Determinants of Relevant Constraints Inhibiting Farmers’ Adoption of Climate Change Adaptation Strategies in South Africa

Abstract

The adverse effects of climate change on agricultural productivity are on the increase. This study employed both descriptive statistics and the multivariate probit model to estimate factors constraining the adoption of climate change adaptation strategies among smallholder farmers in the study area. The empirical results of the multivariate probit model showed that a lack of knowledge of climate change constraints was influenced by smallholder farmers’ age, gender, off-farm activity, susceptibility and membership in farmer-based organizations. Thus, to improve the adaptive capacity of farmers, government and development partners should work together to improve the conditions under which farmers can gain access to climate change information and suitable agricultural credit as well as policy incentives aimed at lowering the stringent conditions of borrowing in the agricultural sector.

Keywords

Adaptation strategies constraints multivariate probit model smallholder farmers

Background information

Climate change (CC) has brought substantial welfare loss, especially to smallholder farmers, in many developing countries (Khan et al., 2020; Komba and Muchapondwa, 2012). Given the known negative impact of CC, it is desirable to minimize its adverse effects. Thus, CC discourse focuses on mitigation and adaptation (Chambwera and Stage, 2010; Mogomotsi et al., 2020). As empirically posited in literature, the adoption of adaptation strategies therefore remains an important option in mitigating the effect of CC and addressing its prevailing challenges on agricultural production (Deressa et al., 2009; Guodaar et al., 2020; Seo, 2011). CC adaptation has emerged as one of the most important responses to environmental development challenges in the 21st century (Boko et al., 2007). The adverse effects of CC on agricultural productivity are on the increase, as posited by Mugi-Ngenga et al. (2016), and, with that, the need for smallholder farmers to devise adaptation measures (Omoyo et al., 2015). Adaptation strategies widely implemented in Africa include soil and water conservation structures, early maturing and drought-resistant improved crops, diversification, tree-planting and agroforestry and development of early warning systems. As opined by the Intergovernmental Panel on Climate Change (IPCC, 2001), adaptation strategies can help farmers mitigate their vulnerability to CC to enable rural communities to adjust better to CC and moderate potential damages, and to cope with adverse consequences. Adaptation will require the involvement of multiple stakeholders, including farmers (primarily), policymakers, extension agents, NGOs, researchers, local communities and private sectors. The limited adaptive capacity of smallholder farmers in sub-Saharan African countries, such as South Africa, places them mainly in a position of susceptibility to the adverse impact of CC.

In South Africa, the growth of smallholder farmers (land reform beneficiaries – LRBs) is stalled by the numerous challenges that they face. These also impede their contribution to food security compared with commercial farmers. Some of the challenges faced by smallholder farmers are linked to a lack of access to land, physical and institutional infrastructure (Kamara et al., 2019; Ncube, 2018; Ortmann and King, 2007). These constraints hinder their propensity to adopt CC adaptation strategies because of the huge capital outlay thereof. Another factor that impedes growth is the high transaction costs that these farmers incur mainly due to poor infrastructure, for example, unreliable markets and poor road networks that are disincentives to production and lead to low productivity. These farmers end up resorting to farm gate or local selling at low prices. Smallholder farmers also face the constraint of high illiteracy and poor technological skills. Most of the targeted LRBs lack business skills and their produce often fails to meet quality standards. In addition to the low quality, the quantity of their products is also low, resulting in failure to compete in the market. This inconsistency in production is worsened by the lack of bargaining power; hence the need for support services to ensure smallholder farmer growth.

Empirical studies related to CC adaptation strategies considered the impact of CC on agricultural productivity (Ajetomobi et al., 2010; Hassan and Nhemachena, 2008; Kurukulasuriya et al., 2006) and the perception of CC (Phuong et al., 2018; Williams et al., 2019). Other studies also analysed the determinants of CC adaptation strategies (Bryan et al., 2011; Mulwa et al., 2017; Ojo and Baiyegunhi, 2019) and the impact that the adoption of CC adaptation strategies has on food security (Kerr et al., 2019; Nunfam et al., 2019). However, there is limited information on the determinants of constraints to CC adaptation strategies, except for Vignola et al. (2015) who profiled the constraints but failed to estimate the factors constraining smallholder farmers’ adoption of CC adaptation strategies. It is against this backdrop that this study sought to empirically estimate the determinants of constraints to the adoption of CC adaptation strategies among LRBs in South Africa.

Knowledge of key socioeconomic factors influencing farmers’ adaptability to CC can play a significant role in policy formulation in order to mitigate the effects of CC on smallholder agriculture (Deressa et al., 2009). In the same vein, knowledge of these socioeconomic factors can also assist policymakers in strengthening adaptation by investing in socioeconomic measures. For instance, education of the household head is hypothesized to increase the probability of adapting to climate variability (Deressa et al., 2009). According to Mugi-Ngenga et al. (2016), the reason could be ascribed to the fact that education increases farmers’ ability to access, process and use relevant information to adapt to the effects of climate variability. Understanding the determinants of the adoption constraints on CC adaptation strategies among smallholder farmers is particularly relevant as a panacea to the implementation of CC adaptation strategies (Leclère et al., 2013).

This study therefore investigates the determinants of smallholder farmers’ constraints to the adoption of CC adaptation strategies. The paper is structured to consist of the following sections: discussions on the study area; survey design and method of data collection; the analytical framework used for this study; trends of CC variables; the empirical results and discussion; and, lastly, the conclusion and policy implication of the study.

