‘Unlock the Complexity’: Understanding the Economic and Political Pathways Underlying the Transition to Climate-Smart Smallholder Forage-Livestock Systems: A Case Study in Rwanda

Abstract

The livestock-dairy sector in Sub-Saharan Africa, particularly in Rwanda, is experiencing rapid growth due to population expansion, urbanisation, and changing food preferences. The unmet local production demands are causing soil and water pollution, competition for biomass, land, and water, but also grassland degradation, biodiversity loss, and increased GHGs emissions. Rwanda has the lowest productivity in the region, largely due to inadequate and poor-quality livestock feed resources. To increase animal productivity, promoting forage species with higher nutritional value and better adaptation to drought-prone and poor-fertility soils could be beneficial. Using a mixed-methods approach, the study explores Brachiaria forage adoption and profitability and analyses policy objectives and measures to overcome adoption barriers and promote the transition from subsistence to market-oriented systems. Results show that Brachiaria, although advantageous from an economic point of view, is characterised by very low adoption rates. Furthermore, access to extension programmes is limited and often not supported by adequate incentives. To overcome such barriers, policy interventions should be harmonised and information and knowledge management prioritised, public and private extension and advisory services (EASs) programmes coordinated, agricultural input subsidies increased, and institutional coordination promoted to enhance climate-smart animal feeding.

1. Introduction

The livestock-dairy sector in Sub-Saharan Africa (SSA) plays a significant role in the region’s economy, contributing to about 35% of the agricultural GDP and supporting rural livelihoods and food security across local communities (Erdaw 2023). It is dominated by smallholder farmers, with the majority of production coming from small-scale, family-owned farms. This sector has been experiencing significant expansion due to various factors, including rapid population growth, urbanisation, and changing food preferences among an increasingly affluent and urbanised population (van Berkum et al. 2017). This phenomenon, also known as the ‘livestock revolution’, is driving a surge in the demand for high-value products, including animal-derived foods such as meat, milk, and eggs (Delgado et al. 2001). In SSA, the per capita yearly consumption of meat and milk (now equal to about 14 kg and 30 litres, respectively, on average) is anticipated to increase to 26 kg and 64 litres by 2050 (Balehegn et al. 2021). As the increasing consumption of animal products is not yet met by equivalent growth in local production, producers need to invest in livestock farming and the related value chains to supply such a growing demand (FAO 2019). However, an uncontrolled expansion of livestock activities and associated markets may have negative implications in terms of soil and water pollution, competition for biomass, land, and water, but also grassland degradation, biodiversity loss, and increased greenhouse gas (GHG) emissions (Gerber et al. 2005; Nijdam et al. 2012; Steinfeld et al. 2006; Enahoro et al. 2019; FAO 2019; Herrero et al. 2014). In such a situation, encouraging forage species with higher nutritional content and improved tolerance to soils prone to drought and low fertility is an effective strategy for raising animal production. In Eastern Africa, Brachiaria is emerging as a nutritious grass with high biomass production and hence the potential to increase cattle productivity (Maina et al. 2020). It can adapt to a variety of soil types and settings, withstand protracted droughts, and thrive in low fertility conditions (Rao et al. 2014). It can also utilise nitrogen more efficiently and sequester carbon thanks to its extensive root system (Subbarao et al. 2009; Arango et al. 2014; Rao et al. 2014; Maina et al. 2020). However, despite the potential benefits of Brachiaria, little is known about its drivers, barriers, impacts of adoption, and economic viability.

The objectives of the present study are to: (i) compare the profitability of farm systems that use Brachiaria grass to those adopting Napier; (ii) investigate the adoption of Brachiaria grass and the obstacles that prevent farmers from using it; and (iii) map and evaluate the primary policy measures aimed at overcoming barriers to sustainable agriculture intensification practices, with a specific focus on extension advisory services’ role, structure, and effectiveness.

The research employs a mixed-methods approach, combining quantitative and qualitative instruments to allow for a holistic understanding of Brachiaria adoption dynamics and identify a policy framework that can promote sustainable and market-oriented livestock-dairy systems. Regarding the study’s novel feature, we make use of an extensive and distinctive set quantitative and qualitative data related to the adoption of SAI practices in mixed crop-livestock systems in two Rwandan districts. This improves the research since there is a lack of information concerning on-farm, cleaner agricultural production practices in the areas under investigation. It also allows us to provide more rigorous data for policymakers as well as additional information about the strategic path for promoting the implementation of climate-smart farming systems.

The study was carried out in Kirehe and Nyamagabe districts, respectively located in the south-eastern and south-west parts of the country. Mixed crop-livestock systems are common among smallholder farms in these two regions, and animals are generally enclosed and fed by cut-and-carried plant foods, including crop residues, and other feeds such as concentrates, while the adoption of Brachiaria forage is particularly low.

