International Journal of Agricultural Education and Extension

Vol. 6(2), pp. 371-378, June, 2020. © www.premierpublishers.org. ISSN: 2167-0432

Research Article

Economic Efficiency Analysis of Smallholder Sorghum

Producers in West Hararghe Zone, Ethiopia
*Azeb Lemma1, Tekalign Diyana2, Girma TekleHana2
1, 2Department of Agricultural Economics Oda Bultum University, Chiro, Ethiopia

The study was aimed at analyzing the economic efficiency of sorghum producing smallholders in
West Hareghe zone. It was based on cross-sectional data of 200 sample sorghum producing
households randomly selected. The estimation of stochastic frontier production function
indicated that labor, DAP fertilizer, area, seed and oxen power affects sorghum yield positively.
The estimated results showed that the mean technical, allocative and economic efficiencies were
78.9%, 38.6% and 33.6% respectively which indicates the presence of inefficiency in sorghum
production in the study area. Among factors hypothesized to determine the level of efficiencies,
frequency of extension contact had positive relationship with technical efficiency and it was
negatively related to both allocative and economic efficiencies, while soil fertility was also found
to significantly influence technical efficiencies positively and experience has positive
relationships with technical efficiency and allocative efficiency and slope significantly affects
technical efficiency negatively. The result also indicated that cultivated land was among
significant variables in determining technical efficiency and economic efficiency of farmers in the
study area. Education was found to significantly determine allocative and economic efficiencies
of farmers positively. The result indicated that there is a room to increase the efficiency of
sorghum producers in the study area. Therefore, emphasis should be given to improve the
efficiency level of those less efficient farmers by adopting and using the best practices of
relatively efficient farmers.

Keywords: Sorghum, Economic efficiency, Cobb-Douglass, Stochastic frontier.

INTRODUCTION

Ethiopia, the country with an area of about 1.12 million agricultural dominance, there were more than seven
square kilometers, is one of the most populous countries million peoples in need of food assistance in the country.
in Africa with the population of 112 million in 2019 with The country is food insecure mainly due to lack of
annual growth rate of 2.6% this growing population improved technology and economic inefficiency in
requires better economic performance than ever before at production. The smallholder farmers, who are providing
least to ensure food security. However, the agricultural the major share of the agricultural output in the country,
sector in the country is characterized by small-scale, commonly employ backward production technology and
subsistence-oriented, an adverse combination of agro limited modern inputs. Hence, being an agriculturally
climatic, demographic, economic and institutional dependent country with a food deficit, increasing crop
constraints, and heavily dependent on rainfall. Ethiopian production and productivity is not a matter of choice rather
agricultural sector contributes 46.4% of the country’s GDP, a must to attain food self-sufficiency (World Food
employs 83% of total labor force and contributes 90% of Programe, 2015).
exports.

Even though Ethiopia is the country with largest grain *Corresponding Author: Azeb Lemma; Department of
producers in Africa it is characterized by large pockets of Agricultural Economics Oda Bultum University, Chiro,
food insecurity and a net importer of grains. Despite Ethiopia. Email: [email protected]

Economic Efficiency Analysis of Smallholder Sorghum Producers in West Hararghe Zone, Ethiopia
Lemma et al. 372