Research methodology

Both qualitative and quantitative methods were used to collect data on smallholder farmers’ choice of CC adaptation strategies. According to Tewksbury and Scheufele (2009), qualitative research was used to seek understanding on the decisions of the smallholder farmer’s adaptation strategies towards CC by looking at first-hand experiences to provide data that was meaningful to the researcher. Qualitative data was collected through face-to-face interviews with smallholder farmers to probe their choice of the CC adaptation strategies by means of an open-ended guided questionnaire. The quantitative research method mainly uses a numerical analysis to reduce data into numbers or percentages, unlike the qualitative method (Crossman et al., 2012). This method uses a close-ended questionnaire. In this study, the quantitative method was used to allow for significant comparison of participants’ responses across study areas because they were all asked identical questions in the same order (Crossman et al., 2012).

The survey questionnaire was prepared in English and then translated into the various local languages (IsiZulu, Setswana, South Sotho and Tshivenda) so that the field workers could obtain accurate information from the farmers, since these languages are used by all of the residents/farming communities in these areas. A multistage stratified random sampling procedure, a combination of purposive and random sampling procedure, was used to identify and select a sample of the districts and smallholder farmers. Data was collected from 183 participants from nine district municipalities in four provinces as depicted in Figure 1, in South Africa, namely Lejweleputswa and Thabo-Mofutsanyane (Free State), eThekwini, uGu, iLembe and Amajuba (KwaZulu-Natal), Mopani and Vhembe (Limpopo) and Dr Kenneth Kaunda (North West). A purposive selection technique was used, based on the prevalence and susceptibility of the districts to climate-related problems and also districts that were recently declared drought disaster areas. A sample household was then selected by using simple random sampling (SRS) with probability proportional to size (PPS) technique. A simplified formula provided by Cochran (1977) was used to determine the required sample size at 95% confidence level, 5% degree of variability and 8% level of precision. Accordingly, 183 samples out of 200 households found analysable were selected for inclusion in the analysis. The primary data used for the study was obtained from a cross-sectional survey of the LRBs (smallholder farmers) in four provinces, namely the Free State, KwaZulu-Natal, Limpopo and North West, which made up the study area.

Figure 1.

District municipalities of the study areas in South Africa.

The fieldwork for this study was carried out from November 2017 to June 2018. The initial field activity was the investigation survey of the study area to establish background information on agroecological condition, livelihood activities, land use systems, natural resource base and development activities being implemented in the context of CC variations. To enrich the investigation survey, interactions were carried out with government officials working in the districts and selected individuals with knowledge of the study localities. Points of concern in this exercise were issues related to CC, variability incidences, development interventions on agriculture and natural resources management, and environmental protection activities designed to avert problems arising from climatic variability. The overall activity in this regard helped the researcher to establish a good picture of the study areas and prepare relevant questions in each data collection tool such as a questionnaire. To enhance selection of the specific districts in the rural areas, the districts were stratified based on similar attributes and representative districts were selected purposely considering the previously mentioned variables. STATA 15 was used for analysis.

Conceptual framework

Climate change adaptation strategies are a form of mitigating option that reduce the exposure of farmers to risk by reducing the marginal effect of climate change on smallholders’ productivity (Fisher-Vanden et al., 2011; Ojo and Baiyegunhi, 2019). The study adopted a utility maximization function in the presence of risk to conceptualize constraints to adoption of adaptation strategies. For this study, the utility to a farmer needs not be defined by higher yields. In the context of adaptation, the utility derived from adopting a practice could be yield stability and the implied reduction in downside risk. A risk-averse farmer maximizes utility by choosing an adaptation strategy if the benefits of adaptation (risk reduction) minus the cost of adaptation are higher than the benefits realized without adaptation.

Empirical model for the multivariate probit model

The multivariate probit (MVP) model was used to analyse the determinants of CC adaptation strategy constraints among smallholder farmers in the study area. This approach had been used in similar studies (Mulwa et al., 2017; Ojo and Baiyegunhi, 2019). Where the covariance matrix of disturbance is unknown, the feasible generalized least square (FGLS) is applied to estimate the parameters and correlation coefficients simultaneously (Zellner and Huang, 1962), while the least square residuals is used to consistently estimate the elements of the covariance matrix of disturbance (Greene, 2000). The MVP model was formulated in line with Lin et al. (2005), using three dummy dependent variables that represented the constraints to the adoption of adaptation strategies used by farmers in the study area to mitigate the effect of CC on their farms. CC adaptation strategy constraint is described as a series of dichotomous variables defining the possible categories of constraint, namely a lack of knowledge of CC constraint, a lack of CC information constraint and a lack of capital constraint. MVP recognizes the correlation in the error terms by simultaneously modelling the effects of a set of covariates on each of the constraints to the adoption of CC adaptation strategies and estimating a set of binary probit models.

In terms of econometric modelling, separate estimations would not capture this correlation and would not exploit the information deriving from the entire set of common regressors (Asfaw et al., 2016). A correlation coefficient with a positive sign is consistent with the unobserved heterogeneity in the discriminatory tendency against farmers (complementarity). However, a coefficient with a negative value is consistent with the interpretation that factors causing farmers to be placed in the constraint category may make them less likely to be placed in another category (substitutability).

