Livestock Research for Rural Development 27 (5) 2015 Guide for preparation of papers LRRD Newsletter

Citation of this paper

A comparative economic analysis of goat production systems in Jordan with an emphasis on water use

J Al-Khaza’leh, C Reiber1, R Al Baqain1 and A Valle Zárate1

Faculty of Agricultural Technology, Al-Balqa' Applied University, P.O. Box 19117, Al-Salt, Jordan,
jkhazaleh1981@yahoo.com
1 Institute of Animal Production in the Tropics and Subtropics, University of Hohenheim, Garbenstr. 17, 70599 Stuttgart, Germany

Abstract

Goat production is an integral part of farming systems in Jordan and can play an important role in poverty alleviation and improved food security in rural households. Water provision is a crucial input for animal productivity. This study aims to assess the economic performance of goat production in different production systems in southern Jordan, with an emphasis on water use. Data were collected from a total of 120 purposely selected goat keepers from June to October 2012 using a survey based on structured questionnaires. Indicators for the economic performance of goat production were gross margin (GM), benefit-cost ratio (BCR) and net benefit (NB).

Results showed that goats were mainly kept to generate cash income from sales of surplus milk and live animals. In both systems (n=114 farmers), feed shared the highest costs (75%) of the total variable costs, followed by transportation of water and feed (12%), labor (9%), veterinary services (4%) and water (0.2%). Total variable costs per flock were significantly (P<0.05) higher in the transhumant than in the sedentary system. The significantly longer distance associated with high fuel and labor costs to fetch water from water source sites in the transhumant system resulted in a significant (P<0.05) higher cost of transportation per flock compared to the sedentary system. The economic performance parameters were similar between production systems. A 28% of farmers in the transhumant system and 11% of farmers in the sedentary system had a negative NB. Production system, flock size, fertility rate, feeding intensity and productivity rate showed a significant (P<0.05) impact on GM1, while only flock size and fertility rate had a significant (P<0.05) impact on GM2 and NB. Further investigations of effects of water use on the profitability and thus on the sustainability of goat farming under more extensive production systems are recommended. Moreover, improvement of goat reproductive management and flock health are required to improve economic efficiency of goat production.

Keywords: economic performance, goat farming, sedentary, transhumant


Introduction

Goat production plays an important role for the livelihoods of farmers in developing countries of subtropical regions such as Jordan. Goats contribute in economy, poverty alleviation and food security of rural households in terms of meat, milk, income, capital storage, savings, an insurance against emergencies and serving cultural purposes (Lebbie 2004; Al-Atiyat and Tabbaa 2009; Al-Atiyat 2014). Goats disseminated all over the world due to their unique characteristics, such as their ability to withstand heat stress, endure prolonged water deprivation and adapt to adverse climatic and geophysical conditions where cattle and sheep can hardly survive (Abdel Aziz 2010). Nevertheless, goat production is often constrained by direct and collateral effects of water and feed shortages (Mbuku et al 2014).

In Jordan, small ruminants are raised under different production systems which were classified based on the degree of herd movement into sedentary, transhumant and nomadic systems (Abu-Zanat et al 2005), and which are associated with varying levels of input and intensity of production (Qtaishat et al 2012). With a population of about 884,864 goats in 2012 (Ministry of Agriculture 2012), they are second behind sheep within Jordan’s agricultural economy. Grazing conditions, rainfall, consumer demand, and livestock imports from neighbouring countries were found to be the major factors that directly affected the population of livestock, especially small ruminants (Al-Assaf 2012).

The value of sheep and goat production was estimated to be 361,622 thousand JD in 2012, which was about 34% of the total livestock production in the country (Department of Statistics 2012a). The share of goats within the total red meat and milk yield of Jordan was 24.8% and 9.6%, respectively (Ministry of Agriculture 2012). This reflects the important role of goats for food security, especially in rural areas.

The economic performance of goat production is an indicator for the livelihood of households (Al Baqain and Valle Zárate 2011). Water provision is the most crucial input for animal productivity. The associated costs of labor and fuel for water fetching and supply for livestock could have a significant impact on the profitability and thus on the sustainability of livestock farming. The economic implications of water use for goat production systems in Jordan have not been investigated. Therefore, this study aims to assess the economic performance of goat production and to identify factors that influence the economic performance in different production systems in the Karak Governorate of Jordan.


Materials and methods

The study area

The study was carried out in the Karak Governorate, located 120 km south-west of Amman, the capital city of Jordan. The estimated human population of the Karak Governorate at the end of 2012 was 249,100 with an area of 3495 km² and a population density of 71.3 (Department of Statistics 2012b). The Karak Governorate is subdivided into 7 administrative units located in different agro-ecological zones, ranging from the rift valley in the west, highlands in the central region, and a semi-desert in the eastern and southeastern regions. The study area has a diversified climate characterized by a long and hot dry season from April to October and a rainy wet season which may extend from November to March. The average temperature ranges from about 4°C in winter to more than 32°C in summer. The temperature increases by between 3 to 7°C towards the south and east, with the exception of some southern highlands (Department of Meteorology 2011).

The study area was subdivided into two agro-ecological zones: the mountain zone (MZ) and the semi-desert zone (SDZ). The purposive selection of the study area was justified to encompass dry climatic conditions and in order to assess the effect of water availability on the performance of different goat production systems. The average annual rainfall was 284 mm for Rabbah (MZ) and 76 mm for Qatraneh (SDZ) stations, during the years 2010-2012 (rainfall data obtained from the Meteorological Department of Jordan). In general, the study area is characterized by availability of different water sources and their distribution varies between zones (Al-Khaza’leh et al 2015).

