spots in addition to having similar traits
to Jenguma.
4 Personal conversation with Imogen Bellwood-Howard (IDS).
IDS PRACTICE PAPER 10
15
The adoption rate of the inoculant technology at the N2Africa operational sites was
(self-)reported as 5,600 farmers. Even when farmers show a preference for cash
investments in inoculants above legume fertiliser or seeds, access to inoculants
remains a bottleneck. For example, the study by the Centre for Agriculture and
Bioscience International CABI (Musebe et al. 2018: 14) indicated that in their
research areas:
There were no farmers using inoculant in either year [2017 and 2018],
citing availability issues. Majority of farmers had learned about the
application of inoculants and its benets, mainly from radio programs,
village-based lm screenings, extension agents and demo plots through
the ADVANCE project. Nevertheless, access and availability had greatly
hindered trialing and adoption of this technology.
Overall, we see that the use of the N2Africa-promoted technologies by smallholder
farmers is still incipient. The diffusion of the technological innovations in soybeans
by farmer-to-farmer interactions is constrained by the lack of cash (fertiliser) and
distribution channels (inoculants), while only the use of improved varieties is at a
large scale, with N2Africa contributing to the offer of certied seeds to farmers to
renew their seeds periodically. Without the government subsidy system (the Planting
for Food and Jobs (PFJ) programme), only the impact pathway related to seeds
is really at scale. The adoption of inoculants seems largely restricted to outgrower
groups or farmer associations that can link demand and supply of inoculant outside
the normal agrovet system (Avea et al. 2016). Fertiliser seems to be used rarely,
though likely more often than inoculants due to availability issues, but mainly by
larger farmers and smallholders that can get it through the input subsidy programme
or outgrower schemes. Outgrower schemes around certied seed production will
use many more technologies than groups that produce soybean as fodder or food
crop, because the requirements and prices are more conducive.
2.3 Critical assumption 3: N2Africa-recommended technologies are
available to smallholders
N2Africa claims that by 2017, it had helped establish ve input distribution centres
working with more than 14 local partners through public–private partnerships
(PPPs) to increase farmers’ access to legume inputs including rhizobium
inoculants, phosphorus fertilisers and certied seeds. Considering reach and scale,
the most prominent partnership is the one with Yara Ghana Ltd.
Yara is a large fertiliser company headquartered in Norway. N2Africa collaborated
with Yara through eld trials to evaluate the benets of triple superphosphate
(TSP) in grain legume production. As a result of successful eld trials, Yara Ghana
produced and distributed a legume-specic fertiliser (Yara Legume) in retail shops
in northern Ghana for purchase by farmers, which resulted in a sale of about
200 tons in 2016. In 2017, according to N2Africa, Yara sold about 2,500 metric
tons of the New Yara blend, packaged in 50kg bags to about 15,000 farmers.
Almost all of this fertiliser (‘90 per cent’, according to Mahamah Abdul-Rahaman
of Yara Ghana Ltd) was sold under the Ghanaian government’s agship PFJ
programme in the 2017 cropping season. Not all fertiliser was effectively
distributed and used in 2017, especially due to delays and problems with quality
IDS PRACTICE PAPER 10
16
in the (imported) soybean seed provided as part of the package, so part of it was
transferred to the next agricultural cycle, reected in lower volumes sold by Yara
in 2018. In that year, Yara only supplied NPK (nitrogen/phosphorus/potassium)
and the supply of legume fertiliser was given to another company called Chemico,
which has been supplying another P-fertiliser (TSP).
Yara can make the Yara Legume fertiliser blend on demand but only when the
demand is large enough. In 2018, the company produced almost 400 metric tonnes
of Yara Legume, far less than the volume sold in the 2017 season (the company
was not selected to provide fertilisers for the PFJ programme in 2018). This demand
is still too low to make it part of the normal supply to the agrovet shops. According
to the interview with Mahamah Abdul-Rahaman of Yara Ghana Ltd:
It entails a lot of logistical costs if we are distributing around the shops.
When farmers do not buy, it will be a big cost for us [Yara] because you
leave it in commission in the shops. And if it’s not sold, it comes back to
us. So we need to get assurance that the products will be bought before...
When somebody orders it, we will produce it... Now the minimum order
[to make the blend] is 45 metric tons.
Yara tries to make it easier for traders to access Yara Legume fertiliser. Yara sees it
still as a pilot, in which it invests in order to generate sufcient demand in the future:
Last year I brought in a consignment of 150 tons [to Tamale] to keep it
closer... So I will do again this year, and those who want can pick up, at
least 4–5 metric tons.
We can conclude that the N2Africa activities have contributed to a process that
has made Yara Legume available to smallholders, but only on a pilot scale – not
yet part of the regular input distribution system. Smallholders therefore still have
only limited access to Yara Legume, and that access has been almost exclusively
provided through the PFJ programme.
We noted earlier that the adoption of the improved seed varieties is already at
scale. At the start of N2Africa, the improved soybean varieties were already
present in the research institutes. N2Africa, however, played a major role in
transforming the multiplication of these varieties in a way that it could support this
scaling process. The presence of N2Africa resolved several formal requirements
for the company Heritage Seeds that allowed it to increase the volume of certied
seed it could produce, especially the regulation that required the presence in the
seed producing farm of an agronomic expert in seed production. Due to N2Africa’s
partnership with Heritage, the government allowed Heritage’s expansion of soy
seed production through outgrower contracts with smallholders. Smallholders
could be used to provide the land and labour required for seed production at
a scale that Heritage could not manage alone. Currently, several organised
smallholder groups multiply seeds, as outgrowers, that Heritage can package and
sell with the formal seed certicate. The distribution system of certied soybean
seeds is in place through the normal agrovet system, and many smallholders buy it
to renew their stock.
The third element of N2Africa technology, inoculants, is not part of the input
package provided by PFJ. Several interviewees indicated that smallholders would
IDS PRACTICE PAPER 10
17
like to use inoculants when available. The N2Africa project formed a PPP with a
private company, Green-ef, for the registration, importation and sale of inoculants
in Ghana. In partnership with N2Africa, Green-ef trained more than 50 agro-inputs
dealers within the regions that the project was working in. The technical training
included understanding what the inoculant does and how to store it properly.
Cooling is recommended to extend shelf life, though this creates a hurdle to
effective distribution in rural areas. Agro-inputs shops located outside the cities
need to take a risk when stocking inoculants that might not be sold directly (the
same year). Therefore, it seems that in remote areas, inoculants will be supplied
only to farmer groups (e.g. outgrower schemes) that can organise demand with
reliable estimates of the volume of inoculants needed and payment modalities that
reduce the risk of default.
Access to inoculants is still largely restricted to groups that are organised in
associations or cooperatives, or directly supported by development NGOs. Many
of these groups are soybean seed producers, with a good internal organisation
for distribution of inputs and cash advances. Adoption at scale, outside these
groups, seems almost absent. The development of inoculants with a longer shelf
life may be one of the necessary conditions for adoption at scale, next to effective
buy-back arrangements between the inoculant producer and the agro-input shops
for unsold packages.
2.4 Critical assumption 4: N2Africa-recommended technologies can
improve yields and incomes
N2Africa has done much research on the effectiveness of the recommended
technologies. There is general agreement that all three components (seeds,
fertilisers and inoculants) are nancially sound investments under average agronomic
conditions. The ‘average conditions’ are, however, not always present, and mask the
reality of unreliable weather and differing soil qualities on smallholder plots. Thus,
even when on average a technology is protable, there are always nancial risks
which explain (partial) non-adoption of one or more of the technologies on offer.
The main issue addressed by N2Africa with the work on certied seeds was the
availability of improved varieties to smallholder farmers. These varieties are less
vulnerable to yield losses caused by poor germination or pod shattering. Pod
shattering makes the timing of the harvest very critical, while this was difcult to
be reconciled with other activities on the farm. The improved varieties give farmers
more exibility in their use of labour during the soybean harvest, considering that
they have to care for multiple crops at the same time. There is convincing evidence
in the interviews with key informants that N2Africa helped to establish community
seed multiplication systems in a way that can produce sufcient improved seeds
for the expanding area of soybean production.
As described earlier, farmers indicated that they get yield benets of improved
varieties for several cropping cycles, without the need to buy certied seeds every
year. Farmers will still benet from the improved varieties using early generations
of self-saved seed (or obtaining these from neighbouring farmers). More research
on the yield effects of certied seeds compared with rst- and second-generation
farmer-saved seeds could help to understand this practice, and verify whether
IDS PRACTICE PAPER 10
18
the recommendation to buy certied seed every season is valid. When this
research shows that early generation seeds maintain most of their characteristics,
strengthening the capabilities of farmers to select and save seeds, and promoting
(early generation) seed exchanges in rural areas (Richards 2007) could be
promising pathways to create (and attribute) more impact at scale from the seed
multiplication efforts.
The training material on inoculant use states that ‘There is a yield increase that
more than pays for the input costs of inoculants itself’. Most farmers and technical
staff interviewed corroborated this statement. However, the data on this are
somewhat contentious, with authors using widely different values for the price
of inoculants and the average yield effect.5 Moreover, Ulzen et al. (2018: 31)
writes that:
The agronomic approach adopted for determining responsive and
non-responsive sites indicated that a large majority of the elds were
non-responsive to P and / or I. Only 17–40% of the study elds in the
Northern region were responsive while 6–17% were responsive in the
Upper West region.
This suggests that the application of inoculants is not always cost-effective. Also,
Samuel Adjei-Nsiah of N2Africa indicated that there is no research yet to determine
whether the use of inoculants in preceding seasons could have legacy effects and
make it less opportune to apply it year after year on the same plot. There is little
known about the legacy effects of rhizobium that has established itself in the soil.6
Van Heerwaarden (pers. comm. 2019), based on data from 2011 from northern
Ghana, shows an average effect of inoculant use of 197kg ha−1, with a condence
interval (95 per cent) of 106–287kg ha−1. This shows that overall, inoculants are a
(cost-)effective technology, though it would be benecial if N2Africa were to collect
more and better data on the yield effects and cost-effectiveness of the partial
adoption of the promoted technologies (Ronner et al. 2016). When communicating
with farmers/trainers, N2Africa could use an interval instead of a point estimate
for the average yield and income effects that can be expected from applying the
technology. The (enhanced) N-xation by soybeans is available for subsequent
crops. Therefore, the benecial effects of soybean seeds and inoculants will be a
higher yield of the subsequent crop. There is, however, lack of research evidence
on these subsequent yield effects.