The rest of this paper is structured as follows. The background of the study is presented in Section 2. Data and methodology are presented in Section 3. Results are illustrated in Section 4 and discussed in Section 5. Conclusions are reported in Section 6.

2. Background of the Study

2.1. Challenges and Opportunities for Implementing Brachiaria in East Africa

Over 70% of African agricultural GHG emissions come from the livestock sector, with emissions per unit of livestock product in East Africa four times higher than the global average (Tubiello et al. 2014; Pressman et al. 2018). Moreover, 58% of land degradation and soil erosion in the SSA regions is caused by overgrazing, which is a result of rising human populations and increased cultivation of food crops (Balehegn et al. 2021). Low productivity of livestock risk exacerbating these hazards and related implications (Herrero et al. 2013, 2014; Enahoro et al. 2019). As a result, increasing farm animal productivity through sustainable and climate-smart practices may be viewed as a strategic option to address food security and climate change concerns, as well as social and economic issues. However, several factors, including inadequate infrastructure, lack of market access, lack of access to technology and knowledge, lack of control over animal diseases, and environmental concerns, significantly hinder the productivity of smallholder livestock production (Erdaw 2023). This is especially true in East Africa, which is characterised by the SSA’s greatest livestock population, but where the livestock productivity is 15% lower than the rest of the region because of infrastructural inefficiencies and climate-induced stresses that are putting additional strain on the local food systems (Rahimi et al. 2021). In Rwanda, the livestock-dairy sector experiences the lowest productivity in the region, despite its pivotal role in the local economy and in improving the nutrition and livelihood of rural people (Mutimura and Ghimire 2021). One of the major factors contributing to such inefficiency is the inadequate and poor-quality livestock feed resources used at the farm level. Overgrazing, poor pasture management, and drought have led to a decline in natural grass land, resulting in decreased productivity and low nutritive values for crop residues, which are often insufficient for animal maintenance (Mutimura and Everson 2012). Napier grass, the country’s most widely cultivated forage, faces threats from smut and stunting diseases, causing yield losses ranging from 5% to 90% (Lusweti et al. 2004). This poses a risk to the livestock-dairy sector’s ability to reach its full potential, meet the anticipated future increase in demand for animal products, join complex and high-value markets, and cope with the projected climate changes. In such a context, promoting forage species with higher nutritional value and better adaptation to drought-prone and poor-fertility soils is a viable strategy for increasing animal productivity. In this regard, the Rwandan government has been introducing, assessing, and promoting improved and climate-smart forages like Brachiaria, whose production and utilisation is a tried-and-true method for increasing availability of high-quality feed and improving animal productivity (Mutimura and Ghimire 2021). It is one of the most important tropical forage grasses of African origin and is widely cultivated in Latin America due to its large foliage biomass production, its root systems that improve soil structure, its capacity to fix atmospheric carbon, and its resilience to drought and low fertility soils (Tesfai et al. 2019). In addition, numerous studies have demonstrated that Brachiaria is distinguished by a high nutritional value for cattle feed, resulting in an increase in daily milk production ranging from 30 to 100% in the long term (Ghimire et al. 2015; Mutimura et al. 2018). This means that the adoption of such forage in animal feeding can have a potential positive effect in terms of food security but can also promote the transition from subsistence to market-oriented smallholder production systems. Despite the described benefits, this climate-smart forage is characterised by low adoption rates among smallholder farms due to their reliance on traditional systems, limited awareness, limited availability of agricultural inputs such as quality planting materials and fertilisers, and lack of technical assistance and supportive policies and incentives (Ondabu et al. 2017; Njarui et al. 2016). In such a scenario, creating an enabling environment for agricultural development is regarded as a key element for climate-smart agricultural innovations to take place. However, the effectiveness of such a condition depends on its combination with investments in extension services as well as a policy framework aimed at removing the barriers to the implementation of climate-smart strategies on a large scale. Extension and advisory services (EAS) are offered by public-private partnerships, non-governmental organisations, and farmer-based organisations (Birner et al. 2009). These services use various training and education approaches, including participatory methods like farmers’ field schools and informal channels like farmer-to-farmer communication (Branca et al. 2022b). However, the greater decentralisation of public agricultural extensions and a lack of recognition and involvement of private EASs providers may underserve resource-poor farmers.

