Determinants of Nepalese hog plum (Choerospondias axillaris Roxb.) production in Sindhupalchok, Nepal

Abstract

Nepali hog plum (Choerospondias axillaris Roxb.) is a multipurpose agroforestry fruit tree grown in hilly regions of Nepal for livelihood and income generation in rural households. The research was conducted in 2019 to assess the determinants of the Nepalese hog plum/Lapsi sub-sector in Sindhupalchok district. Three major lapsi-producing areas, namely, Chautara Municipality, Indrawati Rural Municipality and Melamchi Municipality were selected for research with a sample size of 150 producers. In addition, focused group discussions with producers/collectors and rapid market appraisal with traders were used as survey tools. Chi square test and one way ANOVA were used to analyze and compare mean among the geographical area. Indexing and multiple regression model were also used to analyse problems and factors affecting household income. The average variable cost involved in Lapsi production was 4.93 Nepalese Rupees (NRs.) (0.042 United States Dollar (USD)) per kilogram whereas the total cost for producing fresh Lapsi fruits was NRs. 5 (0.042 USD) per kilogram. Benefit-cost ratio was found 2.79 at the farm level. The major problems faced by producers were non-bearing trees followed by lack of improved production technology and low production. Likewise, the major marketing problem identified were low bargaining capacity followed by low selling price and lack of market and collection center. Significant factors affecting the income were gender of household, adoption of value addition practices, attitude towards this business, involvement of household members, literacy, gender-based decision, type of house and experience.

Keywords:

• benefit-cost ratio
• Lapsi
• marketing
• production
• value chain actors

1. Introduction

Agriculture is one of the critical factors of Nepal economy. About 60% of country population depends on Agriculture (AITC, 2020). In Nepal, agriculture and forestry sector contributes 26.50% share in national GDP (CBS, 2019). Being the himalayan country, diversified agro-ecological condition makes Nepal suitable for varieties of agricultural production. Majority of fruits (e.g., Apple, Mango, Litchi, Nepali Hog plum, Mandarin, Apricot) can be grown for import substitution, export promotion and tourism sector development (NHPC, 2017; Misha et al., 2020). The total production of fruits in the country is 1,249,764 Mt with productivity 10.50 Mt/ha in 2019/2020 (MoAD, 2018). Nepali hog plum (Choerospondias axillaris, Roxb.), commonly known as Lapsi, is a deciduous and subtropical agroforestry fruit tree. It is also considered as a source of income for subsistence farmers in Nepal (Pandit et al., 2014). Lapsi trees were cultivated in 301 village development committees covering 29 hill districts from east to west. Lapsi is found from 850 meters above sea level (masl) to 1900 masl. It was reported that over 40,000 trees were at fruit bearing stage and more than 450,000 new trees were planted in various districts of Nepal in 2001 (Poudel, 2003). It starts bearing fruits from 3 years of plantation and continues production up to 20 years. It has further potential for value addition. It can be used to produce pickle, powder, and candy. The quality candy and activated carbon can be exported to earn foreign currency (Joshi, 2017). It was selected as One Village One Product (OVOP) in Bhaktapur and One District One Product (ODOP) in Prabhat district. However, the highest production and productivity was reported in Sindhupalchok i.e. 3360 metric ton (Mt) and 12 Metric ton/hectare (Mt/ha) respectively (MoAD, 2018). Generally, most of this fruit from Sindupalchok was collected by collectors and sold to the trader in Saghna, Bhaktapur (DADO, 2018). The major areas of distribution includes Sindupalchok, Kathmandu, Lalitpur and Kavre. Also, medium intensity of planting is found on Kaski, Parbat and Bhojpur. They are then sold to local as well as regional markets (Poudel, 2003).

The current production and supply don’t meet the market demand for quality products due to traditional cultivation practices, lack of proper knowledge and poor technology on production (SAC, 2014). Similarly, another major problem that limits cultivation is the risk associated with non-bearing, as only female trees produce fruits, normally after 7–10 years of planting (Labh & Shakya, 2016; Shakya & labh, 2016; Shakya & Labh, 2018). Another major cause for high market price and food insecurity is poor infrastructure and high transportation cost (Akter & Azad, 2014) and lack of the policy that support the producer (Joshi, 2016).

Nepali hog plum is a potential agroforestry tree species for income generation and nutrient supplementation in the middle mountains of rural Nepal (Chhetri & Gauchan, 2007; APP, 1995). Leaves are often used as a fodder for goats. Female trees are more preferably used for the timber as farmer considers it to be stronger and more attractive in color. Fruits are rich in vitamin C. Similarly, the seed stones inside the fruit are utilized as wood fuel for post-harvest processing (Poudel, 2003; Shrestha et al., 2012).

It is a relatively new crop for scientific research and often ignored both from agricultural and forestry sector. The promotion and upliftment of this sector is necessary to encourage existing producers and increase the new producers. The broad objective of this study was to assess the determinants of Nepali hog Plum (Lapsi) production in Sindhupalchok, Nepal. Furthermore, producing and marketing problems were also the part of the study.

2. Materials and methods

2.1. Study site and sub sector

The study was carried out in Sindhupalchok district (Bagmati Province) covering two municipalities (Melamchi and Chautara) and one Rural Municipality (Indrawati). This district was selected because it had the highest recorded production in Nepal (MoAD, 2018). Sindhupalchok is a high-hill district which occupies 1.73% of total land area of the country. It lies 27°27’-28°13ʹN latitude and 85°27’-86°06ʹE longitude. Its altitude varies from 747masl to 7085masl whereas average temperature ranges from 32.5°C to 5°C with annual rainfall of 1615 mm. This district is surrounded by Dolakha, Ramechap and Tibet in the east, Nuwakot and Rasuwa in the west, Rasuwa and China in the North and Kavreplanchwok, Kathmandu and Ramechap in the south. The total population of district was 287,797 (DADO, 2018). Sindhupalchok was one of the major district affected by earthquake occurred in Baisakh 12 and 19, 2075 (25 and 29 April 2015). 3552 people were killed and 109 people were lost along with destruction of structures, houses, and agricultural land (DADO, 2018).