MVP also establishes the relationship between the constraints to adaptation strategies and the accounting for potential correlations between unobserved disturbances (Danso-Abbeam and Baiyegunhi, 2017). Since the utility could not be observed, it was depicted as a function of observable components, as expressed in equation (1)

Q_{i k}^{*} = β_{k} X_{i k} + α_{k} S_{i k} + ε_{k} where (k = 1, \dots, m)

(1)

$Q_{i k} = 1 i f Y_{i k}^{*} > 0$ and 0 if otherwise

where $Q_{i k}^{*}$ , the latent variable that represents the unobserved characteristics, is associated with $k^{t h}$ , representing CC adaptation strategies, while $Q_{i k}$ denotes the binary dependent variables, and $(k = 1, \dots, m)$ represents the various practices used by smallholder farmers in the study area. A smallholder farmer was given a value of 1 if he or she was constrained in adopting any of the adaptation strategies, and 0 otherwise. $X_{i k}$ is a vector of the explanatory variables used in the model. According to Wooldridge (2003), $S_{i k}$ denotes susceptibility of smallholder farmers to CC, such as experience of drought and flood that account for unobserved heterogeneity. $β_{k}$ and $α_{k}$ are conformable vectors that are estimated. The error term, $ε_{k}$ , has multivariate normal distributions, with zero means, unitary variance and an n × n correlation matrix (Mulwa et al., 2017), where $ε_{k} \approx M V N (0, Π)$ . The covariance matrix, $Π$ , is provided by equation (2)

Π = | \begin{matrix} 1 & ρ_{12} & ρ_{13} & \dots & ρ_{1 m} \\ ρ_{21} & 1 & ρ_{23} & \dots & ρ_{2 m} \\ ρ_{31} & ρ_{32} & 1 & \dots & ρ_{3 m} \\ ⋮ & ⋮ & ⋮ & 1 & ⋮ \\ ρ_{m 1} & ρ_{m 2} & ρ_{m 3} & \dots & 1 \end{matrix} |

(2)

where $ρ$ represents the unobserved correlation between the stochastic components of the error terms in the model. As shown in equation (2), the elements at the off-diagonal represent the correlation between the stochastic error terms of the different constraints facing the farmers ( $ρ_{21}$ , $ρ_{12}$ , $ρ_{31}$ and $ρ_{13}$ ) in the variance-covariance matrix (Teklewold et al., 2013). The assumption of the unobserved correlation between the stochastic component of the $k^{t h}$ and $m^{t h}$ type of constraints indicates that equation (1) provides an MVP model that represents the joint constraints to adoption of a particular adaptation strategy.

Results and discussion

Descriptive statistics of smallholder farmers in the study area

This section reports the description of both dependent and explanatory variables included in the model estimations. The dependent variables are the adoption and intensity of adaptation strategies employed. This study draws its empirical specification from the studies of determinants of the adoption of CC adaptation strategies (Abdulai and Huffman 2014; Mulwa et al., 2017; Ojo and Baiyegunhi, 2019). The description of the explanatory variables and their respective means are presented in Table 1. The socioeconomic characteristics such as gender, age, educational attainment, household size and number of years in crop farming were included in the model to control for household heterogeneity. These variables have been hypothesized to potentially influence the constraints to the adoption of adaptation strategies. Out of the 183 responses, about 61% were from males, while 39% were from females. The average age of the sampled farmers was 43 years, suggesting that the majority of the farmers in the sample were in the productive age bracket. The majority (about 60%) of the respondents had attained at least a primary level of education.

Table 1.

Definitions and summary statistics of variables used in the model.

Variables	Description of variables	Mean	Std. Dev.
Dependent variables
Lack of knowledge	1 if constrained by lack of knowledge, 0 if otherwise	89.20	85.13
Lack of information	1 if constrained by lack of information, 0 if otherwise	77.30	74.20
Lack of capital	1 if constrained by lack of capital, 0 if otherwise	72.34	69.45
Explanatory variables
Age	Age of HH head (years)	43.83	12.68
Gender	1 if HH head is male, 0 if female	0.61	0.49
Educational level	Years of education of HH head	59.96	52.55
Farming experience	Years of household experience in farming	10.54	4.71
Access to extension	1 if HH has access to extension, 0 if otherwise	0.33	0.47
Non-farm income	1 if HH engages in any off-farm activity	0.38	0.49
Credit	1 if HH has access to credit, 0 if otherwise	0.45	0.50
Land reform	1 if beneficiary of land reform, 0 if non-beneficiary	0.32	0.47
Access to training	1 if HH has access to agricultural training, 0 if otherwise	0.48	0.50
ICT_Radio	1 if HH has access to information through radio, 0 if otherwise	0.48	0.50
ICT_Mobile phone	1 if HH has access to information through mobile phone, 0 if otherwise	0.36	0.48
ICT_Television	1 if HH has access to information through television, 0 if otherwise	0.25	0.43
ICT_Neighbour	1 if HH has access to information through neighbour, 0 if otherwise	0.50	0.50
ICT_Family members	1 if HH has access to information via family members, 0 if otherwise	0.36	0.48
Access to irrigation	1 if HH has access to irrigation, 0 if otherwise	0.57	0.50

HH: household; ICT: information and communication technologies.

According to Mulwa et al. (2017) and Khanal et al. (2019), one of the challenges in choosing adaptations in smallholder farming is the proper identification of true adaptation strategies. This is the case because many changes in agricultural management and technology vary across countries (Khanal et al., 2019). As opined by Below et al. (2012) and Deressa et al. (2009), farmers practise different adaptation strategies depending on the varying climatic, social, economic and institutional factors. CC adaptation strategies employed by the smallholder farmers to mitigate the effect of CC in the study area are presented in Table 2. These adaptation strategies include mulching, varying planting date, soil and water conservation, use of improved planting materials, tree planting and planting early maturing varieties. Others include reducing the livestock number, migration and insurance.

Table 2.

The distribution of climate change adaptation strategies used by the farmers.

Adaptation strategies	Percentage
Insurance	25
Varying planting date	49
Mulching	36
Improved variety	56
Reduce livestock number	50
Tree planting	50
Soil and water conservation	54
Early maturing variety	45
Migration	52

In response to the question on their experiences of high temperatures, the majority of farmers (66%) experienced high temperatures, as indicated in Figure 2. This is in agreement with the information from the South Africa Weather Services (SAWS) in Figure 3, which shows that the temperatures in the four provinces have been very high in the last 10 years. These results show that the majority of farmers in the study area correctly perceived CC in terms of temperature rise. This could mean that the farmers are more likely to adapt to CC based on their experience (Singh et al., 2019).