2.2. Study Area

The study here illustrated was conducted in two districts of Rwanda: Nyamagabe and Kirehe (Figure 1). Nyamagabe is the southern province; it occupies 1090 km² and has a population of 333,587. Kirehe is the eastern province; it occupies 1225 km² and has a population of 229,468. The topography in Kirehe district is flat land surrounded by hills and water bodies, with moderate levels of soil erosion. Nyamagabe receives higher annual rainfall (1636 mm) with respect to Kirehe (750 mm) and is characterised by lower average temperatures (16.5 °C) if compared to the other district (characterised by an average temperature equal to 21 °C). In Nyamagabe, soils are clayey and acidic, with high levels of aluminium. Soils in Kirehe are characterised by a sandy loam texture with lower amounts of organic matter. Mixed crop-livestock systems are prevalent in both districts. In Nyamagabe, crops and livestock are well integrated, and terracing, due to its steep slopes, is common. In this district, the average farm size is less than 0.5 ha. In Kirehe, the current farming systems are characterised by monocropping of maize, intercropping of cereals with legumes, mixed crop-livestock systems, and agroforestry. The average land holding here is equal to 0.7 ha. The major challenges to the adoption of sustainable agriculture intensification (SAI) systems in the selected districts are less fertile soils and heavy rain, which negatively affect crop and livestock productivity in Nyamagabe, whereas drought stress and small land holdings are the major constraints to improving crop and livestock production in Kirehe district. Other challenges include feed shortages for livestock, limited access to agricultural technologies, and market availability.

3. Materials and Methods

3.1. Methodology

The methodology applied in the present study consists of the following analytical steps: (i) a marginal analysis comparing the profitability of Brachiaria grass to the conventional fodder used at the farm level (Napier); (ii) a descriptive analysis examining Brachiaria’s adoption rate and determining the associated barriers; (iii) an intervention logic diagram identifying and evaluating policy objectives and strategies to overcome obstacles and limitations faced by farmers in adopting SAI systems; and (iv) a VENN diagram investigating the role and impact of EASs services. Figure 2 shows the linkage between paper objectives, methods employed, and sources of information.

3.1.1. Brachiaria Financial Viability through Marginal Analysis

A marginal analysis at the farm level was conducted to compare the profitability of forage-livestock systems adopting Brachiaria with respect to those that use Napier grass. In the analysis, the value of gross production (which includes fodder and milk produced by the farm), operating costs, and labour costs were calculated using financial pricing. The following indicators were then computed to assess and compare the financial viability of the two different farm-level feeding systems (Branca et al. 2016, 2021): (i) gross margin (revenues minus operating costs), (ii) net margin (revenues minus total production costs), and (iii) return to family labour (gross margin to amount of family labour employed ratio). Such financial analysis was realised using household survey data.

3.1.2. Quanti-Qualitative Analysis of Brachiaria Adoption

A descriptive analysis was realized to determine the adoption rate of Brachiaria in the research area and the barriers to its adoption. The survey data were used to conduct such an analysis. The results were then evaluated, assessed, and discussed utilising information gathered from FDGs carried out in the two districts.

3.1.3. Policy Analysis through Intervention Logic Diagram

An Intervention Logic diagram was here defined to map the relationships between the policies (their instruments/measures and their specific, intermediate, and global objectives), the institutional environment (public and private EASs providers), and the barriers related to the adoption of SAI systems (also based on Brachiaria) in the study area. This diagram was realised based on the information gathered from both FDGs and MAP meetings held in the two selected districts. The intervention logic diagram, as defined by the European Commission (2016), illustrates the logical link between the problem that needs to be tackled (or the objective that needs to be pursued), the underlying drivers of the problem, and the available policy options (or the actions actually taken) to address the problem or achieve the objective. Such an instrument is used in prospective impact assessments and evaluations.

The Intervention Logic diagram here illustrated consists of different sections that are interlinked. Section 1 is related to the identification and categorization of the institutional environment. Section 2 is dedicated to the definition of weaknesses and adoption barriers to innovations. Finally, Section 3 aims to: (i) identify and rank the most relevant policies related to innovation adoption; (ii) identify the most important instruments and measures as relevant tools to implement policy programs; and (iii) identify the most crucial global, intermediate, and specific goals that the policy can achieve.

3.1.4. EAS Analysis through Venn Diagram

A Venn diagram was here used to identify and map out the key institutions involved in extension agricultural services and to assess their importance and their links with the others (Gentle and Maraseni 2012). It allows us to draw the relevance of institutions through the size of circle representing each institution as follows: large circles are used for important organizations and smaller for less relevant ones.

Venn diagram information is typically gathered through focus group talks, with participants asked to rank the various institutions according to importance in the local context (Branca et al. 2022b). When a farmer receives the same service from two different institutions, two circles intersect, resulting in an overlapping area proportional to the number of households supplied by both organisations.

For the purpose of the present study, we draw Venn diagrams using household survey data. To assess the quality of extension services provided by different organisations, we examined the frequency of farm visits and farmer related satisfaction. As a result, the Venn diagram was created considering institution coverage in terms of households supplied, weighted by the number of visits realized and the level of farmer satisfaction.

