Vegetable cultivation in Eastern Nepal: resource use efficiency and socio-economic drivers of adoption

Abstract

Resource use efficiency and socioeconomic factors influencing the adoption of vegetables for cultivation were analyzed using cross-sectional data from randomly selected 120 vegetable producers in the Udyapur District of eastern Nepal. A field survey was conducted between October and November 2019. Five major vegetables of the district based on the area of cultivation, namely potato, cauliflower, radish, cabbage, and tomato, were taken for the study. The benefit-cost ratio was highest for cauliflower (2.46), followed by tomato (2.20), radish (2.01), cabbage (1.70), and potato (1.36). The Cobb-Douglas production function revealed that the cost of human labor and the cost of manure and fertilizer had a significant positive impact on income from vegetables, whereas the cost of seed had a significant negative impact on income from vegetables. The resource use efficiency analysis shows that the costs incurred for human labor, tillage, seed, irrigation, and packaging/marketing were overused. In contrast, the costs of manure and fertilizer are underused. The Logit model was used to study various factors affecting the adoption of vegetables for cultivation in the study area, which revealed the relationship and extent of the impact of various socio-economic factors on the adoption of the five different vegetables.

Keywords:

• Cobb-Douglas
• resource use efficiency
• Logit
• adoption

REVIEWING EDITOR:

• Manuel Tejada, Universidad de Sevilla, Spain

SUBJECTS:

• Agriculture
• Horticulture
• Economics

1. Introduction

Agriculture forms the foundation of the Nepalese Economy, with a 24.12% share in the National Gross Domestic Product (GDP) (NRB, 2023), while households engaged in agriculture account for 62% and the farming population accounts for 66.7% of the total population of the country (NSO, 2023). According to the findings of the World Bank, economic growth driven by agriculture in developing nations is more impactful in alleviating poverty compared to growth stemming from other sectors (World Development Report 2008, 2007). The agriculture sector also has a decisive effect on ensuring access to adequate food and alleviating poverty in Nepal, and the vegetable subsector is an important and growing subsector.

Vegetables are grown by 1.97 million farm holdings in an area of 1,07,733 hectares and the majority of the farm holdings are smallholders with a land area of less than 0.5 hectares (NSO, 2023). The National Sample Census of Agriculture 2021/22 also reported a significant increase in the area used for vegetables over the past 10 years. The diversity in agroecology within Nepal has created the comparative advantage of producing seasonal and off-season vegetables; thus, the agricultural development strategy (2015–2035) has prioritized the sub-sector (Adhikari & Pokharel, 2020).

In developing countries, vegetable farming plays a crucial role in enhancing income, alleviating poverty, and strengthening food and nutrition security in developing countries (Shrestha et al., 2022). Vegetable growers in Nepal are mostly smallholders (NSO, 2023) and need to specialize in production with high productivity within a small area (Gurung et al., 2016). Smallholders exploit very low capital-to-labor ratios, and family labor is substantial (Rapsomanikis, 2015).

Producers have shifted rapidly from growing annual staple food crops such as rice, wheat, and maize, etc. to vegetable crops in Nepal (NSO, 2023). Attraction to vegetable farming has increased because of its ability to generate income quickly, even from a small plot of land, and in a relatively short timeframe (Gurung et al., 2016; Rai et al., 2019). With an increase in demand for vegetables, there is an increasing trend in the adoption of improved and hybrid varieties along with improvements in cultivation practices, thereby showing a tendency towards commercial farming (Begho, 2022). However, rural farm households in Nepal face limitations due to low literacy rates, slow adoption of technology, and inefficient resource utilization, resulting in higher production costs, thereby reducing the competitiveness of local products with imported vegetables in the market (Shrestha et al., 2016).

The per capita consumption of vegetables is increasing and internal production has not been able to meet domestic demand. However, the volatility and price sensitivity of the vegetable market in Nepal pose challenges for locally produced vegetables, which, despite higher production costs, contend with competitively priced imports from India benefiting from lower production costs and higher subsidization. The average quantity of import of vegetables from the fiscal year 2009/10 to 2018/19 was 0.51 million metric tons valued at an average of NRS 14185 million, which indicates the notable increase in import quantity (278.87%) and value (602.94%) over the years (Adhikari & Pokharel, 2020). Compared to imports, the export quantity and value are almost negligible, accounting for only 28,458 metric tons, with an average value of NRS 2207 million in fiscal years 2009/10 to 2018/19 (Adhikari & Pokharel, 2020). These data demand the necessity of scaling up production along with the increasing competitive capacity of Nepalese farmers.