Data collection and sampling procedure
Questionnaire survey

A survey was conducted in the study area from beginning of June to end of October 2012. A multistage sampling technique was applied to select goat keepers for the interview. A list of all goat keepers in each administrative unit (Liwa) of the study area was obtained from officials of the Ministry of Agriculture. Goat keepers with less than 5 adult goats were excluded from the list to ensure coverage of the performance of the goats per farm. From each agro-ecological zone, two administrative units were purposively selected at the first stage based on the proportional contribution of the number of goat keepers in each zone to the total number for all the zones. By taking into account the available logistics and time, three villages were randomly selected from each of the four purposively selected administrative units, resulting in a total of 12 villages at the second stage. At the third stage, systematic random sampling was used to select goat keepers for interviewing from the list of goat keepers compiled from each village. The number of goat keepers selected was proportional to the contribution of the number of goat keepers in the unit to the total for all the surveyed units (Table 1). Farmers who did not have time or were unwilling to take part in the interview were replaced with the next candidate farmer on the list.

Table 1. Distribution of the population and number of goat-keeping households sampled across administrative units within each agro-ecological zone
Agro-ecologicalzone Administrative Unit
(Liwa)
Population of
goat keepers
Percent of
total
Number sampled
Mountain Qasabat Al-Karak 609 28 34
Mountain Al-Qasr 444 21 25
Semi-desert Al-Qatraneh 317 15 18
Semi-desert Al-Mazar Al-Janubiy 777 36 43
Total 2147 100 120
Note: data obtained from Ministry of Agriculture offices in the study area, 2012.

Before conducting the full scale survey, a pre-test was undertaken with 4 farmers, and adjustments in the questionnaire were made accordingly. The structured questionnaire contained questions regarding socio-economic characteristics of households, flock management and dynamics, water accessibility and utilization, animal productivity, input and output parameters and annual production costs and revenues. Data was collected from farmers by asking them to recall information retrospectively from the previous year (from June 2011 to June 2012).

Three goat production systems were identified in the study area, namely nomadic, sedentary and transhumant (semi-sedentary) systems that were distinguished mainly based on the degree of herd movement. The majority of farmers selected for the study practice the transhumant production system (Table 2). The transhumant farmers have a permanent base and move seasonally to take advantage of grazing on the vegetation during late winter and early spring and grazing on stubbles, barley, or wheat crop failures in cultivated areas during summer. In the sedentary (agro-pastoral) systems, animal movement is restricted to certain distances around the farm base and is characterized by more supplementary feeding and greater access to the water. The nomadic production systems of the present study characterized by high movement, less access to water, less supplementary feeding, and longer grazing were omitted from statistical analyses due to small sample size (5% of the total sample size).

Table 2. Distribution of the number of sampled households across production systems and agro-ecological zones
Production system Agro-ecological zone
Mountain Semi-desert Total
Sedentary 20 18 38
Transhumant 36 40 76
Nomadic 3 3 6
Total 59 61 120
Economic performance of goat production

The economic performance of goat production was evaluated on the basis of gross margin (GM), net benefit (NB) and benefit-cost ratio (BCR) referring to the goat species only, irrespective of whether the flock composition of household was mixed of sheep and goats. All monetary values of inputs and outputs are given in Jordanian Dinar (JD) with the exchange rate of 1 JD = 1.4 USD at the time of the study (year 2012). The total costs and total revenues were calculated on yearly basis (from June 2011 to June 2012) and seasonal and annual variations in input availability and prices were considered. The costs, revenues and economic parameters per flock were divided by the number of goats in the flock to calculate the costs, revenues and economic parameter per head.

The economic performance was calculated from the following equations 1-5:

GM1 = CR – VC (1)
GM2 = (CR+NV) – VC (2)
BCR1 = CR/VC (3)
BCR2 = (CR+NV)/VC (4)
NB = (CR+NV+BC+BI) – (VC+IR) (5)

Where:
CR is the total cash revenue
VC is the total variable cost
NV is the non-marketable value of production
BC is the benefit of capital
BI is the benefit of insurance
IR is the value of the interest rate on the fixed capital
(adapted equations from sources: Ayalew et al 2003; Lemke et al 2007; Ogola et al 2010; Al Baqain and Valle Zárate 2011).

The cash revenues calculated included the sale of animals and sale of milk and milk products. The non-marketable value of production comprised the values from meat and milk consumed by households or used for social and religious festivities. The monetary value of milk sold was calculated by multiplying the prevailing price of milk by the amount of milk (kg) produced per day and farm from goats milked after weaning, and by the respective number of days in lactation. The monetary value from the sale of milk products was calculated with the respective price multiplied by the amount sold during the year. The monetary values of manure and hair were not included on the revenue side because they were not marketed and only used by few farmers. The farm gate price for a unit of each product was used for the calculation of the home consumption.

Benefit of capital (BCg) was calculated as indicated in equation 6 below:

BCg = NGg x f (6)
(adapted equation from source: Al Baqain and Valle Zárate 2011).

Where:
NGg is the monetary value of number of goats based on the respective farm gate prices of gth flock.
f is the financing factor (3.8%) (Central Bank of Jordan 2012).

Benefit of insurance (BIg) was based on the goat flock dynamics during the one year period and was calculated from the equation 7 below:

BIg = AFg x I (7)
(adapted equation from source: Al Baqain and Valle Zárate 2011).

Where:
AFg is the monetary value of the average of gth flock.
I is the insurance factor (3.8%) (Central Bank of Jordan 2012).

The variable costs included were feed (purchased feed and fees for stubble grazing), water, veterinary service, transportation and hired labor costs. The cost of purchased feed was calculated by multiplying the feed-use per day (kg) by the number of days of feeding, and the prevailing price paid by the farmers. The government sold feed to farmers at subsidized rate e.g. equivalent to about 10 kg of barley per head of goat per month. The fee of grazing on stubbles, barley or wheat crop failures varied depending on the size of rented area and quality of crop. The fee of grazing on stubble was calculated by multiplying the fee per 1 dunum (du) (1000 m²) by total size of rented area, based on the proportion of goats in the flock (assessed by the ratio of goat number to total number of animals). The cost of water was calculated by multiplying the seasonal varying quantity of daily water consumption (L) by the respective price per M³ of water paid by the farmers.

Information on veterinary costs (based on charges for treatment and vaccination of animals by the veterinarian) and on transportation costs including transporting of water, feed and goats to market was given directly by the farmers. Hired labor costs were calculated as follows: monthly payment x number of persons hired x months of hiring x the ratio of goat number to the total flock number. The interest rate of annual costs of housing and machinery or equipment was added to the fixed cost.