2.2. Sample and sampling technique

It consists of 150 samples i.e., 50 producers were selected randomly from each Municipality and Rural Municipality. Six rapid market appraisals (RMA) were conducted with Lapsi trader (2 RMA in each Municipality) with snowball sampling method. This method is applied when there are limited number of respondents. Further it is also called referral sampling methods (Naderifar et al., 2017).

Similarly, a meeting was done with two entrepreneurs in study area. The selected site and market area was visited by the surveyor to gather information about the scenario of production and also interacted with the producers, local collectors, forest officer, agricultural officer, entrepreneurs, industries and enablers.

2.3. Questionnaire survey

The actors (producers, local traders, middleman) involved were enumerated with open and close-ended questions that aided in collection of some useful data regarding the various operations which not only helped to design research methods, but also helped to set a baseline and an end line for the study. The market analysis was done by collecting data from local trader and middlemen.

2.4. Key informants’ interview

Key informants such as the local leaders, Agriculture Knowledge Center (AKC) officer, forest officers, collectors, traders were interviewed regarding the present scenario of cultivation in the area, current production, price trend and value addition.

2.5. Rapid Market Appraisal (RMA) and case studies

The RMA could provide an up-to-the-moment snapshot of the market status for a given product or service within the study area; the size of the area depending on the scope of the study, i.e. at the local, national or regional level. A rapid market appraisal was conducted in the study site, and it helped to knew about the existing market situation, opportunities, and constraints. The detailed study about successful producer and entrepreneurs in and around the area, was carried out to learn about the package of practice and production practices.

2.6. Data and data types

2.6.1. Primary data

Primary data are those which are collected afresh and thus happen to be original in character. These were collected via questionnaire survey, key informants’ interview, and rapid market analysis.

2.6.2. Secondary data

Secondary data are those data which have already been collected by someone else and passed through the statistical process. These data were collected from the publications of Agriculture Knowledge Center (AKC), Department of Forestry (DOF), and various governmental and non-governmental project reports, articles, journals, and books.

2.6.3. Technological approach

The data was analyzed by using SPSS (Version 16), STATA (Version 12.1) and MS -Excel.

2.6.4. Socio demographic and economic variables

The continuous variables like age, family size, landholding, dependency percentage (socio demographic), production, production per tree (economic), were analyzed by using t test while categorical variables like education level, migration status, family type, ethnicity (socio demographic), training, price, satisfaction, membership in organization (institutional), type of house, cooking system and decision on business extension of business (framework) were analyzed by using chi-square test.

2.6.5. Benefit cost analysis

Benefit-cost ratio compares the monetary cost and benefit of taking the actions. The benefit-cost analysis was obtained by calculating the total fixed cost, variable cost, and gross return from the Lapsi cultivation. The cost of production was calculated by the addition of fixed and variable cost in the production process. The gross return was calculated by obtaining income from product sale. Cost benefit analysis measures the societal value by quantifying the cost and benefits (Koopmans & Mouter, 2020; Nicol & Coen, 2016).

Hence, benefit cost ratio was calculated by using formula:

$\mathrm{B}\mathrm{e}\mathrm{n}\mathrm{e}\mathrm{f}\mathrm{i}\mathrm{t}\mathrm{c}\mathrm{o}\mathrm{s}\mathrm{t}(\mathrm{B}\mathrm{C})\mathrm{r}\mathrm{a}\mathrm{t}\mathrm{i}\mathrm{o}=\hspace{0.17em}\frac{\mathrm{G}\mathrm{r}\mathrm{o}\mathrm{s}\mathrm{s}\mathrm{R}\mathrm{e}\mathrm{t}\mathrm{u}\mathrm{r}\mathrm{n}}{\mathrm{T}\mathrm{o}\mathrm{t}\mathrm{a}\mathrm{l}\mathrm{C}\mathrm{o}\mathrm{s}\mathrm{t}}$

Where,

$\mathrm{G}\mathrm{r}\mathrm{o}\mathrm{s}\mathrm{s}\mathrm{R}\mathrm{e}\mathrm{t}\mathrm{u}\mathrm{r}\mathrm{n}=\mathrm{T}\mathrm{o}\mathrm{t}\mathrm{a}\mathrm{l}\mathrm{q}\mathrm{u}\mathrm{a}\mathrm{n}\mathrm{t}\mathrm{i}\mathrm{t}\mathrm{y}\mathrm{o}\mathrm{f}\mathrm{L}\mathrm{a}\mathrm{p}\mathrm{s}\mathrm{i}\mathrm{f}\mathrm{r}\mathrm{u}\mathrm{i}\mathrm{t}\left(\mathrm{L}\mathrm{a}\mathrm{p}\mathrm{s}\mathrm{i}\mathrm{p}\mathrm{r}\mathrm{o}\mathrm{d}\mathrm{u}\mathrm{c}\mathrm{t}\right)\mathrm{s}\mathrm{o}\mathrm{l}\mathrm{d}\times \mathrm{P}\mathrm{r}\mathrm{i}\mathrm{c}\mathrm{e}\mathrm{p}\mathrm{e}\mathrm{r}\mathrm{u}\mathrm{n}\mathrm{i}\mathrm{t}\mathrm{k}\mathrm{i}\mathrm{l}\mathrm{o}\mathrm{g}\mathrm{r}\mathrm{a}\mathrm{m}\mathrm{L}\mathrm{a}\mathrm{p}\mathrm{s}\mathrm{i}\left(\mathrm{L}\mathrm{a}\mathrm{p}\mathrm{s}\mathrm{i}\mathrm{p}\mathrm{r}\mathrm{o}\mathrm{d}\mathrm{u}\mathrm{c}\mathrm{t}\right)$