The diverse agroecological conditions and climatic variation available in Nepal have provided a great opportunity to produce vegetables throughout the year with a comparative advantage, but the sub-sector needs to maintain productivity and efficient utilization of the available resources (Shrestha et al., 2016). However, the lack of sufficient studies on vegetable production has limited the progress of the vegetable sector, despite various national-level initiatives that have been undertaken. There still exists a gap between the demand and supply of agricultural inputs, such as vegetable seeds, pesticides, farm machinery, and infrastructure.

Udayapur district is located in eastern Nepal and has an elevation ranging from 360 to 2310 m above sea level. The major vegetables cultivated in the district in terms of area are potato, cauliflower, radish, cabbage, and tomato. The vegetable subsector has great potential and comparative advantages in the district. Climatic variability, along with well-adapted local varieties, as well as increasing demand for fresh vegetables, indicates impressive growth and trade potential of the vegetable sub-sector. Farmers’ decisions for specific vegetable production could be affected by various biophysical and socioeconomic factors, which are also responsible for income generation from the vegetable sub-sector.

On such grounds, resource use efficiency and profitability of vegetable production were evaluated along with the assessment of socio-economic factors affecting the adoption of vegetables for cultivation and identification of problems concerning both the production and marketing of vegetables within the Udayapur district of eastern Nepal.

2. Materials and methods

2.1. Study site and sample

Udayapur district (Triyuga and Chaudandigadi municipalities) was purposively selected for study purposes as one of the research areas under the Nepal Agriculture Services Development Programme (NASDP), a bilateral project between the government of Nepal and the government of Switzerland. Ethical approval for the study was obtained from the Institutional Review Board of the Agriculture and Forestry University, Nepal via the Directorate of Research and Extension of the same University. The five vegetables with the highest area coverage in the Udayapur district (potato, cauliflower, radish, cabbage, and tomato) were selected for the study. According to the 2011 National Population and Housing Census conducted by the Central Bureau of Statistics, a total of 30,003 households were reported in the selected two municipalities, and approximately 82% (approximately 24,602 households) were reported to be involved in agriculture. To construct our sampling frame, we assumed that each farming household is cultivating at least one type of vegetable, resulting in a sampling frame of 24,602 vegetable-cultivating households. The sample size was determined by using the following formula given by Yamane (1967). $$n=\frac{N}{1+N\mathrm{*}{e}^2}$$

Where n is the sample size, N is the total number of households cultivating at least one of the vegetables in the study area and e is the precision level (10% was used). The sample size was set at 120, slightly exceeding the calculated sample size. Vegetable Producers were chosen through a simple random sampling from the selected municipalities, and cross-sectional data from 120 vegetable producers were collected using a semi-structured questionnaire through a household survey conducted from October to November 2019. Focus group discussions and key informant interviews were conducted with the concerned stakeholders. All participants were fully informed about the study′s objectives, and verbal consent was obtained from each participant before gathering any information as per the approved proposal from the Directorate of Research and Extension of the Agriculture and Forestry University, Nepal. The collected quantitative data were analyzed using Stata (version 12.1) and Statistical Package for Social Sciences (SPSS) software (version 16).

2.2. Benefit-cost analysis

The examination of vegetable cultivation costs focused on variable expenses associated with production, excluding fixed costs considering the short duration of the production cycles. Variable costs covered expenditures on labor, chemical fertilizers, organic manures, pesticides, irrigation, seeds, irrigation, packaging materials, and other related expenses. The total variable production cost was determined by aggregating all costs associated with variable inputs. The gross margin of the primary vegetables grown in the research area was computed as the difference between gross returns (total revenue) and total variable costs.

2.3. Cobb-Douglas production function and resource use efficiency

The Cobb-Douglas production function was employed to depict the technological relationship between the inputs utilized and the resultant output as follows: (1) $$Y=\alpha X_1{}^{b1}{X}_2{}^{b2}{X}_3{}^{b3}{X}_4{}^{b4}{X}_5{}^{b5}{X}_6{}^{b6}{\text{e}}_{\text{u}}$$(1)

Or, (2) $$\begin{array}{l}\text{Ln\hspace{0.17em}Y}=\text{\hspace{0.17em}Ln\hspace{0.17em}}\alpha +b_1\text{\hspace{0.17em}Ln\hspace{0.17em}}{X}_1+b_2\text{\hspace{0.17em}Ln\hspace{0.17em}}{X}_2+b_3\text{\hspace{0.17em}Ln\hspace{0.17em}}{X}_3+b_4\text{\hspace{0.17em}Ln\hspace{0.17em}}{X}_4\\ +b_5\text{\hspace{0.17em}Ln\hspace{0.17em}\hspace{0.17em}}{X}_5+b_6\text{\hspace{0.17em}Ln\hspace{0.17em}\hspace{0.17em}}{X}_6+u\end{array}$$(2) here Y = Total income from vegetable production (NPR/ha) X₁ = cost of Human labor used (NPR/ha) X₂ = cost of tillage (cost of bullock + tractor cost) (NPR/ha) X₃ = cost of seed/seedling used (NPR/ha) X₄ = cost of manure and fertilizer used (NPR/ha) X₅ = Cost of irrigation (NPR/ha) X₆ = Cost of packaging and marketing (NPR/ha) e = Error term u = random disturbance term Ln = natural logarithm α = intercept and,b₁ to b_{6 =} elasticity coefficients.