3.2. Data and Source of Information

The study here presented was realized under the Horizon 2020 project “Innovations in technology, institutional and extension approaches towards sustainable agriculture and enhanced food and nutrition security in Africa (Innovafrica)”. In this context, qualitative and quantitative data were collected through a survey focusing on smallholders in Kirehe and Nyamagabe districts in Rwanda, as well as focus group discussions (FGDs) and Multi-Actor Platform (MAP) meetings involving livestock-dairy value chain actors.

3.2.1. Field Survey, Questionnaire Structure and Sample Description

The study used data obtained from a field survey carried out from January to February 2018 in Kirehe and Nyamagabe districts. The study adopted a multi-stage approach for selecting farmers to participate in the research, which included identifying the relevant farmer population, involving both adopters and non-adopters of Brachiaria, selecting clusters representing lower administrative units, and randomly selecting farmers for interviews. A total of 616 farms were surveyed in the study area (308 farms in each district). The sample size was defined using the following Cochran’s sample formula:

n₀ = Z²pq/e²

where: Z represents the standard deviation of the 95% confidence interval characterising a normal distribution; p is the estimated proportion of farms adopting mixed livestock-forage systems; q is equal to (1 − p), and e represents the desired level of precision.

Table 1. Structure of the HH questionnaire.

A	General information
B	Demographic characteristics and assets
C	Fodder/pasture production and management
D	Milk production, processing and marketing
E	Crop production/Seed System
F	Climate change and variability, adaptation and coping strategies
G	Membership in agricultural associations and access to extension and advisory services
H	Food and nutrition security
J	Access agricultural input and credit
K	Household decision making

Source: Authors’ elaborations.

reports the structure of the questionnaire that allows us to collect the following information: smallholders’ socio-demographic and economic profiles, household’s structure and decision making, challenges in implementing innovations in the agriculture sector, access to agricultural input and credit, food and nutrition security, membership in agricultural associations, and access to extension and advisory services.

Table 2. Sample description and statistics.

Variables	Description	Mean	St.Dev.
Demographics
Household head, gender	1 if female, 0 if male	0.781	-
Household head, age	Age of the household head (years)	50.394	12.822
Household head, education	1 if household head attended at least primary school, 0 otherwise	0.760	-
Economic assets
On-farm income	Income generated by agricultural activities (USD)	36.742	43.539
Off-farm income	Income generated by non-agricultural activities (USD)	12.955	29.929
Credit access	1 if household has access to credit, 0 otherwise	0.391	-
Physical assets
Land area	Size of the agricultural land (ha)	0.804	0.987
Livestock owned	Number of animal heads owned	1.567	0.857
Local breed	1 if household own local breed, 0 otherwise	0.243	-
Exotic breed	1 if household own exotic breed, 0 otherwise	0.102	-
Cross breed	1 if household own cross breed, 0 otherwise	0.684	-
Social assets
Group participation	1 if household is member of farmer’s associations, 0 otherwise	0.357	-
EASs access	1 if household has access to EASs, 0 otherwise	0.128	-
Other
Milk sales	1 if household sells the milk produced, 0 otherwise	0.193	-
Feed shortage experience	1 if household experienced feed shortages, 0 otherwise	0.911	-
Climate change	1 if household perceived climate changes, 0 otherwise	0.974	-

Source: Authors’ elaborations.

illustrates the major characteristics of the sample analysed. The bulk of the sampled farmers interviewed are middle-aged men with at least a primary school education. The selected rural households have average on-farm and off-farm monthly incomes, respectively equal to about 13 and 40 USD, while just about 39% of them have access to credit. In terms of physical assets, sampled farms have an average land area of 0.8 hectares and own an average of two milking cows, with 68% owning cross-breeding cattle heads, while 24% and 10% hold local and exotic breeding animals, respectively. Considering milk production, only roughly 19% of farms sell milk and animal-derived products, while the rest of them are still subsistence-oriented. In terms of involvement in networks, associations, and extension programmes, around 36% of sampled farms belong to farmers’ groups, with just about 13% having access to EASs. Finally, in terms of environmental and climate shocks, it is worth noting that the majority of farmers interviewed have encountered climate stress in the previous five years, as well as a livestock feed shortage in the past 12 months.

3.2.2. Focus Group Discussion

A FGD meeting was conducted in 2019 in the selected study sites with the support of a semi-structured questionnaire to obtain in-depth information about weaknesses, opportunities, and barriers characterizing the livestock-dairy value chain as well as details on existing policies and extension advisory services supporting its growth. The participants in the meeting included farmer organisations and cooperatives (i.e., the IAKIB cooperatives), input suppliers (agro-dealers), extensionists, market intermediaries and collectors, international traders, food processors, and food retailers.

Table 3. Structure of the Focus Group Discussion’s questionnaire.