Statistical analysis

Descriptive statistics were used for the analysis of continuous and categorical variables to summarize the socio-economic characteristics of households and flock ownership. The variable costs, revenues and economic parameters data did not follow normal distribution. Therefore, the nonparametric Wilcoxon-Mann-Whitney test was used to test presence of significant differences of the mean values of variables between transhumant and sedentary systems. Indices were calculated to provide an overall ranking of purposes of keeping goat breeds. Ranks were based on the choices of priority characteristics (i.e. 5= highest importance) according to the formula: index = sum of [5 for rank 1+ 4 for rank 2+,… , + 1 for rank 5 for specific purposes] divided by the sum [5 for rank 1+ 4 for rank 2, etc.] for all purposes of keeping goats.

The general linear model (GLM) in SAS 9.3 (SAS Institute 2012) was used to evaluate the effect of the different explanatory variables on the economic success, i.e. GM and NB of goat production. The data were tested for normality and homogeneity of error variances prior to model fitting. Outliers with high deviation from the mean were removed for the analysis of GM1, GM2 and NB (n = 4 outliers), to get a normal distribution for the residuals. The main effect of the two agro-ecological zones was included in the model to account for any potential climatic differences between the two zones. Moreover, the type of production system (sedentary and transhumant) was included in the model to account for potential differences between systems in e.g. transportation costs. The economic success was also assumed to be influenced by goat breed composition of flock (through, e.g., differences between goat breeds in their performance). Crop residue use was regarded as a proxy for access to grazing in cultivated areas. Feeding intensity (in kg per head per day) was used as an indicator of feed intake. The watering frequency per day over the year was used to index the level of water availability and intake. The age of the household head was used as proxy for experience of farmers in goat farming and management. To estimate reproduction efficiency and health management of flock the following definitions were used: flock fertility rate was calculated as the number of does kidding at least once divided by the number of does in the flock during the year and expressed as percentage, while the flock productivity rate was calculated as the number of kids born alive divided by the does kidding during the year. The correlation between fertility rate and productivity rate was not significant (P>0.05) and rather low (r = 0.15). Therefore, they were included into the model simultaneously.

The full model used for GM1, GM2 and NB was:
yijklmnop = µ + Zi + Sj + Lk + Bl + Cm + Fn + Wo + a + h + f + p + eijklmnop (model 1)

where; yijklmnop is the GM and NB per flock, µ is the general intercept, Zi is the effect of agro-ecological zone (i= mountain, semi-desert), Sj is the effect of the production system (j= sedentary, transhumant), Lk is the effect of hired labor use (k= 0, 1), Bl is the effect of goat breed composition of flock (l= all Mountain Black, ≥50% Mountain Black, <50% Mountain Black), Cm is the effect of crop residue use (m= 0, 1), Fn is the effect of feeding intensity in kg per head per day (n = ≤1kg, >1kg), Wo is the index of watering frequency over the year (o= ≤2 times, >2 times), a is the age of household head in years, h is the effect of goat flock size treated as a continuous covariate, f is the fertility rate treated as a continuous covariate, p is productivity rate treated as a continuous covariate, and eijklmnop is the residual error, assumed to be normally distributed, N (0, δ²e).

All possible two-way interactions with production system were also included in the full model. Beginning with the full model, including all the explanatory variables, backward elimination was applied to eliminate factors that did not significantly (P>0.05) contribute to predicting variation in the response variables until a reduced model was obtained for each of the economic parameters.

The final reduced model for GM1 was:
yjnp = µ + Sj + Fn + h + f + p + ejnp (model 2)

The final reduced model for GM2 was:
yp = µ + h + f + ep (model 3)

For NB, the final reduced model was:
yp = µ + h + f + ep (model 4)

All variables in the reduced models are as defined previously.


Results

Socio-economic characteristics of household

The majority of households (95% in both systems) were headed by males with a mean age of 54 and 52 years in the transhumant and sedentary systems, respectively (Table 3). The family had on average 9 persons in the transhumant and 8 in the sedentary system. In general, farmers of the transhumant system had slightly larger (P>0.05) landholdings (40 du) than farmers of the sedentary system (34 du). The percentage of farmers who leased land for grazing of livestock was apparently higher (67.1%) in the transhumant than in the sedentary systems (47.4%) but the difference between systems in the mean rented land for grazing was only marginally significant (P>0.05). The operations in both systems are based mostly on family labor. In the transhumant system, the percentage of farmers who lived in block house, tent and both (block house and tent), were 0%, 61% and 39%, respectively compared to 79%, 16% and 5% in the sedentary system. In the transhumant system, the one-way mean distance and the corresponding time to the water sources were significantly (P<0.001) higher with 6.9 km and 25.9 minutes, compared to 2.8 km and 9.9 minutes in the sedentary system.

Herd size, composition and livestock ownership

The herd size and livestock composition per system are shown in Table 3. The majority of farmers interviewed kept goats in mixed livestock systems rather than in goat-specific systems. Most flocks in the transhumant (74%) and 50% in the sedentary systems were composed of sheep and goats. Goats comprised about 41% and 58% of the total flock size for the transhumant and sedentary systems, respectively. The percentage of the farmers who raised sheep was higher in the transhumant than in the sedentary systems and the average number of sheep was marginally higher (P>0.05) in the transhumant than in the sedentary systems. In the transhumant and sedentary systems, 39% and 16% of farmers, respectively, owned donkeys. Although the percentage of the farmers who raised camels was higher in the transhumant than in the sedentary system, the average number of camels was significantly higher (P<0.05) in the sedentary than in the transhumant system. None of the surveyed farms raised cattle.