Furthermore, the ratio of the Marginal Value Product (MVP) to Marginal Factor Cost (MFC) was computed based on the regression coefficient of each unit. MVP is the value of the incremental unit of output resulting from the additional unit of input, and MFC is the increase in the cost of inputs due to the purchase of additional units of inputs and is equal to one in this context because both the dependent and explanatory variables were transformed to a monetary unit. Thus, marginal value product served as gauze for resource use efficiency (RUE). It was computed by multiplying the average value product of a specific resource (AVPs) by its production elasticity (b_i).

By definition, $$\frac{\partial y}{\partial X_1}=\text{\hspace{0.17em}marginal\hspace{0.17em}\hspace{0.33em}value\hspace{0.33em}product\hspace{0.33em}}{X}_1\left({\text{MVPX}}_1\right)$$

And, $\frac{Y}{X_1}$ = average value product of X₁ (AVP_X1).

Thus, MVP_X1 can be derived by multiplying its production elasticity (β₁) by AVP at the geometric mean level of both Y and X₁.

For ith resource, $${\mathrm{\text{AVPX}}}_{\mathrm{i}}=\frac{\overline{Y}}{\overline{X_i}}$$ where $\overline{Y}$ = value of Y at the geometric mean level and $\overline{X_i}$ = value of X_i at the geometric mean level. $$\mathit{\text{RUE}}\left(r\right)=\frac{\mathit{\text{MVP}}}{\mathit{\text{MFC}}}$$

Decision rule:r = 1, Efficient use of the resourcer > 1, underutilization of the resourcer < 1, Overutilization of the resource.

The relative percentage change in MVP was calculated as: $$\mathrm{D}=\left(1-\frac{1}{\overline{r}}\right)\times 100$$ where D is the absolute value of the percentage change in the MVP for each resource.

2.4. Factors affecting the adoption of vegetable cultivation

Logistic regression (logit model) was used to study the various factors affecting vegetable production. The effects of factors such as age of household head, gender of household head, ethnicity,¹ education status, migration status, number of economically active household members, Livestock Standard Unit² (LSU), membership in agriculture-related organizations, input availability, and loan accessibility were analyzed for the number of farmers producing potato, cauliflower, radish, cabbage, and tomato. The selection of these independent variables was based on the research findings of Subedi et al. (2019); Kersting and Wollni (2012), regarding the adoption of potatoes and vegetables, respectively, for cultivation, as well as the assumptions made by the authors. The description of the variables used in the Logit model is presented in Table 1.

Table 1. Description of variables used in the Logit model.

Variables	Description of variables
Dependent variables in the Logit model
Potato production	Potato production dummy,1 if the farmer produces potato, otherwise 0
Cauliflower production	Cauliflower production dummy,1 if the farmer produces cauliflower, otherwise 0
Radish production	Radish production dummy,1 if the farmer produces radish, otherwise 0
Cabbage production	Cabbage production dummy,1 if the farmer produces cabbage, otherwise 0
Tomato production	Tomato production dummy,1 if the farmer produces tomato, otherwise 0
Independent variables
Age of HHH	Age of the household head (in years)
Gender of HHH	Gender of household head (1= Male, 0 = Female)
Ethnicity	Ethnicity of respondent (1 = Brahmin/Chhetri, 0 = otherwise)
Education	Education status of household head (1 = literate, 0 = Illiterate)
Migration status	Migration status in household (1 = Migrated abroad, 0 = otherwise)
Economically active members	Economically active members with in household
Livestock holding	Livestock holding measured in terms of Livestock standard unit cattle equivalent of household
Membership in organizations	Member of agriculture related organizations (1 = Yes, 0-No)
Input availability	Availability of input in vegetable cultivation (1 = Available, 0- otherwise)
Loan access	Loan access for vegetable farming (1 = Access, 0- otherwise)

The empirical model used as:

• Vegetable production (Yes = 1, otherwise 0) = β₀ + β₁ Age of HH + β₂ Gender of HH + β₃ Ethnicity + β₄ Education + β₅ Migration status + β₆ Economically active members + β₇ Livestock holding + β₈ Membership in organizations + β₉ Input availability + β₁₀ Loan access.

