EFFECT OF SELECTED INSECTICIDE ON WHITEFLY (Bemisia tabaci) INFESTING BRINJAL PLANTS 686 © 2025 AESS Publications. All Rights Reserved. Income inequality in phezukomkhono mlimi farming households: A case study of Nkomazi Local Municipality, South Africa Themba Andries Samboa James W Oguttub Tulisiwe P Mbombo-Dwebac a,b,cDepartment of Agriculture and Animal Health, University of South Africa, Florida, South Africa.  sambota9@gmail.com (Corresponding author) Article History ABSTRACT Received: 9 June 2025 Revised: 22 September 2025 Accepted: 20 November 2025 Published: 28 November 2025 Keywords Farm Gini coefficient Income distribution Lorenz curve Poverty Wealth. The South African government has developed several policies and intervention programmes to address food insecurity, poverty, and income inequality. One such programme is the Phezukomkhono Mlimi (PKM) initiative. However, empirical evidence demonstrating the impact of government programmes like the PKM on poverty and income inequality remains limited. This study examined income inequality and its determinants. A structured questionnaire was used to collect data from 230 respondents. The data were analyzed using descriptive statistics, the Gini coefficient, and a quantile regression model. The results indicated a high level of income inequality, with a Gini coefficient of 0.48. The majority (60.87%) of households derived their income from farming activities. Other sources of income included retirement funds (10.43%) and salaries (7.39%). The findings suggested a high unemployment rate within the study population. Age, education, household size, and farm size were identified as determinants of income inequality in the study area. Although the income inequality observed was lower than provincial and national levels, government intervention through the PKM programme has not significantly impacted income inequality in the study area. Considering this, policy interventions to improve equitable access to education, family planning, and farm size are recommended. Contribution/Originality: There is a dearth of studies that investigate income inequality and its determinants among rural farming households in South Africa. This is the first study to quantify the level of income inequality among PKM farming households and to identify predictors of income inequality among farming communities in a rural setting. DOI: 10.55493/5005.v15i4.5774 ISSN(P): 2304-1455/ ISSN(E): 2224-4433 How to cite: Sambo, T. A., Oguttu, J. W., & Mbombo-Dweba, T. P. (2025). Income inequality in phezukomkhono mlimi farming households: A case study of Nkomazi Local Municipality, South Africa. Asian Journal of Agriculture and Rural Development, 15(4), 686–695. 10.55493/5005.v15i4.5774 © 2025 Asian Economic and Social Society. All rights reserved. 1. INTRODUCTION Goal 1 of the Sustainable Development Goals of the United Nations is to eliminate poverty in all its forms everywhere by 2030 (Rahman, 2017; United Nations, 2015). However, many rural areas, especially in developing countries, remain vulnerable to income inequality and poverty, and as a result, there is a need to help rural farmers diversify and sustain production for increased social, economic, and environmental benefits (Akpan, Okon, Udo, & Akpakaden, 2020; Kadigi, 2021; Sambo, 2021). This is especially crucial for smallholder farmers in rural areas because it can enhance their food supply, income, and wealth (Akpan et al., 2020). The significant gap in income inequality between rural and urban populations in developing countries has been attributed to the dependence of rural societies on low-paid agricultural activities (Waris, Khan, Urrahman, Hronec, & Suplata, 2023). Inequality is responsible for Asian Journal of Agriculture and Rural Development Volume 15, Issue 4 (2025): 686-695 https://orcid.org/0000-0002-2966-4244 https://orcid.org/0000-0001-6810-4437 https://orcid.org/0000-0001-8395-3322 mailto:sambota9@gmail.com https://doi.org/10.55493/5005.v15i4.5774 Asian Journal of Agriculture and Rural Development, 15(4) 2025: 686-695 687 © 2025 AESS Publications. All Rights Reserved. several problems such as uneasiness, violence, and corruption. Given the high income inequality in South Africa, the government has been preoccupied with the fight against poverty and inequality. As a result, the government has developed several policies and intervention programmes to help combat poverty and its impact on society, especially the formally disadvantaged groups. One such programme is the Phezukomkhono Mlimi (PKM) programme, introduced in 2005 (Sambo, Oguttu, & Mbombo-Dweba, 2025; Sambo, 2021). Available evidence on the PKM programme shows that the programme has had an impact on food security among the participating households. Particularly, it has been able to decrease the severity of food insecurity (Sambo, 2021). Based on this, it is hypothesized that the PKM programme has the potential to ameliorate inequality in the local population. However, empirical evidence showing the impact of government support programmes on poverty and income inequality is lacking, more so in the context of programmes like the PKM. Such evidence is needed by policymakers and other stakeholders to formulate suitable policies, plans, and strategies to achieve sustainable development in these areas. It is worth noting that the levels of income inequality