$TotalCost=Variablecost+FixedCost\left(PlantingandFarmyardmanure\right)$

Marketing margin

Marketing margin provides the demonstration of the performance of the particular industry or market structure and efficiency (Onogwu et al., 2018). The difference between the farm gate price and retailer’s price is the marketing margin which was calculated by using formula:

Marketing margin = Retailer’s price - Farm gate price

Producer’ share

Producers’ share is the price received by farmer expressed as a percentage of the retail price, i.e., the price paid by consumers which was calculated by using following formula:

$\mathrm{P}\mathrm{r}\mathrm{o}\mathrm{d}\mathrm{u}\mathrm{c}\mathrm{e}{\mathrm{r}}^{\prime }\mathrm{s}\mathrm{s}\mathrm{h}\mathrm{a}\mathrm{r}\mathrm{e}\left(\mathrm{F}\mathrm{a}\mathrm{r}\mathrm{m}\mathrm{g}\mathrm{a}\mathrm{t}\mathrm{e}\mathrm{p}\mathrm{r}\mathrm{i}\mathrm{c}\mathrm{e}\right)=\frac{\mathrm{P}\mathrm{r}\mathrm{o}\mathrm{d}\mathrm{u}\mathrm{c}\mathrm{e}{\mathrm{r}}^{\prime }\mathrm{s}\mathrm{p}\mathrm{r}\mathrm{i}\mathrm{c}\mathrm{e}}{\mathrm{R}\mathrm{e}\mathrm{t}\mathrm{a}\mathrm{i}\mathrm{l}\mathrm{e}{\mathrm{r}}^{\prime }\mathrm{s}\mathrm{p}\mathrm{r}\mathrm{i}\mathrm{c}\mathrm{e}}\times 100$

Income analysis

Income analysis was done by multiple regression analysis models and cross checking of variable was done by checking collinearity by calculating variance inflation factor (VIF) value.

$Ln\left(Y\right)=a+b_1{X}_1+b_2{X}_2+b_3{X}_3+b_4{X}_4+b_5{X}_5+b_6{X}_6+b_7{X}_7+b_8{X}_8+b_9{X}_9+b_{10}{X}_{10}+b_{11}{X}_{11}+b_{12}{X}_{12}+b_{13}{X}_{13}+b_{14}{X}_{14}$

where,

Y = Gross income from Lapsi (NRs)

X₁ = Age of household head

X₂ = Gender of Household (1 = Male, 0- Otherwise)

X₃ = Type of family (1 = Nuclear, 0 = Joint)

X₄ = Sorting (1 = Yes, 0 = Otherwise)

X₅ = Mada (1 = Yes, 0 = Otherwise)

X₆ = Training (1 = Yes, 0 = Otherwise)

X₇ = Attitude towards business (1 = Positive 0 = Otherwise)

X₈ = Female number in the family

X₉ = Family member involved in Lapsi production

X₁₀ = Literate population in a family

X₁₁ = Decision (1 = Male, 0 = Otherwise)

X₁₂ = Ethnicity (1 = Brahmin/Chhettri, 0 = Other)

X₁₃ = Type of house (1 = Cemented House, 0 = Otherwise)

X₁₄ = Experience in Lapsi production (years)

a = intercept b1, b2 … …, b14 = coefficient for respective variables

2.7. Problem ranking

Problem ranking provides an indication of ranking severity of a problem based on the response from the producer. Production and marketing problems were categorized into groups and ranking were provided and analyzed with scaling and ranking techniques.

$\mathrm{I}=\hspace{0.17em}\sum {\frac{SiFi}{N}}^{\hspace{0.17em}}$

where,

I = Index of severity

∑ = Summation

S_i = Scale value at i^th severity

F_i = Frequency of severity given by the respondents

N = Total number of respondents

Most severe = 1, High Severe = 0.8, Medium Severe = 0.6, Less severe = 0.4 and Least severe = 0.2

3. Results

3.1. Economic characteristics

The total landholdings and number of trees were found non-significant. Among the surveyed area, producer from Melamchi (7.68 Ropani or 0.39 ha) had higher average landholding than Indrawati (7.54 Ropani or 0.38 ha) and Chautara (6.88 Ropani or 0.35 ha). The average number of Lapsi trees was 4.21 which was found higher in Chautara (5.20) as compared in Indrawati (3.88) and Melamchi (3.54) (Table 1).

Table 1. Landholding, number of trees, production, and production per tree

Variable	Chautara (n = 50)	Indrawati (n = 50)	Melamchi (n = 50)	Overall mean (N = 150)	F value
Total landholding (Ropani)	6.88 (0.35 ha)	7.54 (0.38 ha)	7.68 (0.39 ha)	7.37 (0.37 ha)	0.490
No of Tree	5.20	3.88	3.54	4.21	1.441
Productions (Kg)	627.80	431.28	397.88	485.65	5.687***
Production per tree (Kg/Tree)	115.28	106.15	107.33	109.59	6.124***

*** indicates 1% level of significance, Figures in parenthesis denotes hectares, ha: hectares

The average production in the study area was 485.65 kg. Lapsi production was higher in Chautara (627.80 kg) among Indrawati (431.28 kg) and Melamchi (397.88 kg) which was statistically significant at 1% level. The average production was 109.59 kg/tree where higher production per tree was recorded in Chautara (115.28 kg/tree) as compared in Melamchi (107.33 kg/tree) and Indrawati (106.15 kg/tree) showing statistically significant at 1% level (Table 1).

