Analysis of technical, allocative and economic efficiency of broiler production using closed house system in Malang District of East Java Indonesia

S Pakage, B Hartono¹, Z Fanani¹ and B A Nugroho¹

Faculty of Animal Husbandry University of Papua Manokwari West Papua Indonesia
stpakages@yahoo.com
¹ Faculty of Animal Husbandry University of Brawijaya Malang East Java Indonesia

Abstract

The general objective of the study was to calculate the production efficiency of broiler farmers using the closed house system in three Sub-districts of Malang East Java, Indonesia. Cross-sectional data were used to analyze the factors affecting production and production costs, factors causing technical inefficiency and technical, allocative and economic efficiency. Sampling technique used in this study is total sampling. Data collected from 55 respondents The Stochastics Frontier Analysis was employed to measure production efficiency.

Analysis showed that the factors positively affecting production of broilers are DOC, feed, and medicine. Variables of Day Of Chick (DOC), feed, and electricity costs are influential to the production cost. The diversity of broiler production levels caused by differences in the level of technical efficiency among farmers is at 0.63% with the average of 0.929. Estimation of the effects of technical inefficiency that all variables are statistically determinants has no real effect. However, the level of formal education and gender contributed to improving technical inefficiency. Allocative efficiency levels ranged from 0.881 to 1.310 with an average of 1.008. These results indicate that farmers have been allocatively efficient. The diversity of level of production costs caused by the influence of economic efficiency equals to 0.255% with an average of 0.931.

Keywords: broiler, cost function, production function, stochastic frontier

Introduction

Livestock farming is part of agriculture which produces meat as a protein source that is of paramount importance to humans. The significance of broiler chicken meat has led to an increase in the population of broiler chickens in Indonesia. In 2010, the population of broiler reached 1,115,108,029 thousand heads, an increase of 5% from the previous year. Broiler meat production in 2010 amounted to 1,184,366.15 thousand tons, an increase of 2.8% from the total production of chicken meat in the previous year. According to the Ministry of Agriculture, in 2010 the national meat consumption reached 7.75 kg /capita / year. As many as 49% or as much as 3.80 kg / capita / year of national meat consumption are from chicken meat consumption. Being a country with the fourth largest population in the world, Indonesia is an enormous market, causing an increase of the demand for broiler meat. These data indicate that broiler farming business has potential for development.

Development of broiler farms is inseparable from efforts to pull the due maintenance is quite easy with the relatively short production cycle. It is also that the broiler is able to grow faster so that it can produce meat within 4-7 weeks. However, the development of broiler chicken farm is also influenced by weather conditions. In a fairly extreme weather condition, when scorching heat is followed by sudden rain or vice versa, temperature and humidity of the enclosure change drastically resulting in high mortality rates and a decline in production if not alleviated.

Broiler production using closed house system is one of technological innovations which attempts to accommodate the fairly extreme weather changes, and is expected to minimize the adverse effects of the environment or climate change outside of the cage. The intended use of the closed house system cage is to create a controlled microclimate inside the cage, improve productivity, land and labor efficiency, and create environmentally friendly environment. Development of closed house system requires high capital outlay; however, a number of broiler breeders in Malang of East Java, Indonesia, have already been using this housing system.

The capability of broiler breeders using closed house system in addressing environmental issues and resource allocation decision varies, due to the diversity of farmers ability, regional potential and business environment. Differences in the ability of farmers will have an impact on output generated due to a combination of the use of a number of factors of production. Combination of the use of production factors in broiler business is an absolute requirement to generate profit. Broiler farmers' income is influenced by a combination of the use of production factors such as Day Old Chick (DOC); feed; Medicine (drugs, vitamins and vaccines); labor; electricity and fuel. Studies how farmers broilers use the closed house system allocates production factors to obtain the maximum production at minimum costs in order to obtain high profits need to be done. Therefore, it is necessary to examine factors affecting the production function, cost function and the achievement of level of technical, allocative and economic efficiency.