Figure 2.

High temperatures experienced by farmers.

Figure 3.

Annual maximum temperatures in the study area.

In the case of rainfall, as indicated in Figure 4, the majority of smallholder farmers (63.3%) in the study area confirmed having experienced too much rainfall in the past 10 years. This was supported by the SAWS results presented in Figure 5, showing the mean annual rainfall. The SAWS data shows that most parts of KwaZulu-Natal, North West and Free State experienced high volumes of precipitation in the past 10 years. The farmers perceived the floods and rainfall correctly, as also found by Bhushan et al. (2018).

Figure 4.

High rainfall experienced by farmers.

Figure 5.

Mean annual rainfall in the study area.

Figure 6.

Drought experienced by farmers.

A high percentage of farmers have experienced more drought than ever in the past 10 years. This can be confirmed in Figure 5, which shows rainfall patterns of the study area. The experience of drought impact among smallholder farmers is expected to positively influence the adaptation to CC by farmers as opined by Singh et al. (2019). These results agree with those of SAWS, which shows the rainfall patterns in the four provinces of the study area in the past 10 years. These results are in agreement with the results from the studies by Bhushan et al. (2018) and Singh et al. (2019), which also show that a vast majority of smallholder farmers correctly perceived the impact of droughts on their agricultural production.

Therefore, the LRBs in the study area face high temperature and high rainfall as well as recurrent drought, thus making planning for CC adaptation strategies complex.

As posited by Kichamu et al. (2018) and Taneja et al. (2019), smallholder famers face several challenges in adopting CC adaptation strategies such as inadequate technical knowledge, low financial resources and inadequate land size. The results of this study show constraining factors to the farmer’s adaptive capacity in Table 3. However, this table also shows that about 90% of the farmers indicated a lack of climate information as a major constraint to the adoption of CC adaptation strategies. In the same vein, 74% and 77% of the farmers indicated a lack of capital and knowledge of agricultural production, respectively, as important constraints to the adoption of adaptation strategies. The results of this study coincide with that of Asfaw et al. (2019) who posited that the major challenges of smallholder farmers to adapt to changing climate were financial constraint, a lack of affordable technologies, a lack of knowledge, limited access to reliable information and early warning. This study is also in line with the studies of Kichamu et al. (2018) regarding the understanding of farmers’ perceptions of risks associated with CC and the adaptation strategies among smallholder farmers in Eastern Kenya.

Table 3.

Distribution of constraints to adaptation by smallholder farmers.

S/no	Reasons for not adopting adaptation	Yes (%)	No (%)
1	Lack of information	89.20	10.80
2	Lack of capital	73.90	26.10
3	Lack of knowledge	77.30	22.70
4	Shortage of farming land	60.80	39.20
5	Not observing the climate-related problems	63.10	36.90
6	Giving less emphasis to climate change problems	68.80	31.30

Participating farmers were asked to state the reasons for not adopting adaptation strategies. The results were summarized and are presented in Figure 7. The top three constraints were a lack of information, a lack of capital and a lack of knowledge, which are crucial for adaptation although all of them were significant constraints. Despite the fact that earlier in this study the farmers showed correct perceptions of CC, they have however stated that they lack knowledge and information about CC, which has hindered their adaptation strategies. Research has also shown that smallholder farmers’ adaptation strategies are often constrained by their lack of knowledge, information and financial constraints (Deressa et al., 2010; Ojo and Baiyegunhi, 2019; Ojo et al., 2019). The three dominant constraints to the adoption of CC adaptation strategies were used as the outcome variables in the MVP model to ascertain the determinants of constraints to the adoption of CC adaptation strategies among the smallholder farmers in the study area.

Figure 7.

Constraints to adaptation experienced by farmers.

The results of the correlation matrix according to the MVP regression are reported in Table 4. The results support the use of the MVP to study the relationship among the dominant constraints to adopting CC adaptation strategies (lack of CC knowledge, lack of CC information and lack of capital). As indicated in Table 4, the likelihood ratio test (Chi² (3) = 162.57; P > 0.0293) of the independence of the error terms in the different constraint equations was rejected. Therefore, the study accepted an alternative hypothesis of interdependence among the different constraints to the adoption of CC adaptation strategies, the results of which consequently justified the use of the MVP model. All the pair-wise coefficients were positively correlated, demonstrating complementarity among the constraints. The results show that the joint probability of being constrained in the adoption of CC adaptation strategies was 49%, while not being constrained was 18%. The linear predictions of the results show that the likelihood of being constrained by a lack of CC knowledge, a lack CC information and a lack of capital are 46%, 34% and 54%, respectively.

Table 4.

Correlation matrix of the constraints to the adoption of adaptation strategies obtained by means of the MVP model.

	Lack of knowledge	Lack of information	Lack of capital
Lack of knowledge		0.4325 (0.189)**	0.154 (0.091)**
Lack of information			0.256 (0.229)
Likelihood ratio test	Chi² (3) =	(Chi²) 162.57
P-value	0.0293**
Joint probability (success)	49.047
Joint probability (failure)	18.760
Linear predictions
Lack of knowledge	0.457
Lack of information	0.337
Lack of capital	0.540

MVP: multivariate probit.

represents significance level at 5%.

The results in Table 5 show that age of the household is statistically significant and inversely related only to a lack of knowledge of farmers about CC. This implies that, as the age of a farmer increases, the probability of being constrained by knowledge of CC reduces. This could be attributed to the fact that the farming experience of older farmers might possibly increase their propensity to easily perceive the impact of CC compared with their less experienced counterparts. In the same vein, the results revealed that female-headed households have a higher likelihood of being constrained in adoption of CC adaptation strategies due to a lack of knowledge of CC compared with male-headed households. This could be ascribed to the general situation in South African communities where women lack access to productive resources, especially land for farming. Women who have access to land obtained it through their husbands; hence they are most likely to be constrained in the adoption of CC adaptation strategies. The results of this study conform to that of Awotide (2012).