2.5. Problems on production and marketing

To identify primary issues in production and marketing, an index was created based on response frequencies. Problems in production and marketing were assessed using a five-point Likert-Scale ranging from most serious, serious, moderate, low, and very low or no problem at all using scores of 1.00, 0.80, 0.60, 0.40, and 0.20, respectively (Sharma, 2019). The following formula was utilized to compute the index for the severity of production and marketing problems encountered by producers and traders, respectively. The priority index for each variable was determined through a weighted average mean to derive sound conclusions and facilitate decision making. The formula for calculating the index of influence is as follows (Shrestha et al., 2014): $$I_{\mathit{\text{inf}}}=\sum \frac{s_i{f}_i}{N}$$

Where,

I_inf = index of influence s_i = scale value f_i = frequency of influence given by respondents

N = total number of respondents.

3. Results and discussion

3.1. Socio-demographic characteristics and asset holding

The average age of respondents in the study area was 47.09 years, average years of schooling were 5.25 years and average age of the household head was 49.78 years. The average number of economically active household members was 3.26. Of the 120 sampled households, 79.2% of household head were male and 20.8% of household head were female. Regarding ethnicity, 63.3% of the households were Janajati, 30.8% were Brahmin/Chhetri, and 5.8% were Dalits. The average land holding was 15.34 Kattha³ and the average area under vegetable cultivation was 5.73 Kattha. The average Livestock Standard Unit (LSU) was 2.55. Seventy percent of households were involved in at least one agriculture-related organization, such as farmers’ groups, savings and credit cooperatives, and agriculture cooperatives; 77.5% had received services from service providers, mostly for cultivation practices and disease pest management.

3.2. Decision for cultivation

In the study area, farmers decide to cultivate vegetables based on various information systems. The basis for the decision to cultivate vegetables included consultation with neighbors (38.3%), followed by market information (28.3%), last year’s price (24.2%), suggestion from government bodies (6.7%), suggestion from cooperatives (1.7%), and suggestions from NGOs/INGOs (0.8%) (Table 2).

Table 2. Information systems for the decision to cultivate vegetables.

Variables	Overall (N = 120)
Market information	34 (28.3)
Last year′s price	29 (24.2)
Consultation with neighbor	46 (38.3)
Suggestions from cooperatives	2 (1.7)
Suggestions from government bodies	8 (6.7)
Suggestions from NGO/INGOs	1 (0.8)

Notes: Figures in parentheses indicate percentages.

Source: Field survey, 2019.

3.3. Market outlet

The major market outlet in the study area was the village market (82.5%), followed by the farm gate (14.2%), and the outside market (3.3%) (Table 3).

Table 3. Market outlet of vegetables in the study area.

Variables	Overall (N = 120)
Farm gate	17 (14.2)
Village market	99 (82.5)
Market outside village	4 (3.3)

Notes: Figures in parentheses indicate percentages.

Source: Field survey, 2019.

3.4. Gender role in decision making and involvement

The decision for vegetable cultivation was made jointly by males and females (50.8%), followed by males (41.7%) and females (7.5%). In the study area, gender involvement in vegetable production included both males and females (50%), followed by only males (34.2%) and only females (15.8%) (Table 4).

Table 4. Gender role in vegetable production in the study area.

Variables	Overall (N = 120)
The decision for vegetable cultivation
Male	50 (41.7)
Female	9 (7.5)
Both	61 (50.8)
Involvement in vegetable cultivation
Male	41 (34.2)
Female	19 (15.8)
Both	60 (50.0)

Source: Field survey, 2019.

3.5. Food sufficiency status

Food sufficiency status was studied in the study area. The overall context represents food sufficiency in descending order as 9–12 months (35%), more than a year (33.3%), 6–9 months (20%), 3–6 months (10%), and less than 3 months (1.7%) (Table 5).

Table 5. Food sufficiency status in the study area.

Variables	Overall (N = 120)
Less than 3 months	2 (1.7)
3–6 months	12 (10.0)
6–9 months	24 (20.0)
9–12 months	42 (35.0)
More than a year	40 (33.3)

Source: Field survey, 2019.

3.6. Benefit-cost ratio

Assessing the profitability of vegetable farming through benefit to cost ratio revealed cauliflower (2.46) as the most profitable crop, followed by tomato (2.20), radish (2.01), and cabbage (1.70). Potato was found to be a comparatively less profitable vegetable crop, with a benefit-cost ratio of 1.36 (Table 6).

Table 6. Benefit-cost ratio of vegetables in the study area.

Vegetable crops	Minimum	Maximum	Mean	Std. Deviation
Potato (n = 101)	0.54	2. 92	1.36	0.38
Cauliflower (n = 97)	0.81	4.23	2.46	0.78
Radish (n = 36)	1.24	3.26	2.01	0.54
Cabbage (n = 47)	0.74	2.78	1.70	0.42
Tomato (n = 54)	1.00	5.28	2.20	0.72

Source: Field survey, 2019.