Research Methods

Studi area and data

This research was conducted in the Sub-districts of Pangelaran, Dampit, and Bantur Malang District of East Java, Indonesia. The three Sub-districts were selected purposively because these areas are developing area of broilers by using closed house system. The sampling technique used was total sampling because the number of broiler farmers who use closed house system is still small. This is because the broiler farms that use the closed house system was newly developed in this area. Data collected from 55 respondents. The data collected are specified in the empirical model below:

Empirical modeling

The Stochastic Frontier Analysis (SFA) method is used in modeling the technical, allocative, and economic efficiency of broiler farmers using close house system. In the stochastic frontier model, the output is assumed to be bound by a stochastic production function. In the case of Cobb-Douglas, the model can be expressed as follows:

Deviation (vi - ui) comprises two parts: (1) symmetric component that allows random diversity of the frontier among observations and captures the effect of measurement error, random shock, and so on, and (2) one-sided component of deviation which captures the effect of inefficiency. This model was introduced by Aigner et al (1977), Meeusen and van den Broeck (1977), and later developed, among others, by Schmidt and Lovell (1979) and Jondrow et al (1982).

Stochastic frontier production function model applied to estimate the technical efficiency at the broiler farm level is mathematically described as follows:

Yi* = f(Xi;β) + εii=1, 2 .n

Where Yi is the output, Xi denotes the actual input variables, β is a parameter of the production function which value is not yet known, and ε is the error term that consists of two components, namely:

εi = Vi - Ui3

The first error component, the error tem Vi is symmetrical and assumed to be identical, independent and normally distributed N (0, σ2v). While the second is the error term Ui, which is independent to the Vi and normally distributed N (0, σ2u). This error term allows actual production function to be under the frontier production function. According to Jondrow in Ogundari and Ojo (2006) estimation of technical efficiency is shown by the average inefficiency distribution (Ui) with certain ε value; the inefficiency formula is written as follows:

Where λ = σu / σv, and σ2 = σ2u + σ2v,while f and F each indicates standard normal density and cumulative distribution function calculated from εi λ / σ.

Technical efficiency in farming is defined as the actual output condition (Yi) to Frontier output (Yi *) by using the available technology, which is derived from equation (3) mentioned above in order to obtain:

TE value lies in the interval of 0 to 1 or 0 ≤ TE ≤ 1, and if TE = 1, the farming is in an efficient condition.

Frontier stochastic cost function model used to estimate the economic efficiency at the broiler farmers level is described as follows:

Ci = g (Yi,Xi; α ) + εii = 1, 2, … n

Where Ci is the total production cost, Xi denotes the actual input costs, α is the parameter of the function costs, and ε is the error term that consists of two components, namely:

εi = Vi+Ui,

Here the first error component, error tem Vi, is a symmetric error tem and assumed to be identical, independent and normally distributed N (0,σ² v). While the second is the error term Ui, which is independent to the Vi and normally distributed N (0, σ2u). Economic efficiency can be estimated using the following equation:

The efficiency value of the equation for the cost function is referred to as the economic efficiency (EEI) at each observation i. C * is the cost under ideal conditions or conditions in which efficiency is achieved (Full Efficient), while C is the actual cost based on observations. The balance between costs under ideal conditions (C *) and the actual cost based on observations (C) will determine the inefficiency coefficient. There is no inefficiency effect (Ui = 0) in the observation unit if Ci* = Ci. Such condition indicates that the costs required are relatively low, with index of economic efficiency value on the observation i equals to 1 or EEi = 1. If Ci * <Ci, inefficiency (Ui> 0) and index EEI> 1. Economic efficiency value is between 0 and 1.

Frontier version 4.1c computer program has the capacity to estimate the cost efficiency (CE) in which the result is the opposite of the equation 8 (Coelli et al 1998) or, in other words, the economic efficiency (EE) is the opposite of cost efficiency (CE). Thus, according Ogundari and Ojo (2007), economic efficiency (EE) at the farmer level can be estimated using the following equation:

EE = 1/CE 

Economic efficiency is the result of the technical efficiency (TE) multiplied by allocative efficiency (AE) at each observation. Equation economic efficiency according to Kumbhakar and Lovell (2000) is as follows:

EEi = TEi x AEi

According to Martin and Taylor, 2003 based on equation 9 above, if the value of technical efficiency (TE) and the value of economic efficiency (EE) are known, the amount of allocative efficiency (AE) can also be determined. The value of allocative efficiency (AE) is not necessarily less than one or equals to one or 0 <AE <1 (Ogundari and Ojo, 2006). Thus, the value of allocative efficiency (AE) is obtained using the following equation:

AE = EE/TE

Model specification

Factors affecting broiler production are: Day Old Chicks (DOC), feed, medicine, electricity, fuel, and labor. Cobb-Douglas functional form is mathematically formulated as follows:

Where

Y = Production of broilers per production period (kg / pp), X1 = Numbers of DOC per production period (head / pp), X2 = Amount of feed per production period (kg / pp), X3 = Amount of medicine per production period (kg / pp), X4 = Total electricity used per production period (kwh / pp), X5 = Number of Fuel per production period (liters / pp), X6 = Numbers of labor used per production period (person / pp), β0 = constant, β1-β6 = suspected non-fixed input variable parameter, Ln = Natural logarithm e = 2.718, Vi = Error occurred due to random sampling, Ui = Effects of occurred technical efficiency.