in farming households may vary significantly between farming households, the size of agricultural land, and the farmers’ economic characteristics (Akpan et al., 2020; Rahman, 2017). To the best of our understanding, these characteristics have not received adequate attention among scholars. Furthermore, studies that employ cross-sectional data while focusing on small-scale sites and locations, such as the Nkomazi Local Municipality (NKLM), are lacking. In addition, where cross-sectional data is applied, most researchers tend to ignore the effects of variation in household, farming, and economic characteristics as well as the existing transforming structures and processes. Moreover, available evidence suggests that extremely skewed income inequality is one of the major sources of, and positively correlates with, poverty (Akpan et al., 2020). Since the South African government views agriculture as one of the major instruments in fighting poverty, it is important to understand the extent of income disparity and poverty levels as well as the factors that influence the manifestation of poverty among farmers who benefit from the PKM programme. Unfortunately, there is no evidence of studies that have investigated income inequality in PKM farming households in the NKLM. Hence, this study investigates the nature and determinants of income inequality in PKM farming households in the NKLM, South Africa. 2. MATERIALS AND METHODS 2.1. Study Area The study was carried out in the NKLM, the smallest of the four municipalities within the Ehlanzeni District Municipality of the Mpumalanga Province, South Africa. The NKLM is bordered by the City of Mbombela Local Municipality (to the west), Kruger National Park (to the north), Mozambique (to the east), and the Kingdom of Eswatini (to the south). It is linked with Mozambique via the national road, the N4, while two provincial roads, the R570 and R571, connect it with the Kingdom of Eswatini. The NKLM covers an area of 4,785 km² and is inhabited by approximately 591,928 people (Statistics South Africa, 2022). Agriculture, mining, and tourism are the major contributors to the economy of the NKLM. The number of households practicing agriculture and the unemployment level are also high in the study area (Statistics South Africa, 2011). 2.2. The Target Population A list of farming households (N=543) that benefited from the PKM programme during the 2018/19 production season (i.e., April 2018 to March 2019) in the study area was requested from the Department of Agriculture, Rural Development, Land and Environmental Affairs (DARDLEA). The 543 farming households constituted the target population. 2.3. Sampling Method and Sample Size A random sample of 230 respondents was drawn from the sample frame of 543 farming households that benefited from the PKM programme using simple random sampling. The 230 respondents were determined using the formula described by Adam (2020) for determining sample size from a finite population, as described below. 𝑛 = 𝑁 1+𝑁(𝑒2) (1) Where: n = sample size; N = target population; e = margin error set at 0.05. Thus, 𝑛 = 543 1+543(0.052) = 543 1+543(0.0025) = 543 1+1.36 = 543 2.36 = 230.08 ≈ 230 2.4. Study Design and Data Collection A cross-sectional, questionnaire-based study design was adopted to achieve the objectives of this study. Quantitative data were collected between 1 February and 24 March 2020 using a structured questionnaire, which was administered during face-to-face interviews by four trained enumerators. The questionnaire solicited demographic and socio-economic information from the respondents. 2.5. Data Analysis The data were inputted into Microsoft Excel and subsequently transferred to the Statistical Package for the Social Sciences (SPSS version 28) for further analysis. Descriptive statistics, the Gini coefficient, and the quantile regression model were employed to analyze the data. 2.5.1. Descriptive Statistics The socio-economic characteristics of the respondents were summarized and presented as tables and figures. Asian Journal of Agriculture and Rural Development, 15(4) 2025: 686-695 688 © 2025 AESS Publications. All Rights Reserved. 2.5.2. Gini Coefficient The Gini coefficient was used to measure income inequality among the farming households in the study area. The Gini coefficient is a tool widely used to measure inequality, particularly income or wealth distribution. It helps to understand how income or wealth is distributed across a population, providing insight into the level of fairness or disparity in a society. The closer the Gini coefficient is to 0, the more equal the distribution of wealth, and the closer it is to 1, the more unequal the distribution of wealth (Bellu & Liberati, 2006). According to Gini (2005), the Gini coefficient can be presented as the area between the line of equality and the Lorenz curve divided by the total area under the line of equality in the graphical presentation of the Lorenz curve. Furthermore, it can be expressed or specified mathematically using the following equation. 𝐺 = 1 − ∑ (𝑊𝑖−𝑊𝑖−1)(𝑍𝑖+𝑍𝑖−1) 𝑛 𝑖=1 (2) Where, G is the Gini coefficient, Wi is the cumulated proportion of the population variable, for i = 0..., n, with W0 = 0, Wn = 1 and Zi is the cumulated proportion of the income variable, for i = 0....n, with Z0 = 0, Zn = 1. Wi and Zi should be arranged in an increasing order such that Wi > Wi – 1 and Zi > Zi – 1, respectively. 