The total income from Lapsi and Lapsi income per tree was found significant at 1% level. The average income of household in study area was NRs. 251,365.13 (2,094.71 USD) where higher household income was of producer from Chautara (NRs.287,895.00 equivalent to 2,399.12 USD) than the Melamchi (NRs.234670.00 equivalent to 1955.58 USD) and Indrawati (Rs.231,530.40 equivalent to 1,955.58 USD) with no significant differences (Table 2).

Table 2. Household income, Lapsi income and Lapsi income by geographical area category

Variable	Chautara (n = 50)	Indrawati (n = 50)	Melamchi (n = 50)	Overall mean (N = 150)	F value
Household income (NRs.)	287,895.00 (2399.125)	231,530.40 (1929.42)	234,670.00 (1955.58)	251,365.13 (2094.71)	1.641
Income from Lapsi (NRs.)	12,379.00 (103.15)	16,684.40 (139.03)	5004.00 (41.7)	11,355.80 (94.63)	12.977***
Income per tree from lapsi (NRs.)	2188.50 (18.23)	3732.02 (31.10)	1265.65 (10.55)	2395.38 (19.96)	30.881***
Livestock Standard Unit (LSU)	1.6917	1.6634	1.6390	1.6647	0.028

*** indicates 1% level of significance, figures in parenthesis indicates United States Dollars (USD), LSU is Livestock Standard Unit, 1 cattle/buffalo = 0.9 LSU, 1 goat/sheep = 0.1 LSU, 1 poultry = 0.015 LSU and 1 horse = 0.8 LSU

The average income from Lapsi in the study area was NRs.11,355.80 (equivalent to 94.63 USD). The Lapsi income represent the income received by selling the fresh Lapsi, Lapsi seed and Mada to the traders after retaining a portion of harvest for household consumption. The household level Lapsi income of the farmers of Indrawati was NRs.16,684.40 (equivalent to 139.03 USD) which was higher as compared to the household level Lapsi income of the farmers of Chautara (NRs.12,379.00 or 103.15 USD) and Melamchi (NRs. 5,004.00 or 41.7 USD). Average income per tree in the study area was NRs. 2395.38 (19.96 USD) per tree. Among the surveyed area higher Lapsi income per tree was highest in Indrawati (NRs.3732.02/tree) (31.10 USD) followed by Chautara (NRs. 2188.50/tree) and Melamchi (NRs. 1265.65 /tree; Table 2).

3.2. Livestock

Livestock is an important part of Nepalese agriculture. They are used as source of nutrition (milk, meat, eggs) for humans, organic fertilizer for plants (urine, dung), raw material (bone, skin) for industries and draft power for agriculture. The livestock standard Unit (LSU) was calculated by multiplying livestock with livestock standard coefficient. The average livestock standard unit of study area was 1.6647. The greater LSU was found in Chautara (1.6917) followed by Indrawati (1.6634) and Melamchi (1.6390) with no significance difference (Table 2).

3.3. Market and Price

The marketed volume and price of fresh Lapsi at farm-gate in study area were found significant. There were only 113 households involved in marketing at farm gate level. The average marketed volume at farm gate was 355.80 kg. Among the producers, highest average market volume was in Chautara (415.50 kg) followed by Melamchi (369.90 kg) and Indrawati (221.30 kg) which was significant at 5% level. This result is due to higher involvement of Indrawati in Mada production instead of selling fresh Lapsi (Table 3).

Table 3. Farm gate volume and price by geographical area

Variable	Chautara (n = 40)	Indrawati (n = 23)	Melamchi (n = 50)	Overall mean (N = 113)	F value
Fresh Lapsi
Farm gate Lapsi (Volume, kg)	415.50	221.30	369.90	355.80	3.093**
Farm gate Lapsi (Price, NRs.)	17.95 (0.15)	17.61 (0.15)	11.80 (0.10)	15.16 (0.13)	80.413***
Mada	Chautara (n = 3)	Indrawati (n = 30)	Melamchi (n = 0)	Overall mean (N = 33)	F value
Farm Gate Mada (Volume, kg)	91.67	112.47		110.58	0.244
Farm Gate Mada (Price, NRs.)	233.33 (1.94)	213.33 (1.78)		215.15 (1.79)	13.349***
Seed	Chautara (n = 1)	Indrawati (n = 20)	Melamchi (n = 0)	Overall mean (N = 21)	F value
Farm Gate Seed (Volume, kg)	60.00	144.50		140.48	1.997
Farm Gate Seed (Price, NRs.)	5 (0.04)	5.65 (0.047)		5.62 (0.046)	1.167

*** and **represent 1% and 5% levels of significance respectively, figures in parenthesis represents USD

In addition, average marketed price of fresh Lapsi in farm gate was NRs 15.16 per kg where Chautara (NRs 17.50/kg) received higher price followed by Indrawati (NRs 17.61/kg) and Melamchi (NRs 11.80/kg) which was significant at 1% level. The price of fresh Lapsi in Melamchi was very low as compared to other places due to involvement of wholesalers from outside the locality. In Chautara and Melamchi local traders were involved in trading (Table 3).

The volume of Mada at farm gate in study area was found non-significant. There were only thirty-three households (excluding Melamchi) involved in marketing of Mada at farm gate level. Average marketed volume of Mada at farm gate was 110.58 kg. Among the producer’s higher average market volume was in Indrawati (112.47 kg) than Chautara (91.67 kg; Table 3). In addition, the average marketed price of Mada in farm gate was NRs 215.15/kg (1.79 USD/kg). Interestingly, it was found that price of Mada in Indrawati, (NRs. 213.33/kg) (1.78 USD/kg) having high production, received less price as compared to Chautara (NRs.233.33/kg) (1.94 USD/kg) which was significant at 1% level. The price of Mada was higher in Chautara due good road facility which cost low transportation cost for marketing outside the district as well as presence of industry at the study area (Table 3).