Technical inefficiency effects method used is based on the technical inefficiency effects model developed by Battese and Coelli (1995). The u_i variable is to calculate the effects of technical inefficiency. Distribution of value parameter of technical inefficiency in this study is as follows:

μ_i=δ₀+δ₁Z₁ + δ₂Z₂ +δ₃Z₃+δ₄Z₄+δ₅Z ₅

Where:

μ_i = technical inefficiency effects; Z₁ = Age of farmer (years); Z₂ = Education Level of breeder (years), Z₃ = Business experience (years), Z₄ = Numbers of dependents (people), Z₅ = Dummy gender 0 = Female 1 = Male, δ₀ = constant, δ ₁-δ₅ = Variable inefficiency parameter.

Factors affecting the cost function of broiler farming business are: cost of Day Old Chicks (DOC), the cost of feed, medicine cost, electricity cost, fuel cost, labor costs and production. The mathematical formula is as follows:

Where:

Ci = cost of broiler production per production period (IDR/ pp); W1 = Cost of DOC per head (IDR/ head); W2 = Cost of feed per kilogram in one production period (IDR/ kg); W3 = cost of medicine per unit in one production period (IDR/ unit); W4 = Cost of electricity used per production period (IDR/ pp); W5 = cost of fuel per liter in one production period (IDR / liter); W6 = Cost of labor used per production period (IDR/ day); Y = Total broilers production per production period (kg / pp); a0= Constant; a1- a6= assumed input variables parameter; Ui = Effects of occurred economic efficiency.

Estimation of the factors that affect the production function and stochastic frontier cost functions was conducted using the Frontier program version 4.1c.

Result and Discussion

Analysis of factors a ffecting production

The analysis showed that the input variables affecting broiler production in study area is Day Old Chicks (DOC), feed and medicine (drugs, vaccines, and vitamins). The implication is that if the DOC is increased by 10%, the production will upturn by 5.66% and if the variable of feed is added by 10%, the production of broilers will go up by about 4.644% (Table 1). Besides there will be an increase of broiler production of 0.248% when the variable of medicine (drugs, vaccines, and vitamins) is added by 10% (assuming ceteris paribus). Variables which do not affect the broiler production are electricity, fuel and labor (Table 1). A positive correlation obtained shows that the electricity assists broiler to eat and drink at night, although it does not significantly influence the results of the production. The addition of the number of fuel and labor in fact decrease the broiler production.

The value of Return To Scale (RTS) was obtained by summing the coefficients of the variables included within the model. Based on the analysis result, the value of the RTS is 1.032. This value indicates that the production of broilers is on stage II (decreasing positive return to scale) meaning that if all inputs are simultaneously increased by 10%, broiler production will increase by 10.316%.

Analysis of factors affecting production costs

Partial test results show that the variables that significantly affect the cost of production of broilers at α = 0.01 of confidence intervals are variables of expenses of DOC,feed costs and the cost of electricity (Table 2). It implies that if one of the inputs is upturned, the cost of broiler production will increase with the assumption of ceteris paribus. Thus, the usage of these variables must be taken into account by the broiler farmers. Variables of cost of medicine / cost of drugs, vaccines and vitamins, fuel costs, and the production variable do not affect the cost of production. Labor costs do not affect the total cost of production of broilers.

The value of gamma (γ) obtained is 0.99, indicating that the variation of errors caused by economic efficiency is 99.99%. The difference between the actual cost of production andthe possibility of maximum production costs among farmers of 99.99%is triggered by the differences in economic efficiency and 0.11% due to stochastic effects such as measurement error. The LR test values obtained are far greater than the value of χ2 = 3.842, showing that nearly all of the variation in the output of the frontier production can be considered as the result of the level of attainment of cost efficiencies related to the managerial matters in the management of broiler farm business. In the study by Nchinda and Thieme (2012) in domestic poultry in South Artibonite Department of Haiti obtained gamma value of the function cost at 0.582 while the study by Ogundari et al (2006) on corn farmers in Ondo State Nigeria got gamma value of the cost function at 0.805. These gamma values are lower compared to the value obtained in the current study.