2.5.3. Quantile Regression Model The quantile regression model was employed to identify the determinants of income inequality among the PKM farming households in the study area. Quantile regression is an alternative to linear regression when the assumptions for linear regression are not met. This statistical modeling method has recently gained significant attention in income inequality and poverty analysis studies (Debebe & Zekarias, 2020; Garza-Rodriguez, Ayala-Diaz, Coronado-Saucedo, Garza-Garza, & Ovando-Martinez, 2021). This method provides a more detailed view of how variables affect various parts of the population (Cook & Manning, 2013). Quantile regression is particularly useful because it describes how explanatory variables (or predictors) affect different parts of the distribution of the dependent variable in this case, income. Instead of just indicating how a variable impacts the average (as is the case in traditional regression), it shows how these variables influence the outcomes at different quantiles such as the 10th percentile and the 50th percentile (median) (Xu, 2023). This is useful because the effect of the predictor variables can vary across the income spectrum, providing a more complete picture of the data and capturing subtle distributional effects that would otherwise be missed in standard analysis (Cook & Manning, 2013). Based on the equation described by Garza-Rodriguez et al. (2021), the quantile regression model is specified as follows: Qln(𝑌𝑖|𝑋𝑖) = ∑ 𝑞|𝑌𝑖 − 𝑋′𝑖𝛽𝑞| 𝑁 𝑖:𝑌𝑖≥𝑋′𝛽 + ∑ (1 − 𝑞)|𝑖 − 𝑋′𝑖𝛽𝑞| 𝑁 𝑖:𝑌𝑖≤𝑋′𝛽 (3) Q(ln(Yi)∣Xi): This represents the quantile of the log of total income per adult equivalent (the dependent variable, Yi) for the ith household, conditional on the values of the explanatory variables (X). Where: Xi: Represents a column vector of realizations for kth explanatory variables (predictors such as age, education, farming experience, or household size, etc.) that may influence the household's income. β: A column vector of unknown parameters corresponding to the explanatory variables, which the model seeks to estimate. These parameters represent the effect of each predictor at the specified quantile of the income distribution. 0≤q≤1: The quantile of interest. For instance, q=0.5 would represent the median, while q=0.25 would represent the 25th percentile. The quantile regression was applied to assess relationship between the dependent variable and explanatory variables at the 10th, 25th, 50th, 75th, and 90th quantiles of the income distribution. Quantile regression is well-suited for this purpose as it examines the distribution of income at different points instead of focusing solely on the mean (Bottai et al., 2014). In this case, the quantiles used in regression allow for insight into the factors that influence income at the extremes of the income distribution, in addition to those at quantiles between those extremes (Leeds, 2014). Quantile regression differs from ordinary linear regression in the sense that it does not involve any supposition about the distribution of the regression residuals. Quantile regression is not affected by outliers or skewness in the distribution of the outcome variable, giving more statistical efficacy when outliers exist. Moreover, inference on quantiles can allow transformation of the outcome variable without the glitches experienced in ordinary linear regression (Bottai et al., 2014; Leeds, 2014). 2.5.4. The Conceptual Framework Table 1 graphical representation of the dependent, independent, and presumed relationships with the independent/explanatory variables included in the quantile regression model. Table 1. Description, measurement, and hypotheses of explanatory variables used in the quantile regression model. Variables Type of variable Measurements Hypothesis Age Continuous Years - Years of schooling Continuous Years - Household size Continuous Years + Farming experience Continuous Years - Gender Dummy Male/Female ± Living with disability Dummy Yes/No ± Marital status Dummy Married/Otherwise + Participating in off-farm activities Dummy Yes/No - Farm size Dummy ≤3ha / 4-5ha / 6-8ha / >8ha ± Asian Journal of Agriculture and Rural Development, 15(4) 2025: 686-695 689 © 2025 AESS Publications. All Rights Reserved. 2.6. Ethical Consideration Prior to the commencement of the data collection exercise, approval to conduct this study was obtained from the DARDLEA and the Ethics Committee of the College of Agriculture and Environmental Science at the University of South Africa (Ref #: 2019/CAES_HREC/178). 