3.7. Marketing margin and producer’s share of major vegetables

The marketing margin of cauliflower was the highest (22), followed by potato (18), tomato (15), radish (15), and cabbage (14), whereas the producer’s share of tomato had the highest value, followed by cauliflower, potato, cabbage, and radish (Table 7).

Table 7. Marketing margin and producer’s share of major vegetable.

Vegetable	Cost (NPR/kg)	Farm gate price (NPR/kg)	Retailers price (NPR/kg)	Marketing margin (NPR/kg)	Producers (share %)
Potato	21	27	45	18	60
Cauliflower	17	38	60	22	63
Radish	11	20	35	15	57
Cabbage	13	21	35	14	60
Tomato	20	40	55	15	73

Source: Field survey and KII, 2019.

3.8. Cobb-Douglas production function

The Cobb–Douglas production function was employed to examine the impacts of variable inputs on overall income generated from vegetable production. The response variable was income, while the explanatory variables were the cost of human labor, tillage, seed, manure and fertilizer, irrigation, and marketing. The cost of human labor includes the labor costs from tillage to intercultural operations and harvesting. The cost of tillage includes the bullock labor cost and the cost of the tractor for tillage. The Cobb-Douglas production function model was well-fitted, as the p-value for the f-test was found to be 0.000. Further, a multicollinearity test for the independent variable was performed using the Variance Inflation Factor (VIF), whose value was found to be 1.075. Hence, there was no multicollinearity between the independent variables. The Cobb-Douglas production function showing the impact of various inputs on income from vegetable cultivation is shown in Table 8.

Table 8. Cobb-Douglas production function showing the impacts of various inputs on income from vegetable cultivation.

Ln vegetable income	Coef	St. Err	t-Value	p-Value
Ln human labor cost	0.222*	0.126	1.76	0.081
Ln tillage cost	0.018	0.028	0.64	0.522
Ln seed cost	−0.150***	0.032	−4.75	0.000
Ln manure fertilizer cost	0.129**	0.051	2.53	0.013
Ln irrigation cost	−0.044	0.057	−0.76	0.447
Ln packaging & marketing cost	0.009	0.006	1.41	0.161
Constant	10.705***	1.501	7.13	0.000
No. of Observation	120
F (6, 113)	6.96***
R-squared	0.270
Adjusted R-squared	0.2312
VIF	1.075

Note: ***, **, and * represent significance at the 1%, 5%, and 10% levels, respectively.

Source: Field survey and KII, 2019.

Among all the considered variables, the cost of human labor had a significant positive impact on the income of vegetables at the 10% level, the cost of seed had a significant negative impact on the income of vegetables at the 1% level, and the cost of manure and fertilizer had a significant positive impact on vegetable income at the 5% level. When the cost of human labor increased by 1%, income from vegetables increased by 0.22%. When the cost of seeds increased by 1%, income from vegetables decreased by 0.15%, and when the cost of manure and fertilizer increased by 1%, income from vegetables increased by 0.12%. The R-squared value is 0.27. Hence, 27% of the total variation in income is due to the considered variables.

The findings of Kunwar and Maharjan (2019) revealed the opposite result and stated that a one percent increase in human labor decreases the gross return by 1.5%. Better management could have increased the cost of human labor, and ultimately, production and income from vegetables increased in our context. This is in accordance with the findings of Bajracharya and Sapkota (2017), which stated that a 1% increase in the cost of human labor would increase the overall income of potatoes by 0.63%. In addition, Ghimire and Dhakal (2014) reported an increase in the yield of cauliflower production with an increase in the use of hired labor. Negative relation between cost of seed and income from vegetable was consistent with findings of Kunwar and Maharjan (2019) which stated that with one percent increase in cost of seed of tomato the gross return decreases by 1.63% but was found in contrast with findings of Bajracharya and Sapkota (2017) in case of potato production. The positive relationship between the cost of manure and fertilizer and income from vegetables was found to be consistent with the findings of Kunwar and Maharjan (2019) in the case of tomatoes, and with the findings of Bajracharya and Sapkota (2017) in the case of potatoes, which revealed a 0.19% increase in income from potatoes with a 1% increase in cost of FYM. Tolno et al. (2016) revealed the positive impact of fertilizer on potato production in Guinea. Similarly, Ghimire and Dhakal (2014) found a significant impact of organic manure on cauliflower productivity. Similarly, Akter et al. (2012) revealed a significant impact of manure on income generated from the cultivation of tomatoes.