Table 5.

Estimates for the determinants of constraints to the adoption of climate change adaptation strategies.

	Lack of knowledge			Lack of information			Lack of capital
	Coef.	Std. Err.	P-value	Coef.	Std. Err.	P-value	Coef.	Std. Err.	P-value
Age	−0.023	0.011	0.036**	0.012	0.017	0.495	−0.017	0.012	0.152
Sex of the household head	−0.447	0.244	0.067***	−0.134	0.416	0.748	0.278	0.274	0.311
Access to extension service	−0.037	0.384	0.922	−2.138	0.557	0.000***	−1.677	0.467	0.000***
Non-farm activity	−0.596	0.303	0.049**	−1.529	0.508	0.003***	0.253	0.337	0.454
Education status	0.007	0.080	0.928	0.187	0.133	0.159	0.136	0.087	0.119
Farming experience	−0.017	0.026	0.516	−0.045	0.043	0.298	−0.025	0.030	0.400
Susceptibility	1.130	0.265	0.000*	−0.117	0.422	0.782	−0.266	0.281	0.344
Owned TV_Comm	−0.495	0.338	0.144	0.949	0.583	0.104	−0.564	0.360	0.117
Owned Radio_Comm	0.183	0.279	0.512	−1.217	0.477	0.011**	0.789	0.286	0.006***
Membership in FBO	−0.845	0.277	0.002*	−0.397	0.474	0.403	0.140	0.299	0.639
Access to credit	0.312	0.321	0.331	−0.861	0.586	0.142	−0.548	0.355	0.123
Trainings	−0.123	0.287	0.669	−1.438	0.403	0.000***	0.980	−0.283	0.001***
Constant	0.381	0.636	0.549	−3.560	1.119	0.001***	−0.083	0.656	0.899

, ** and *** represent 10%, 5% and 1% significance level, respectively

FBO: farmer-based organization; Owned TV_Comm and Owned Radio_Comm: ownership of television and radio for communication, respectively.

Institutional and policy variables such as agricultural extension services are important factors in agricultural information dissemination to the farmers (Danso-Abbeam and Baiyegunhi, 2017). Extension visits, measured by the number of visits in a production year, are hypothesized to positively influence the adoption of CC adaptation strategies because farmers gain better access to information from extension agents who help link farmer groups to climate information (Anang and Amikuzuno, 2015; Muhongayirea et al., 2013). Access to extension services is statistically significant and positively related to both a lack of CC information and a lack of capital constraint conditions. This implies that the probability of being constrained by a lack of CC information and capital reduces when farmers have contact with extension agents. This is in consonance with Olomola and Gyimah-Brempong (2014) who posited that contact with an extension agent and the presence of collateral security positively and significantly affect the likelihood of farmers’ access to capital. Information received by farmers via agricultural extension agents facilitates their decision on how and when to use CC adaptation strategies such as improved seeds, soil and water conservation, conservation agriculture and changing of cropping calendar. This is in line with the studies of Deressa et al. (2009), Bryan et al. (2011) and Khanal et al. (2019) who indicated a positive and significant relationship between extension access and farmers’ adoption of CC adaptation strategies.

Many empirical studies (Asfaw et al., 2018; Below et al., 2012) had documented that supply-side policy variables, such as membership in a farmer-based organization (FBO), have significant positive effects on farmers’ propensity to adopt a number of adaptation strategies. The coefficient of membership in FBOs is negatively signed and statistically significant in influencing the constraint condition of a lack of knowledge of CC. This implies that the involvement of farmers in FBOs reduces the probablity of being constrained by a lack of CC knowledge. This could be attributed to that fact that farmers share information on farming practices, markets and other production-related issues in groups that enhance their skills and knowledge of farming (Ahmed and Melesse, 2018). This is consistent with the studies of Ghimire and Huang (2015), Baiyegunhi et al. (2018) and Ojo et al. (2019) suggesting that information about new technologies is better shared among farmers in associations.

The estimates related to policy and institutional variables indicate that farmers who had training, perhaps through demonstration farms, are more likely not to lack information about the effect of CC on agricultural production. This is evident from the result of this study which shows that training is negatively signed and statistically significant with being constrained by both a lack of CC information and a lack of capital. Demonstration farms increase farmers’ awareness and knowledge of CC and the ways of mitigating its impact as they are offered opportunities to engage in on-farm trials which improve their farm management skills. This is in line with Danso-Abbeam and Baiyegunhi (2017) who, in their study, posited that farmers who received other agricultural-related training, such as farm financial management, had a higher propensity to adopt fertilizer and insecticides than farmers who did not receive such training.

According to Deressa et al. (2009), farm and non-farm income represent wealth. It is empirically hypothesized that the adoption of agricultural technologies requires sufficient financial wellbeing (Knowler and Bradshaw, 2007). The adoption of CC adapatation strategies is not constrained by a lack of knowledge of and information about CC among farmers who are involved in off-farm activities. This is evident from the negative and statistically significant influence of off-farm activities on a lack of knowledge of and information about CC. It is not suprising, as farmers with alternative secondary sources of income are in a better position to invest in innovative farm technologies; they can afford to plant trees on the limited available land; they can afford the cost of irrigation; and they use agronomic practices such as soil conservation and different crop varieties. This is in line with the study of Ojo and Baiyegunhi (2019) who also found a positive relationship between non-farm income and the adoption of CC adaptation strategies among rice farmers in south-west Nigeria.