3. RESULTS 3.1. Socio-Economic Characteristics of the Respondents The socio-economic characteristics of the respondents presented in Table 2 indicate that most of the farmers were above 56 years of age (66.96%) and were females (61.30%). The highest proportion was married (46.52%), followed by single (11.74%), divorced (6.96%), and widowed (34.78%). Slightly more than half (53.48%) had a household size of 3 to 6 people. Up to 13.48% of the respondents had disabilities. Regarding education, just under half (48.70%) of the respondents did not attain Grade 12, and just above one-third (39.13%) had no formal education. Very few had attained Grade 12 (7.83%) and tertiary (4.35%) education. Most respondents had smaller farms of three hectares or less in size (61.74%), and had access to extension services (70.87%) and did not participate in off-farm activities (62.61%). Table 2. Socio-economic characteristics of the respondents (n=230). Variables Frequency (F) Percentage (%) Age in years ≤35 10 4.35 36-45 17 7.39 46-55 49 21.30 56-65 54 23.48 > 65 100 43.48 Gender Male 89 38.70 Female 141 61.30 Persons with disability Yes 31 13.48 No 199 86.52 Marital status Married 107 46.52 Single 27 11.74 Divorced 16 6.96 Widowed 80 34.78 Household size < 3 79 34.35 3-6 123 53.48 > 6 28 12.17 Level of education No formal education 90 39.13 Less than grade 12 112 48.70 Grade 12 18 7.83 Tertiary education 10 4.35 Farming experience <25 years 125 54.65 ≥25 years 105 45.35 Farm Size (Hectares) ≤3 142 61.74 4-5 58 25.22 6-10 21 9.13 >10 9 3.91 Participation in off-farm activities Yes 86 37.39 No 144 62.61 Access to extension services Yes 163 70.87 No 67 29.13 3.2. The Different Sources of Income for the Farming Households that participated in the study Table 3 presents the different sources of income for the respondents. The majority (60.87%) of the households derived their income from farming activities such as sales of farm produce, leasing of land, and leasing of farm Asian Journal of Agriculture and Rural Development, 15(4) 2025: 686-695 690 © 2025 AESS Publications. All Rights Reserved. equipment and machinery. Other sources of income included receiving social grants (56.52%) and receiving pension payouts (10.43%). A mere 7.39% of the respondents earned wages and salaries, suggesting a high unemployment rate in the study area. Only 37.39% of the respondents generated income from other non-farming activities. Table 3. Sources of income of the respondents. 1Source of income Frequency (F) Percentage (%) Farm income 140 60.87 Social grant 130 56.52 Retirement funds 24 10.43 Wages and salaries 17 7.39 Remittances 23 10.00 Other non-farming activities 86 37.39 Note: 1Mutiple responses were possible. 3.3. Income Distribution and the Gini Coefficient of the Study Population Table 4 presents the income distribution of households in the study area. The results indicate that the majority (90.43%) of households had a per capita monthly income of less than R1,500.00, which accounts for 50.11% of the total income. This was followed by an equal proportion (4.35%) of households earning a per capita monthly income of R1,501.00 to R2,500.00 and R2,501.00 to R5,000.00. However, the latter income bracket (R2,501.00 to R5,000.00) contributed a higher percentage (18.22%) to the total income than the former income bracket (R1,501.00 to R2,500.00), which contributed 15.95%. The proportion of households with an annual income of R5,001.00 to R10,000.00 and those earning above R10,000.00 were both 0.43%, but the contribution of the latter income bracket to the total income was twice as high (10.02%) compared to the former (5.69%). Table 4. Income distribution of households (n=230). Per capita income per month Number of households (F) Percentage of distribution (%) Percentage of income contribution (%) Less than R1 500 208 90.43 50.11 R1 501-R2 500 10 4.35 15.95 R2 501-R5 000 10 4.35 18.22 R5 001-R10 000 1 0.43 5.69 Above R10 000 1 0.43 10.02 Total 230 100 100 Table 5 shows the determination of the Gini coefficient based on the formula derived by Gini (2005). The results show that the study population had a Gini coefficient of 0.48, which indicates some degree of inequality. Table 5. Measurement of the Gini coefficient of households. Income distribution Cumulative % of population Cumulative % of income (Wi-Wi-1) (Zi+Zi-1) (Wi-Wi-1)(Zi+Zi-1) 0 0 0 0.00 0 0.00 Less than R1 500 0.98 0.50 0.98 0.50 0.49 R1 501-R2 500 0.99 0.66 0.00 1.16 0.01 R2 501-R5 000 0.99 0.84 0.00 1.50 0.01 R5 001-R10 000 1.00 0.90 0.00 1.74 0.01 Above R10 000 1.00 1.00 0.00 1.90 0.01 𝐺 = 1 − ∑ (𝑊𝑖−𝑊𝑖−1)(𝑍𝑖+𝑍𝑖−1) 𝑛 𝑖=1 =1-0.52=0.48 Total = 0.52 Figure 1 presents the determination of the Gini coefficient based on the Lorenz curve. The Lorenz curve was plotted based on the cumulative percentage of the population (on the X-axis) and the cumulative percentage of income (on the Y-axis). Graphically, the Gini coefficient (0.48) is reflected as the quotient of the area (A), which is between the Lorenz curve and the line of equality, and the total area under the line of equality (A+B). Asian Journal of Agriculture and Rural Development, 15(4) 2025: 686-695 691 © 2025 AESS Publications. All Rights Reserved. Figure 1. Lorenz curve. 3.4. Determinants of Income Inequality Among the Households Table 6 shows the results of the quantile regression analysis. Variables such as age, education, household size, and farm size had a statistically significant effect on income inequality in this study. The direction and magnitude of the effect of the predictor variables on the outcome variable (per capita income) are shown by the coefficients of the quantile regression analysis. The age of the household head had a negative effect on the per capita income of the household in almost all quantiles, but it was significant at the median (50th) and 75th quantiles. Meanwhile, education had a positive impact on the income level of households at all quantiles except the 10th quantile. However, the impact of education was only significant at the 25th quantile. Furthermore, the magnitude of the impact of education increased with the increase in the quantile levels. Household