Technical efficiency analysis

The value of gamma (γ) obtained in broiler farms whose pattern is closed house system is equal to 0.984 and gives a very significant effect on the confidence level of α = 0.01. Such value indicates that the variation of errors due to the technical efficiency is 98.4%, or the difference between the actual production and the possibility of maximum production 98.4% is triggered by the differences in the technical efficiency and 1.6% is due to variables outside the control or error in the measurement. It is also supported by the value of the LR test of the one-sided error obtained in broiler farmers whose pattern is highly explicit closed house system. The LR test values obtained from the broiler farms with closed house system is 66.672 and such value of the LR test is much greater than the value of χ2 = 3.842. It shows that almost all of the variations in the output of the frontier production can be considered as the result of technical efficiency level of attainment related to the managerial matters in the management of broiler farm business. The value of technical efficiency is within the range of 0.73 to 0.98 of which average is 0.93. The distribution of the value of the technical efficiency is shown in Table 3.

The present results show that the level of technical efficiency achieved per individual farmers varied. The variation is caused by the differences of managerial ability, especially in the term of setting, formulating and using factors of production to yield a number of products. If the average farmer in the sample area were to reach the TE level of its most efficient counterpart, then the average farmer could experience a cost saving of 5.10 percent (1- [0.93 / 0.98] x100). The same condition applies to the farmers whose technical efficiency is on the lowest level. If farmers are able to achieve the highest efficiency, they will be able to save about 25.51% (1- [0.73 / 0.98] x100).

Analysis of the sources of the inefficiency

Estimate of the effects of the technical inefficiency of the Stochastic Frontier production function in broilers farms using closed house system pattern shows that statistically all determinant variables included in the model are not significant (Table 4). However, some of the variables included in the model of technical inefficiency have both positive and negative directions. Findings result is positive sign indicates that the higher the level of formal education of farmers will increase also the technical inefficiency and male farmers could not decrease the technical inefficiency. In addition, male farmers are not shown to reduce the level of the technical inefficiency presumably because the broiler farms using closed house system pattern use farming equipment that can be operated by both men and women. The positive variables indicate that these factors have contributed to the increment of the technical inefficiency.The age of farmers gives negative effect; therefore, such results do not correspond to the initial hypothesis saying thatthe older the farmers, the higher the inefficiency will be. The difference [which difference?] is could bedue to the fact that the equipment of the closed house system is operatable for a wide range of ages. These findings are similar to those by Kabede (2001) and Todsadee et al (2012) who reported that age is negatively correlated to technical inefficiency.Thefactor of farm business experience has no significant effect tothe negative correlation on the technical inefficiency. These findings are similar to the hypothesis that the more experience, the more ability in enhancing everything related to the farm business management in order to increase the production of broilers. Similarly, Udoh and Etim (2009) and Ezeh et al (2012) found that the experience factor was positively correlated to the technical efficiency. The variable of the number of the broiler farmers’dependents indicates that the greater the number of family dependents, the lowerthe level of technical inefficiency will be due to the role of family members in the farm business. The average number of the farmers dependents in the closed house system of the study area is about 3.4 people (3 to 5 people). In addition, the study by Ezeh et al (2012) showed that the number of dependents was positively correlated to the technical efficiency.

Analysis of economic efficiency

The results of the analysis of economic efficiency in Table 5 show the inefficiency effects of the broiler production costs. The value of gamma (g) which is 0.99 significant at 1% level implies that about 99% of the total production is triggered by differences in cost efficiency of the farmers.The economic efficiency predicted is the opposite of the cost efficiencies of which value is varied among farmers. The value of economic efficiency ranged from 0.860 to 0.999 of which average was 0.931. The range of the economic efficiency value indicates that the farmers were quite efficient in using production factors by the lowest possible cost for gaining the highest possible yield of production. If the average farmer is able to achieve the highest possible level of economic efficiency as the most efficient farmers do, the average farmer will able to save costs by 7% (1- (0.93/ 0.99) x100). The same calculation applied for the most inefficient farmer economically provides data in which he/she will gain efficiency level by 14% (1- (0.86 / 0.99) x100).