5 According to Wellspring (2019), at current soybean prices (US$0.42–0.58 per kg-1) the net gain for
farmers in Ghana with 0.5 ha soybean with inoculants is estimated at 300kg/ha−1 and provides a net
benet of US$30–45. This calculation seems at odds with reality, mainly because the yield effect used
is much higher than reported in other studies (Adjei-Nsiah et al. 2019; 2018; van Heerwaarden et al.
2018; Lamptey et al. 2014), which are between 150 and 200kg/ha−1. Simple arithmetic with those
lower yield estimates suggests an evaporation of the net benet: 100kg less yield represents a value
of US$42–58. Adjei-Nsiah et al. (2018) argue that inoculants are a good investment even with lower
prices and lower net-yield effect, but use a cost of rhizobium inoculant of US$12.50 ha−1. The cost
of inoculants needed for proper inoculation in this study is, however, lower than suggested by the
interviews (estimated to be close to US$20 ha–1.
6 Interestingly, the presence of local rhizobium strands in soils is used in some studies to explain
disappointing test results of the inoculant trial (Ulzen et al. 2016).
IDS PRACTICE PAPER 10
19
Instead, there is ample evidence about the positive yield effects of mineral fertiliser
on soybean yields. N2Africa’s fertiliser trials with varying phosphorus sources
and dosage convinced Yara to increase the amount of phosphorus in their NPK
blend, developing the special fertiliser Yara Legume. Yara refrained from offering
TSP fertiliser to farmers, though initially this was recommended by N2Africa,
but instead included other micro-nutrients largely due to the experiments and
recommendations from its headquarters in Norway. The Yara marketing manager
in charge of northern Ghana was explicit in mentioning the positive inuence of
N2Africa in this process of developing Yara Legume, though he also indicated that
these trials also tted quite neatly in the research and development (R&D) trials
that Yara routinely plans to develop and test new products for specic crops.
The interviews with N2Africa staff showed that there are different perspectives
regarding the optimal timing of application of phosphorus fertiliser. Perhaps
agronomic reasons prescribe application before the planting, but there are strong
risk management reasons to wait until the plants have established themselves.
The risk of losing the investment made in inputs due to failing crops is quite high in
agricultural conditions with droughts, late rains or short growing periods. Moreover,
farmers’ rst priority (not just for time use but also in the importance attached to the
crop) is the application of fertiliser to maize. This situation creates a context where
it is unlikely that many farmers will adopt fertiliser application to soybeans at scale.
Even where Yara Legume is commercially available in rural agro-input shops,
at the moment of deciding whether to buy, most of farmers’ scarce cash would
already have been spent on the inputs for earlier and more important crops.
It is widely recognised that N2Africa played a key role in convincing regional and
national policymakers to include the new fertiliser blend, Yara Legume, as one of
the options for farmers in the PFJ inputs subsidy programme. The explicit mention
of soybeans in the eligible crops for the programme created a large market for
the certied soybean seeds and opened a distribution venue for the Yara Legume
fertiliser. Nevertheless, within their mixed farming activities, most of the smallholders
register in the input subsidy programme to obtain inputs for maize, not soybean.7
2.5 Conclusion
There is convincing evidence of a fair contribution of N2Africa to a process of
technological upgrading of soybean production that, as yet, still only weakly
inuences the increase in soybean production in northern Ghana. Most of the
growth in soybean production seems to result from crop area expansion by
smallholder farmers, with relatively low yields and limited use of external inputs,
especially fertiliser and inoculants. Smallholders cultivate soybeans as part of a
mixed cropping system in which they prioritise cash use for external inputs for
maize, not the N2Africa-promoted technologies for soybeans.
Reviewing the hoop tests in relation to the overall contribution claim, we draw the
following conclusions.
7 Using NPK fertilisers on maize may have a P residual effect for soybean in rotation. Less than through
direct application, but also less risky – costly and with a higher chance of being protable.
IDS PRACTICE PAPER 10
20
1. Did the soybean expansion take place?
Yes. Though data are unreliable, most interviewees believe that there
was an expansion of soybean production in northern Ghana. This is
considered as a positive development.
2. Was N2Africa support used in the process?
Yes. The seed multiplication in particular is likely to have had a positive
effect on soybean yields of a large number of smallholder farmers. These
farmers use certied and early-generation seed of the improved varieties.
3. Was N2Africa necessary to speed up or scale faster?
Yes. The seed multiplication work has had a positive systemic effect for
the availability of quality seed for smallholder farmers. The impact of the
work on inoculants and fertiliser is, however, still fairly limited. Uptake
seems largely restricted to groups of farmers that are self-organised or
benet from NGO support, where the demand and supply of inoculants
and fertiliser is being registered and communicated to the agro-input
shops, and where uptake is facilitated by pre-harvest loans.
N2Africa played a facilitating role in opening up the PFJ programme to
legumes. Legume fertilisers are increasingly available but still in a ‘pilot
distribution mode’ by Yara, not part of its regular distribution system. The
quantities that are demanded are too small to make it part of the regular
stock in the agro-input shops. The PFJ has been the most important
driver of demand for the legume fertiliser. N2Africa has contributed to
the availability of other P-fertiliser in the PFJ. However, the PFJ creates
a somewhat articial market that may collapse when the subsidies are
discontinued.
Due to the relatively short shelf life of inoculants, these are only available
through networks and partnerships and not yet a normal part of the basket
of options available to farmers in remote, rural areas.
The positive effects of the three promoted technologies (certied seeds,
inoculants, P-fertiliser) are well-demonstrated, though data on costs
and nancial risks are sometimes inconsistent. More evidence on the
cost-effectiveness of partial adoption strategies could help farmers to
choose their options from the basket of available technologies (Ronner
2018; Ronner et al. 2016).
4. Was N2Africa a necessary causal factor for the expansion to take place?
No. N2Africa cannot claim to be the trigger that started this expansion.
The main trigger is the demand generated by the processing industries
that started to be located in the area. At most, N2Africa played a
facilitating role in a wide number of partnerships, involving many projects
and initiatives, that accompany the efforts of smallholders to benet from
the increasing demand for soybeans in Ghana. Without N2Africa, this
growth of soybean acreage would have started anyhow, and will continue,
but likely with an even lower adoption of fertiliser and inoculants.
IDS PRACTICE PAPER 10
21
3 Case 2 – Verifying the contribution
claims of N2Africa related to the
adoption of legume inoculants by
smallholders in Ethiopia
This second case study examines and discusses the quality and quantity of
evidence supporting the claim that N2Africa contributed substantially to the
expansion of supply of, demand for and use of legume inoculants by smallholder
producers in Ethiopia. The contribution claims are documented in N2Africa’s annual
reports and the programme’s impact evaluation report (Ampadu-Boakye, Ronner
and Kanampiu 2018a; Ampadu-Boakye et al. 2018b; Wolde-Meskel 2019a; Dontsop
and Ampadu-Boakye 2019). The work in Ethiopia is considered an example of
positive impact by N2Africa, because it shows a contribution to the development of
a new input-supply sector for legume inoculant technologies in the country.
Ethiopia joined N2Africa during the project’s second phase (2014–19). The
overarching project has a clear objective to increase the availability and uptake of
nitrogen xation inputs and methods by smallholder farmers. In particular, N2Africa
Box 3.1 N2Africa in Ethiopia
Ethiopia joined the N2Africa project in phase 2 (2014–19) as a ‘core country’.
N2Africa in Ethiopia focused on improving production technologies and value
chain linkages for four legumes: common bean, soybean, faba bean and
chickpea. These legumes are cultivated principally by small-scale producers
in Ethiopia. The production technologies promoted by N2Africa in the
country included improved and locally suitable varieties, inorganic fertilisers
(particularly phosphorus), and rhizobium inoculants that stimulate nitrogen
xation. The project also encouraged farmers to adopt recommended ‘best
management practices’ such as intercropping, crop rotations and regular
spacing of plants. Key activities included scientic eld trials in several
locations to identify high-yielding legume varieties and effective inoculant
strains, local adaptation trials, and demonstration trials to introduce the
technology to farmers.
The project was designed to be sensitive to gender issues in legume
production and to promote women’s empowerment. N2Africa in Ethiopia
was coordinated through the International Livestock Research Institute
(ILRI) from Addis Ababa, through a series of partnership agreements with
research institutes, extension and training organisations, farmers’ cooperative
unions, private sector input manufacturers and suppliers, and grain buyers.
Organisationally, the project operated through seven regional public–private
partnership (PPP) clusters.
Sources: Various N2Africa project documents, annual reports and website.
IDS PRACTICE PAPER 10
22
in Ethiopia has focused on expanding the supply of rhizobium inoculants (mainly
through domestic production and providing a regulatory framework for eventual
imports) and increasing demand for the inoculants among smallholders as a way to
increase the yields and productivity of legumes under Ethiopian farming conditions
(see Box 3.1).
The intervention logic that provides the framework for the N2Africa contribution
story in Ethiopia is shown in Figure 3.1. The gure makes it possible to trace the
causal logic of N2Africa’s contribution through four interlinked impact pathways,
each comprising a number of activities, which may be summarised in simplied
form as follows (Table 3.1).8
Table 3.1 The N2Africa Ethiopia programme
Pathway 1 Developing and promoting packages of improved chickpea inoculant
technologies for different regions of Ethiopia
Activity 1.1 – Identifying best-t combinations of legume variety ×
inoculant strain × site [research activity]
Activity 1.2 – Adaptation trials and demonstrations [promotion,
extension activity]
Pathway 2 Stimulating private sector involvement in inoculant supply (production
and potentially import; distribution and marketing)
Activity 2.1 – Capacity building of Menagesha Biotech Industry
(MBI) (inoculant producer) [increasing quantity, improving quality of
input production and supply, strengthening distribution systems]
Activity 2.2 – Working with local agrodealers [improving distribution
and marketing]
Pathway 3 Building technical capacity of key actors in the legume sector
Activity 3.1 – Creation of PPP platforms for each regional cluster
[networking, platform for cooperation and communication]
Activity 3.2 – Capacity building of National Agricultural Research
System (NARS), universities, extension services, agrodealers
[research, training activity]
Pathway 4 Inuencing development of supportive policy frameworks
Activity 4.1 – Engaging with government ofcials and policymakers
[advocacy and communications activity]
Activity 4.2 – Developing draft protocols and standards [technical
advisory activity]
Source: Authors’ own.