The average number of goats was significantly (P<0.05) higher in the transhumant than in the sedentary system. Three different goat breeds were identified in the study area: Mountain Black (MB), Dhaiwi and crossbred goats, mainly from the Damascus and MB breeds (Table 3). The average number of MB was significantly (P<0.05) higher in the transhumant than in the sedentary system. Goat keepers in the transhumant system had larger flocks (92 head) with about 56% MB, 42% crossbred and 2% Dhaiwi goats compared to the flocks in the sedentary systems (60 head) which were composed of about 48% MB, 51% crossbred and 1% Dhaiwi goats. The average Tropical Livestock Unit (TLU) was significantly (P<0.05) higher in the transhumant than in the sedentary system. The fertility rate, productivity rate and mortality rate per flock were similar between production systems. In the present study, the perinatal mortality (abortion, neonatal) accounted for 98.9% of the total losses of goats. The correlation between mortality rate and productivity rate was significant and negative (P<0.001), (r = -0.38).

Table 3. Socio-economic characteristics of goat keepers and performance traits of goats by production system
Variables Production system P-value
Transhumant
(n=76)
Sedentary
(n=38)
n LSM SE n LSM SE
Age of household head (years) 76 53.9 1.5 38 51.7 2.2 0.407
Household size (no. of members) 76 9.2 0.5 38 8.0 0.7 0.213
Land resources (du*)
- Owned land 29 40.2 8.8 26 34.0 9.3 0.629
- Rented land for grazing 51 194.1 30.4 18 86.8 51.2 0.076
Distance from homestead to water source (km) 76 6.9a 0.6 38 2.8b 0.8 <0.001
Distance from homestead to water source (minutes) 76 25.9a 1.7 38 9.9b 2.4 <0.001
Number of livestock by species
- Goats 76 91.7a 9.0 38 60.3b 12.8 0.047
- Sheep 56 176.5 25.4 19 89.1 43.5 0.087
- Donkey 30 1.5 0.2 6 1.3 0.4 0.678
- Camel 12 4.6a 2.2 2 18.5b 5.4 0.036
Total number of animals (TLU) 76 23.2a 2.4 38 11.6b 3.4 0.006
Number of goats by breed
- Mountain Black 75 52.1a 5.3 34 32.1b 7.9 0.037
- Dhaiwi 22 4.7 0.6 7 3.3 1.0 0.230
- Crossbred 71 41.6 5.3 35 33.6 7.5 0.384
Goat flock performance**
- Fertility rate (%) 76 88.4 1.6 38 86.7 2.2 0.531
- Flock mortality (%) 76 17.2 1.6 38 20.1 2.2 0.284
- Productivity rate 76 1.03 0.03 38 1.00 0.1 0.588
LSM: least squares mean; SE: standard error of the mean;
* local unit of land: 1 du (dunum) = 1000 m²; TLU: Tropical livestock unit was calculated as 0.1 TLU = 1 head of goat or sheep; 0.5 TLU = 1 head of donkey; 1.0 TLU = 1 head of camel. LSMs in the same row with different superscript letters differ significantly at P<0.05.
**Fertility rate: number of does kidding at least once divided by the number of does in the flock during the year, productivity rate: number of kids born alive divided by the does kidding during the year, mortality rate: total deaths of goats in a 12 month period divided by the total number of goats in the flock at the beginning of the year.
Purposes of keeping different goat breeds

Figure 1 shows the different purposes of keeping goat breeds and the ranking of the importance of these purposes for each breed. The goat keepers kept MB and crossbred goats in descending order of importance for cash income, milk production for home consumption, savings, meat production for consumption and manure, while the top ranked purpose of keeping of Dhaiwi breed was milk production followed by cash income, meat production, savings and manure.

Figure 1. Purpose of keeping goat breeds and the ranking of the importance of these purposes.
Index based on rank weights of the choices of priority characteristics (i.e. 5= highest importance).
Water transporting responsibility and methods of water collection

Responsibility for water transportation in the study area showed that, children, housewives, householders and hired labor were involved in the daily fetching of goat water, particularly during the dry season. In the transhumant system, household heads were predominantly responsible for fetching water for goats followed by hired labors and boys, while in the sedentary system (in case of no water access through the pipe network and/or water was not running in the pipe), hired labor were predominantly responsible for fetching water for goats followed by household heads and boys. Water was mainly transported in water storage tanks pulled by tractor in both production systems. Due to cultural and social reasons and because most of the females had no driving license, males predominantly transported drinking water for livestock.

Overview on annual production costs and revenues
Total costs

An overview of variable and fixed costs per flock and head in the different production systems is shown in Table 4. Variable costs varied according to the size of the flock and level of production. The weighted averages of the total variable and fixed costs in both systems (n=114 farmers) were 5,090.1±4,978.6 and 118.7±161.0 JD per flock and 93.9±57.7 and 3.6±8.3 JD per head, respectively. The average of the total variable costs per flock was significantly higher in the transhumant than in the sedentary system, reflecting the significant differences in the costs of feed, grazing fees, veterinary services and transportation between the two systems. These significant differences of variable costs can be ascribed mainly to differences in flock size between systems except for grazing fees, which were also significantly different at per head level. In the two production systems (n=114 farmers), the highest share per flock (75%) and per head (74%) of the total variable costs was for feed including cost for grazing fees.

Table 4. Variable costs (JD*/year) per flock and head of the different production systems for the 2011/2012 season
Production system
Transhumant
(n=76)
Sedentary
(n=38)
Per flock Per head Per flock Per head
Mean Median Mean Median Mean Median Mean Median
Variable costs
Feed 3,994.7 2,494.5a 62.3 51.5 2,652.4 1,744.5b 64.2 58.5
Grazing fees 328.8 100.0a 6.9 2.5a 138.4 0.0b 4.2 0.0b
Water 8.4 0.0 0.2 0.0 10.6 0.0 0.2 0.0
Veterinary** 216.6 100.0a 3.8 2.0 106.2 50.0b 3.0 2.0
Transportation*** 687.8 500.0a 14.8 11.0 454.9 300.0b 13.2 9.5
Hired labor 459.7 0.0 5.9 0.0 516.1 0.0 9.1 0.0
Total variable costs 5,696.0 4,193.0a 93.9 77.5 3,878.5 2,146.5b 94.0 76.0
Fixed cost
Interest of housing 30.6 20.0a 0.4 0.0a 55.1 65.0b 2.0 1.0b
Interest of machinery/equipment 108.9 0.0a 3.5 0.0a 22.2 0.0b 0.9 0.0b
Total fixed cost 139.4 65.0 4.0 1.0 77.3 65.0 2.9 2.0
Mean is arithmetic mean;
*JD = Local currency ([1JD = 1.4 USD] in 2012.); Pairs of variables per flock and per head in the transhumant and sedentary systems with different superscripts are statistically significant, (Wilcoxon-Mann-Whitney test, P<0.05);
**Veterinary costs included charges on treatment and value of purchased drugs and vaccinations;
***Transportation of water, feed and goats.