A	Please, identify the main constraints characterizing the livestock-dairy value chain
B	Please, identify the main strengths characterizing the livestock-dairy value chain
C	Please, identify the main opportunities characterizing the livestock-dairy value chain
D	Please, identify the main threats characterizing the livestock-dairy value chain
F	What are the main policies at national and subnational level and programmes supporting the livestock-dairy value chain
G	To what degree is the subnational/national/international agricultural policy conducive for farmers to adopt innovation?
H	What is the national policy as regards Agricultural Extension and Advisory Services?
I	Who are providing agricultural advice to farmers, and what are the capacities of these different actors

Source: Authors’ elaborations.

reports the main sections of the questionnaire used to guide the FGDs.

3.2.3. Multi-Actor Platforms (MAP)

Multi-actor platforms (MAP) meetings were held during June–July 2019 and January–March 2020. MAP meetings are one of the innovative institutional approaches aimed at facilitating the adoption of the new technologies. They are seen as the vehicle for agricultural innovation and development (van Mierlo and Totin 2014; van Paassen et al. 2014; Hermans et al. 2017). Within the scope of the current research, such meetings bring together people from the public sector, farmers ‘organisations, cooperatives (i.e., the NDFFR cattle cooperative), NGOs, and small and medium enterprises (SMEs). They aimed at activating a process of interactive learning, empowerment, and collaborative governance that enables stakeholders to be collectively innovative and resilient when faced with emerging risks, crisis, and opportunities in a complex and changing environment. During these meetings, participants were guided by a semi-structured questionnaire in order to: (i) determine the implementation level of policy instruments aimed at promoting the sustainable transition of the livestock-dairy systems; (ii) identify instruments, measures and specific, intermediate, and global objectives related to the identified policies; and (iii) discuss and validate the selected policies’ effectiveness, as well as the influence of the institutional environment.

4. Results

4.1. Brachiaria Financial Viability

According to the theory of expected utility, a farmer’s decision to adopt or not adopt a technology such as Brachiaria grass, given the risk and unknown prospects, is based on the maximisation of profits (Maina et al. 2019). In this regard, it is critical to understand the levels of performance attained by farmers in forage-livestock production systems, as well as the financial viability of Brachiaria-based systems with respect to those based on conventional forages. Figure 3 illustrates the difference in terms of grass biomass (kg/ha) and livestock productivity (litres of milk/head/day) between Brachiaria and Napier. In terms of biomass produced, results show that the productivity of Brachiaria per hectare of cultivated land is higher with respect to Napier grass. Additionally, the data indicate that using Brachiaria as animal feed increases milk production.

Table 4. Comparative financial analysis of Brachiaria and Napier-based livestock-dairy systems (in USD).

Financial Budget (in USD)	Conventional (Napier)	Climate-Smart (Brachiaria)
Gross value of production
Forage	402.70	657.48
Fresh Milk	915.03	1098.04
Total revenue	1317.73	1755.51
Operating input costs
Seeds/Planting material	1.92	0.77
Top Dress Fertilizer	0.16	0.00
Basal Fertilizer	8.13	21.57
Animal manure	0.05	0.00
Additional concentrated feed/	496.25	234.62
Water	175.97	175.97
Veterinary services	24.11	24.11
Sub-total operating costs	679.58	457.04
Labour costs
Application of manure/fertilizers	3.92	2.80
Cutting pasture/fodder from farm	18.60	19.96
Land preparation	17.03	16.03
Planting of pastures/fodders	5.22	3.26
Source/buy seeds/planting material	1.93	4.34
Transport forage	10.26	18.80
Weeding pastures/fodders	20.60	18.64
Animal management	43.99	43.99
Sub-total labour costs	121.56	127.84
Sub-total production costs	801.14	584.88
Gross Margin	638.15	1298.47
Net Margin	516.59	1170.63
Return to family labour	4.16	7.72

Source: Authors’ elaboration.

shows the financial budget of representative smallholder farms characterised by an average size of 0.8 hectares and adopting, alternatively, the two aforementioned forage systems. In terms of costs, it is interesting to notice that forage-livestock systems based on the production and use of Brachiaria have lower operating costs and somewhat higher labour costs than forage systems depending on Napier. On the other hand, labour expenses incurred by farms adopting Brachiaria are higher than conventional forage-livestock systems, although by only 8% percent.

In terms of profitability, forage-livestock systems based on Brachiaria are more virtuous, with gross margins, net margins, and returns to family labour that are roughly double then those of traditional and Napier-based production systems.