N2Africa documentation and key informants emphasised a number of key claims
about the project’s impacts. In particular, they stated that the project helped to
increase the volume and improve the quality of legume inoculant production in
the country, and boosted the volumes of inoculants distributed and sold to small-
scale farmers. By identifying and promoting well-performing combinations of
legume varieties and inoculants for specic production areas, and demonstrating
8 Note: These four impact pathways, identied for the purpose of this Contribution Analysis, correspond
loosely to, but cut across, the four pillars of the N2Africa project, namely: capacity building, input
supply, dissemination, and market access.
IDS PRACTICE PAPER 10
23
the improved yield performance of these technology packages to farmers,
N2Africa claims to have raised awareness among both small-scale farmers and
other actors in the legume sector of the effectiveness and economic viability of
these technologies, thus stimulating demand. We reviewed the evidence behind
the contribution claim: ‘N2Africa has contributed to the increase in production,
distribution, uptake and expansion of market demand for legume inoculants
in Ethiopia’.
Figure 3.1 The Theory of Change of N2Africa’s contribution to the
expansion of inoculant technologies in Ethiopia
Source: Authors’ own.
To investigate N2Africa’s contribution, we veried whether the claimed changes
took place and what contribution N2Africa had made to each of these. We used
an approach informed by process tracing (Punton and Welle 2015; Beach and
Pedersen 2013). Process tracing implies a structured way of deriving causal
inference by critically reviewing the process and sequence of change. We veried
the following three critical assumptions in the Theory of Change:
Critical assumption 1: N2Africa support increased the demand for legume
inoculants by smallholders.
Critical assumption 2: N2Africa support improved quality and availability of
legume inoculants in Ethiopia.
Critical assumption 3: N2Africa enhanced the potential of inoculant
technologies as a viable commercial opportunity for small-scale producers
and other stakeholders.
N2Africa has produced, collated and published abundant evidence in documentary
and other forms which attests to its activities, including handbooks and
guidelines on experimental protocols and best practices, baseline studies, policy
assessments and situation reviews, planning documents and progress reviews,
and so on. This report seeks to understand whether this activity contributed to the
achievement of the positive outcomes claimed.
IDS PRACTICE PAPER 10
24
3.1 Critical assumption 1: N2Africa support increased the demand for
legume inoculants by smallholders
N2Africa claims that demand for and use of rhizobium inoculants by Ethiopian
legume growers has increased strongly as a result of the project’s efforts. The
project partners undertook a range of activities to investigate the right technologies
to recommend, to improve the supply of the recommended inputs, and to
encourage farmers to take them up, along with new cultivation techniques. A
survey of farmers conducted by N2Africa at the end of the project found high levels
of awareness, usage and intention to continue using the project’s key technologies,
singly or in combinations of two, three or four elements. However, high levels
of willingness to buy were matched by concern that the inputs might not be
accessible, particularly once the project had ended (Wolde-Meskel 2019b; Dontsop
and Ampadu-Boakye 2019).
Beginning at the research end, the rst phase of the N2Africa project had already
demonstrated that interactions among variety × inoculant × location were critical.
Research was done by N2Africa to identify effective combinations of legume
varieties, rhizobium strains and management practices for different cultivation
environments. The resulting combinations were called ‘best t options’. Project
researchers also assessed the availability and accessibility of legume varieties
and related technologies, and analysed the sale and utilisation of legumes (Farrow
et al. 2019).
Various informants described how the N2Africa project provided additional
resources and focus to allow an increase in the intensity of promotional activities,
to encourage legume farmers to take up new varieties, inoculants, fertilisers, and
cultivation methods such as intercropping. Previously, research institutes and
extension organisations had lacked the resources to promote these technologies
effectively (interview 8 and others, see Annexe). Various interviewees, representing
different organisations involved in N2Africa, afrmed that the introduction of
inoculants constituted the most important new contribution made by N2Africa
(interview 1 and others).
For example, a representative of Balegreen Spice and Grain Development
(Balegreen), an agri-business company and commercial farmer, stated that neither
his company nor the farmers it works with had really heard of inoculants before the
N2Africa project introduced them. Retrospectively, he acknowledged that inoculant
had been available in principle from the government’s National Soil Testing Centre
(NSTC), but neither he nor the farmers understood the benecial potential of the
technology (interview 4). An interviewee from an agricultural input dealership that
was involved in N2Africa’s Southern PPP cluster, Mirko Agrodealer, also noted that
he had encountered inoculants for the rst time when he met with a representative
of MBI; prior to that encounter, he had known nothing about them. MBI
demonstrated the product and he agreed to take on the dealership (interview 1).
An informant from the Bureau of Agriculture conrmed that no farmers in his area
were using inoculant until the N2Africa project came along (interview 13).
All three of these informants concurred that demand for and uptake of inoculants
grew rapidly from nothing. According to the representative of Balegreen, both he
and his outgrowers used inoculants initially to indulge the N2Africa staff, without
IDS PRACTICE PAPER 10
25
a condent expectation of a good result. However, the positive effect on yield
was very remarkable. As a result, he said, the smallholder farmers who produce
chickpea for his company are not only using inoculant on his instructions but
are now beginning to use inoculants on their own initiative (interview 4). Mirko
Agrodealer began distributing inoculant for MBI in the Southern PPP cluster in
2015, with just 50 packets, but by 2019 the company was handling 5,000 packets
(of which around 600 remained in stock at the time of our interview in May 2019).
The Mirko representative explained that inoculants were particularly in demand
among smallholders in highland areas, where there was sufcient moisture. In
lowland areas, where there could be moisture stress, farmers did not always see
any advantage from using inoculants; however, he opined that the farmers did
not mind paying for the inoculants as the additional cost was small in proportion
to the cost of seed. As a standalone product, he reckoned that inoculant could
only be protable if dealing in bulk, but with the small volumes he was handling
at the moment, he regarded inoculant supply as more of a ‘social service’ than
a commercially viable business. He saw inoculants as an item with commercial
potential, and was interested in bundling the inoculant with seed as a means of
product promotion (interview 1).
A farmer and chairman of a producer cooperative explained that the N2Africa
approach had helped to raise farmers’ awareness about inoculants, as well as
seed varieties and seed production and cultivation methods. He claimed that all
the farmers in the Northern PPP cluster area of N2Africa were now using inoculant
and had also adopted the recommended kabuli type of chickpea alongside the desi
types they used to grow in the past. Although the investment and effort involved
in using the new technologies was greater, the extra effort was rewarded at the
end of the season (interview 12). A representative of the Bureau of Agriculture in
Gondar afrmed that farmers in his area had also adopted both inoculants and the
new kabuli chickpea variety, Arerti (interview 13).
An informant from one of the universities involved in N2Africa stated that smallholder
farmers had believed that pulses could be grown without external inputs, but
demonstrations of the impacts of inoculants and legume fertilisers in combination
with improved varieties convinced them that these inputs were valuable. In fact, the
same informant argued that even researchers from the public agricultural research
institutes had typically grown pulses without external inputs, and that their behaviour
had also changed as a result of N2Africa’s activities (interview 6).
An informant from Catholic Relief Services (CRS) stated that, through his
organisation, 20,000 farmers had used inoculant technology, rising from only
200 in 2015 when CRS rst formed its relationship with N2Africa. The woredas
(districts) covered by CRS’s Farmer-to-Farmer project overlapped only partly with
one of N2Africa’s PPP cluster areas, so that much of the impact from CRS’s reach
was additional to N2Africa’s reach (interview 3).
In spite of these positive assessments of the impact of N2Africa on raising
awareness and increasing demand for inoculants and other technologies,
MBI’s own data (collated by N2Africa) suggest that actual uptake of inoculants
is far below 100 per cent of farmers in the project’s cluster areas. Meanwhile,
multiple informants expressed concern about the ongoing sustainability of these
developments now that the project has come to an end.
IDS PRACTICE PAPER 10
26
Various interviewees agreed that the inoculant business has huge potential to grow
in Ethiopia and to have a positive impact, but at present it is fragile, ‘like an infant’
(e.g. interview 8). A representative of Tsehay Farmers’ Cooperative Union – part of
the Northern PPP cluster – claimed that their union alone had 120,000 members
seeking inoculants, yet N2Africa’s data on the distribution and sales of inoculants
across the entire project imply that usage in the whole of the Northern cluster is
far short of that (see below). This informant explained that the problem had to do
with logistical issues and matching supply to demand in a timely fashion. He said
that farmers typically left it until planting time to place their inoculant orders with the
union, but the company (MBI) could not supply the product at such short notice.
Neither the union, nor the local agrodealers, nor MBI was willing to stockpile the
product close to the market in anticipation of demand. The union’s traditional priority
in terms of bulk purchasing on behalf of its members was to procure chemical
inputs; for MBI, the problem was that the company lacked the nancial and other
resources to produce inoculant speculatively and stockpile it in large quantities, not
forgetting the issues of storage and shelf life (interviews 2, 10 and others).
Various informants expressed concerns over whether the project’s achievements
would be sustainable now that the project had come to an end. One agrodealer
pointed out that, within the framework of the N2Africa project, MBI had offered
him a 50 per cent discount on the wholesale price of inputs, and allowed his
company to take 25 per cent of their stock on a sale-or-return basis, reducing his
risk (interview 1). Another interviewee felt that N2Africa had achieved a big impact
in just a few years, and that the momentum would continue, but necessarily at a
slower pace unless the government or another inuential actor steps in to support
the production and supply of inoculants and other inputs (interview 3). There must
be some doubt about whether the progress achieved to date can be sustained.
N2Africa claims that its activities built demand for and use of inoculants and other
legume production technologies, not only through its work in support of inoculant
producers but also through its creation of the PPP regional cluster model of
organisation and related efforts to build the capacity of various players involved in
the legume sector (Wolde-Meskel 2019b). Multiple interviewees afrmed that the
PPP organisational model was a novelty that made a big difference. Informants
said that the PPPs were a platform for communication and cooperation that had
not existed before. In particular, the PPP clusters were reported to have connected
for the rst time various organisations that had not collaborated or cooperated
previously, including private sector enterprises, public agricultural research institutes,
universities, farmers’ cooperative unions and the Bureau of Agriculture. This also
meant working across the entire value chain – from research and input production
via agrodealers and farmers’ cooperatives to legume growers and export buyers.