Farmers who have access to climate information or have more information have a higher probability of more adaptation (Thi and Chaovanapoonphol, 2014). Access to weather information is also vital in helping farmers to plan ahead against any unexpected outcome on their farms in an effort to reduce shock effects (Oyekale and Oladele, 2012). Access to information and communication technologies (ICT) through the use of radio is negatively signed and statistically significant in constraining smallholder farmers to adopt CC adaptation strategies in the study area. This is attributed to the fact that the use of ICT tools in general and mobile phones in particular can help tackle market failures that smallholder farmers face due to a lack of access to market information (Ndambiriet al., 2014; Okello et al., 2012). The availability of better climate and agricultural information helps farmers make comparative decisions among alternative crop management practices, and hence choose the ones that enable them to cope better with changes in climate (Baethgen et al., 2003; Jones, 2003).

The adoption of CC adapatation strategies is constrained by a lack of information about CC among farmers that are susceptible to its impact. This is evident from the positive and statistically significant influence of susceptibility on a lack of information about CC. The implication of the result is that smallholder farmers who are constrained as a result of a lack of information are likely to be vulnerable to the impact of drought in the study area. This is in consonance with the study of Hann et al. (2011) who opined that early warning systems and community preparedness plans may help communities prepare for extreme weather events such as drought. Seasonal weather forecasts distributed through local farming associations may help farmers time their plantings and prevent diversion of scarce water resources for irrigation.

Conclusion and policy recommendations

CC adaptation has emerged as one of the most important responses to environmental development challenges of the 21st century. The adverse effects of CC on agricultural productivity are on the increase, thus the need for smallholder farmers to devise adaptation measures. However, the limited adaptive capacity of smallholder farmers in sub-Saharan African countries, such as South Africa, mainly positions them to be susceptible to adverse impacts of CC. This study employed both descriptive statistics and the MVP model to estimate factors constraining the adoption of CC adaptation strategies among smallholder farmers (LRBs) in the study area. The result of the study shows that a high percentage of smallholder farmers are constrained in adopting CC adaptation strategies as a result of a lack of knowledge, a lack of information on CC and a lack of capital. About 90% of the farmers indicated a lack of climate information as a major constraint to the adoption of CC adaptation strategies. In the same vein, 74% and 77% of the farmers indicated a lack of capital and a lack of knowledge of agricultural production, respectively, as important constraints to the adoption of adaptation strategies. The likelihood ratio test of independence of the error terms in the different constraint equations was rejected. Therefore, the study accepted an alternative hypothesis of interdependence among the different constraints to the adoption of CC adaptation strategies. Hence, the results justified the use of the MVP model. The empirical results of the MVP model showed that a lack of smallholder farmers’ knowledge of the CC constraint was influenced by their age, gender, off-farm activity, susceptibility and membership in FBOs. Simultaneously, a lack of information about CC was influenced by access to extension agents, non-farm activity, access to radio and agricultural trainings. In the same vein, a lack of capital constraint was influenced by access to extension services and radios, and on-farm demonstrations in the form of training. Thus, to improve the adaptive capacity of smallholder farmers, government and development partners should work together to improve the conditions of access of farmers to adequate CC information and timely agricultural credit, including policy incentives aimed at lowering the stringent conditions of borrowing in the agricultural sector, and to extension agent services, including on-farm training, as these are all important aspects in the adoption of CC adaptation strategies. It could further be recommended that policies that enhance and strengthen institutional support, such as improved access to extension services and farmers’ education, be employed as valuable tools in improving the productivity of smallholder farmers in South Africa.

Footnotes

Funding

The authors received no financial support for the research, authorship, and/or publication of this article.

ORCID iDs

AA Ogundeji

TO Ojo

Author biographies

KT Thinda is a PhD student and the Director, Regional Shared Service Centre, agriculture land reform and rural development, Free State, South Africa.

AA Ogundeji is a Senior Lecturer at the Department of Agricultural Economics, University of the Free State P.O Box 339 Bloemfontein 9300 South Africa.

JA Belle is a Senior Lecturer and Acting Director at Disaster Management Training and Education Centre for Africa at the University of the Free State P.O Box 339 Bloemfontein 9300 South Africa.

TO Ojo is a Post-Doctoral Fellow at Disaster Management Training and Education Centre for Africa at the University of the Free State P.O Box 339 Bloemfontein 9300 South Africa. He is also a Lecturer at the Department of Agricultural Economics, Obafemi Awolowo University, Nigeria.

References

Abdulai

Huffman

(2014) The adoption and impact of soil and water conservation technology: An endogenous switching regression application. Land Economics 90(1): 26–43.

Ahmed

Melesse

(2018) Impact of off-farm activities on technical efficiency: Evidence from maize producers of eastern Ethiopia. Agricultural and Food Economics 6(1): 3.

Ajetomobi

Abidun

Hassan

(2010) Economic impact of climate change on irrigated rice agriculture in Nigeria. In: AAAE Third Conference/AEASA 48th Conference, Cape Town, South Africa, 308-2016-5082, AAAE, pp.1–20.

Anang

Amikuzuno

(2015) Factors influencing pesticide use in smallholder rice production in Northern Ghana. Agriculture, Forestry and Fisheries 4(2): 77–82.

Asfaw

Simane

Bantider

, et al. (2019). Determinants in the adoption of climate change adaptation strategies: Evidence from rainfed-dependent smallholder farmers in north-central Ethiopia (Woleka sub-basin). Environment, Development and Sustainability 21(5): 2535–2565.

Asfaw

Simane

Hassen

, et al. (2018) Variability and time series trend analysis of rainfall and temperature in north-central Ethiopia: A case study in Woleka sub-basin. Weather and Climate Extremes 19: 29–41.

Asfaw

Di Battista

Lipper

(2016) Agricultural technology adoption under climate change in the Sahel: Micro-evidence from Niger. Journal of African Economies 25(5): 637–669.