4.2. Brachiaria Adoption Rate and Barriers

Although the adoption of Brachiaria allows farmers to benefit from undoubted improvements in terms of productivity and profitability, this climate-smart forage remains scarcely used. As illustrated in Figure 4, the majority of farms surveyed are unfamiliar with Brachiaria. Out of the 20% of farmers who know about this improved forage, only 10% actually use it on their farms. As illustrated in Figure 5, household survey data indicates that the main reason why farmers do not use Brachiaria is connected with the lack of access to seeds. This result was also confirmed by information gathered during the FDGs held in the two districts under investigation. In relation to the general constraints characterizing the livestock-dairy value chain, participants were asked to focus on the barriers to the adoption Brachiaria grass in the study area. Specifically, the following issue arose during the meeting: “What could be the main constraints to the adoption of Brachiaria grass faced by local farmers and other value-chain actors?”. In both districts, what emerged from the discussion was that lack of information and technical knowledge, as well as lack of seeds and high input costs, are the main barriers to the adoption of Brachiaria. Furthermore, during the meeting’s discussion, the lack of recognition of Brachiaria among the surveyed farmers was attributed to the fact that this improved forage is relatively new in Africa (despite the aforementioned African origins). The barriers directly related to the use of Brachiaria were also associated with threats related to the variability of climate conditions, the increasing incidence of pests and diseases, and the diminishing fertility of the soil. On the other hand, as highlighted by the focus groups’ participants, widespread adoption of Brachiaria might be viewed as both a development opportunity and a strategy to be combined with increased utilisation of improved highly productive dairy cows (an initiative supported by the one cow per family national policy).

4.3. Policy Analysis and EAS Investigation

4.3.1. Intervention Logic and Effectiveness of Policy Objectives and Measures

Based on the benefits associated with Brachiaria and its low adoption rate in Rwanda, this study examines policy and strategies put in place to promote the adoption of such climate-smart technology and to improve the livestock-dairy value chain as a whole. Figure 6 shows the Intervention Logic diagram, which maps the relationship between policies (their instruments and measures and their specific, intermediate, and global objectives), the institutional environment, and the barriers hindering the transition to the adoption of climate-smart innovations and market-oriented agricultural systems.

The barriers identified by the participants during the FGDs include land-related issues such as fragmented arable land, limited land size, erosion, and unfertile soil. Acidic soil is one of the major limits to agricultural productivity in many parts of Sub-Saharan Africa and is included in the list of obstacles that farmers encounter in their activities. However, while this element is considered hazardous for the growth of food crops, it appears to be well-suited to the production of Brachiaria. Other major barriers include a lack of mechanisation and inputs, poor infrastructure, a high Brachiaria seed price, and their scarce availability. Policy measures to overcome such barriers, promote innovation, and execute national policies were identified during the FGDs and then addressed in terms of implementation level and effectiveness during the MAP sessions. They include subsidies on livestock inputs, enhancing smallholders’ access to inputs at reasonable prices, improving livestock nutrition, and boosting animal productivity. Another policy measure adopted to support the local livestock-dairy systems consists of improving farmers’ skills in the use of inputs through training programmes. This element is strongly connected with the access to extension services, whose provision is guided by the National Agriculture Sector Extension Policy (NASEP). MAP members highlighted that extension services serve as a focal point for the processing of information on markets, inputs, credits, and producer coordination. They also confirmed that farmers have access to extension programmes through farm visits mainly provided by research institutes and the public sector, or through farmer-to-farmer systems. The involvement of the private sector, agro-dealers, and NGOs in the provision of extension services remains marginal, preventing a real diversification of the services and incentives provided. In this regard, as illustrated in the figure (institutional environment section), government closely collaborates in policy implementation with private actors through institutional tables for public and private coordination (Agricultural Sector Working Group, Sector-Wide Approach Group, and Sub Sector Working Groups).

Specific, intermediate, and global objectives have been linked with the selected policies’ instruments and measurements. The two specific objectives associated with the improvement of livestock nutrition and animal productivity are directly related to the adoption of Brachiaria grass as animal feed. In response to the question of whether such objectives have been reached or are still attainable, the MAP members said that “such goals are not yet reached because of poor livestock management and nutrition. Farmers continue to feed their cows in the traditional way, with the majority of animals fed using only Napier grass. They also lack experience in feeding and caring for their livestock. The primary impediment here is the scarcity and high cost of forage crops, concentrate feeds, and minerals that farmers may employ on their farms. They lack new approaches for feed management and conservation (hay and silage)”. Consider the specific aim connected to the enhancement of farmers’ skills in the utilisation of inputs, MAP members confirmed that “such an objective is achievable because of the availability of extension services from the government, researchers, and NGO’s. There is also a lot of training for farmers on the use of inputs and good practices in agriculture. However, we still have a long way to go to reach the expected target in the whole country.” Achieving the aforementioned specific objectives linked to the livestock sector and innovation adoption would make it possible to accomplish a number of intermediate productivity goals in addition to the more general objectives of enhancing sustainable agriculture and food and nutrition security.