Our informant from GUTS Agro Industry mentioned that the PPP cooperative
platform helped his company to mitigate the risks of operating in an environment
where regulation and enforcement of contracts could be unreliable. The PPP
approach strengthened relations among stakeholders along the value chain in
order to build trust, mutual commitment and condence (interview 5).
Various informants expressed the desire that this form of communication and
mutual engagement should continue beyond the end of the project, but with
varying degrees of condence that this would happen. Some informants expressed
IDS PRACTICE PAPER 10
27
concern that, without funding, the cooperative PPP approach would not survive;
others felt that the rapid turnover of staff in government organisations was
disruptive to the continuity of relationships and working methods (interview 3
and others). However, one informant also noted that a separate project, Feed
the Future (funded by the United States Agency for International Development
(USAID)), had taken up the effort to promote legume inoculants in his region and
was mobilising the existing N2Africa PPP platform for this purpose (interview 4).
A representative of the Bureau of Agriculture explained that, through N2Africa, he
had learned about new approaches to agricultural extension, especially the new
method of working with individual farmers on demonstrations and scaling up from
there, rather than the conventional approaches to ‘mass mobilisation’ that relied
principally on spreading information. He noted that the novel approaches used by
N2Africa saved the time and increased the impact of extension workers, enabling
them to support whole woredas much more effectively than they had been able to
previously (interview 13).
Another novel approach was the N2Africa project’s method of organising eld
trials and demonstrations on farmers’ own land. For this purpose they introduced
‘small packs’, comprising sufcient seed and inoculant for a small 20m × 20m trial
plot, packaged along with information and guidance for farmers on how to set up
a trial planting alongside a control plot, so that any differences in performance
could be compared. One informant regarded this as a cost-effective way to reach
farmers more effectively than using traditional mass-exposure approaches, which
would typically involve setting up larger demonstrations that would be managed by
professional technicians (interview 11).
3.2 Critical assumption 2: N2Africa increased the quality and
availability of legume inoculants in Ethiopia
N2Africa reports that the production, distribution and sale of legume inoculants
in Ethiopia increased six-fold, seven-fold and 13-fold respectively between
2013 and 2016 (Wolde-Meskel et al. 2018). Production rose from 29,000 125g
packets of inoculant in 2013 (the year before N2Africa in Ethiopia commenced)
to 207,445 packets in 2018 (the project’s nal year) (Wolde-Meskel 2019a;
Ampadu-Boakye et al. 2018a; 2018b; 2017). Each 125g packet is designed to
inoculate seed for 0.25 ha of land, suggesting that production in 2018 would
have been sufcient for 52,000 ha, if fully taken up by legume farmers. However,
distribution and sales lagged considerably behind production in each year. In
2013, the inoculant manufacturer says that it managed to distribute 20,000 packets
and sold 10,000, out of the total production of 29,000 packets. In 2018, N2Africa
reported that nearly 145,000 packets were distributed and nearly 138,000 were
actually sold, suggesting that inoculants were used on about 34,500 ha of land
planted with legumes in that year (Wolde-Meskel 2019a).
N2Africa claims that its legume technologies, including inoculants, lead to higher
yields and production compared to national averages. Project documents argue
that the inoculants are affordable and their effect on productivity is large enough
to represent good value for the additional cost (Wolde-Meskel et al. 2018;
Wolde-Meskel 2019b).
IDS PRACTICE PAPER 10
28
Some research on rhizobium inoculants for the production of various legumes was
carried out in Ethiopia before the N2Africa project began there (Samago, Anniye
and Dakora 2018; Degefu, Wolde-Meskel and Rasche 2018; Beshir et al. 2015;
Argaw and Mnalku 2017). Inoculants were produced in Ethiopia by the two public
sector institutes, the NSTC and the Hollota Agricultural Research Centre (HARC),
but on a small scale, principally for research purposes, and with both a limited
capacity and an unclear mandate to scale up production on a commercial footing
(interview 2). A new private enterprise, Menagesha Biotech Industry plc (MBI), was
established in 2012 to produce and market rhizobium inoculant products in Ethiopia,
but as a start-up company it struggled to secure nance and its capacity was very
small. During its rst two operational years (2013 and 2014), MBI managed to
produce nearly 100,000 packets of inoculant without making a prot. Developing a
market for its products was a struggle, as it lacked effective distribution channels.
Multiple interviews and other sources conrm that, at that time, there was no other
production or import of legume inoculants in Ethiopia and that market demand
was almost entirely absent, due to very low levels of awareness among farmers
and agri-input dealers alike. Awareness of inoculant technologies was beginning
to emerge in various quarters, but various informants stated that they had known
very little about rhizobium inoculation for legumes before coming into contact with
N2Africa (e.g. interviews 8, 12 and others). For example, a representative of CRS
described how one of its in-country volunteers, an agronomist, had advised the
organisation in his end-of-assignment report that farmers should use rhizobium
inoculants to improve N-xation in common bean. Prompted by his advice, CRS
made contact with N2Africa and through the project with MBI, and began to
procure and distribute inoculant packets to farmers. The intervention proved so
benecial that CRS incorporated inoculant promotion routinely into its activities,
putting new legume growers in contact with MBI (interview 3).
Against this background, it is evident that the N2Africa project carried out research
into the performance of legume varieties and rhizobium inoculants through a
concentration of effort on the four legumes selected for attention in the country
(Wolde-Meskel et al. 2018). Alongside this research, N2Africa offered essential
technical support to MBI to prepare a business plan and submit a successful
application to the Alliance for a Green Revolution in Africa (AGRA) under its
Scaling Seeds and Technologies Partnership (SSTP). With this support, MBI
secured a grant of just under US$300,000, which the company used to purchase
manufacturing equipment and expand its capacity to produce chickpea inoculant
(Ampadu-Boakye et al. 2017). The AGRA grant also nanced the multiplication and
delivery of improved chickpea varieties and fertilisers to 90,000 growers in the major
chickpea areas of Ethiopia, alongside training of farmers and other stakeholders in
the chickpea value chain. N2Africa served as the chair of the AGRA–MBI project’s
steering committee and provided direct support to MBI’s marketing effort. This
investment can properly be understood as additional funding that was leveraged by
N2Africa, since the AGRA grant was additional to N2Africa’s own budget and was
only available to private sector enterprises (interviews 2 and 7).
While it is evident – from project documents, scientic publications (e.g. Wolde-Meskel
et al. 2018) and testimony from interviews – that N2Africa stimulated and facilitated
research to identify high-yielding combinations of crop varieties and inoculants,
IDS PRACTICE PAPER 10
29
it is also relevant to note that the project’s research effort was not necessarily
decisive in selecting the combinations of cultivars and inoculant strains that would
be promoted for different crops and growing areas. For example, the project
introduced no new varieties or inoculants for soybeans. This is because soybean
is an introduced crop in Ethiopia, for which commercial cultivars have been newly
imported, and where there is no presence of locally specic rhizobium strains
that have evolved a specialised symbiotic relationship with particular varieties
of soybean. Some other legumes have a much longer history of cultivation in
Ethiopia, and the N2Africa-sponsored research did achieve new scientic insights
into the genetic diversity and geographical distribution of rhizobium strains that
have evolved alongside particular varieties. One such crop is chickpea, for which
at least one effective inoculant strain had already been identied in Ethiopia before
N2Africa began its work.
According to Birhan Abdulkadir, one of the coordinators of the project, N2Africa
identied and recommended for commercialisation two new inoculant strains
that were not already present: one for faba bean and one for common bean. An
informant from the Northern PPP cluster noted that his region was one where
experiments had identied a particular inoculant strain suitable for faba bean,
which was supplied to MBI. However, the company was reluctant to produce and
market the new strain, as it considered the potential customer base too small a
market niche to be protably served (interview 11).
There is evidence that N2Africa made an important contribution in supporting the
development of MBI. A company representative explained that the partnership
with N2Africa beneted MBI in several important ways. In addition to the securing
of the AGRA grant, the company extended its networks and raised its prole
among regional research institutes and universities, the Bureau of Agriculture
(the government agricultural extension service) and input distributors. MBI also
made contact through N2Africa with an international NGO, SNV Netherlands
Development Organisation, which it regarded as a very important contact for future
collaboration. Through N2Africa, MBI was also able to accelerate the process of
validating the effectiveness of inoculant strains for different legumes and regions,
and it was also helpful for the company to have the quality of its products veried
independently by the research institutes involved in the N2Africa PPP clusters. The
company had also learned, from participating in the project, how to organise and
facilitate farmer eld days and demonstrations. This informant said that, thanks
to N2Africa, the company’s production, distribution and sale of inoculants had
increased substantially. In short, the edgling company’s journey to becoming a
stable and protable business had been made easier and quicker, reducing the
‘years of struggle’ that they were facing on their own (interview 2).
While N2Africa’s technical assistance to MBI has contributed to the improvement of
quality in the production and distribution of inoculants, interviews revealed that there
remains considerable anxiety over the risk that shortcomings in manufacturing,
distribution and storage of inoculant products may undermine quality standards
and reduce farmers’ condence in inoculant technologies (e.g. interview 6).
Rhizobia are living organisms that require careful handling. Recent research has
found that the performance, and particularly the shelf life of some commercially
available inoculants in East Africa, falls below the standards advertised by their
IDS PRACTICE PAPER 10
30
manufacturers (Balume et al. 2015). Maintaining the quality of inoculant products,
both at the manufacturing stage and throughout the supply chain, is an urgent
concern for MBI. A company representative estimated that it would take only two
consecutive years of poor results for farmers to lose condence in the inoculants
(interview 2). N2Africa has sought to train distributors, extension workers and
farmers to handle the product well. N2Africa representatives have also persistently
urged the Ethiopian government to create and enforce adequate standards for
the industry, not only through its encounters with government ofcials but also by
drafting a manual of policies and procedures for inoculant quality manufacturing,
inspection and monitoring. This draft document has been submitted to the Ethiopian
government, but has yet to be endorsed and implemented.
N2Africa staff also believe that their example and their success in generating
demand among farmers for rhizobium inoculant helped to stimulate the NSTC
to increase its own production and distribution of inoculants, from around
11,000 packets produced in 2010–11 to nearly 90,000 packets in 2015–16
(unpublished NSTC data passed on by N2Africa). However, evidence that this
increase was attributable in part to N2Africa is circumstantial rather than denitive.