size had a negative and significant effect on income across all quantiles. Its magnitude also increased with the increase in quantile levels. Regarding farm size, a positive effect on household income was observed across all quantiles except the 10th quantile in the first two levels of farm size, namely, in three (3) or fewer hectares and between 4 and 5 hectares. However, their association was only significant at the 25th quantile. The third level of farm size (6-8 hectares) had a positive effect on household income across all quantiles, but was only significant at the 25th and 50th quantiles. The magnitude of the effect increased with higher quantiles in all farm size levels. Asian Journal of Agriculture and Rural Development, 15(4) 2025: 686-695 692 © 2025 AESS Publications. All Rights Reserved. Table 6. Quantile regression analysis of the determinants of income inequality among households. Variables 10th Quantiles 25th Quantiles 50th Quantiles 75th Quantiles 90th Quantiles Coe1 t2 p3 Coe t p Coe t p Coe t p Coe t p Age -0.02 -0.04 0.97 0.48 0.81 0.42 -5.18 -2.16 0.03* -14.02 -2.50 0.01* -5.10 -0.52 0.60 Years of schooling -2.13 -1.56 0.12 6.83 4.18 0.00* 11.72 1.80 0.07 29.28 1.92 0.06 41.19 1.55 0.12 Household Size -53.97 -31.67 0.00* -51.67 -25.38 0.00* -54.49 -6.69 0.00* -88.68 -4.67 0.00* -143.30 -4.33 0.00* Farming experience -0.25 -0.73 0.47 -0.63 -1.56 0.12 -0.84 -0.52 0.60 0.77 0.21 0.84 -9.66 -1.47 0.14 Gender (Male) 19.96 1.72 0.09 -7.20 -0.52 0.60 70.97 1.28 0.20 77.41 0.60 0.55 292.61 1.30 0.20 Disabled (Yes) -1.47 -0.10 0.92 -18.29 -1.02 0.31 -61.17 -0.85 0.40 -72.23 -0.43 0.67 -42.23 -0.14 0.89 Married (Yes) -10.47 -0.91 0.36 6.73 0.49 0.63 22.90 0.42 0.68 171.92 1.34 0.18 64.99 0.29 0.77 Farm size (≤3 ha) -8.56 -0.30 0.76 78.31 2.32 0.02* 237.03 1.76 0.08 436.30 1.39 0.17 572.33 1.04 0.30 Farm size (4-5ha) -1.64 -0.06 0.96 78.28 2.23 0.03* 243.02 1.73 0.08 482.82 1.48 0.14 763.19 1.34 0.18 Farm size (6-8ha) 42.87 1.38 0.17 129.70 3.49 0.00* 469.72 3.16 0.00* 531.33 1.54 0.13 766.34 1.27 0.21 POFA4 (Yes) 10.63 0.99 0.32 -0.46 -0.04 0.97 89.51 1.74 0.08 98.87 0.82 0.41 308.65 1.48 0.14 Note: 1Coefficient. 2t-ratio. 3Statistically significant p-value. 4Participation on off-farm activities. *Statistically significant at p=0.05. Asian Journal of Agriculture and Rural Development, 15(4) 2025: 686-695 693 © 2025 AESS Publications. All Rights Reserved. 4. DISCUSSION Widening income inequality among households is a concern for most governments (Teka, Woldu, & Fre, 2022), policy makers and the research community (Anyiam et al., 2023). In this study, the total Gini index was 0.48, which indicates high inequality in the study population. This is concerning given the association between income inequality and poverty, food insecurity (Abdulsalam, Abdulwahab, Zakari, & Adamu, 2024; Haini, Musa, Wei Loon, & Basir, 2023), and the fact that income inequality is a deterrent to overall economic growth (Sulla, Zikhali, & Cuevas, 2022). Among farming communities in particular, income inequality compromises livelihoods and farming activities (Abdulsalam et al., 2024). With regard to the existence of income inequality in farming households, the results of the current study are consistent with previous studies (Dib, Alamsyah, & Qaim, 2018; Kadigi, 2021; Priscilla, Singh, & Vatta, 2021; Teka et al., 2022). However, in comparison, previous studies reported higher Gini coefficients of more than 50% (Kadigi, 2021; Priscilla et al., 2021; Teka et al., 2022). Furthermore, the Gini coefficient reported here is also lower than the provincial income inequality and the national level. According to the Mpumalanga Provincial Treasury (2024), the Gini coefficient in the Mpumalanga Province, where this study was conducted, was 0.59 in 2022. The observed difference could be attributed to variations in household characteristics across these studies. For example, the present study included only rural farming households, whereas the determination of the provincial Gini coefficient encompassed households with a variety of characteristics. Reports have shown that, in general, rural areas (Sulla et al., 2022) and smaller cities (Alimi, Maré, & Poot, 2018) have lower inequalities when compared to urban areas. While the income inequality registered in this study is considerably lower than the provincial rate, it is still a cause for concern. This is because, according to previous studies (Abdulsalam et al., 2024; Luebker, 2010; Zantsi, Greyling, & Vink, 2019), a Gini coefficient of 0.35 is considered high. It is noteworthy to mention that this Gini coefficient is comparable to the one observed in a study conducted in the Eastern Cape, South Africa, among emerging farmers, where a Gini coefficient of 0.48 was also observed (Zantsi et al., 2019). This confirms the findings of previous studies, which concluded that while income inequality exists in rural areas, it tends to be lower than that observed in urban areas (Sulla et al., 2022). Age had a negative effect on income inequality and was significant at the 50th and 75th quantiles. This suggests that an increase in the proportion of the population with a higher age results in a decrease in income inequality. This finding contrasts with the findings by Debebe and Zekarias (2020), who demonstrated that income among households in southern Ethiopia increased with age, thus