4.3.2. Extension Advisory Services: Structure, Providers and Effectiveness

Considering the relevance of EAS services arose during the MAP meeting discussions, the study focused on examining the structure and the effectiveness of such services through the definition of a VENN diagram. The following delivery options were considered: formal governmental extension services, activities promoted by non-governmental organizations, services marketed by agro-dealers or other private firms, innovation disseminated by research centres, and knowledge shared by individual farmers (farmer-to-farmer) or farmer groups (cooperatives). Furthermore, the following extensions services were examined: fodder management, fodder selection, land preparation, feed conservation, feed management, climate early warning, credit, insurance, and milk processing.

As illustrated in Figure 7, most farmers have no access to any form of extension. Extension service system reaches only 12% of the sample. Additionally, the government, research institutions, and farmer-to-farmer programmes are the major sources of EASs services for smallholders, with a small percentage of farms receiving them from NGOs, private providers, and agrodealers. Furthermore, in the case study area, no farm has received extension services from cooperatives or community-based organizations. As highlighted during the MAP meetings, farmers mainly have access to services such as fodder management and selection and land preparation while the extension services related to the other farming activities are not available.

Considering the comparative weights of the various options, Figure 7 shows highlights the absence of overlapping circles, indicating that households receive the EASs from a single provider.

5. Discussion

5.1. Theoretical Relevance of the Results

The results of the comparative marginal analysis show that forage-livestock systems using Brachiaria yield double the profitability of traditional and Napier-based systems, primarily due to its improved productivity and higher market value, offering revenue opportunities through baled hay sales and extra milk production (Tesfai et al. 2019). This finding is in line with the literature, confirming that Brachiaria outperformed Napier in terms of profitability, primarily as a result of increased productivity (Maina et al. 2019). This can be because Brachiaria grass tends to be resistant to drought, infertile soils, and several diseases affecting traditional varieties in Eastern Africa (Ghimire et al. 2015; Maass et al. 2015). Furthermore, as illustrated by Schiek et al. (2018), in Rwanda, the adoption of Brachiaria reported a 30% increase in milk production and a 20% increase in meat production. In this regard, Tesfai et al. (2019) found that dairy cattle fed with Brachiaria grass and supplemented with legumes reported a higher daily milk yield than those based on Napier grass, while cows solely fed with Brachiaria brizantha cv. Piata produced 33% more milk then cows with a diet only based on Napier. Forage-livestock systems using Brachiaria have lower operating costs and higher labour costs compared to Napier systems. This is due to the low usage of inputs such as top dress fertilisers and animal manure. However, basal fertiliser expenditures are higher in Brachiaria-based systems. In this regard, despite Brachiaria’s greater resilience to infertile soils, the use of fertilisers has been demonstrated to improve yield and quality of the output, resulting in higher fresh and dry mass production, fibre, and crude protein content (do Valle Pereira et al. 2016). Furthermore, as illustrated by Mutimura and Ghimire (2021), the grass harvested under a cut and carry system (which is common in Rwanda) increases soil fertility quickly if no fertilizer or manure is added to the land under Brachiaria grass production. In terms of livestock management, Brachiaria systems are characterised by lower costs for the use of concentrates and supplements for animal feeding. Indeed, as illustrated in the literature, producing and properly storing high-quality forages can help lower the cost of feeding concentrates and supplements, particularly during the dry season (Guyader et al. 2016; Sokupa et al. 2023).

Although the cultivation of Brachiaria would allow farmers to benefit from undeniable increases in productivity and profitability, this climate-smart fodder is rarely exploited due to knowledge gap and difficulties in accessing seeds. This result is in line with findings found in previous studies (Njiru et al. 2023; Tesfai et al. 2019; García de Jalón et al. 2017) and confirms that one of the primary impediments to the use of Brachiaria is connected to seed economic and physical accessibility. In this regard, it is important to consider that the germplasm currently in use has been derived from cultivars that have been specifically selected or developed for acidic soil in Latin America (Ghimire et al. 2015). A limited amount of seeds has been imported from this region at a cost that frequently exceeds USD 20 per kilogramme, a price that is unaffordable to the majority of Eastern African smallholder farmers (Ibid.). Furthermore, as illustrated by Mutimura and Ghimire (2021), in Rwanda, the lack of seeds is also due to the troubles farmers face in producing seeds for commercial uses. This is due to the fact that such grass cannot generate enough seeds in locations near the equator.. Indeed, environmental conditions significantly impact seed production, especially in humid lowland tropics near the equator. Species that thrive in these environments often do not produce seeds, and those that do, like Brachiaria decumbens cv Basilisk, have poor yields (Hare et al. 2015). In this regard, it was demonstrated that altitude and latitude significantly impact flowering and seed setting in Brachiaria hybrids, with successful seed production in Brazil and Thailand occurring in latitudes 200–220S and 700–1000 m above sea level (Hare et al. 2015; Maass et al. 2015; Kamidi et al. 2016). The lack of seeds and the commercial sector’s reluctance to deal with an unorganised and dispersed demand for such input have been major impediments to the expanded usage of Brachiaria in Eastern Africa (Ghimire et al. 2015; Cheruiyot et al. 2020). Moreover, import limitations and a lack of defined procedures for importing seeds deter the private sector from marketing Brachiaria in the region (Vijay et al. 2015).