3.3 Critical assumption 3: N2Africa strengthened the potential
of inoculant technologies with legumes as a viable commercial
opportunity for small-scale producers and other stakeholders
N2Africa claims that its activities have expanded scientic knowledge on rhizobium
inoculants, demonstrated the productivity of inoculants for legume production,
revealed a large unmet market demand for legume inoculants, and demonstrated
the protability and productivity of inoculant technologies in conjunction with
legume cultivation in Ethiopia. In this area of claim-making, N2Africa is asserting
that its work in the country has helped to establish the beginnings of a substantial
commercial expansion of inoculant production and use, and improvement of
legume production. The potential market demand for appropriate strains of legume
inoculant could be over 6 million packets, based on a cultivated area under
legumes of 1.6 million ha (Wolde-Meskel 2019b).
Alongside MBI, other private sector enterprises involved in N2Africa related that they
were also grateful for the project’s support in building their businesses. For example,
the representative of an agricultural input dealer involved in one of the PPP clusters
said that N2Africa had had a profound effect on his fortunes, raising him from a
‘nobody’ to a substantial businessman. He noted in particular that the Ethiopian
government routinely focuses on farmers’ cooperative unions, whereas N2Africa had
created an opportunity for private sector entrepreneurs, such as himself (interview 1).
This area of claims expands the interest of this report beyond inoculants to
embrace N2Africa’s focus on the improvement of legume production in general,
and its whole value chain approach. Another private sector informant, from GUTS
Agro Industry, appreciated that the N2Africa approach of working throughout
the value chain had made a positive difference to its business. As a result of the
project, the company had: expanded into chickpea; got their outgrowers to use
inoculants that are increasing production; developed new legume-based baby
food products; and entered into a new business – legume-based animal feeds –
IDS PRACTICE PAPER 10
31
both as a way to make use of by-products from their existing production and as a
channel to reprocess and re-direct baby food products that are reaching the end
of their shelf life. This meant that the company was now involved in a number
of complementary activities and product lines, enabling it to construct a more
sustainable business (interview 5). As a company involved in contract farming
of legumes as well as food and feed production, it beneted not only from the
PPP model of collaboration with other stakeholders, but from improvements in
the quality of both legume seed and post-harvest residues that could be used for
livestock feed (Dejene et al. 2018; Belete et al. 2019).
A representative of Balegreen, another agribusiness company, opined that
N2Africa had helped to put chickpea cultivation in the Bale zone of Oromia region
on the road towards becoming a viable and sustainable business. The major new
elements introduced by N2Africa were inoculants, as well as the PPP model of
cooperation among various stakeholders. He noted that his rm rst ventured into
chickpea cultivation in 2012, but the yields were poor at rst and the company
had difculty in nding a market for the crop because the region was not known
for chickpea production. In 2013 the company continued chickpea cultivation
with support from an ACDI/VOCA9 project, which introduced a different chickpea
variety and linked the producers to a buyer. During this period, Balegreen also
experimented with mechanical harvesting and threshing. The yield improved but
was still quite small. Yields improved further after N2Africa eld trials, in which
Balegreen got involved, identifying another new chickpea variety (Habru), which
proved to be more disease resistant and responsive to inoculants. It also has a
larger grain, which is in export demand, and N2Africa linked the producers to an
export buyer. The new inoculants, in combination with the Habru variety, produced
remarkable yields – even better (according to this informant) than wheat. As
a consequence of these steps, the Balegreen representative considered that
N2Africa had helped establish chickpea as an emerging commercial opportunity
for growers in Bale zone, having reached a coverage of 172,000 ha in the
current season (2019) and with the potential to expand even further in Bale and
neighbouring Arsi zone (interview 4).
The same informant argued that nitrogen-xing leguminous crops were vital in his
area in order to improve the sustainability of agriculture as a whole. He pointed out
that continuous cultivation of wheat would be unsustainable unless nitrogen could
be replaced in the soil. He also felt that, thanks to N2Africa’s demonstration of the
viability of chickpea cultivation in the area, the government and other stakeholders
were now displaying a more positive attitude towards chickpea; however, the
government’s chief focus remained on grain crops such as teff and barley
(interview 4).
Notwithstanding these positive impacts, N2Africa cannot claim credit for fully
establishing chickpea technologies and chickpea cultivation on a sustainable
footing. The Balegreen informant praised the project for helping to build
momentum, but the momentum is not yet enough to be self-sustaining. He noted
that N2Africa had provided some inputs free of charge, which is unsustainable,
9 ACDI/VOCA is an international development non-prot organisation based in Washington DC, USA.
See www.acdivoca.org/.
IDS PRACTICE PAPER 10
32
particularly for a commercial enterprise like his. He estimated that the impact of
N2Africa in Bale had been very local and Balegreen would not be able to produce
further scaling by itself. However, the Feed the Future project had taken up the
reins, retaining the PPP cluster model established by N2Africa and continuing the
focus on chickpeas and inoculant technologies (interview 4).
3.4 Conclusion
There is clear evidence that N2Africa made an important, catalytic contribution
to the development of a viable commercial production system and a functional
value chain for legume inoculant production, distribution and sale in Ethiopia, and
particularly the uptake of inoculants by small-scale legume growers. Reviewing the
hoop tests, we conclude as follows.
1. Did the expansion of inoculants production and use take place?
Yes. Both documentary evidence and the testimony of key informants
conrms that the volumes of production, distribution, sale and use of
inoculants have increased signicantly, from a very low base. Demand
among farmers for reliable inoculant products has also increased.
However, all stakeholders agreed that the inoculant market remains far
below its estimated potential.
2. Was N2Africa support used in the process?
Yes. It is very clear that N2Africa’s support was crucial, especially in two
key respects. First, it played a direct role in securing funds to upgrade and
expand inoculant production at MBI. Second, the PPP regional cluster
model brought stakeholders into contact with one another and enabled
them to work together in ways that had not existed before the project.
3. Was N2Africa necessary to speed up or scale faster?
Yes. While actors in the system already aspired to develop and expand a
market for legume inoculants, and to use new biofertilisers to improve and
increase legume production before N2Africa began, the project played
a catalytic, facilitating role that substantially helped to accelerate these
processes. In particular, N2Africa’s support was key in enabling MBI to
prepare and submit a winning proposal to AGRA and the Bill & Melinda
Gates Foundation (BMGF), through which they won a grant to expand
their production facilities. There is also good evidence that the PPP
model of collaboration among multiple stakeholders helped to accelerate
progress in legume inoculant production, distribution and uptake.
4. Was N2Africa a necessary causal factor for the expansion of inoculant
supply and use to take place?
Likely, yes. While it is clear that the expansion of inoculant production
and use would have been slower and less assured in the absence of
N2Africa, it is very plausible that these developments might have been
delayed indenitely or prevented entirely without N2Africa’s intervention.
The AGRA/BMGF to MBI might have occurred without N2Africa’s
IDS PRACTICE PAPER 10
33
intervention, but the company acknowledged that it had relied very heavily
on N2Africa’s expertise to compile the successful bid. Similarly, the PPP
model was hailed as a positive innovation that was introduced uniquely by
N2Africa. Hypothetically, N2Africa’s interventions might have been made
by a different agency, programme or project, and in fact there were and
are other organisations seeking to improve legume production systems.
However, as far as we are aware, no other organisation was actually
undertaking such work during the period when N2Africa was underway.
The question remains whether the capacity of MBI and other key players
has attained a stable critical mass to expand the market using its own
resources, and whether the productive collaboration of partners involved
in N2Africa Ethiopia will continue now that the project has come to an end.
Several of the key informants interviewed for this analysis expressed the
desire to sustain the relationships they had developed with other players
within the framework of the N2Africa PPPs, but also a concern that the
incentive and capacity to collaborate would disappear without N2Africa to
coordinate. Various informants called for greater support and commitment
from government, but various other factors were also mentioned, such as
access to capital and the need to invest in training and capacity building,
particularly in key areas such as quality control, handling and storage of
inoculants, and training farmers.
It is worth observing that N2Africa was not the only actor that has sought to
develop the legume sector and promote legume technologies. There was
some interest in legumes and inoculants before N2Africa came along and
some of its work is being taken up by other projects now that N2Africa has
ended. However, the available evidence does support the conclusion that
N2Africa’s strategic and targeted approach has made a strong contribution
to the development and strengthening of legume production in Ethiopia.
Finally, it is worth observing that this report has focused on N2Africa’s major
claim to have contributed on the input and production side of the legume sector,
particularly the improvement of capacity in the production of effective inoculants,
the production and supply of improved seeds and fertilisers, and work with farmers
to improve productivity on farms. However, these priorities cannot be pursued
exclusively without ensuring that there are channels to market, driven by the
demand of end-users domestically and internationally. Although some activities
were carried out that focused on the downstream links of the value chain (including
linking farmers to export buyers) and some work on domestic consumption and
nutrition (e.g. food preparation training), the expansion and strengthening of the
production system presented sufcient challenges in their own right and appear
to have absorbed much of the project ofcers’ attention. In this area, some major
challenges remain, including securing investment for the further expansion of
inoculant production, improving quality, expanding distribution networks, and
resolving bottlenecks in cashow and logistics.
IDS PRACTICE PAPER 10
34
4 Discussion: the value of
Contribution Analysis
A notable feature of this evaluation study was the degree of engagement and
cooperation between the small evaluation team and the staff and stakeholders who
had been involved in implementing the N2Africa project, especially the respective
project teams in Ghana and Ethiopia. The study was not commissioned to be a
detached and independent evaluation for the benet of an external audience, but
as an exercise in accompanied learning and reection for the benet of the project
members themselves.
Key objectives of the evaluators were not only to evaluate the project as rigorously
as possible, but also to demonstrate the methodology of Contribution Analysis and
to show how it could be used by the N2Africa team themselves to examine and
reect on their own activities and achievements. The underlying motivation was to
explore a way to learn effectively from the implementation of a project, in order to
be able to extract and apply lessons to future activities. From this perspective, it
was important for the project participants to learn about what worked well and what
fell short of expectations; to reect on the realism of the assumptions embodied
in the project’s Theory of Change and its specic impact pathways; to be rigorous
in identifying the kinds of evidence that could underpin convincing impact claims;
and to gather and collate this information so that a truthful and convincing impact
narrative could be developed.
