The Trajectory of Climate Induced Shocks on Livelihoods and Responses: A Case of CSA Adopters' and Non Adopters' in Nyando Basin, South Western Kenya

Journal of Environmental Hazards

ISSN: 2684-4923

Open Access

Research - (2021) Volume 5, Issue 4

The Trajectory of Climate Induced Shocks on Livelihoods and Responses: A Case of CSA Adopters' and Non Adopters' in Nyando Basin, South Western Kenya

Josephine Njogu1*
*Correspondence: Josephine Njogu, College of Agriculture and Veterinary Sciences, University of Nairobi, Kenya, Tel: 0721259255, Email:
1College of Agriculture and Veterinary Sciences, University of Nairobi, Kenya


Climate aggravated stressors like crop failure, pests and diseases, hunger, high food prices and death of livestock or family members have increased in frequency, intensity and magnitude. These have, aggravate ding rural poverty while and threatening the sustainability of rural livelihoods. Yet Hitherto, many small scale farmers fail to adopt what appear to be relatively simple agronomic or management practices like climate smart agriculture (CSA) which can help them cope. The study focused on the effect of related household shocks to climate change on livelihoods, and factors influencing choice of a coping strategy of smallholder farmers in the Nyando Basin. The population of the study comprised all smallholder farmers that were registered with CCAFS, Kenya. Data was collected in a year-long panel survey of financial diaries (FDs) from 124 households between March 2019 and February 2020. The households were classified into adopters and non-adopters of climate smart agriculture (CSA). Descriptive statistics and multivariate probit regression were used to determine common types of shocks experienced, their trends, and choice of coping strategies. The results showed that most shocks arose from pests and diseases (48.3%) including sickness/death of family member (38.1%) with non-adopters of CSA being affected more than adopters. Market for farm produce was affected by low output prices (81.3%), no market (100%) over the year. Low production of crops was due to pests and diseases (93.9%), high input prices (91.5%) and asset disputes mainly land related (73.3%). Regression analysis indicated that characteristics of the household head, climate smart village location, and county were significant factors influencing choice of a coping strategy. The data indicates that policy recommendation be addressed according to County based selfhelp capacity.


Climate smart agriculture (CSA) • Covariate shocks • Idiosyncratic shocks • Adopters and non-adopters of CSA • Nyando Kenya


Weather variation has resulted in adversity and market fluctuations causing stress on individual households and the population at large in an area at any particular time. Subsequent effects of climate change has brought pests and diseases to the plants and animals [1]. Climatic events, mainly floods and droughts cause shocks and ensue stress to household members. Floods cause extensive soil wearing resulting in gullies, destruction of transport infrastructure, interruptions in school terms, destruction to property and increased incidences of water borne diseases (GoK, 2012). In Nyando basin Kenya, floods and droughts destroy crops/livestock and force agricultural activities to be practiced on a limited period within a year and at worst devastate livelihoods. Loss of livelihoods, reduction in purchasing power and loss of land value in the flood plains lead to economically vulnerable communities. The effects of repeated low agricultural yields are felt outside the production area as food prices increase due to limited supply. Although floods could result in long-term gains to agricultural production by recharging water reservoirs, especially in drier inland areas and through reviving soil fertility by silt deposition, their overall impact is negative. They devastate households and at times force relocation from one's home, loss of possessions and interference to business and social affairs causing stress. To some people, the psychological impacts of these events could be long lasting [2]. In particular, flood destruction to roads, rail networks and key transport points lead to noteworthy impacts on regional and national economies [3]. According to the Kenya Flood Mitigation Strategy (KFMS, 2009), floods recur in Nyando basin and an estimated 5000 people are affected in the area every year with grave impacts on the social, economic, environmental, physical and psychological welfare of people and even on the political and institutional levels.

Devereux and McKune (2014) in a study in Nyando indicated that Kisumu side was predisposed to drier conditions and endured less rain with seasonal variability. The rivers in this location frequently dry up, necessitating farmers to walk long distances in search of water. The average land size here is one acre with only one main planting season and poor crop yields as reported in recent years [4]. However, Kericho side of Nyando is wetter with better access to water, has two cropping seasons and larger land sizes (on average 5-10 acres). Farmers here have more livestock – mainly traditional and crossed breeds – and are more reliant on livestock products for income than on crop production.

As of 2011, 81% of the families in Nyando experienced at least two hunger months, while 17% had incidences of up to four hunger months (periods household faced difficulty in getting food from any source) [5]. Kimtai (2019) found out that those affected by death shock in Nyando basin relied on contributions from relatives and neighbors, loans from finance institutions for the funeral ceremonies, or stayed put and did nothing. Those affected by pests and diseases shocks used agro chemicals or sought assistance from extension providers to cope. Thorlakson and Neufeldt (2012) found that farmers in Nyando sold their farm implements and consumed seeds they had reserved for planting to cope with drought conditions. They also relied on relatives, churches and NGO’s, borrowing, selling of livestock, moving to higher grounds of Kericho to seek wage activities, and scaling down on the quantity and quality of meals. To cope with these changes small-scale farmers have devised their own adaptation strategies integrated with CSA technologies.

In 2011, CCAFS, ILRI, amongst other NGOs initiated CSA activities in the region to assist farmers mitigate effects of climate change especially floods and drought. CSA practices in Nyando area include agro-forestry, aquaculture, greenhouse, water catchment and use of improved seed varieties through different farm practices and systems in the villages [6]. The red Masai sheep and galla goats have also been introduced to improve breeds in the region.