The second highest share of variable costs was for transportation. The transportation costs accounted for 12% per flock and 15% per head of the total variable costs in both systems (n=114 farmers). Eighty percent of farmers had their own vehicles (a pickup, tractor, tanker, or truck) for transportation. Twenty-eight percent of goat keepers in the transhumant and 26% in the sedentary systems hired external permanent and seasonal labor, mostly Syrians, in addition to few labors belonging to the Al-A’zazmeh tribe, mainly for shepherding, feeding and watering the flock. The hired labor costs accounted for 9% per flock and 7% per head of the total variable costs in both systems (n=114 farmers). The cost for veterinary services accounted for about 4% per flock and per head of the total variable costs in both systems (n=114 farmers). The cost for water per se was very low, accounting for 0.2% per flock and per head of the total variable costs in both systems (n=114 farmers) and was similar between production systems. Fixed costs were generally very low. The fixed costs per flock and head over the year associated with housing were significantly (P<0.05) lower in the transhumant system than in the sedentary system. In contrast, fixed costs associated with machinery were significantly (P<0.05) higher per flock and head in the transhumant system compared to the sedentary system.

Total revenues

The major goat production revenues are from the sales of kids in the two production systems (Table 5). The average of the total revenues was similar between production systems at flock and head level. In the two production systems (n=114 farmers), meat and milk including milk product sales accounted for 68% and 32% per flock of the total sales. Goat meat sales per flock and head did not differ significantly between the two production systems. Milk and milk product sales per flock were marginally significant higher in the transhumant than in the sedentary system (P=0.055). Only the values of goat meat consumption per flock were significantly higher in the transhumant than in the sedentary system. In contrast, the values of milk and milk products per flock and head consumed at home were significantly higher in the sedentary than in the transhumant system. The insurance and finance function per flock were significantly higher in the transhumant than in the sedentary system.

Table 5. Composition of revenues (JD*/year) per flock and head of the different production systems for the 2011/2012 season
  Production system
Transhumant
(n=76)
Sedentary
(n=38)
Per flock Per head Per flock Per head
Mean Median Mean Median Mean Median Mean Median
Revenues
Cash revenue
Sale of goats (meat) 3,903.7 2,000.0 59.7 47.5 3,181.6 1,500.0 77.9 57.5
Sale of milk and milk products 1,977.2 1125.0 30.2 28.0 1,248.3 986.5 24.3 24.0
Non-marketable value of production**
Meat 974.3 800.0a 21.3 16.0 724.2 500.0b 21.7 14.0
Milk and milk products 145.7 0.0a 4.3 0.0a 311.4 0.0b 14.2 0.0b
Finance and insurance 924.7 636.0a 13.7 13.0 618.2 437.0b 14.1 14.0
Total revenue 7,925.5 5,250.0 129.3 107.5 6,083.2 3,644.5 152.2 139.5
Mean is arithmetic mean; SD: standard deviation;
*JD = Local currency (1JD = 1.4 USD in 2012); Pairs of variables per flock and per head in the transhumant and sedentary systems with different superscripts are statistically significant (Wilcoxon-Mann-Whitney test, P<0.05);
**Includes the value consumed by households or used to meet social and religious activities.
Economic efficiency of goat production

The analysis of data by Wilcoxon-Mann-Whitney test showed that no significant differences were found per flock and head for all economic parameters between production systems (Table 6). Standard deviations were generally very high. With 55%, farmers in the transhumant system had more often negative values of GM1 compared to 47% in the sedentary system. Still, about 30% of farmers in both systems had negative values of GM2. The averages of GM1 and GM2 per flock and head were generally lower in the transhumant than those in the sedentary systems (P>0.05).

The averages of BCR1 and BCR2 values per flock in the transhumant system were generally lower than those in the sedentary system (Table 6). Twenty-eight percent of the farmers in the transhumant systems conducted goat production inefficiently (NB>0) compared to 11% in the sedentary system. The averages of NB values per flock and head were similar between systems (P>0.05).

Table 6. Economic success parameters of goats (JD*/year) per flock and head under different production systems for the 2011/2012 season: gross margin (GM), net benefit (NB) and benefit-cost ratio (BCR)
Variables Transhumant (n=76) Sedentary (n=38)
Median Range Median Range
GM1 (flock) -150.00 -6,777 to 11,817 -81.00 -3,347 to 7,580
GM1 (head) -3.50 -276 to 207 0.00 -103 to 229
GM2 (flock) 943.50 -5,886 to 14,317 792.50 -2,747 to 11,403
GM2 (head) 17.00 -216 to 325 23.00 -70 to 243
BCR1 0.95 0.00 to 4.30 1.00 0.00 to 4.70
BCR2 1.25 0.20 to 5.20 1.30 0.50 to 4.90
NB (flock) 1,349.50 -4,840 to 15,969 1,226.50 -1,144 to 12,516
NB (head) 29.50 -204 to 341 34.50 -55 to 257
*JD = Local currency (1JD = 1.4 USD in 2012);
GM1 = CR – VC; GM2 = (CR+NV) –VC; BCR1 = CR/VC; BCR2 = (CR+NV)/VC; NB= (CR+NV+BC+BI) – (VC+IR). CR = total cash revenue, VC = total variable cost, NV = non-marketable value of production, BC = benefit of capital, BI = benefit of insurance, IR = interest rate on the fixed capital.
Factors affecting economic success of goat production

Results of the reduced model from the analysis of variance for the effect of different explanatory variables on the GM1, GM2 and NB of goats are presented in Table 7. GM1 of goat production was significantly affected by production system, feeding intensity, flock size, fertility rate and productivity rate, while GM2 and NB were significantly affected by flock size and fertility rate.