Analysis of allocative efficiency

Based on the allocative efficiency analysis results in Table 6, it can be seen that there is a difference among farmers. Allocative efficiency values ranged from 0.88 to 1.31 of which average is 1.01. The value of allocative efficiency suggests that if the average farmer in the sample is able to reach the level of allocative efficiency of the most efficient farmers, the average farmer will be able to save 23% of costs (1- (1,01 / 1,31) x100). The same calculation applied to a farmer who is not efficient provides data in which they can save cost by 33% (1- (0.88 / 1.31) x100). Based on the distribution value of allocative efficiency that 47,3% are in the interval from 0.90 to 0.99. The lowest value achieved allocative efficiency suggests that farmers in the study area is quite efficient in producing broiler. This shows that the breeder has efficient in minimizing the cost to produce broilers.

Conclusions

•  Variables affecting broiler production applying pattern of closed house system are DOC, feed, medicine (drugs, vaccines and vitamins) and input variables such as the cost of DOC, feed and electricity which give highly significant effect on the cost of production.
  
  
• The determinant variables which gave positive influence on the technical inefficiency are the business experience and the number of dependents. However, age, sex and age education level had no negative effect on the technical inefficiency.
  
  
•  Broiler farms using closed house system pattern in Malang district have not been technically and economically efficient despite its high level achievement of the average technical efficiency. The average level of technical and economic efficiency was 0.93 and 0.92, respectively. On the other hand, both groups of farmers have been efficient allocatively.

References

Aigner D J, Lovell C A K and Schmidt P 1977 Formulation and estimation Stochastic Frontier Production Function Model. Journal of Econometrics 5(6): 21-37.

Battese G E and Coelli T J 1995 A model for technical inefficiency effects in stochastic frontier production for panel data. Empirical Economics 20: 325–345.

Coelli T, Prasada R and Battese G E 1998An introduction to efficiency and productivity analysis.Kluwer Academic Publishers, Boston.

Ezeh C I, Anyiro C O and Chukwu J A 2012 Technical efficiency in poultry broiler production in Umuahia Capital Territory of Abia State, Nigeria. Greener Journal of Agricultural Sciences 2(1): 001-007.

Jondrow J, Lovell C A K, Materov I and Schimidt P 1982 On estimation of technical inefficiency in the stochastic frontier production model. Journal Econometrics 19(1): 233-238.

Kabede T A 2001 Farm household technical efficiency: A Stochastic frontier analysis. A study of rice producers in Mardiwatershed in the western development region of Nepal. Thesis. Department of Economics and Social Sciences, Agricultural University of Norway. Norway.

Kumbhakar S C and Lovell C A K 2000 Stochastic frontier analysis: an econometric approach. Cambridge University Press, Cambridge.

Martin M and Taylor T G 2003 A comparison of the efficiency of producers under collective and individual modes of organization. Colombian Economics Journal 1(1): 246-266.

Meeusen W and Van den Broeck J 1997 Efficiency estimation from Cobb-Douglas Production Functions with composed error. International Economic Review18:435-444.

Nchinda V P and Thieme O 2012 Technical efficiency of family poultry husbandry in Artibonite and South Departments of Haiti. Livestock Research for Rural Development 24 (12) 2012 Article#218. Retrieved April 2, 2015 from http://www.lrrd.org/lrrd24/12/nchi24218.htm

Ogundari K, Ojo S O and Ajibefun I A 2006 Economies of scale and cost efficiency in small scale maize production: Empirical evidence. Nigeria Journal of Social Science 13(2): 131-136.

Ogundari K and Ojo S O 2007 An examination of technical, economic and allocative efficiency of small farms: The case study of cassava farmers in Osun State of Nigeria. Bulgarian Journal of Agricultural Science 13:185-195.

Schmidt P and Lovell C A K 1979 Estimating technical and allocative inefficiency retative to stochastic production and cost frontiers. Journal of Econometrics 19(3): 343 – 366.

Todsadee A, Hiroshi K, Kamol N and Koh-en Y 2012 Production efficiency of broiler farming in Thailand: A Stochastic frontier approach. Journal of Agricultural Science 4(12):221-231.

Udoh E J and Etim N A 2009 Measurement of farm level efficiency of broiler production in Uyo Akwa Ibom State, Nigeria. World Journal of Agriculture Sciences 5(s):832-836.

Received 2 June 2015; Accepted 15 July 2015; Published 1 September 2015

Go to top