Contribution Analysis and process tracing use multiple data sources to evaluate
impact similar to a courtroom session, and do not rely on one method of data
collection to test whether an intervention works. It collects the evidence that
supports the presence of the causal step and the evidence that could falsify it,
and, as a result, gives an evaluation of the evidence presented. Ideally the three
functions (attorney, barrister, judge) are taken up by separate persons or groups
of people. However, real-world constraints made it necessary for us to take the
three functions in one. The conrming evidence was largely presented by N2Africa
teams in the initial stage of the research. To look for disconrmative evidence, we
explored the wider literature and interviewed persons that could be expected to
have a more critical stance to the intervention. Finally, as judges, we reected on
both sources of evidence.
Contribution Analysis is meant to give a critical look, not to work towards a
communication product that only presents the good sides of an intervention.
The selection of cases took place during a workshop of all N2Africa country
coordinators based on an inventory of cases where the participants expected to
have contributed to impact at scale. The choice of overtly successful cases for the
pilot of the approach made it difcult to nd disconrmative evidence, especially
in the Ethiopian case. That is good news for N2Africa but also implies that it
makes it quite difcult to show the main strength of the process tracing approach –
searching explicitly for evidence to conrm or discard alternative explanations than
N2Africa support. Generally, this external critical scrutiny helps to reconsider some
causal links or rene the understanding of the causal process. Another challenge
IDS PRACTICE PAPER 10
35
was the differences in the extent to which the contribution claim specied higher-
level outcomes in the Theory of Change. When an intervention makes a bold claim,
like in Ghana, disconrming or contested evidence might be easier to nd than
when the contribution claim is more modest and realistic, as in the Ethiopian case.
A key benet of Contribution Analysis is that when implemented carefully, it is
capable of revealing outcomes and impact pathways that were unexpected. This
makes it possible, for example, to reveal how the activities of other stakeholders
may have helped the project to achieve its goals or expand its reach; or to
discover some positive impacts of a project that were not anticipated in the original
plan. Of course, it can also show how and why a project has fallen short of its
objectives. In the case of N2Africa, key lessons include the fact that the project
managed to achieve some positive impacts, but that there is doubt over whether
its achievements will prove to be sustainable now that the project has come to an
end. This, of course, is a very common predicament in the eld of development
projects. Contribution Analysis may make a positive impact here, if it helps
project funders to see how heavily outcomes typically rely on stable institutional
commitments, building strong relationships, and investing in initial successes.
IDS PRACTICE PAPER 10
36
Annexe: List of interviews conducted
for this Contribution Analysis
Ghana
Interview
number
Informant name Organisation
1. Joshua Nyaaba Evangelical Presbyterian Development and Relief
Agency (EPDRA), Yendi, Manager
2. Mahamah Abdul-Rahaman Yara Ghana Limited, Commercial Director,
Northern Ghana and Burkina Faso
3. Jalil Zakaria 2SCALE, Tamale, Country Team Leader
4. Nana Osei Benji X’Mart Marketing Links, Managing director
5. Sachibu Mohammed Green-EF Eco-Business Village Ltd, Chief
Executive Ofcer (CEO)
6. Samuel Adjei-Nsiah N2Africa, Country Coordinator
7. Eric Doe N2Africa, Business Development Ofcer
8. Abdul-Rashid Zakaria Urban Agriculture Network, Executive Director
9. Zakaria Sumani Iddrisu Heritage Seeds, Managing Director
10. Multiple (7 farmers) Taaganoba/Tibigangso Farmers Union
Ethiopia
Interview
number
Informant name Organisation
1. Mirko Shibru Mirko Agrodealer
2. Dejene Woldemariam Menagesha Biotech Industry (MBI)
3. Biruk Tesfaye Catholic Relief Services (CRS)
4. Million Bogale Balegreen Spice and Grains Development plc
(Balegreen)
5. Engidu Legesse GUTS Agro Industry plc
6. Tulu Degefu Hawassa University
7. Yonas Sahlu Alliance for a Green Revolution in Africa –
Scaling Seeds and Technologies Partnership
8. Kie Degefa Oromia Agricultural Research Institute, Bako
Agricultural Research Centre
9. Muleta Assefa Bureau of Agriculture, Bako
10. Endalkachew Abie Tsehay Farmers’ Cooperative Union
11. Yonas Worku Amhara Agricultural Research Institute
(ARARI–Gonder)
12. Wubetu Ayele Farmer, Dembia
13. Aragaw Tefara Bureau of Agriculture, Dembia
IDS PRACTICE PAPER 10
37
References
Adjei-Nsiah, S.; Alabi, B.U.; Ahiakpa, J.K. and Kanampiu, F. (2018) ‘Response of
Grain Legumes to Phosphorus Application in the Guinea Savanna Agro-Ecological
Zones of Ghana’, Agronomy Journal 110.2: 1–8
Adjei-Nsiah, S.; Kumah, J.F.; Owuso-Bennoah, E. and Kanampiu, F. (2019)
‘Inuence of P Sources and Rhizobium Inoculation on Growth and Yield of
Soybean Genotypes on Ferric Lixisols of Northern Guinea Savanna Zone of
Ghana’, Communications in Soil Science and Plant Analysis 50.7: 853–68
Amanor, K. (2019) Mechanised Agriculture and Medium-Scale Farmers in Northern
Ghana: A Success of Market Liberalism or a Product of a Longer History? APRA
Working Paper 23, Brighton: IDS
Ampadu-Boakye, T.; Ronner, E. and Kanampiu, F. (2018a) N2Africa Annual Report
2017, Wageningen: Wageningen University & Research
Ampadu-Boakye, T.; Ronner, E. and Kanampiu, F. (2018b) N2Africa Annual Report
2018, Wageningen: Wageningen University & Research
Ampadu-Boakye, T.; Ronner, E. and Kanampiu, F. (2017) N2Africa Annual Report
2016, Wageningen: Wageningen University & Research
Argaw, A. and Mnalku, A. (2017) ‘Effectiveness of Native Rhizobium on Nodulation
and Yield of Faba Bean (Vicia Faba L.) in Eastern Ethiopia’, Archives of Agronomy
and Soil Science 63.10: 1390–1403
Avea, A. et al. (2016) ‘Do NGOs and Development Agencies Contribute to
Sustainability of Smallholder Soybean Farmers in Northern Ghana – A Stochastic
Production Frontier Approach’, Sustainability 8.5: 465
Balume, I.K.; Keya, O.; Karanja, N.K. and Woomer, P.L. (2015) ‘Shelf-Life of
Legume Inoculants in Different Carrier Materials Available in East Africa’, African
Crop Science Journal 23.4: 379–85
Beach, D. and Pedersen, R.B. (2013) Process-Tracing Methods: Foundations and
Guidelines, Ann Arbor MI: University of Michigan Press
Befani, B. and Mayne, J. (2014) ‘Process Tracing and Contribution Analysis:
A Combined Approach to Generative Causal Inference for Impact Evaluation’,
IDS Bulletin 45.6: 17–36, hps://bullen.ids.ac.uk/index.php/idsbo/arcle/view/139
(accessed 4 November 2019)
Belete, S. et al. (2019) ‘Inoculation and Phosphorus Fertilizer Improve Food-feed
Traits of Grain Legumes in Mixed Crop-livestock Systems of Ethiopia’, Agriculture,
Ecosystems & Environment 279: 58–64
Belkin, A. and Tetlock, P.E. (1996) Counterfactual Thought Experiments in World
Politics: Logical, Methodological, and Psychological Perspectives, Princeton NJ:
Princeton University Press
Beshir, H.M.; Walley, F.L.; Bueckert, R. and Tar’an, B. (2015) ‘Response of Snap
Bean Cultivars to Rhizobium Inoculation under Dryland Agriculture in Ethiopia’,
Agronomy 5.3: 291–308
IDS PRACTICE PAPER 10
38
Degefu, T.; Wolde-Meskel, E. and Rasche, F. (2018) ‘Genetic Diversity and
Symbiotic Effectiveness of Bradyrhizobium Strains Nodulating Selected Annual
Grain Legumes Growing in Ethiopia’, International Journal of Systematic and
Evolutionary Microbiology 68.1: 449–60
Dejene, M. et al. (2018) ‘Variations in Seed and Post-harvest Residue Yields
and Residues Quality of Common Bean (Phaseolus Vulgaris L.) as a Ruminant
Feedstuff’, Animal Feed Science and Technology 244: 42–55
Dontsop, P. and Ampadu-Boakye, T. (2019) Impact Evaluation Report of N2Africa
Project, Wageningen: Wageningen University & Research
Dotse, S.K. and Badu, A.-E. (2018) N2Africa Success Story: Putting Nitrogen
Fixation to Work for Smallholder Farmers in Northern Ghana, Accra: International
Institute of Tropical Agriculture
Farrow, A. et al. (2019) ‘From Best Fit Technologies to Best Fit Scaling:
Incorporating and Evaluating Factors Affecting the Adoption of Grain Legumes in
Sub-Saharan Africa’, Experimental Agriculture 55: 226–51
Hecht, S.B. (2005) ‘Soybeans, Development and Conservation on the Amazon
Frontier’, Development and Change 36.2: 375–404
Hoppenbrouwers, M. (2018) ‘Usage of Agricultural Technologies for Soybean and
Groundnut’, MSc thesis DEC-80433, Wageningen: Wageningen University
Lamptey, S.; Ahiabor, B.; Yeboah, S. and Asamoah, C. (2014) ‘Response of
Soybean (Glycine Max) to Rhizobial Inoculation and Phosphorus Application’,
Journal of Experimental Biology and Agricultural Sciences 2.1: 72–77
Mayne, J. (2011) ‘Contribution Analysis: Addressing Cause and Effect’, in K. Forss,
M. Marra and R. Schwartz (eds) Evaluating the Complex: Attribution, Contribution,
and Beyond, Piscataway: Transaction Publishers
Mayne, J. (2001) ‘Addressing Attribution Through Contribution Analysis: Using
Performance Measures Sensibly’, Canadian Journal of Program Evaluation 16.1:
1–24
Mohammed, A-R.S.; Al-hassan, S. and Jatoe, J.D. (2018) ‘An Overview of
Constraints to Soybean Production in the Northern Region of Ghana’, UDS
International Journal of Development 5.1: 32–40
Musebe, R.; Njuge, R. and Silvestri, S. (2018) Gender and the Legume Alliance:
Integrating Multi-media Communication Approaches and Input Brokerage: A Report
of Focus Group Discussions in Ghana, London: CAB International
Punton, M. and Welle, K. (2015) Straws-in-the-wind, Hoops and Smoking Guns:
What can Process Tracing Offer to Impact Evaluation? CDI Practice Paper 10,
Brighton: IDS
Richards, P. (2007) ‘How Does Participation Work? Deliberation and Performance
in African Food Security’, IDS Bulletin 38.5: 21–35, hps://bullen.ids.ac.uk/index.