decreasing the income gap. Similarly, Liu, Ibrahim, and Chin (2025) observed a negative relationship between population aging and income inequality. In support of this, Abdulsalam et al. (2024) argue that as household members get older, they can find income sources which ultimately contribute to the overall income of the household. In the case of agricultural households, the observed trend could be explained by the fact that, generally, agriculture among smallholder farmers is labor-intensive; therefore, productivity may be negatively affected due to age. An increase in the number of years of schooling corresponds with a higher education level. Education was expected to play a significant role in reducing income inequality in this study. However, an increase in the number of years of schooling of the household head translated into an increase in income inequality in this study. Previous studies have found the relationship between education and income inequality to be ambiguous at times. For instance, Nabassaga, Chuku, Mukasa, and Amusa (2020) found a non-linear, inverted U-shape relationship between education and income inequality. In another study, Moyo, Mishi, and Ncwadi (2022) found that the effect of education on income inequality may be either positive or negative, depending on equal access to education. The results of the present study are comparable with the findings of other studies. For example, in Nigeria, Aminu, Wei, Arowolo, and Ibrahim (2021) found that education had an increasing effect on income inequality among smallholder arable crop farmers. In Nekemte Town, Ethiopia, Teshome, Sera, and Dachito (2021) found that variations in education contributed negatively to income inequality. This could be attributed to the fact that household heads who have attained higher education levels are more likely to adopt better farming practices and hence earn more than households with lower education. Therefore, it is not surprising that in a study population like the one observed in this study, where there are wide discrepancies in education, there are variations in income, thus resulting in high income inequality. Similar assertions were made by Aminu et al. (2021), who attributed this to the fact that education augments the skills base and consequently positively influences income-earning capacity. Thus, increasing the income gap if there is inequality in access to education (Moyo et al., 2022). The fact that household size had a statistically significant negative effect across all quantiles is consistent with findings of previous studies, which demonstrated that the size of the household is significantly associated with income inequality (Nebebe, & Appa Rao, 2016; Shin, 2020; Waris et al., 2023). Therefore, the findings reported here imply that having many large households contributes to a decrease in income inequality in the area. Some studies have reported similar results. For example, it has been noted that an increase in the size of the family is likely to lead to more income earners per household, thereby increasing the overall income of the household (Debebe & Zekarias, 2020; Rahman, 2017). Therefore, the effect of household size on income inequality might be informed by the dependency ratio or working-age group in the households and the area where the study is conducted. For example, in cities and towns where jobs are more readily available compared to rural areas, household size might significantly decrease income inequality. In support of this, Abdulsalam et al. (2024) argue that in a large-sized household where the dependency ratio is low, most of its members can find income sources, which ultimately contribute to the overall income of the household. Asian Journal of Agriculture and Rural Development, 15(4) 2025: 686-695 694 © 2025 AESS Publications. All Rights Reserved. Farm size was another significant determinant of income inequality. In this study, farm size was captured as a categorical variable with three levels: ≤3ha, 4-5ha, and 6-8ha. All three levels of farm size had a positive association with income inequality, except for the first quantile of the ≤3ha and 4-5ha levels. The magnitude of the impact of farm size increased with the increase in the quantiles. This shows that an increase in farm size positively impacts income. This implies that households with smaller farms have lower incomes than those with larger farms, increasing the gap between the poor and rich households, hence higher inequality within those lower categories. This could be attributed to the fact that households with larger cultivated lands tend to improve their earning capacity (Aminu et al. (2021). 