Awotide

(2012) Assessment of women’s participation in cooperative societies and its determinants in Yewa North local government area of Ogun state, Nigeria. Asian Journal of Agriculture and Rural Development 2: 344–350.

Baethgen

Meinke

Gimenez

(2003) Adaptation of agricultural production systems to climate variability and climate change: Lessons learned and proposed research approach. In: Climate Adaptation.net Conference: Insights and Tools for Adaptation: Learning from Climate Variability, Washington, DC, Island Press, 18–20.

10.

Baiyegunhi

LJS

Hassan

Ortmann

(2018) Impact of integrated striga management (ISM) technology on maize productivity in northern Nigeria: A treatment effect approach. African Journal of Science, Technology, Innovation and Development 10(3): 335–344.

11.

Below

Mutabazi

Kirschke

, et al. (2012) Can farmers’ adaptation to climate change be explained by socio-economic household-level variables? Global Environmental Change 22(1): 223–235.

12.

Bhushan

Swami

Sharma

, et al. (2018) Indigenous knowledge management for disaster mitigation and sustainable development in the eco community villages of India. SSRN Electronic Journal. DOI: 10.2139/ssrn.3297971.

13.

Boko

Niang

Nyong

, et al. (2007) Africa climate change 2007: Impacts, adaptation and vulnerability. In: Parry

Canziani

Palutikof

, et al. (eds) Contribution of Working Group II to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change. Cambridge: Cambridge University Press, 433–467.

14.

Bryan

Ringler

Okoba

, et al. (2011) Coping with climate variability and adapting to climate change in Kenya: Household and community strategies and determinants. Washington, DC: IFPRI.

15.

Chambwera

Stage

(2010) Climate Change Adaptation in Developing Countries: Issues and Perspectives for Economic Analysis. Available at: www.iied.org/pubs/display.php?o=15517llED

16.

Cochran

(1977) Sampling Techniques. 3rd ed. New York: John Wiley & Sons.

17.

Crossman

Bryan

Summers

(2012) Identifying priority areas for reducing species vulnerability to climate change. Diversity and Distributions 18(1): 60–72.

18.

Danso-Abbeam

Baiyegunhi

(2017) Adoption of agrochemical management practices among smallholder cocoa farmers in Ghana. African Journal of Science, Technology, Innovation and Development 9(6): 717–728.

19.

Deressa

Hassan

Ringler

, et al. (2009) Determinants of farmers’ choice of adaptation methods to climate change in the Nile Basin of Ethiopia. Global Environmental Change 19(2): 248–255.

20.

Deressa

Ringler

Hassan

(2010) Factors affecting the choices of coping strategies for climate extremes. The Case of Farmers in the Nile Basin of Ethiopia, IFPRI Discussion Paper, 1032–1045. Washington, DC: International Food Policy Research Institute.

21.

Fisher-Vanden

Wing

Lanzi

, et al. (2011) Modeling climate change adaptation: Challenges, recent developments and future directions. Mimeographed paper, Boston University.

22.

Ghimire

Huang

(2015) Household wealth and adoption of improved maize varieties in Nepal: A double-hurdle approach. Food Security 7(6): 1321–1335.

23.

Greene

(2000) Econometric Analysis. 4th ed. Upper Saddle River: Prentice-Hall.

24.

Guodaar

Asante

Eshun

, et al. (2020) How do climate change adaptation strategies result in unintended maladaptive outcomes? Perspectives of tomato farmers. International Journal of Vegetable Science 26(1): 15–31.

25.

Hanna

Kjellstrom

Bennett

, et al. (2011) Climate change and rising heat: Population health implications for working people in Australia. Asia Pacific Journal of Public Health 23(Suppl. 2): 14–26.

26.

Hassan

Nhemachena

(2008) Determinants of African farmers’ strategies for adapting to climate change: Multinomial choice analysis. African Journal of Agricultural and Resource Economics 2: 83–104.

27.

IPCC (Intergovernmental Panel on Climate Change) (2001) Climate Change: The Scientific Basis [online]. Available at: http://www.ipcc.ch (accessed June 2006).

28.

Jones

(2003) The European Miracle: Environments, Economies and Geopolitics in the History of Europe and Asia. 3rd ed. Cambridge: Cambridge University Press.

29.

Kamara

Conteh

Rhodes

, et al. (2019) The relevance of smallholder farming to African agricultural growth and development. African Journal of Food, Agriculture, Nutrition and Development 19(1): 14043–14065.

30.

Kerr

Kangmennaang

Dakishoni

, et al. (2019). Participatory agroecological research on climate change adaptation improves smallholder farmer household food security and dietary diversity in Malawi. Agriculture, Ecosystems & Environment 279: 109–121.

31.

Khan

Lei

Shah

, et al. (2020) Farm households’ risk perception, attitude and adaptation strategies in dealing with climate change: Promise and perils from rural Pakistan. Land Use Policy 91: 104395.

32.

Khanal

Wilson

Hoang

, et al. (2019) Impact of community-based organizations on climate change adaptation in agriculture: Empirical evidence from Nepal. Environment, Development and Sustainability 21(2): 621–635.

33.

Kichamu

Ziro

Palaniappan

, et al. (2018) Climate change perceptions and adaptations of smallholder farmers in Eastern Kenya. Environment, Development and Sustainability 20(6): 2663–2680.

34.

Knowler

Bradshaw

(2007) Farmers’ adoption of conservation agriculture: A review and synthesis of recent research. Food Policy 32(1): 25–48.

35.

Komba

Muchapondwa

(2012) Adaptation to climate change by smallholder farmers in Tanzania. Economic Research Southern Africa (ERSA), Working paper 299. Environmental-Economics Policy Research Unit (EPRU), School of Economics, University of Cape Town, Rondebosch.