php/idsbo/arcle/view/818 (accessed 4 November 2019)
Ronner, E. (2018) ‘From Targeting to Tailoring: Baskets of Options for Legume
Cultivation Among African Smallholders’, PhD dissertation, Wageningen University
IDS PRACTICE PAPER 10
39
Ronner, E. et al. (2016) ‘Understanding Variability in Soybean Yield and Response
to P-fertilizer and Rhizobium Inoculants on Farmers’ Fields in Northern Nigeria’,
Field Crops Research 186: 133–45
Samago, T.Y.; Anniye, E.W. and Dakora, F.D. (2018) ‘Grain Yield of Common Bean
(Phaseolus Vulgaris L.) Varieties is Markedly Increased by Rhizobial Inoculation
and Phosphorus Application in Ethiopia’, Symbiosis 75.3: 245–55
Shepherd, A. et al. (2005) Economic Growth in Northern Ghana: Revised Report
for DFID Ghana, London: Overseas Development Institute
Ton, G. (2017) ‘Contribution Analysis of a Bolivian Innovation Grant Fund:
Mixing Methods to Verify Relevance, Efciency and Effectiveness’, Journal of
Development Effectiveness 9: 120–43
Ulzen, J.; Abaidoo, R.C.; Ewusi-Mensah, N. and Masso, C. (2018) ‘On-Farm
Evaluation and Determination of Sources of Variability of Soybean Response
to Bradyrhizobium Inoculation and Phosphorus Fertilizer in Northern Ghana’,
Agriculture, Ecosystems & Environment 267: 23–32
Ulzen, J.; Abaidoo, R.C.; Mensah, N.E.; Masso, C. and Abdelgadir, A.H. (2016)
‘Bradyrhizobium Inoculants Enhance Grain Yields of Soybean and Cowpea in
Northern Ghana’, Frontiers in Plant Science 7: 1770
van Heerwaarden, J. et al. (2018) ‘Soyabean Response to Rhizobium Inoculation
Across Sub-Saharan Africa: Patterns of Variation and the Role of Promiscuity’,
Agriculture, Ecosystems & Environment 261: 211–18
Wellspring (2019) Inoculant ‘Go to Market’ Strategy (draft) – Inception Report,
Bristol: Wellspring
Wolde-Meskel, E. (2019a) N2Africa Annual Report 2018 Ethiopia, Wageningen:
Wageningen University & Research
Wolde-Meskel, E. (2019b) ‘A Walk Through the N2Africa-Ethiopia Project
Implementation and Partnership Journey: Experiences on Legume Research,
Dissemination and Input-Output Market Access’, N2Africa – Ethiopia Project
Closing Workshop, International Livestock Research Institute, Addis Ababa, 3 May
Wolde-Meskel, E. et al. (2018) ‘Additive Yield Response of Chickpea (Cicer
Arietinum L.) to Rhizobium Inoculation and Phosphorus Fertilizer Across
Smallholder Farms in Ethiopia’, Agriculture, Ecosystems & Environment 261:
144–52
IDS PRACTICE PAPER 10
40
This page is intentionally left blank
Article
Full-text available
Background: The Foodgrains Bank has an established record working in agriculture and food security with resource-constrained, marginalized farmers in sub-Saharan Africa. The three outcome areas of the Scaling-Up Conservation Agriculture in East Africa (SUCA) program were the adoption of conservation agriculture systems, an enabling institutional environment, and the promotion of enabling policies. These program areas were expected to yield intermediate outcomes that, together, would lead to the ultimate outcome of improved food security and sustainable livelihoods for smallholder farming households in East Africa. This case study reports on the endline evaluation of the five-year program. Purpose: To illustrate the overlap between utilization-focused evaluation (UFE) and collaborative approaches to evaluation (CAE). The case study profiles an agricultural intervention and explores how the evaluation design accommodated the systemic nature of the program. Setting: Scaling-Up Conservation Agriculture in East Africa (SUCA) was a five-year program of the Canadian Foodgrains Bank implemented from 2015 to 2020 to expand the size and scope of Foodgrains Bank's work in conservation agriculture in East Africa. The program supported local partners with a target of 50,000 male and female farmers practicing a minimum of two of three conservation agriculture principles, and of improving food security and sustainable livelihoods for 18,000 of these farmers' households across three countries. Research Design: The Foodgrains Bank was directly involved in the evaluation design through the definition of evaluation uses and key evaluation questions. Eleven implementing partners in East Africa were involved in primary data collection and some initial analysis. Data Collection and Analysis: A mixed-methods approach was used, combining quantitative, qualitative, and participatory / visual data collection tools. A robust, intersectional gender lens was applied to the data collection instruments in the form of gender-disaggregated data collection and gender-focused questions across most data collection instruments. Findings: The collaborative process confirmed a sense of ownership by the primary evaluation users over the evaluation design. The evaluation design combined outcome and learning uses that took advantage of the implementing organizations' commitment to learning. The findings demonstrated the value of the program and inspired a framework called CA Plus that illustrates the multidisciplinary approach underlying the program's success.
Article
Full-text available
Grain legumes play an important role as source of food and feed in smallholder mixed systems. They also contribute to soil fertility improvement through biological nitrogen fixation. Although rhizobium inoculation and phosphorus fertilizer are known to improve grain yield of legumes, information is limited on the effect of this practice on the yield and fodder quality of the haulm. This study was conducted to evaluate the effects of rhizobium inoculation (I) and phosphorus fertilizer (P) on yield and nutritional quality of grains and haulms of grain legumes (faba bean, chickpea, common bean and soybean) on farm across diverse agroecological locations in the Ethiopian highlands. The crops were subjected to four treatments [+I, +P, −I + P and a negative control (-P-I)] at multiple locations on farm during the main cropping season in 2016. Yield data was recorded during grain harvesting, and subsequently representative samples of grains and haulms were collected and analyzed for quality variables. Effects of the treatments were significant (P < 0.05) with 30% increase on grain yield for all studied crops and 28% increase on haulm dry matter yield for faba bean, common bean and soybean. Crude protein (CP) and in vitro organic matter digestibility (IVOMD) values of faba bean, common bean and soybean haulms were higher (P < 0.05); and neutral detergent fiber (NDF) and acid detergent fiber (ADF) contents were lower (P < 0.05) for the treatments than the control. The haulm CP content and IVOMD of chickpea also responded positively (P < 0.05) to the treatments. The current results demonstrated the possibility of improving both yield and quality of grains and haulms of grain legumes with the application of efficient rhizobium inocula and P fertilization. This practice offers an opportunity for smallholders in the crop-livestock system to improve the food-feed traits of grain legumes with minimal input and environmental footprint.
Article
Full-text available
Soybean yields on smallholder farms in sub-Sahara Africa (SSA) are far below the potential yield thus creating a huge yield gap. Interventions are thus needed to bridge this yield gap and ascertain the factors influencing the yield variation. This study evaluated the on farm response of soybean to rhizobia inoculation and or mineral P fertilizer in Northern and Upper West regions of Ghana in a single non-replicate trial using four treatments: no input (control), TSP fertilizer (P), rhizobia inoculant (I) and TSP plus inoculant (P + I). In addition, the study sought to develop a robust approach for determining responsiveness and non-responsiveness using agronomic and economic indices. The results showed that the average grain yield of plots that received P or I were higher than control plots. Higher grain yield responses were however, obtained by the plots that received combined application of P and Bradyrhizobium inoculant. Grain yield response in the Northern region was higher than in the Upper West region. Response to P and or I were highly variable within and between locations. The cumulative rainfall and some soil factors including soil nitrogen, phosphorus, soil type, organic carbon, pH and texture explained about 42-79% of these variations in soybean grain yield. The agronomic approach for determining responsive and non-responsiveness revealed that 17-40 % and 6-17% of the locations within the Northern and Upper West regions, respectively were responsive to P fertilization and/ or Bradyrhizobium inoculation. However, the economic approach indicated that 64-75% and 14-24% of the locations within the Northern and Upper West regions, respectively were responsive to P fertilization and Bradyrhizobium inoculation. The results imply that rhizobia inoculation is an effective strategy for increasing soybean yield and improving livelihood of smallholder farmers.
Article
Full-text available
This paper investigates the constraints that impede soybean production in the Northern Region of Ghana. The paper describes the trend in soybean production in the Northern Region from 2001 to 2010. It identifies and ranks the constraints to soybean production in the region. Ms Excel was used to plot soybean production figures from 2001 to 2010 sourced from SRID, MoFA. The Kendall’s coefficient of concordance (W) model was used to rank the identified constraints impeding soybean production. The results indicate that output of soybean increased from about 36,852MTs in 2007 to about 99,408MTs in 2010 in the study area (about 169.7%). Inadequate farm credit, inadequate rainfall and lack of improved planting materials are the most serious constraints hindering soybean production. The paper recommends that there should be timely delivery of credit in the right amount requested by soybean farmers by the Ministry of Food and Agriculture, SADA and other stakeholders. Quality planting materials should be made accessible and affordable to farmers by Agricultural stakeholders in Ghana.