In this study shocks were categorized into two namely; according to how widespread their impact/ prevalence was on households; villages and beyond. Idiosyncratic shocks were experienced within households while covariate shocks affected many households in a village and beyond. The effects of shocks are diverse with different impacts on various households, villages and populations. The study sought to understand if there were differences in effects from weather related households shocks on livelihoods of adopters and nonadopters of climate smart agricultural practices and strategies undertaken in Nyando basin.

Materials and Methods

Study area and sample size

The study was conducted in the CCAFS defined Nyando block that cuts across Kisumu and Kericho counties comprising the upper and lower Nyando basin. A stratified random sample of 122 households from the 2017 CCAFS end line survey of adopters and non adopters of CSAs was used according to [7]. The villages had similar characteristics in terms of climate, soils and main agricultural practices. The survey was conducted in September 2019 and March 2020 as mid-term and end line data. This was carried out within a financial diaries panel survey that took 52 weeks. 4,786 respondents were targeted and included household head, spouse and any child or members contributing to welfare of the household ( 1).

Using a pre-tested standardized questionnaire, purposively selected farmers were asked what their experiences were in the last six months. The data were analyzed using both descriptive and inferential statistics. Descriptive statistics were used primarily to analyze farmers' socioeconomic attributes. The probit multivariate model was used to determine the factors behind the choice of coping strategies and to analyze the effects of shocks on farmer’s livelihoods- incomes and assets losses.

Econometric model

The multivariate probit model is sui in probability analysis. After a shock, a household decides how to cope and takes a coping action/strategy considered accessible and achievable. At times, a household may refrain from adopting a coping action. The model allows for coexistence of several different strategies for one type of shock. It also relaxes the assumption of independence of irrelevant alternatives (IIA) assumed in the logit model [8]. Similar studies that explore the association between shock and coping strategy find multivariate probit appropriate for making J − different choices at a point in time [9,10]. It simultaneously captures the influence of the set of explanatory variables on each of the different coping strategy choices while allowing for the relationships between the choices of different options [10].

Model specification

The multivariate probit approach insync models the effect of a set of explanatory variables on each different coping strategies and at the same time allowing for the possible relationship between unobserved disturbances, as well as the association among the adoptions of diverse strategies (Mulwa et al., 2017). The equation in a system of multiple coping choices can be shown as

Սij = XijΒj + εij ( j = 1,2,3,..., s) (1)


Where Սij is the utility that the household i gets by choosing strategy j from the given set of options. In the MVP model where the adoption of several coping strategies is possible, the error terms ε jointly follow a multivariate normal distribution (MVN) with zero conditional mean and variance normalized to unity [11]. The model is specified as follows,


Where the dependent variable adapt: Yi is observed and takes the value 1 if a coping strategy is taken and; not adapt; Yi =0 and a set of explanatory variables (X for socio-demographic, V for village, S for shock and I for institutional) characteristics and the error term Σ i1.


Definition and measurement of variables

In this study using the multivariate approach, the dependent variable is categorized into two coping strategies. namely; 1. lifestyle adjust =(labourbased strategies, reduced food consumption) and 2 Welfare adjust = ( dissaved, borrowed, sold assets, received support) from which small scale farmers adopters and non-adopters of CSA made a choice.

The explanatory variables were selected from those which were cited in literature as likely to influence choice of a livelihood coping strategy and availability of variables in the data set. These were mainly household sociodemographic characteristics- land size, gender, age and household size; village characteristics- CSA adopters, distance to market, County; shock characteristics- income loss, asset loss; institutional characteristics- group membership, savings. The explanatory variables and how they were measured are summarized in 1.

Conclusion and Recommendations

Households cope differently to diverse shocks. Generally, the results of this study imply that shocks experienced by rural households in emerging market economies direct to losses in income and assets and therefore have repercussion for their exposure to poverty status. Shocks’ experience within non- adopters were greater than for adopters of CSAs within climate smart villages (CSVs). 84.9% of non-adopters respondents reported shocks as a result of pests and diseases in crops and livestock, 75.8% due to little rainfall, 42.4% sickness/ death of family member and 33.3% increase in input prices. Similarly adopters of CSA experienced shocks from pests and diseases 77.7%, little rain 62.4%, 30.6% sickness/death of family member, 12.9% increase in input prices. However, they suffered more shocks from high food prices than non-adopters at 45.5%. Market for farm produce was affected by low output prices (81.3%), no market (100%) over the year. Low production of crops was due to pests and diseases (93.9%), high input prices (91.5%) and asset disputes – mainly land (73.3%). Variables that influenced livelihood loss from idiosyncratic shocks were age resulting to lifestyle change and welfare change- gender, CSAs adopt, County. For covariate shocks variables that led to lifestyle change were age; welfare adjustment- County. The coping strategies employed included receiving support, doing nothing, borrowing, labor-supply based strategies and de-saving which involved also selling of household assets.

Therefore policies formulation on strategies to advocate for the uptake of healthcare and social insurance systems could be made more responsive and target-oriented. For example, community health funds and provision of shelters during droughts and floods may be developed to provide support for medicine and nutritive food to the vulnerable groups namely the aged, disabled persons, women and children.

Boost access to market information at village levels and self-help capacity of social safety nets at village level targeting rural households should be given priority over wide-range based government support schemes. This will enable small holder farmers access subsidized high yielding crop varieties and affordable credit. Because the often lengthy, complicated and corrupt procedures for subsidies and compensation from government authorities inhibit the effectiveness of such measures. Future area of study to focus on assessing what impacts growth in incomes portfolios amongst CSA small holder farmers including money borrowing from mobile phone lenders and beter gains/ loss.


arrow_upward arrow_upward