5.2. Policy Implications

The policy analysis’s findings show that the most important policy initiatives that promote innovation and allow overcoming the identified adoption barriers are those that support livestock input subsidies, improve smallholders’ access to inputs at fair prices, and train farmers on how to use inputs and improve livestock nutrition. The effectiveness of these programmes, however, may be hampered by a lack of funding and difficulties determining the best ways to move from high-level commitments to local action, depending on the executive ability and calibre of sector governance (Badiane et al. 2018; Haug et al. 2021; Branca et al. 2022b). Considering the policy measures related to the improvement of farmers’ skills and knowledge through extension services, García de Jalón et al. (2017) suggest that investing in factors that enhance social capital, such as extension services or agricultural associations, can provide a cost-effectiveness ratio in terms of adoption improvement that is much lower than investing in infrastructure. As highlighted during the MAP meetings, in Rwanda the extension delivery methods are becoming more pluralistic with the widespread use of mobile phones and information and communication technology (ICT). Furthermore, farmers can get information from different government institutions, including the Ministry of Agriculture and Animal Resources (MINAGRI) and the Rwanda Agriculture Board (RAB). One of the RAB’s responsibilities is “to provide agricultural extension services in accordance with agricultural and animal husbandry needs” (MacNairn and Davis 2018). It is decentralised at the village level in order to work with farmers together with extensionists from the local government. This enables famers to access information on input supply, market, and credit available in the area. However, as illustrated in Section 4.3.2, the extension advisory services remain inadequate and reach only 13% of farms. Furthermore, households generally receive the EASs from a single provider. According to Beaman et al. (2021), this aspect might negatively influence the adoption of such climate-smart technology since farmers are more likely to embrace new technologies when they are exposed to a wide range of extension sources. Farmers have access to extensions programmes through farm visits mainly provided by research institutes and public sector, or through farmer-to-farmer systems. The involvement of the private sector, agro-dealers, and NGOs in the provision of extension services remains marginal, preventing a real diversification of the services and incentives provided.

Despite the efforts made in recent years, inadequate financial resources allocated to policy, a low extensionist/household rate, the private sector not being perceived as an extension provider, and insufficient training on leadership skills for cooperative members are some of the most relevant barriers to policy implementation (Haug et al. 2021; Branca et al. 2022a).

6. Conclusions

The study examines Rwanda’s transition from a traditional to a sustainable livestock dairy sector, focusing on climate-smart solutions and policy instruments to promote sustainable innovations. The methodology uses converging and triangulation of quantitative and qualitative data sources to provide a comprehensive understanding of the complexities of developing policy interventions for efficient, profitable, and sustainable livestock-dairy smallholder systems. The study suggests that drought-resistant grass like Brachiaria could help farmers overcome feed shortages and stabilize production, leading to more continuous arrangements with buyers and improved bargaining positions. However, Brachiaria has low adoption rates, with only 2% of farmers using it. This is due to limited farmer awareness, access to extension programs, and the availability of expensive and difficult-to-produce agricultural inputs like seeds. Policy interventions should focus on promoting sustainable seed production systems, maximizing seed availability for smallholder farmers, and developing cost-effective harvesting, threshing, and storage structures.

Considering the current policy framework, Rwanda is implementing market-driven agricultural sector transformation initiatives, but their effectiveness is hindered by inadequate financial capacity, ineffective public extension service systems, limited private involvement, poor market coordination, and limited investor engagement in market infrastructure. Therefore, promoting new technologies should focus on clustering injection sites within a network (Beaman et al. 2021; Branca et al. 2022b). In this regard, a stable political environment and proper legislative actions are systemic characteristics that can promote agricultural innovation and rural revival through EAS systems customised to effectively communicate knowledge to rural households through diversified sources (Kosec and Resnick 2019; Norton and Alwang 2020).

Finally, it might be useful to harmonize and improve existing policies to promote the development of smallholder-friendly contexts through: (i) review current legal frameworks to remove inadequacies, conflicts, and overlaps in the various institutions charged with implementation of the various policies (e.g., a better coordination between the Ministry of Agriculture and the Ministry of Environment on the use of marshland could have an effective impact on the low land availability); (ii) increase budget allocation accompanied with adequate accountability mechanisms to ensure judicious use of the funds target policy instruments; (iii) provision of adequate and qualified staff to implement the policies; and (iv) creating awareness of and establishing mechanisms for sensitization of communities on existing policies to ensure adherence.

Future research on the effectiveness of policies promoting Brachiaria adoption and profitability should be expanded to other geographical and institutional settings, identifying parallels, or new critical elements underlying climate-neutral mixed crop-livestock systems.