Table 7. Analysis of variance for factors affecting economic success parameters of different production systems per flock
  GM1 GM2 NB
Factors DF F Value DF F Value DF F Value
Production system 1 4.10* - - - -
Feeding intensity 1 5.24 * - - - -
Flock size 1 13.94 *** 1 16.08*** 1 49.80***
Fertility rate 1 13.63 *** 1 17.19 *** 1 17.50 ***
Productivity rate 1 5.01 * - - - -
0.30 0.23 0.38
F Value 8.95 *** 15.85*** 32.27***
*P<0.05,
***P<0.001;
R²= R-Square; DF: degree of freedom; GM1 = CR – VC; GM2 = (CR+NV) –VC; BCR1 = CR/VC; BCR2 = (CR+NV)/VC; NB= (CR+NV+BC+BI) – (VC+IR).CR = total cash revenue, VC = total variable cost, NV = non-marketable value of production, BC = benefit of capital, BI = benefit of insurance, IR = interest rate on the fixed capital; four outliers with high deviation from the mean were removed for the analysis of GM1, GM2 and NB to get a normal distribution for the residuals.

GM1 was significantly positively correlated with productivity rate. GM1, GM2 and NB were significantly positively correlated with flock size, fertility rate (Table 8). An increase of the flock’s productivity rate by 0.1 is associated with an expected increase of the GM1 by about 2 JD per year. Additionally, an increase of the flock’s size by one head is associated with an expected increase of the GM1, GM2 and NB by 16, 19 and 33 JD, respectively. A 1% increase in flock’s fertility rate will increase GM1, GM2 and NB by 55, 66 and 67 JD, respectively.

Table 8. The relationship between flock size, productivity rate and fertility rate of goat production and GM1, GM2, NB (JD/flock per year)
Variables Estimate SE T value P-value
GM1
Intercept -8093.90 1460.33 -5.54 <0.001
Flock size 16.19 4.33 3.73 <0.001
Fertility rate 55.41 15.01 3.69 <0.001
Productivity rate 15.95 7.12 2.24 0.027
GM2
Intercept -5622.36 1443.67 -3.89 <0.001
Flock size 18.51 4.62 4.01 <0.001
Fertility rate 65.93 15.90 4.15 <0.001
NB
Intercept -5856.07 1456.58 -4.02 <0.001
Flock size 32.87 4.66 7.06 <0.001
Fertility rate 67.12 16.04 4.18 <0.001
GM1 = CR – VC; GM2 = (CR+NV) –VC; NB= (CR+NV+BC+BI) – (VC+IR). CR = total cash revenue, VC = total variable cost, NV = non-marketable value of production, BC = benefit of capital, BI = benefit of insurance, IR = interest rate on the fixed capital.

Table 9 shows that the GM1 values were significantly (P<0.05) higher in the sedentary than in the transhumant systems. The GM1 values were significantly (P<0.05) higher at farms with feed intake intensity ≤1kg than at farms with feed intake intensity >1kg.

Table 9. Gross margin (JD/flock per year) by production system and feeding intensity

Source of variance

n GM1 (Mean ± SD)
Production system #
Transhumant 73 -12.5±2,545.3a
Sedentary 37 420.2±2,226.8b
Feeding intensity #
≤1kg 64 691.9±2,165.0a
>1kg 46 -644.5±2,609.6b
Mean is arithmetic mean;
#P<0.05; GM1 = CR – VC.


Discussion

Differences in the landholdings, herd movement, livestock herd size and composition between production systems reflect that the farmers followed different livelihood strategies. Flock fertility rate reported in this study indirectly indicates the barren rate (12%) which is higher than that (6.6%) reported in the same study area by Aldomy et al (2009). According to Aldomy et al (2009), the unbalanced nutrition, infectious disease and poor reproductive management could be the major reasons for barrenness of does. The high perinatal mortality rate was the major reason for the loss of goats in the present study. Therefore, the flock productivity rate was significantly negatively associated with flock mortality rate. The percentage of flock mortality rate (18%) in the present study is in accordance with that of 16% new born mortality rate in the Karak Governorate in 2012 (Department of Statistics 2012a). Though flock mortality rate was negatively associated with productivity rate (1.02 kid born alive/doe per year), it was compensated because some of does gave birth to twins and triplets. Zaitoun et al (2004) reported that in the near-south region where the current study was conducted, the MB breed had the highest twinning ability followed by crossbred and Dhaiwi.

The result of the purposes for keeping goats as ranked by the goat owners (Figure 1) was in line with the result of the revenue shares (Table 5) reflect characteristics of livelihood oriented production systems. Al-Khalidi et al (2013) reported different reasons to keep small ruminants in Jordan, among them, that they are kept as an additional or main source of income. The highest rank given for cash which includes milk sales in this study is consistent with results from Tabbaa and Al-Atiyat (2009). They showed that the overall rank order of the breeding objectives for goat breeds in Jordan were milk production (for MB, Dhaiwi and crossbred goats), followed by breed unique morphological traits, meat production and good health status. Milk production was ranked highest by farmers in their study due to less fluctuation of milk prices in the market compared with meat and in turn stable source for cash income.

Feed cost including cost for grazing fees represented the major part (75%) of total variable costs, which is not surprising under conditions of forage shortage in rangelands of Jordan (Al-Khaza’leh et al 2015). Although the government provides farmers a subsidy that is equivalent to about 10 kg of barley per head of goat per month, the level of feed cost was still high and was considered as one of the major constraints for goat production (Al-Khaza’leh et al 2015). Feed cost including cost for grazing fees per head (74%) in this study was much higher compared to results from Hamadeh et al (2001) who reported that feed costs including cost for pasture fees per head per year ranged from 28% to 46% of the total input costs in small ruminant production systems ranging from semi-nomadic to settled in marginal areas of Lebanon. Abu Zanat and Tabbaa (2004) reported that drought and stopping fodder subsidies adversely affected the enterprises of small ruminants. In the present study, most of the goat keepers sold the goat kids at times when cash was needed, purchasing feeds for the remaining animals in their flock or shortage of natural forage resources.