5. CONCLUSION AND RECOMMENDATIONS The study examined the extent and determinants of income inequality among farming households using the Gini Coefficient and quantile regression model. The Gini coefficient in this study area was higher than 0.35, indicating that income inequality persists despite the participation of farming households in the PKM program.This situation has the potential to predispose farming households to adverse effects such as food insecurity, poverty, and weakened livelihoods. Age, education, household size, and farm size were identified as factors with a significant impact on income inequality among respondents. Therefore, policies aimed at reducing income inequality among farming households should consider these key factors. Strategies to promote equitable access to education are necessary. For example, educational support programs targeting disadvantaged children and children with uneducated parents should be prioritized. Other equity strategies include providing free and mandatory primary and secondary education for all children, ensuring access for excluded groups and learners with special educational needs, increasing resource allocation to schools, and creating equitable learning and teaching environments. Providing comprehensive sexual and reproductive health education to both adults and school-going adolescents could aid in family planning, thereby reducing large family sizes. The positive impact of farm size in this study emphasizes the need for larger agricultural lands as a means to increase the earning capacity of agricultural households. Successful implementation of land reform programs and leasing vacant productive land could help increase farm size and production, directly contributing to reducing poverty and income inequality. However, agricultural land alone is insufficient; support for agricultural households from both public and private sectors to maximize land use is also essential. Funding: This work was supported by the University of South Africa’s Masters and Doctoral Support Program. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. Institutional Review Board Statement: The Ethical Committee of the College of Agriculture and Environmental Science, University of South Africa, South Africa granted approval for this study on 19 November 2019 (Ref #: 2019/CAES_HREC/178). Transparency: The authors state that the manuscript is honest, truthful, and transparent, that no key aspects of the investigation have been omitted, and that any differences from the study as planned have been clarified. This study followed all writing ethics. Competing Interests: The authors declare that they have no competing interests. Authors’ Contributions: All authors contributed equally to the conception and design of the study. All authors have read and agreed to the published version of the manuscript. REFERENCES Abdulsalam, R. Y., Abdulwahab, B. A., Zakari, Y., & Adamu, Y. (2024). Poverty status and income inequality among farming households in Dutse Local Government Area of Jigawa State, Nigeria. Journal of Agriculture and Environment, 20(1), 13-23. https://doi.org/10.4314/jagrenv.v20i1.2 Adam, A. M. (2020). Sample size determination in survey research. Journal of Scientific Research and Reports, 26(5), 90–97. https://doi.org/10.9734/jsrr/2020/v26i530263 Akpan, S. B., Okon, U. E., Udo, U. J., & Akpakaden, I. S. (2020). Analysis of income inequality and poverty incidence among oil palm farmers in Akwa Ibom State, Nigeria. Ife Journal of Agriculture, 32(2), 102-117. Alimi, O. B., Maré, D. C., & Poot, J. (2018). More pensioners, less income inequality? The impact of changing age composition on inequality in big cities and elsewhere. In Modelling aging and migration effects on spatial labor markets. In (pp. 133-159). Cham: Springer International Publishing. Aminu, R. O., Wei, S., Arowolo, A. O., & Ibrahim, S. B. (2021). Socioeconomic determinants of income inequality among smallholder arable crop farmers in Nigeria. The Journal of Developing Areas, 55(3), 175-188. https://doi.org/10.1353/jda.2021.0059 Anyiam, K. H., Ben-Chendo, G. N., Ibeagwa, O. B., Obi-Nwandikom, C. O., Isaiah, G. I., Igwe, L. U., . . . Oddi, C. J. (2023). Analysis of income inequality amongst rice farmers in Ohaozara Local Government Area, Ebonyi State. The International Journal of Agriculture, Management and Technology, 7(1), 566–571. Bellu, L. G., & Liberati, P. (2006). Inequality analysis: The Gini index. Italy: Food and Agriculture Organization. Bottai, M., Frongillo, E. A., Sui, X., O'Neill, J. R., McKeown, R. E., Burns, T. L., . . . Pate, R. R. (2014). Use of quantile regression to investigate the longitudinal association between physical activity and body mass index. Obesity, 22(5), E149-E156. https://doi.org/10.1002/oby.20618 Cook, B. L., & Manning, W. G. (2013). Thinking beyond the mean: A practical guide for using quantile regression methods for health services research. Shanghai Archives of Psychiatry, 25(1), 55–59. Debebe, S., & Zekarias, E. H. (2020). Analysis of poverty, income inequality and their effects on food insecurity in Southern Ethiopia. Agriculture & Food Security, 9(1), 16. https://doi.org/10.1186/s40066-020-00269-3 Dib, J. B., Alamsyah, Z., & Qaim, M. (2018). Land-use change and income inequality in rural Indonesia. Forest Policy and Economics, 94, 55-66. https://doi.org/10.1016/j.forpol.2018.06.010 https://doi.org/10.4314/jagrenv.v20i1.2 https://doi.org/10.9734/jsrr/2020/v26i530263 https://doi.org/10.1353/jda.2021.0059 https://doi.org/10.1002/oby.20618 https://doi.org/10.1186/s40066-020-00269-3 https://doi.org/10.1016/j.forpol.2018.06.010 Asian Journal of Agriculture and Rural Development, 15(4) 2025: 686-695 695 © 2025 AESS Publications. All Rights Reserved. Garza-Rodriguez, J., Ayala-Diaz, G. A., Coronado-Saucedo, G. G., Garza-Garza, E. G., & Ovando-Martinez, O. (2021). Determinants of poverty in Mexico: A quantile regression analysis. Economies, 9(2), 60. https://doi.org/10.3390/economies9020060 Gini, C. (2005). On the measurement of concentration and variability of