The marketed volume and price of Lapsi seed (dried) at farm gate were found non-significant. There were only 21 households involved in marketing of Lapsi seed at farm gate level. The average marketed volume of Lapsi seed at farm gate was 140.48 kg. The producer’s highest average market volume was in Indrawati (144.50 kg) than Chautara (60 kg). In addition, average marketed farm gate price was NRs. 5.62/kg where Indrawati (NRs 5.65/kg) received higher price than Chautara (NRs 5/kg; Table 3).

The average volume of fresh Lapsi sold to consumer was 6.67 kg at an average price of NRs. 21.67 (0.18 USD/kg). Chautara (7.5 kg) and Indrawati (5 kg) were only involved in Lapsi trading at average rate of NRs. 22.50/kg (0.187 USD/kg) and NRs.20/kg (0.16 USD/kg) respectively (Table 4).

Table 4. Marketing by volume and price in different geographical area

Variable	Chautara (n = 2)	Indrawati (n = 1)	Melamchi (n = 0)	Overall mean (N = 3)
Consumers (Fresh Lapsi)
Volume, kg	7.50	5.00		6.67
Price, NRs.	22.50 (0.187)	20.00 (0.16)		21.67 (0.18)
Local traders (Fresh Lapsi)	Chautara (n = 40)	Indrawati (n = 23)	Melamchi (n = 0)	Overall mean (N = 63)
Volume, kg	415.13	221.08		344.28
Price, NRs.	17.95 (0.15)	17.82 (0.15)		17.90 (0.15)
Local traders (Mada)	Chautara (n = 3)	Indrawati (n = 30)	Melamchi (n = 0)	Overall mean (N = 33)
Volume, kg	91.67	112.47		110.58
Price, NRs.	233.33 (1.94)	213.33 (1.77)		215.15 (1.79)
Middlemen (Fresh Lapsi)	Chautara (n = 3)	Indrawati (n = 0)	Melamchi (n = 50)	Overall mean (N = 53)
Volume, kg	423.33		369.90	372.92
Price, NRs.	18.33 (0.15)		11.80 (0.10)	12.17 (0.10)

Figures in parenthesis represents USD, 1 USD = 120NRs.

The Chautara and Indrawati were involved in trading fresh Lapsi with local trader. The average volume of fresh Lapsi sold to local traders was 344.28 kg at average price NRs.17.90/kg (0.15 USD/kg). Average Lapsi volume and rate was higher in Chautara (415.13 kg at NRs.17.95/kg (0.15 USD/kg)) as compared to Indrawati (221.08 kg at NRs.17.82/kg(0.15USD/kg); Table 4).

There were only thirty-three households (excluding Melamchi) involved in marketing of Mada with local trader. The average marketed volume of Mada with local trader was 110.8 kg. Between producers, highest average market volume was found in Indrawati (112.47 kg) than Chautara (91.67 kg; Table 4).

In addition, average marketed price of Mada with local trader was NRs. 215.15/kg (1.79 USD/kg). Reversely the volume of Chautara (NRs.233.33/kg) (1.94 USD/kg) received higher price than Indrawati (NRs. 213.33/kg) (1.77 USD/kg. The price of Mada was higher in Chautara due good road facility reducing transportation cost. It is added by the presence of industry at the study area (Table 4).

The involvement of middlemen was found only in two places of study area i.e. Chautara and Melamchi. The producer from Chautara (423.33 kg at NRs.18.33 /kg) sold greater volume at higher price than Melamchi (369.90 kg at NRs. 11.80 /kg; Table 4).

The price was set by the trader most of the time (95.30%). The farmers from Melamchi were not involved in price setting where farmers from Chautara (8%) and Indrawati (6%) had minimal involvement. 86.7% of farmers were satisfied with price while 5.3% show neutral response and 8% were dissatisfied with price they got by selling Lapsi and Lapsi product. Among the study area, farmers from Chautara (94%) responded to satisfaction followed by Indrawati (84%) and Melamchi (82%). In addition, higher dissatisfaction was found in Melamchi (10%) but there was no significant difference on the level of price satisfaction (Table 5).

Table 5. Price setting and price satisfaction by geographical area category

Variable	Chautara (n = 50)	Indrawati (n = 50)	Melamchi (n = 50)	Total (N = 150)	Chi square value
Who sets price?
Farmers themselves	4 (8)	3 (6)	0 (0)	7 (4.7)	3.896
Traders	46 (92)	47 (94)	50 (100)	143 (95.3)
Price satisfaction?
Satisfied	47 (94)	42 (84)	41 (82)	130 (86.7)	4.977
Neutral	0 (0)	4 (8)	4 (8)	8 (5.3)
Dissatisfied	3 (6)	4 (8)	5 (10)	12 (8)

Value in parentheses indicate percentage

Table 6 reflects the market margin from fresh Lapsi was found highest in Melamchi (NRs. 44.96/kg) followed by Indrawati (NRs.38.43/kg) and Chautara (NRs.38.11/kg) respectively with average market margin NRs. 40.88/kg. The producer’s share was found highest in Chautara (32.03%) followed by Indrawati (31.42%) and Melamchi (21.05%) respectively with an average of 27.05%.

Table 6. Market margin and producer’s share by geographical area

Place	Market Margin (NRs/kg)	Producer’s share (%)
Chautara	38.11 (0.32)	32.03
Indrawati	38.43 (0.32)	31.42
Melamchi	44.96 (0.37)	21.05
Average	40.88 (0.34)	27.05

Figures in parenthesis indicate the values in United States Dollar

The result demonstrated that majority (68.7%) of households were deprived from training. Among the three-study area, greater number of farmers from Indrawati (46%) received training followed by Chautara (26%) and Melamchi (22%) and result was statistically significant at 5%. 71.3% of surveyed households have access to credit and finance. 76% of households were engaged in social group or organization including higher percentage in Chautara (84%) followed by Melamchi (70%) and Indrawati (68%) with no significance difference. 81.3% of the respondent had knowledge on agri-insurance where highest percentage was recorded in Chautara (88 %) followed by Indrawati (80 %) and Melamchi (76%) with no significant difference (Table 7).