Article
Full-text available
Grain legumes (cowpea, peanut, and soybean) play important roles in household food and income security in smallholder farming systems in the Guinea Savanna agro-ecological zones of Ghana. However, yields are low, rarely exceeding 600 kg ha⁻¹, prompting the need to evaluate responses of grain legumes to P fertilizer applications for two seasons. Conducting P studies is critical to help farmers adopt economic-based recommendations. Treatments evaluated in 2015 for the three crops were (i) farmers’ practice (no input and planted by farmer); (ii) control (no input and planted by researcher), and (iii) triple super phosphate (TSP) fertilizer. However, for soybean, an additional two treatments (inoculant only and inoculant plus TSP fertilizer) were included. In 2016, the treatments were the same, except on-farm demonstrations were not conducted on cowpea. The demonstrations were laid out in a Randomized Complete Block Design with each demonstration representing a replicate within a region. On average, P-fertilizer application increased yields by 296; 527, and 390 kg ha⁻¹ for cowpea, peanut, and soybean grains, respectively. On average over the two seasons, P-fertilizer increased yield by 9.85; 13.00, and 17.56 per kg ha⁻¹ kg⁻¹ P applied for cowpea, soybean, and peanut, respectively, and these applications were cost effective. Peanut showed little response to P in the Upper East Region compared with a greater response in the Northern and Upper West Regions, suggesting that benefits from P-fertilizer for peanut may be location-specific. On average, rhizobium inoculation increased grain yield by 157 kg ha⁻¹ across the three regions and significantly positive effects of inoculation were observed in both seasons. Our results show that substantial increases in grain legume yield may be achieved by applying P fertilizers, but farmers cannot afford them because of their relatively high cost. Planting adapted and improved varieties and using rhizobium inoculants may provide the most economically viable and low risk options for increasing yields of grain legumes in the savanna agro-ecological zones of Ghana.
Article
Full-text available
A field experiment was conducted to assess plant growth, symbiotic performance and grain yield of common bean in response to rhizobial incoculation and phosphorus application at Galalicha in Southern Ethiopia during the 2012 and 2013 cropping seasons under rain-fed conditions. The treatments consisted of 2 released common bean varieties (Hawassa Dume and Ibbado), 3 levels of Rhizobium inoculation (uninoculated, inoculated with strain HB-429 or GT-9) and 4 levels of phosphorus application (0, 10, 20 and 30 kg P ha⁻¹) using a split-split plot design with four replications. Here, phosphorus levels, Rhizobium inoculation and common bean varieties were assigned as main, sub- and sub-sub treatments, respectively. The results revealed marked varietal differences in plant growth, grain yield and symbiotic performance. Of the two common bean varieties studied, Hawassa Dume generally showed superior performance in most measured parameters in 2013. Rhizobium inoculation significantly (p ≤ 0.05) increased plant growth, symbiotic performance and grain yield. Applying Rhizobium strain HB-429 to bean crop respectively increased plant growth, %Ndfa, amount of N-fixed and grain yield by 19, 17, 54 and 48% over uninoculated control. Similarly, the application of 20 kg P ha⁻¹ to bean plants respectively resulted in 36, 20, 96 and 143% increase in plant growth, %Ndfa, N-fixed and grain yield when compared to the control. These results clearly indicate that plant growth, symbiotic performance and grain yield of common bean can be significantly increased by Rhizobium inoculation and phosphorus fertilization in Ethiopia. Rhizobium inoculants are a cheaper source of nitrogen than chemical fertilizers and when combined with moderate phosphorus application can markedly increase grain yield for resource-poor farmers.
Article
Full-text available
Improving bacterial nitrogen fixation in grain legumes is central to sustainable intensification of agriculture in sub-Saharan Africa. In the case of soyabean, two main approaches have been pursued: first, promiscuous varieties were developed to form effective symbiosis with locally abundant nitrogen fixing bacteria. Second, inoculation with elite bacterial strains is being promoted. Analyses of the success of these approaches in tropical smallholder systems are scarce. It is unclear how current promiscuous and non-promiscuous soyabean varieties perform in inoculated and uninoculated fields, and the extent of variation in inoculation response across regions and environmental conditions remains to be determined. We present an analysis of on-farm yields and inoculation responses across ten countries in Sub Saharan Africa, including both promiscuous and non-promiscuous varieties. By combining data from a core set of replicated on-farm trials with that from a large number of farmer-managed try-outs, we study the potential for inoculation to increase yields in both variety types and evaluate the magnitude and variability of response. Average yields were estimated to be 1343 and 1227 kg/ha with and without inoculation respectively. Inoculation response varied widely between trials and locations, with no clear spatial patterns at larger scales and without evidence that this variation could be explained by yield constraints or environmental conditions. On average, specific varieties had similar uninoculated yields, while responding more strongly to inoculation. Side-by side comparisons revealed that stronger responses were observed at sites where promiscuous varieties had superior uninoculated yields, suggesting the availability of compatible, effective bacteria as a yield limiting factor and as a determinant of the magnitude of inoculation response.
Article
Full-text available
This study was initiated to evaluate the effect of locally isolated Rhizobium on nodulation and yield of faba bean at Haramaya, Ethiopia for three consecutive years. Ten treatments comprising of eight effective isolates of rhizobia, uninoculated, and N-fertilized (20 kg N ha⁻¹) were laid out in a randomized complete block design with three replications. The result of the experiment indicated that all inoculation treatments increased nodule number and dry weight over the control check in all cropping seasons. The result, however, showed the non-significant effect of Rhizobium inoculation on shoot length, number of tiller per plant and 100 seed weight in all cropping season. Inoculating HUFBR-15 in 2011 and NSFBR-30 in 2012 and 2013 gave the highest grain yields (4330, 5267 and 4608 kg ha⁻¹), respectively. These records were 75, 48 and 5% over the uninoculated treatment of respective years. Over the season, NSCBR-30 inoculation resulted in the highest nodulation and grain yield production as compared to the other treatments. In general, isolates from central Ethiopia were better than those isolated from eastern Ethiopia and TAL-1035 in enhancing faba bean production at Haramaya site. Therefore, NSFBR-30 is recommended as a candidate isolate for faba bean biofertilizer production in eastern Ethiopia soils.
Article
Soybean is becoming an important cash crop in northern Ghana. Yet the yields are low due to use of low yielding varieties and limited use of inputs. Greenhouse and field experiments were carried out to evaluate the effects of two phosphorus (P) sources and Rhizobium inoculation on growth, nodulation, P uptake, and yield of three soybean genotypes on Ferric Lixisols of the Guinea savanna zone of Ghana. The P sources were triple superphosphate (TSP) and Morocco phosphate rock (MPR), while the genotypes were TGx 1448-2E, TGx 1904-6F, and TGx 1955-4F. The greenhouse experiment was conducted at the University of Ghana, Legon in a completely randomized design. The field experiment which was carried out in the Upper East region of Ghana was laid out in a split-split plot design with four replicates. In both the greenhouse and field experiments, application of TSP at 30 kg P ha⁻¹ resulted in significantly higher growth and P uptake in shoot compared with MPR and control. Soybean genotypes showed significant differences in growth, nutrient uptake, and grain yield in both the greenhouse and the field experiments. Rhizobium inoculation increased nodule number and dry weight but did not increase grain yield. The genotype TGx 1955-4F appears to show greater potential for increasing productivity of soybean in low P soils in northern Ghana.
Article
Common bean is widely grown as a food legume and the post-harvest crop residues (CR) (i.e. haulm + pod wall (HPW)) are valuable as ruminant feedstuffs. The yields and constituents indicative of nutritive value for ruminants of the HPW from a wide range of common bean genotypes (G) were examined at 4 trial sites in Ethiopia during the 2013 main cropping season to assess the extent of genetic variation among G for simultaneous improvement of both HPW attributes and seed yield. Attributes measured were seed and HPW yields and the amounts of the morphological components, their concentrations of total nitrogen (N), neutral detergent fibre (aNDFom) and acid detergent fibre (ADFom), and the dry matter digestibility (DMD). The constituents were measured using near infrared spectroscopy (NIRS) and calibrations based on a large set of reference tropical forages and CR (including common bean), and were validated against other CR reference samples. These CR quality attributes were very well predicted with R²v and RPDv ranging from 0.90 to 0.98 and 3.13–7.36, respectively. There was considerable variation in yields of HPW and seed, and in the proportions and attributes of the HPW fractions among the common bean G. Trial site means for yields of HPW and seed ranged from 0.74 to 2.54 t/ha and 0.79–2.62 t/ha, respectively while for N, aNDFom and ADFom concentrations and DMD of HPW ranged from 7.7 to 11.4 g/kg DM, 648–739 g/kg DM, 502–585 g/kg DM, and 467–570 g/kg DM, respectively. Environment (E), as represented by site, generally affected the yields of HPW and seed (P < 0.001) and nutritive value of the HPW fractions (P < 0.05) as feedstuffs. Seed yield was positively correlated with HPW yield both within and across trial sites (r = 0.92; P < 0.0001), but in general seed yield was not related to the N concentration. Across all sites, seed yield was positively correlated (r = 0.68; P < 0.0001) with haulm DMD. Although this correlation may be due to variation associated with E rather than G, it is nevertheless important in that selection for higher seed yield is likely to also increase metabolisable energy (ME) content of the HPW. There were G x E interaction effects on yields of HPW (P < 0.0001) and seed (P = 0.011), but these were generally less important than E effects which explained 52–58% of the variation. In conclusion the study demonstrated that it is possible to identify genotypes such as ECAB0081 which combine high yields of both seed and HPW, and with HPW attributes which improve their quality as ruminant feedstuffs.
Article
Vigna unguiculata, Vigna radiata and Arachis hypogaea growing in Ethiopia are nodulated by a genetically diverse group of Bradyrhizobium strains. To determine the genetic identity and symbiotic effectiveness of these bacteria, a collection of 36 test strains originating from the root nodules of the three hosts was investigated using multilocus sequence analyses (MLSA) of core genes including 16S rRNA, recA, glnII, gyrB, atpD and dnaK. Sequence analysis of nodA and nifH genes along with tests for symbiotic effectiveness using δ(15)N analysis were also carried out. The phylogenetic trees derived from the MLSA grouped most test strains into four well-supported distinct positions designated as genospecies I-IV. The maximum likelihood (ML) tree that was constructed based on the nodA gene sequences separated the entire test strains into two lineages, where the majority of the test strains were clustered on one of a well-supported large branch that comprise Bradyrhizobium species from the tropics. This clearly suggested the monophyletic origin of the nodA genes within the bradyrhizobia of tropical origin. The δ(15)N-based symbiotic effectiveness test of seven selected strains revealed that strains GN100 (δ(15)N=0.73) and GN102 (δ(15)N=0.79) were highly effective nitrogen fixers when inoculated to cowpea, thus can be considered as inoculants in cowpea production. It was concluded that Ethiopian soils are a hotspot for rhizobial diversity. This calls for further research to unravel as yet unknown bradyrhizobia nodulating legume host species growing in the country. In this respect, prospective research should also address the mechanisms of symbiotic specificity that could lead to high nitrogen fixation in target legumes.