3.9. Estimation of resource use efficiency (RUE)

The Cobb Douglas production function was further used to analyze the efficiency of resources such as human labor, tillage, seed, manure fertilizer, irrigation, and marketing costs in the income generation from vegetable cultivation. The efficiency ratio was calculated to determine the efficiency of resources used, and it was found that the efficiency ratios for human labor cost, tillage cost, seed cost, irrigation cost, and marketing cost were less than one, meaning that those costs were overused. In contrast, the efficiency ratio for manure and fertilizer costs was found to be underused because the efficiency ratio for the cost of manure and fertilizer was found to be greater than one. The D value indicated that manure and fertilizer should be increased by 40. 93% for the optimum allocation. Moreover, and human labor, tillage, seed, irrigation, and marketing costs should be decreased by 28.11%, 124.11%, 142.59%, 154.55%, and 53.79%, respectively, for the optimum utilization of resources (Table 9).

Table 9. Estimation of resource use efficiency of vegetable cultivation using Cobb-Douglas production function.

Ln vegetable income	Coef	MVP	MFC	r = MVP/MFC	D = (1-1/r)*100	Efficiency
Ln labor cost	0.222	0.78	1	0.78	−28.11	Overused
Ln tillage cost	0.018	0.44	1	0.44	−124.33	Overused
Ln seed cost	−0.150	−2.34	1	−2.34	142.59	Overused
Ln manure fertilizer cost	0.129	1.69	1	1.69	40.93	Underused
Ln irrigation cost	−0.044	−1.83	1	−1.83	154.55	Overused
Ln packaging & marketing cost	0.009	0.65	1	0.65	−53.79	Overused

Source: Field survey and KII, 2019.

The human labor, tillage, seed, irrigation, and packaging and marketing costs were overused. Tolno et al. (2016) also reported the overuse of seeds and labor in potato production in Guinea. Ghimire and Dhakal (2014) also reported the overutilization of human labor but underutilization of seeds for cauliflower production in Dhading, Nepal. In contrast, the efficiency ratio for manure and fertilizer cost was found to be underused and inconsistent with the findings of Tolno et al. (2016) in the case of the cultivation of potatoes in Guinea and Ghimire and Dhakal (2014) in the case of cauliflower production in Dhading, Nepal.

3.10. Factors determining the production of major vegetables at the household level using the Logit model

Table 10 presents a summary of the Logit model results.

Table 10. Factors determining the production of major vegetables using the Logit model.

Variables	Potato production (Yes = 1)	Cauliflower production (Yes = 1)	Radish production (Yes = 1)	Cabbage production (Yes = 1)	Tomato production (Yes = 1)
	dy/dx	dy/dx	dy/dx	dy/dx	dy/dx
Age of HHH	0.002	0.001	0.005	0.011**	0.002
Gender of HHH (Male = 1)	−0.073	0.221*	0.070	0.176	0.467***
Ethnicity (Brahmin/Chhetri = 1)	0.005	−0.013	−0.063	0.056	0.215*
Education (Literate = 1)	−0.084	0.194*	0.039	0.327**	0.221
Economically active members	0.029	−0.006	0.013	0.047	0.041
Livestock holding (LSU)	−0.008	−0.007	0.044	0.042	−0.067*
Membership in organizations (Yes = 1)	0.201**	0.062	−0.007	0.287**	0.221*
Migration (Yes = 1)	−0.080	−0.159	−0.011	−0.007	0.044
Input availability (Yes = 1)	−0.086	0.368***	−0.068	0.177	0.024
Loan access (Yes = 1)	−0.051	0.079	0.007	−0.238**	0.203*
No. of observation	120	120	120	120	120
LR chi^2 (I2)	14.55	24.71	8.33	30.82	37.49
Prob > chi^2	0.149	0.006	0.597	0.001	0.000
Log likelihood	−45.15	−46.47	−69.13	−64. 92	−63.82
Pseudo R^2	0.139	0.211	0.057	0.192	0.227
Predict (%)	88.46%	86%	28.59%	34.50%	42.88%

Notes: ***, **, and * indicate significance at the 1%, 5%, and 10% levels, respectively; dy/dx resembles the marginal effect after Logit regression.

Source: Field survey and KII, 2019.

3.10.1. Adoption of potato for cultivation

Variables such as the age of the household head, ethnicity being Brahmin/Chhetri, number of economically active members in the household, and membership in agriculture-related organizations have a positive impact on the chance of adoption for potato cultivation. Instead, the gender of the household head being male, education status being literate, livestock holding, members migrating abroad, local availability of inputs, and access to loans had a negative impact on the chance of adoption for potato cultivation. This is in contrast with the findings of Subedi et al. (2019), who found that the adoption of potatoes for cultivation increases with an increase in input (seed) availability. This may be due to the lower profitability of potatoes over other crops in the study area, where parcels of landholding for cultivation might have been occupied by another commodity. Households that were members of agriculture-related organizations had an increased chance of cultivating potatoes by 20.1% and was found to be significant at the 5% level. The group approach for providing subsidies for potato cultivation by different tiers of government in the area might be the major cause of the increased chance of cultivation by members of farmer groups and cooperatives. The model had a pseudo R² of 13% and a log-likelihood of -45.15. The model with potato production as the dependent variable was statistically non-significant.