Table 7. Social characteristics of households by geographical area category

	Chautara (n = 50)	Indrawati (n = 50)	Melamchi (n = 50)	Total (N = 150)	Chi square value
Training received	13 (26)	23 (46)	11 (22)	47 (31.3)	7.684**
Access to credit/finance	34 (68)	38 (76)	35 (70)	107 (71.3)	0.848
Membership status in social group or organization	42 (84)	34 (68)	38 (76)	114 (76)	3.509
Knowledge about agri-insurance	44 (88)	40 (80)	38 (76)	122 (81.3)	2.459

Figures in parentheses indicate percentage, **represents 5% level of significance

3.4. Framework variables

Table 8 explains the house type, cooking system, decision on production and marketing, climate change effect and willing to extension of business by study area category. The willingness to extend the business was statistically significant at the 1% level. 58.7% of household show positive response in extending business where highest response was in Chautara (72%) followed by Indrawati (62%) and Melamchi (58.7%). The highest percentage was recorded in Chautara due to access of good road, market and price as compared to other study areas. A study showed that 53.3% of houses in the study area was traditional. Every house had access to electricity. Most households used both gas and firewood (68.7%) for cooking while 12.7% used gas and 18.7% used firewood. Study revealed that both male and female (69%) were involved in decision making on their household followed by male only (38%) and female only (16%). In addition, most farmers responded that there was no impact of climate change in agriculture while 28% responded the impact of climate change.

Table 8. House type, electricity access, cooking system, decision, climate change effect and interested in extending by geographical area category

Variable	Chautara (n = 50)	Indrawati (n = 50)	Melamchi (n = 50)	Total (N = 150)	Chi square value
House Type
Traditional Mud House	23 (46)	29 (58)	28 56)	80 (53.3)	1.661
Cemented Modern House	27 (54)	21 (42)	22 (44)	70 (46.7)
Electricity Access	50 (100)	50 (100)	50 (100)	150 (100)
Cooking System
Gas	8 (16)	7 (14)	4 (8)	19 (12.7)	2.549
Firehood	7 (14)	11 (22)	10 (20)	28 (18.7)
Both	35 (70)	32 (64)	36 (72)	103 (68.7)
Decision Maker of household
Male	20 (40)	18 (36)	19 (38)	57 (38)	1.464
Female	10 (20)	7 (14)	7 (14)	24 (16)
Both	20 (40)	25 (50)	24 (48)	69 (46)
Climate Change Effect	15 (30)	14 (28)	13 (26)	42 (28)	0.198
Interested in extending business	36 (72)	31 (62)	21 (42)	88 (58.7)	9.622***

Figures in parentheses indicate percentage, *** represents the 1% level of significance.

3.5. Cost analysis

Table 9 presents the average variable cost per tree, average variable cost per production, total cost per tree, total cost per production, and benefit cost ratio (including Mada) in the study area. All the variables were statistically significant. The average variable cost was NRs. 523.54 (USD 4.36) which included cost for harvesting, transportation annually per single tree. Among the average variable cost was greater in Chautara (NRs. 600.34 equivalent to USD 5.0) followed by Melamchi (NRs. 497.49 equivalent to USD 4.15) and Indrawati (NRs. 472.81 equivalent to USD 3.94). This result was due to high labor cost at Chautara. Similarly, total cost per tree (NRs. 531.67 equivalent to USD 4.43) shows a similar response as variable cost. The variable cost and total cost per production was NRs. 4.93/kg (USD 0.041/kg) and NRs. 5/kg (USD 0.041/kg) respectively. Lastly, Benefit cost ratio was found 2.79. Benefit cost (BC) ratio is the ratio of income from Lapsi divided by total cost of production including post-harvest cost at household level. Among the three-study area BC ratio was greater in Indrawati (2.82) followed by Melamchi (2.38) and Chautara (3.16).

Table 9. Variable cost per tree, total cost per tree, variable cost per production, total cost per production and benefit cost ratio by geographical area category

Variable	Chautara (n = 50)	Indrawati (n = 50)	Melamchi (n = 50)	Overall mean (N = 150)	F value
Variable cost per tree	600.34 (5.00)	472.81 (3.94)	497.49 (4.15)	523.54 (4.36)	13.561***
Total cost per tree	608.54 (5.07)	480.75 (4.00)	505.73 (4.21)	531.67 (4.43)	13.269***
Variable cost per production	5.31 (0.04)	4.68 (0.039)	4.80 (0.04)	4.93 (0.04)	3.961**
Total cost per production	5.38 (0.04)	4.76 (0.039)	4.89 (0.04)	5.00 (0.04)	3.856**
Benefit Cost ratio (including Mada at HH level)	3.16	2.82	2.38	2.79	10.16***

Note: *** and ** represent the 1% and 5% levels of significance respectively, Figures in parenthesis represents USD

3.6. Production problem

Table 10 reflects major problems faced by producers. It includes nonbearing fruit trees followed by lack of improved production technology and low production (except for Chautara). But in Chautara, the major problem was identified as high labor cost due to shortage of labor. Non-bearing trees was identified as major problem of Lapsi producer too (Poudel, 2003). Lapsi is a unique fruit crop and technology advancement is a major problem. Harvesting and post production process are labor intensive and appropriate technology had not been developed and not much found on other countries.