Households with access to the municipal water network pay the respective price per M³ set by the government according to the amount of water used, whereas households not connected to the water network had to bring water tanks and paid per M³. As an example for the water price, the price set for one specific borehole in the mountain zone, which is only suitable for livestock consumption, was 0.25 JD/M³ while the price for one specific borehole suitable for human consumption in the semi-desert zone was 0.75 JD/M³ (Jordanian Dinar (JD) with an exchange rate of 1JD = 1.4 USD in 2012). Additionally, during the time of water scarcity and high demand, few owners of private boreholes are selling the water to livestock owners with prices ranging from 1.5 to 3 JD/M³. The significantly higher transportation costs in the transhumant systems than in the sedentary system were associated mainly with differences in daily water fetching, particularly during dry the season. This was influenced by remoteness from water sources, fuel cost and hired labor (Al-Khaza’leh et al 2015). It has been reported that under the transhumant system in Syria, longer distances to water sources and markets was among the major factors reducing the efficiency of sheep production (Shomo et al 2010).

The value of inputs used in livestock production (goats and sheep) in Jordan was estimated at 158,253 thousand JD in 2012, water and fuel contributed about 2% and 4% of the total, respectively (Department of Statistics 2012a). The shares of feed (75%), water (0.2%) and labor (9%) costs per flock in this study were in the range of results by Al-Khalidi et al (2013) showing that the shares of the total variable costs per small ruminant herd in Jordan ranged from 51% to 86% for feed, 0% to 3% for water, and 11% to 44% for labor. According to Hosri and Nehme (2006), water costs per head per year accounted for the lowest share of the total input costs in the sedentary and transhumant systems in north Lebanon. Based on data obtained from the Water and Irrigation Office in the study area, the price of water per se for livestock consumption was subsidized by the government and was set to cover the cost of operation and maintenance of the water pumping system. Moreover, water from wadis, springs and some boreholes was taken free of charge by the majority of farmers (75%). Therefore, the cost for water per se was generally very low. The fixed cost invested for housing was very low in the transhumant system because the majority of farmers (92%) in this system used traditional animal houses made of available and cheap materials such as iron fences with nets for sun and rain protection. On the other hand, the number of vehicles owned by farmers used for transportation was higher in the transhumant system. Consequently, the fixed cost invested for machinery/equipment was higher.

The similar cash revenue per flock and head in the production systems could be ascribed to the similar selling prices and quantities of meat and milk. The percentage of meat, milk and milk product sales per flock of this study is almost in agreement with Al-Khalidi et al (2013) who reported that the share of kid and milk sales of small ruminant herd ranged from 26% to 73% and 5% to 24% of the total sales, respectively. In the present study, meat contributed 72% and 66% and milk 28% and 34% to cash income in the sedentary and transhumant systems, respectively. According to Hosri and Nehme (2006), in the sedentary, semi-nomadic and transhumant systems in north Lebanon, meat production was the main source of revenue while in the subsistence and intensive systems, milk production appeared to be the main source of revenue.

The average share of milk and milk product sales per head of this study (31%) is consistent with a previous study by Hamadeh et al (2001) who reported from the marginal regions of Lebanon, that in small ruminant production systems ranging from semi-nomadic to settled, milk output per head per year represented 30% to 47% of the total output revenues. On the other hand, the average share of meat sales per head was higher in this study (69%) compared to the study by Hamadeh et al (2001) in which live lambs and kids contributed 48% to 57% of the total output revenues in the different systems. Goats represented a high capital value for the farmers. Goat flocks are used as a source of finance and insurance for some major expenditure (such as expenditure on education fees or on important social activities).

In this study, the average of BCR1 (1.1) and BCR2 (1.5) in both systems (n=114 farmers) with dual-purpose goat breeds of meat and milk seemed to be economically acceptable. However, nearly half of farmers in the transhumant (55%) and 47% in the sedentary systems conducted goat production inefficiently regarding to GM1, while about one third of farmers (30%) in both systems conducted goat production inefficiently with respect to GM2. Contrary to the results of this study, Metawi (2011) found that all the systems of goat production (transhumant-extensive, semi-intensive and smallholder) in Egypt had positive profits in the financial analysis (which accounted only for cash paid input costs). In our study, if the family labor cost was included as opportunity cost, e.g. the cost of employing someone else to undertake the farm tasks, the cost of labor would have been higher in both systems which would even increase the share of negative values. Hamadeh et al (2001) reported that when the family labor cost was included as an opportunity cost in the economic analysis of small ruminant production systems in Lebanese marginal lands, which ranged from semi-nomadic to settled, there were negative returns in all systems.

The economic efficiency of goat enterprises is influenced by the number of offspring produced (Husein et al 2005). An increase of the flock’s productivity rate had positive impact on the GM1. Because the perinatal mortality was the major reason for loss of goats in the present study, the increase in productivity rate is associated with a decrease in mortality rate. The high mortalities in young animals can severely affect the farmers’ economic returns (Al-Assaf 2012). The significant relationship between trends in fertility rate and large flock size, and economic success parameters (GM1, GM2 and NB) could be ascribed to higher returns from milk and meat production and the higher capital value of goat flocks.


Conclusions

This study found differences in the costs and revenues of goat farming between the sedentary and transhumant systems. NB, GM2 and BCS2 were found to better reflect farmers’ production efficiency than GM1 and BCS1 due to considering the non-marketable values and socio-economic benefits. Regarding the NB, 28% of farmers in the transhumant system and 11% of farmers in sedentary system conducted goat production inefficiently. The sedentary systems produced a slightly higher benefit-cost ratio than the transhumant systems. High feed cost, high barren rate and high perinatal mortality rate were the major problems that affected goat economic production. Therefore, improvement of feed use efficiency, reproductive management and flock health are recommended. Additionally, monitoring the flock and culling infertile does and replacing them are also needed to improve economic efficiency of the flocks. The cost for water per se was not an important cost element in the surveyed production systems. Further investigations are recommended on the effect of water use on the profitability and thus on the sustainability of goat farming under more extensive production systems.