3.10.2. Adoption of cauliflower for cultivation

The results show that the age of the household head, gender of the household head being male, and education status being literate, membership of agriculture-related organizations, input availability, and loan access had a positive impact on the chance of adopting cauliflower cultivation by household. Instead, the ethnicity of households being Brahmin/Chhetri, number of economically active household members, livestock standard units, and members of households that have migrated abroad have a negative impact on the chance of adopting cauliflower cultivation. The gender of the household head being male and education status being literate increased the chance of adopting cauliflower cultivation by 22.1% and 19.4%, respectively, and was significant at the 10% level. Similarly, the availability of agro-inputs at the local level increased the chance of cultivating cauliflower by 36.8% and was significant at the 1% level. Kersting and Wollni (2012) also stated that farmers are more likely to be involved in vegetable farming if they are better educated, participate in community-based organizations, and are supported by government agencies. The model with cauliflower production as the dependent variable had a pseudo R² value of 21.1% and a log-likelihood of −46.27 and the model was significant at the 1% level.

3.10.3. Adoption of radish for cultivation

The results revealed that variables such as age of household head, gender of household head, education dummy, and number of economically active household members, livestock holding, and loan access had a positive impact on the likelihood of adopting radish cultivation by household. Instead, variables such as ethnicity, membership in agriculture-related organizations, migration abroad, and input availability had a negative impact on the chance of adopting radish cultivation. The relatively low profitability of radishes in the study area might be the cause of this relationship. However, the model with radish production as the dependent variable was non-significant.

3.10.4. Adoption of cabbage for cultivation

The factors affecting the adoption of cabbage cultivation were also studied. The study revealed that age of the household head, gender of the household head being male, ethnicity of the household being Brahmin/Chhetri, educational status being literate, number of economically active household members, livestock holding, household members being members of agriculture-related organizations, and local availability of inputs have a positive impact on the likelihood of adopting cabbage cultivation by household. Instead, families having migrated members abroad and had access to loans had a negative impact on the chance of adopting cabbage cultivation. The table also shows that literate household heads, household heads being members of agriculture-related organizations, increased the chance of cultivating cabbage by 32.7% and 28.7%, respectively, and was significant at the 5% level. However, access to loans decreased the chance of cultivating cabbage by 23.8%, which was found to be significant at the 5% level. The model with cabbage production as the dependent variable had a pseudo R² of 19.2% and, log likelihood of -64.92, and the model was significant at the 1% level.

3.10.5. Adoption of tomato for cultivation

The Logit model depicts that households with a male head had a significant positive impact on the decision to cultivate tomatoes. The chance of cultivating tomatoes increased by 46.7% if the household head was male in comparison to female, and the difference was found to be significant at the 1% level. In addition, households with Brahmin/Chhetri ethnicity had an increased chance of cultivating tomatoes by 21.5%, households with membership in agriculture-related organizations had an increased chance of cultivating tomatoes by 22.1%, and households with loan access had an increased chance of cultivating tomatoes by 20.3%, and the difference was statistically significant at the 10% level under those conditions. Livestock holding had a negative effect on the chances of cultivating tomatoes. It was found that if livestock holdings increase by one unit, the chance of cultivating tomatoes by household decreases by 6.7%, and the difference is significant at the 10% level, which might be due to the competitive nature of tomato farming and livestock holding in terms of time and labor utilization. The model with tomato production as the dependent variable had a pseudo R² value of 22.7% and log-likelihood of -63.82, and the model was significant at the 1% level (Table 10).

3.11. Problems during vegetable cultivation

Among the various problems persisting in the study area during vegetable cultivation, the inadequate technical knowledge was the first ranked problem, followed by attack of pests, inadequate quality seeds, inadequate farm labor, and irrigation problems (Table 11). Subedi et al. (2019), Kunwar and Maharjan (2019), and Rai et al. (2019) reported similar problems in the cultivation of various vegetables in different areas of Nepal.

Table 11. Problems during vegetable cultivation.

Problems during vegetable cultivation	Index	Rank
Inadequate technical knowledge	0.83	I
Attack of pest	0.72	II
Inadequate quality seed	0.56	III
Inadequate farm labor	0.49	IV
Irrigation problem	0.41	V

Source: Field survey and KII, 2019.