characters. METRON-International Journal of Statistics, 63(1), 1-38. Haini, H., Musa, S. F. P. D., Wei Loon, P., & Basir, K. H. (2023). Does unemployment affect the relationship between income inequality and food security? International Journal of Sociology and Social Policy, 43(1-2), 48-66. https://doi.org/10.1108/IJSSP-12-2021-0303 Kadigi, R. M. J. (2021). Income inequality in mountain areas: the case of agroforestry farming systems in Uluguru Mountains, Tanzania. Open Journal of Forestry, 11(3), 254-291. https://doi.org/10.4236/ojf.2021.113017 Leeds, M. A. (2014). Quantile regression for sports economics. International Journal of Sport Finance, 9(4), 346-359. https://doi.org/10.1177/155862351400900405 Liu, S., Ibrahim, S., & Chin, L. (2025). Investigating the impact of population aging on income inequality. Hong Kong Journal of Social Sciences, 64(2), 1-10. Luebker, M. (2010). Inequality, income shares and poverty: The practical meaning of Gini coefficients. Switzerland: International Labour Organization. Moyo, C., Mishi, S., & Ncwadi, R. (2022). Human capital development, poverty and income inequality in the Eastern Cape province. Development Studies Research, 9(1), 36–47. https://doi.org/10.1080/21665095.2022.2032236 Mpumalanga Provincial Treasury. (2024). Second adjustments estimates of provincial revenue and expenditure. South Africa: Mpumalanga Provincial Treasury. Nabassaga, T., Chuku, C., Mukasa, A. N., & Amusa, H. (2020). How does educational inequality affect income inequality in Africa? Working Paper Series N° 343: African Development Bank. Nebebe, F. A., & Appa Rao, C. (2016). Determinants of income inequality in urban Ethiopia: A study of South Wollo administrative zone, Amhara national and regional state. International Journal of Applied Research, 2(1), 550-563. Priscilla, L., Singh, O. K., & Vatta, K. (2021). Effects of income sources on inequality among agricultural households in North-east India. Indian Journal of Agricultural Economics, 76(4), 658–671. https://doi.org/10.63040/25827510.2021.04.006 Rahman, S. (2017). Factors influencing household income in poor urban slum settlements in Bangladesh. Journal of Poverty Alleviation and International Development, 8(2), 55–80. Sambo, T. A., Oguttu, J. W., & Mbombo-Dweba, T. P. (2025). Food insecurity and coping strategies of Phezukomkhono Mlimi farming households in the Nkomazi Local Municipality, South Africa. African Journal of Food, Agriculture, Nutrition and Development, 25(2), 25985–26003. https://doi.org/10.18697/ajfand.139.25035 Sambo, T. A. (2021). Analysis of the food security status of Phezukomkhono Mlimi beneficiaries in the Nkomazi municipality. Master’s Thesis. University of South Africa, South Africa. Shin, K.-Y. (2020). A new approach to social inequality: Inequality of income and wealth in South Korea. The Journal of Chinese Sociology, 7(1), 17. https://doi.org/10.1186/s40711-020-00126-7 Statistics South Africa. (2011). Census 2011. South Africa: Statistics South Africa. Statistics South Africa. (2022). Census 2022: Statistical release. South Africa: Statistics South Africa. Sulla, V., Zikhali, P., & Cuevas, P. F. (2022). Inequality in Southern Africa: An assessment of the Southern African Customs Union. United States: World Bank. Teka, A., Woldu, G., & Fre, Z. (2022). Status and determinants of poverty and income inequality in pastoral: Household-based evidence from Afar Regional State, Ethiopia and agro-pastoral communities. In Social protection, pastoralism and resilience in Ethiopia (1st ed.). United Kingdom: Routledge. Teshome, G., Sera, L., & Dachito, A. (2021). Determinants of income inequality among urban households in Ethiopia: A case of Nekemte Town. SN Business & Economics, 1(11), 152. https://doi.org/10.1007/s43546-021-00158-2 United Nations. (2015). Global sustainable development report. United States: United Nations. Waris, A., Khan, S., Urrahman, A., Hronec, M., & Suplata, M. (2023). Measurement of income inequality and its determinants in District Dir Lower. Journal of Financial Risk Management, 12(2), 180-202. https://doi.org/10.4236/jfrm.2023.122010 Xu, M. (2023). Quantile regression model and its application research. Academic Journal of Science and Technology, 8(3), 172–176. https://doi.org/10.54097/vt1qpm59 Zantsi, S., Greyling, J. C., & Vink, N. (2019). Towards a common understanding of ‘emerging farmer’in a South African context using data from a survey of three district municipalities in the Eastern Cape Province. South African Journal of Agricultural Extension, 47(2), 81-93. https://doi.org/10.17159/2413-3221/2019/v47n2a505 Views and opinions expressed in this article are the views and opinions of the author(s), Asian Journal of Agriculture and Rural Development shall not be responsible or answerable for any loss, damage or liability etc. caused in relation to/arising out of the use of the content. https://doi.org/10.3390/economies9020060 https://doi.org/10.1108/IJSSP-12-2021-0303 https://doi.org/10.4236/ojf.2021.113017 https://doi.org/10.1177/155862351400900405 https://doi.org/10.1080/21665095.2022.2032236 https://doi.org/10.63040/25827510.2021.04.006 https://doi.org/10.18697/ajfand.139.25035 https://doi.org/10.1186/s40711-020-00126-7 https://doi.org/10.1007/s43546-021-00158-2 https://doi.org/10.4236/jfrm.2023.122010 https://doi.org/10.54097/vt1qpm59 https://doi.org/10.17159/2413-3221/2019/v47n2a505