36.

Kurukulasuriya

Mendelsohn

Hassan

, et al. (2006) Will African agriculture survive climate change? The World Bank Economic Review 20(3): 367–388.

37.

Leclère

Jayet

De Noblet-Ducoudré

(2013) Farm-level autonomous adaptation of European agricultural supply to climate change. Ecological Economics 87: 1–14.

38.

Lin

CTJ

Jensen

Yen

(2005) Awareness of foodborne pathogens among US consumers. Food Quality and Preference 16(5): 401–412.

39.

Mogomotsi

Sekelemani

Mogomotsi

(2020) Climate change adaptation strategies of small-scale farmers in Ngamiland East, Botswana. Climatic Change 1–20.

40.

Mugi-Ngenga

Mucheru-Muna

Mugwe

, et al. (2016) Household’s socio-economic factors influencing the level of adaptation to climate variability in the dry zones of Eastern Kenya. Journal of Rural Studies 43: 49–60.

41.

Muhongayire

Hitayezu

Mbatia

, et al. (2013) Determinants of farmers’ participation in formal credit markets in rural Rwanda. Journal of Agricultural Sciences 4(2): 87–94.

42.

Mulwa

Marenya

Kassie

(2017) Response to climate risks among smallholder farmers in Malawi: A multivariate probit assessment of the role of information, household demographics, and farm characteristics. Climate Risk Management 16: 208–221.

43.

Ncube

(2018) Constraints to smallholder agricultural production in the Western Cape, South Africa. Physics and Chemistry of the Earth, Parts A/B/C 106: 89–96.

44.

Ndambiri

Mwisukha

Muniu

(2014) Physical factors that influence training and performance of female athletes in sports in teacher training colleges in central region of Kenya. International Journal of Sports Science 4(3): 79–83.

45.

Nunfam

Oosthuizen

Adusei-Asante

, et al. (2019) Perceptions of climate change and occupational heat stress risks and adaptation strategies of mining workers in Ghana. Science of the Total Environment 657: 365–378.

46.

Ojo

Baiyegunhi

LJS

(2019) Determinants of climate change adaptation strategies and its impact on the net farm income of rice farmers in south-west Nigeria. Land Use Policy 95: 103946.

47.

Ojo

Baiyegunhi

LJS

Salami

(2019) Impact of credit demand on the productivity of rice farmers in South West Nigeria. Journal of Economics and Behavioral Studies 11(1): 166–180.

48.

Okello

Kirui

Njiraini

, et al. (2012) Drivers of use of information and communication technologies by farm households: The case of smallholder farmers in Kenya. Journal of Agricultural Science 4(2): 111.

49.

Olomola

Gyimah-Brempong

(2014) Loan demand and rationing among small-scale farmers in Nigeria. IFPRI Discussion Paper, 01403. Available at SSRN: https://ssrn.com/abstract=2545468.

50.

Omoyo

Wakhungu

Oteng’i

(2015) Effects of climate variability on maize yield in the arid and semi arid lands of lower eastern Kenya. Agriculture & Food Security 4(1): 8.

51.

Ortmann

King

(2007) Agricultural cooperatives II: Can they facilitate access of small-scale farmers in South Africa to input and product markets? Agrekon 46(2): 219–244.

52.

Oyekale

Oladele

(2012) Determinants of climate change adaptation among cocoa farmers in southwest Nigeria. ARPN Journal of Science and Technology 2(1): 154–168.

53.

Phuong

LTH

Biesbroek

Sen

LTH

, et al. (2018) Understanding smallholder farmers’ capacity to respond to climate change in a coastal community in Central Vietnam. Climate and Development 10(8): 701–716.

54.

Seo

(2011) An analysis of public adaptation to climate change using agricultural water schemes in South America. Ecological Economics 70(4): 825–834.

55.

Singh

Naresh

Vivek

, et al. (2019) Conservation agriculture effects on aggregates carbon storage potential and soil microbial community dynamics in the face of climate change under semi-arid conditions: A review. Journal of Pharmacognosy and Phytochemistry 8(2): 59–69.

56.

Taneja

Pal

Joshi

, et al. (2019) Farmers’ preferences for climate-smart agriculture: An assessment in the Indo-Gangetic Plain. In: Pal

Kishore

Joshi

, et al. (eds) Climate Smart Agriculture in South Asia: Technologies, Policies and Institution. Singapore: Springer, 91–111.

57.

Teklewold

Kassie

Shiferaw

, et al. (2013) Cropping system diversification, conservation tillage and modern seed adoption in Ethiopia: Impacts on household income, agrochemical use and demand for labor. Ecological Economics 93: 85–93.

58.

Tewksbury

Scheufele

(2009) News framing theory and research. In: Bryant

Oliver

(eds) Media effects – Advances in Theory and Research. 3rd ed. New York: Routledge, 33–49.

59.

Thi

Chaovanapoonphol

(2014) An evaluation of adaptation options to climate pressure on highland Robusta coffee production, Daklak Province, Vietnam. World Journal of Agriculture Research 2(5): 205–215.

60.

Vignola

Harvey

Bautista-Solis

, et al. (2015) Ecosystem-based adaptation for smallholder farmers: Definitions, opportunities and constraints. Agriculture, Ecosystems & Environment 211: 126–132.

61.

Williams

Crespo

Abu

(2019) Adapting to changing climate through improving adaptive capacity at the local level: The case of smallholder horticultural producers in Ghana. Climate Risk Management 23: 124–135.

62.

Wooldridge

(2003) Cluster-sample methods in applied econometrics. American Economic Review 93(2): 133–138.

63.

Zellner

Huang

(1962) Further properties of efficient estimators for seemingly unrelated regression equations. International Economic Review 3(3): 300–313.