3.12. Problems during vegetable marketing

The five major problems during vegetable marketing in the study area include price fluctuation as a first-ranked problem, followed by a greater gap between farm gate and retail price, inadequate storage houses, influence of imports from the Indian market, and no formal agreement for marketing (Table 12).

Table 12. Problems during vegetable marketing.

Problems during vegetable marketing	Index	Rank
Price fluctuation	0.81	I
Greater gap between farm gate and retail price	0.73	II
Inadequate storage house	0.72	III
Influence of import from Indian market	0.42	IV
No formal agreement	0.32	V

Source: Field survey and KII, 2019.

4. Conclusion

The costs of human labor, and manure and fertilizer had a significant positive impact on income from vegetables, whereas the cost of seed had a significant negative impact on income from vegetables. Costs incurred for human labor, tillage, seed, irrigation, and packaging/marketing were overused, whereas the cost of manure and fertilizer was underused; thus, vegetables, being heavy feeders of nutrients, should be supplied with optimum quantities of manure and fertilizers. Cauliflower was the most profitable vegetable crop with the highest benefit-cost ratio and gross margin per unit area, followed by tomato, radish, cabbage, and potato. The education status of the household head being literate, availability of input at the local level, and membership of households in agriculture-related organizations had a positive influence on the cultivation of relatively more profitable vegetable crops, whereas holding livestock could have a competitive relationship with the adoption of vegetable crops, both being labor-intensive. The major problems during vegetable production include inadequate technical knowledge, pest attacks, inadequate quality seeds, shortage of farm labor, and irrigation problems. The problems during marketing include price fluctuation, greater gap between farm gate and retail price, inadequate storage houses, influence of imports from the Indian market, and no formal agreement for marketing. Based on the findings of the study, we suggest that vegetable farmers must carefully assess resource allocation to maintain an efficient production system and prioritize crop profitability while selecting crops for cultivation. Government intervention is advised to support access to quality inputs and ensure smooth market functioning for vegetables.

Authors’ contributions

Both authors contributed to the study design. Sagar Ghimire collected the data, analyzed, and wrote the original draft of the manuscript. Assoc. Prof. Dr. Rishi Ram Kattel advised and provided comments and feedback to finalize this manuscript. All authors approved the manuscript.

Acknowledgements

The authors would like to extend our gratitude to Assoc. Prof. Dr. Shiva Chandra Dhakal and Mr. Bishnu Kumar Bishwakarma for their valuable comments and feedback during the Master’s thesis submission by Mr. Sagar Ghimire in Agriculture and Forestry University, Chitwan, Nepal, from which the major results of the manuscript were extracted. We are grateful to the respondent farmers, enumerators, and stakeholders of the study area, without whom this study would not have been possible. Errors, if any, are entirely our own.

Disclosure statement

The authors declare no competing interests.

Data availability statement

Data that support the findings of the study are available from the corresponding author, (Kattel, R. R.), upon reasonable request.

Additional information

Funding

This research was funded by the Directorate of Research and Extension of Agriculture and Forestry University, Chitwan, Nepal in collaboration with the Nepal Agriculture Services Development Programme.

Notes on contributors

Sagar Ghimire

Sagar Ghimire serves as an agriculture extension officer at the Ministry of Agriculture and Land Management in the Gandaki Province, Nepal. He holds a Master’s Degree in Agricultural Economics from Agriculture and Forestry University, Rampur, Chitwan, Nepal. He is currently dedicated to advancing his academic pursuits through a Doctoral Course in Agriculture Sciences with a major in Agricultural Economics at the University of Tsukuba in Ibaraki, Japan, supported by the Japanese Government (MONBUKAGAKUSHO: MEXT) Scholarship since 2023. His research interests include production economics, food security, agricultural transformation, and rural development.

Rishi Ram Kattel

Rishi Ram Kattel is an academic with extensive practical and field experience in the agricultural sector in Nepal. He is an Associate Professor at the Department of Agricultural Economics and Agribusiness Management and also serves as Director of the Directorate of Research and Extension in Agriculture and Forestry University, Rampur, Chitwan, Nepal. He has expertise in data analysis, econometric modeling, and value chain analysis. His research interests include global value chain analysis, agribusiness management, and agricultural economics.

Notes

1 Ethnicity composition in the study area was Brahmin, Chhetri, Janajati and Dalits. Brahmin and Chhetri are so called upper caste, Janajati are the indigenous ethnic groups and Dalits are the underprivileged ethnic groups in Nepal.

2 Livestock holding is measured in terms of Livestock Standard Unit (LSU) based on cattle equivalent using following equation (Kattel & Upadhyay, 2018).

1 cattle/cow= 0.66 buffalo = 10 goat/lamb = 4 pig = 143 chickens/ducks

3 Kattha is unit of area of land which is equivalent to 338.63 square meters.