Table 10. Lapsi production problem by geographical area category

Problems	Chautara (n = 50)		Indrawati (n = 50)			Melamchi (n = 50)			Sindhupalchok (N = 150)
	Index	Rank	Index		Rank	Index		Rank	Index	Rank
Non-bearing fruit	0.928	I	0.836		I	0.908		I	0.89	I
Lack of improved Production technology	0.728	II	0.804		II	0.7		II	0.74	II
High labor cost	0.632	III	0.32		IV	0.44		IV	0.46	IV
Low production	0.364	IV	0.752		III	0.544		III	0.55	III
Lack Financial support	0.348	V	0.288		V	0.408		V	0.34	V

Note: Problems were ranked from 1 to 0.2 in 5-point scale based on severity (1 = very highly severe, 0.8 = highly severe, 0.6 = moderately severe, 0.4 = less severe, 0.2 = very less severe) identified on household survey.

3.7. Marketing problem

Table 11 reflects the major problem for marketing faced by producers. It includes low bargaining capacity followed by low selling price, lack of market and collection center. Among the studied geographical area, low bargaining capacity was major common problem. Lack of market and collection center was identified as second major problem in Chautara, and low selling price was second major problem in Indrawati and Melamchi. Likewise, lack of market and collection center was found as third major problem in Melamchi whereas low selling price was third major problem in Chautara.

Table 11. Marketing-related problems of Lapsi by geographical area category

Problems	Chautara (n = 50)	Indrawati (n = 50)		Melamchi (n = 50)		Sindhupalchok (N = 150)
	Index Rank	Index Rank		Index Rank		Index	Rank
Low bargaining capacity	0.904 I	0.852	I	0.884	I	0.88	I
Lack of market and collection center	0.736 II	0.352	IV	0.544	III	0.544	III
Low selling price	0.604 III	0.776	II	0.712	II	0.697333	II
Transportation problem	0.384 IV	0.724	III	0.444	IV	0.517333	IV
Certification problem	0.372 V	0.296	V	0.416	V	0.361333	V

Marketing problems were ranked from 1 to 0.2 in 5-point scale based on severity (1 = very highly severe, 0.8 = highly severe, 0.6 = moderately severe, 0.4 = less severe, 0.2 = very less severe) identified on household survey.

3.8. Factors affecting income from Lapsi sub-sector

In Table 12, value of R² reflects that about 68% of the variation in an income from Lapsi was explained by the explanatory variable in the model. The value of adjusted R² represent that when degree of freedom is considered about 65% of the variation in the income is explained by the independent variables in the model. Statistically significant F value indicated that the explanatory variables included in the model were important for explanation of the variation in the dependent variable. The mean VIF value was found 1.66 was less than the recommended value i.e., 10 at the maximum level. Therefore, there was low multi-collinearity in the model.

Table 12. Determinants of income from Lapsi in study area using income function model

Variables	Coefficients	Standard Error	t value	p > |t|
Age of HHH (years)	−0.002	0.005	−0.45	0.656
Gender^a	−0.321**	0.143	−2.250	0.026
Family type^a	−0.195	0.158	−1.24	0.218
Sorting^a	0.026	0.134	0.200	0.843
Mada^a	1.289***	0.163	7.870	0.000
Training^a	0.103	0.140	0.730	0.465
Attitude towards Lapsi business^a	0.399***	0.113	3.500	0.001
Female number in a family	0.026	0.068	0.390	0.697
Members involved in a Lapsi production	0.147*	0.076	1.940	0.055
No of literate member in a family	0.098***	0.035	2.780	0.006
Decision^a	0.360***	0.120	3.000	0.003
Ethinicity^a	0.208	0.141	1.480	0.142
Type of House	0.237*	0.134	1.770	0.080
Experience of Lapsi farming (years)	0.081***	0.015	5.130	0.000
Constant	6.909***	0.394	17.510	0.000
Observation	150
F (14,135)	20.93***
R^2	0.6845
Adj. R^2	0.652

HHH: Household head, ***, ** and * represent the 1%, 5% and 10% levels of significance, respectively.““a”” indicate dummy variable, R²: Coefficient of determination, t: hypothesis test statistic, p: probability value

It describes that having male as a household head, the Lapsi income of the family decreases by about 32.10% as compared to the female headed household which is statistically significant at 5% level. The reason may be due to majority of migration of male members to different countries in search of employment opportunities. Likewise, in case of family types affecting the income from Lapsi production, income for one with nuclear family decreases by 19.5 % as compared to one with joint family because more individuals are involved and giving more time for the Lapsi production. Similarly, if the households are involved in sorting, Lapsi income was increased by 2.60 % which was not significant. Household which prepared Mada resulted in increased income by 128% which was statistically significant at 1% level. The reason for this can be due to value addition product (Mada) having higher product price. Household who participated in training showed that income from Lapsi was increased by 10.30% which was not statistically significant. Similarly, individuals with positive determination to extend the Lapsi business resulted in increase of 39.92% of Lapsi income. It was significant at 1% level of significance. The reason may be due to their determination, continuous effort, and never-ending enthusiasm. Moreover, it was found that involvement of more female number in household yielded 2.6% increase in income. As the number of members increase in the Lapsi farming, the income increases due to proper care and management by the higher number of family member which was found to be statistically significant at 10% level. If the family had higher number of literate members, the income from Lapsi increased by 9.8% which may be due to technical knowledge. Male dominated farms significantly increase the income by 36%. The reason may be due to better field and decision-making experiences. Family living in the cemented house was found to have increased income by 23% than family with traditional mud house which was significant at 1% level. The reason may be due to economically strong background and ability to invest more in Lapsi production. Farmer with more experience of Lapsi farming was found that 8.13% increase in Lapsi income which is significant at 1% level. The good experience leads to good knowledge, better decision making and technically sound individual. The constant is also statistically significant at 1% level.