Acknowledgment

The authors thank the German Academic Exchange Service (DAAD) for providing financial support for this study. We greatly acknowledge the cooperation of the personnel of Ministry of Agriculture and Ministry of Water and Irrigation in Jordan. The authors express also sincere gratitude to all interviewed farmers for their participation in this study.


References

Abdel Aziz M 2010 Present status of the world goat populations and their productivity. Lohmann information 45: 42-52.

Abu Zanat W and Tabbaa M 2004 Effect of drought on feed resources and performance of small ruminants in the Northern Badia of Jordan. Dirasat: Agricultural Sciences 32: 347-354.

Abu-Zanat M M, Migdady H A and Tabbaa M J 2005 Production systems of small ruminants in middle Badia of Jordan. Dirasat: Agricultural Sciences 32: 205-214.

Al-Assaf A 2012 Economic implications of small ruminant diseases in the Northern Area of Jordan. Journal of Food, Agriculture & Environment 10: 323-326.

Al-Atiyat R M 2014 Role of small-scale dairy sector in food security and poverty alleviation. Food, Agriculture and Environment 12: 427-433.

Al-Atiyat R M and Tabbaa M J 2009 Role of livestock in poverty alleviation and food security: A review study. In Proceeding of Jordan Society of Scientific Research. Amman, Jordan.

Al Baqain R and Valle Zárate A 2011 Economic analysis of Bedouin sheep farming in Jordan and the Palestinian territories. Livestock Research for Rural Development. Volume 23, Article#249. Retrieved December 10, 2014, from http://www.lrrd.org/lrrd23/12/baqa23249.htm.

Aldomy F, Hussein N O, Sawalha L, Khatatbeh K and Aldomy A 2009 A national survey of perinatal mortality in sheep and goats in Jordan. Pakistan Veterinary Journal 29: 102-106.

Al-Khalidi K M, Alassaf A A, Al-Shudiefat M F and Al-Tabini R J 2013 Economic performance of small ruminant production in a protected area: a case study from Tell Ar-Rumman, a Mediterranean ecosystem in Jordan. Agricultural and Food Economics 1: 1-8.

Al-Khaza’leh J, Reiber C, Al Baqain R and Valle Zárate A 2015 Drinking water sources, availability, quality, access and utilization for goats in the Karak Governorate, Jordan. Tropical Animal Health and Production 47: 163-169.

Ayalew W, Rischkowsky B, King J M and Bruns E 2003 Crossbreds did not generate more benefits than indigenous goats in Ethiopian smallholdings. Agricultural Systems 76: 1137-1156.

Central Bank of Jordan 2012 Publication and statistics, statistical database, interest rate structure for 2011/2012. Amman. Jordan. Retrieved December 05, 2014, from http://statisticaldb.cbj.gov.jo/index?lang=en.

Department of Meteorology 2011 Climate and Climate Change Division. Amman, Jordan. Retrieved 02, February, 2012, from http://met.jometeo.gov.jo/jometeo/e_main.

Department of Statistics 2012a Statistics by sector, agriculture, surveys, livestock for 2012. Amman, Jordan. Retrieved October 27, 2014 from http://www.dos.gov.jo/dos_home_e/main/index.htm.

Department of Statistics 2012b Statistics by sector, population and housing, demographic statistics for 2012. Amman, Jordan. Retrieved October 27, 2014 from http://www.dos.gov.jo/dos_home_e/main/index.htm.

Hamadeh S K, Bistanji G N, Darwish M R, Abi Said M and Abi Ghanem D 2001 Economic sustainability of small ruminants production in semi-arid areas of Lebanon. Small Ruminant Research 40: 41-49.

Hosri C and Nehme M 2006 Small ruminant production systems in north Lebanon. CIHEAM -Options Méditerranéennes 70:111-116.

Husein M Q, Ababneh M M and Haddad S G 2005 The effects of progesterone priming on reproductive performance of GnRH-PGF2alpha-treated anestrous goats. Reproduction Nutrition Development 45: 689-698.

Lebbie S H B 2004 Goats under household conditions. Small Ruminant Research 51: 131-136.

Lemke U, Kaufmann B, Thuy L T, Emrich K and Valle Zárate A 2007 Evaluation of biological and economic efficiency of smallholder pig production systems in North Vietnam. Tropical Animal Health and Production 39: 237-254.

Mbuku S, Kosgey I, Okeyo M and Kahi A 2014 Economic values for production and functional traits of small East African goat using profit functions. Tropical Animal Health and Production 46: 89-795.

Metawi H R 2011 Economic sustainability of goat production under different production systems in Egypt. CIHEAM -Options Méditerranéennes 100: 185-190.

Ministry of Agriculture 2012 The Agricultural year book. Ministry of Agriculture, Amman, Jordan.

Ogola T D, Nguyo W K and Kosgey I S 2010 Economic contribution and viability of dairy goats: implications for a breeding programme. Tropical Animal Health and Production 42: 875-85.

Qtaishat T, Al-Sharafat A and Majdalawi M I 2012 A comparative economic analysis of sheep production systems: A case study of Jordan. Journal of Food, Agriculture and Environment 10: 690-694.

SAS Institute 2012 Base SAS® 9.3 Procedures Guide: Statistical Procedures, Second Edition. SAS Institute Inc, Cary, NC, USA.

Shomo F, Ahmed M, Shideed K, Aw-Hassan A and Erkan O 2010 Sources of technical efficiency of sheep production systems in dry areas in Syria. Small Ruminant Research 91: 160-169.

Tabbaa J M and Al-Atiyat R 2009 Breeding objectives, selection criteria and factors influencing them for goat breeds in Jordan. Small Ruminant Research 84: 8-15.

Zaitoun I S, Tabbaa M J and Bdour S 2004 Body weight, milk production and lifetime twining rate of the different goat breeds of Jordan. Dirasat: Agricultural Sciences 31: 143-149.


Received 14 February 2015; Accepted 15 February 2015; Published 1 May 2015

Go to top