4. Discussion

The result from the household survey of 150 respondents at Sindhupalchok supported some of the alternative hypothesis. The results show the significant difference in household (HH) age and family size of HH among the geographic area. Both HH age and family size were higher in Chautara followed by Melamchi and Indrawati respectively. Result was different from the annual household survey (CBS, 2017) which might be due to better health facility causing low mortality rate in Chautara and Melamchi Municipalities than Indrawati Rural Municipality with an assumption that municipalities have more access to health facilities than rural municipalities. Economically active member was higher in Municipalities than Rural Municipality which show similar result as annual household survey (CBS, 2017). Similarly, dependency ratio was also higher in Chautara among the Melamchi and Indrawati respectively with different in result (CBS, 2017) where dependency ratio was 59% for urban and 70 % for rural. Although the study area was municipalities, but family size was higher as there were more dependent population in municipalities. It might be due to better health facilities with low mortality resulting in higher life span. In addition, literacy percent was also highest in Chautara followed by Melamchi and lowest in Indrawati which may be due to good education facility in Chautara and Melamchi being municipality than Indrawati being rural municipality.

Majority of household head (HHH) were male (79.3%) than female (20.7%) which show almost similar trend as annual household survey where male HHH and female HHH were 75.20% and 24.8% respectively (CBS, 2017).This is due to Nepal is male dominated society (Bhadra & Shah, 2007). Majority surveyed household were Aadibasi/Janajati (Tamang) followed by Brahmin/ Chhetri and very few were Dalit. The result was in line with the District Agriculture Development Office (DADO) profile. Maximum age of tree noticed was 30 years old in study area and was in consistent with the past report (Gautam, 2004).

The average production per tree was 109.59 kg/tree where higher production per tree was in Chautara (115.28 kg/tree) as compared in Melamchi (107.33 kg/tree) and Indrawati (106.15 kg/tree). The result was due the better management practice and availability resources in for production in Chautara though the average production was found almost half of the production per tree as reported in past studies in Parbhat district (Seber, 2016). Low Production/tree in Sindhupalchok might be due to loss during harvesting and poor package of practice.

The average income per tree in Sindhupalchok (NRs 2,395.38) (USD 19.96) was also found low compared to Parbat where the reported average income per tree was NRs. 3150 (USD 26.25; Gautam, 1997; Seber, 2016) which may be due to low bargaining capacity. Income per tree and total income from Lapsi in Indrawati was highest among rest two and which was due to most of farmers involved in Mada preparation. While income per tree was lowest in Melamchi due lowest price of fresh Lapsi as well as no involvement in Mada preparation. Producer used to sell majority of fresh Lapsi, processed Mada, and seed(dried) from farm gate to local level trader within from village, middlemen outside the village and industry.

Average volume of fresh Lapsi they trade to and from farm and price was statistically different where price was almost same in Chautara and Indrawati due to involvement of local trader while minimal in Melamchi due to involvement of middlemen. The average price was NRs.15.16/kg higher than NRs.7.35/kg (USD 0.06; Gautam, 1997) almost same as reported in previous finding i.e. NRs.13.95/kg (Paudel, 2012) and NRs.15/kg (USD 0.125; Seber, 2016).

The average price of Mada at farm gate level was NRs. 215.15/kg (USD 1.79) which was slightly higher in the Chautara as compared to Indrawati. Higher price was in Chautara may be due low transportation cost in Chautara as being headquarters of the district with good road facility as compared to Indrawati while there is no involvement of producers from Melamchi in Mada preparation. Some of the producer from Chautara sold the fresh Lapsi directly to industry at NRs. 22/kg (USD 0.1833/kg) which was in an average 4 km far from production site by including transportation cost.

Average variable cost per tree, average variable cost per kg, total cost per tree and total cost per kg Lapsi along with BC ratio was found highest in Chautara as compared to Indrawati and Melamchi. Variable cost includes harvesting, transportation and HH level processing cost while total cost includes transplanting, organic manure, pit digging cost and variable cost. Cost of production was highest in Chautara because of processing cost and high labor cost. But, Chautara had more selling price resulting in higher BC ratio. Although the Melamchi did not involve in Lapsi processing, but cost of production was higher than Indrawati due to high labor cost. It was surprising to found that BC ratio (2.79) was slightly lower than the previous research i.e., 3.1 in Bhaktapur district (Joshi, 2016). The result might be due to distance from the major market.

5. Conclusion

The major factors affecting the income from Lapsi at household level were found to be gender of household, value addition (Mada), attitude towards Lapsi business, members involve in Lapsi production, literate member in family, decision, type of house and experiences. Although Lapsi cultivation practice and value-addition practice in the study area was by locally available indigenous knowledge, the BC ratio obtained from analysis was more than two. Hence, it can be grown as cash crop for promoting economy of rural people in hilly region which can be accelerated by the further use of new technology and knowledge. The major problem identified in Lapsi cultivation was difficulty to identify the sex of Lapsi tree in early stage and lack of improved production technology. Similarly, the major problem identified in marketing was low bargaining power of producer and low selling price. Further, a collaborative effort from all the concerned stakeholders is expected to uplift the farming situation of Lapsi in Nepal.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Funding

The authors received no direct funding for this research.

Notes on contributors

Santosh Adhikari

Santosh Adhikari Department of Agricultural Economics and Agribusiness Management, Agriculture and Forestry University, Chitwan, Nepal

Subodh Raj Pandey

Subodh Raj Pandey Department of Agriculture Sciences, Texas State University, Round Rock United States

Padam Lal Bhandari

Padam Lal Bhandari Department of Agricultural Economics and Agribusiness Management, Agriculture and Forestry University, Chitwan, Nepal

Rishi Ram Kattel

Rishi Ram Kattel Department of Agricultural Economics and Agribusiness Management, Agriculture and Forestry University, Chitwan, Nepal

Shiva Chandra Dhakal

Shiva Chandra Dhakal Department of Agricultural Economics and Agribusiness Management, Agriculture and Forestry University, Chitwan, Nepal