According to CSA (2017) within the category of grain zone has a population density of 124.23. While 160,895 or
crops, cereals are the major food crops both in terms of 9.36% are urban inhabitants, a further 10,567 or 0.56% are
the area they are planted and volume of production pastoralists. The topography of West Hararghe is
obtained. They are produced in larger volume compared characterized by steep slopes in the highlands and mid-
with other crops because they are the principal staple highlands and large plains in the lowland areas. The
crops. Cereals are grown in all regions with varying highlands and mid-highlands are normally extensively
quantity as shown in the CSA survey results. Out of the cultivated but only partially protected by soil conservation
total grain crop area, 79.3% (9,588,923.7 ha) was under structures and practices such as grass strips, alley
cereals. The proportion of the crop grain areas for teff, cropping and bench terraces. The Zone is characterized
maize, sorghum and wheat took up 22.6% (about by crop-livestock mixed farming system where livestock in
2,731,111.7 ha), 17% (about 2,054,723.69 ha), 15.9% general and dairy production in particular contribute
(1,923,717.5 ha and 11.9% (1,437,484.7 ha), respectively. significantly to farmer livelihoods used as cash income
Sorghum accounts for an average ten percent of daily generating purpose. The major crops grown in the area are
caloric intake of households living in the eastern and sorghum barley, wheat, and pulses production and
northwest areas of the country. About three-quarters of the sorghum are primarily produced followed by maize. Khat
sorghum grain in Ethiopia is used for making injera (the and Coffee is an important cash crop of this Zone. Over 50
traditional bread, made from teff in more productive areas square kilometers is planted with this crop of coffee and
of the country). Another 20 percent is used for feed and for the climatic condition is conducive to livestock production.
local beer production, with the remainder held for seed.
The entire plant is utilized, with sorghum stalks used for Data Source and Sampling Procedure
house construction and cooking fuel and leaves used for
animal fodder (GAIN, 2017). Both primary and secondary data were used for this study.
Primary data were collected using semi structured
Research institutions claim that it is possible to produce questionnaires in two stages. First a preliminary survey
50-60 qt of sorghum per ha if improved technologies and was conducted through focus group discussion (using
practices are used appropriately. Yet, the average checklist) to obtain general information about the study
productivity level of sorghum in the study areas is about 22 area. Then formal survey data collection was undertaken
qt/ha, which is below minimum potential yield level. with the sampled households and secondary data are
Despite increase in the use of improved inputs especially collected from different published and unpublished
fertilizers, the productivity level is so low. This is an materials.
indication that farmers are not using inputs efficiently. If the
existing production system is not efficient, introduction of So, the study followed the formal survey procedure where
new technology could not bring the expected data collection for quantitative information is gathered
improvements in the productivity of sorghum and other using semi-structured questionnaire and selecting a
crops. Given the existing technology, improvements in the representative sample from a given population. Since the
level of technical efficiency will enable farmers to produce sample selected from a given population is expected to
the maximum possible output from a given level of inputs. represent the population as a whole, homogeneity of the
Hence, improvement in the level of technical efficiency will population is very important. As far as the agro-ecology
increase productivity. Theoretically introducing modern and farming system of the study area is concerned, it is
technologies can increase agricultural output. However, more or less homogenous. Hence, multi stage random
according to Tarkamani and Hardarkar (1996), cited in sampling technique was implemented to draw a
Mustefa (2014) in areas where there is inefficiency, trying representative sample. In the first stage among twelve
to introduce a new technology may not have the expected sorghum producing districts two districts are selected, in
impact and “there is a danger of trying to rediscover the the second stage five kebeles from the districts were
wheel” if the existing knowledge is not efficient. selected randomly based on probability proportional to
size of kebeles in the districts, finally following the
establishment of a sample frame for sorghum growing
MATERIAL AND METHODS farmers in each of the five kebeles, the sample households
were selected using simple random sampling with
Description of the Study Area probability proportional to size. The sample size was
determined by the formula given by (Yemane, 1967).
The study was conducted in West Hararghe zone, Oromia 𝑁
𝑛=
National Regional State of Ethiopia and the capital city of 1 + 𝑁(𝑒 2)
the Zone is located about 326 km east of the capital city of Where: n= Sample size, N= Population size and e= level
Addis Ababa along the main road. Based on the population of precision by taking” e” as 7.5%, total number of
projection value (2018) report, West Hararghe has a household head 10277the sample size would be 200.
population of 1,951,706, an increase of 47.16% over the Accordingly, a total of 200 sorghum producing households
1994 census, of whom 989,861 are men and 961,845 were randomly drawn in 2019 production year.
women; with an area of 15,065.86 square kilometers, the

Economic Efficiency Analysis of Smallholder Sorghum Producers in West Hararghe Zone, Ethiopia
 Int. J. Agric. Educ. Ext. 373

Econometric model analysis Where Ci is the minimum cost of production; ɷ i is a vector

of input prices for the ith firm; Yi* refers to farm output
The Stochastic Frontier Model will be used to estimate the which is adjusted for noise vi and α is a vector of
level of technical, allocative and economic efficiencies of parameters to be estimated from primal function. To
sorghum producers. The model was simultaneously estimate the minimum cost frontier analytically from the
developed by Aigner et al. (1977) and Meeusen and Van production function, the solution for the minimization
Den Broeck (1977). problem is given as:
6

The model is defined as: 

i  F(X i ;  )  i Min ∑ C = ∑ Xj ɷj
x j=1
Where Yi is measures the quantity of output of the ith firm, Subject to Yi∗ = 𝐴̂ ∏ X j 𝛽𝑗
̂
Xi is a vector of inputs used by ith firm, f(Xi ; β) is a suitable
production functionand,is vector of unknown parameters Where 𝐴̂ = Exp(β̂o ), ɷj is input prices,β̂j is parameter
to be estimated. i, is the composed disturbance term estimate of the stochastic production function and Yi*is
which equals (𝑣𝑖 − 𝑢𝑖 ). The random disturbance term vi is input oriented adjusted output level. Then, substituting the
intended to capture the effects of the stochastic noise and input demand equations derived using shepherd`s lemma
assumed to be independently and identically distributed [N and output adjusted for stochastic noise in the
(0,σ2v )] while ui captures technical inefficiency and minimization problem above, the dual cost function can be
assumed to be independent of vi . It also assumed to be written as follows:
μ
independently and identically distributed as half-normal, C(Yi∗ , ɷ, αj ) = HYi∗ ∏ ɷj αj
u~N (0,σ2u ).The variance parameters are expressed asσ2 = j
𝜎𝑣2 +𝜎𝑢2 and λ = σu/σv where, it is the ratio of the standard 1 ̂
𝛽
errors of the non-symmetric to symmetric error term. Where; αj = μβ̂j , μ = (∑ β̂j )-1 and H = (𝐴̂ ∏ 𝛽̂j 𝑗 )-µ
μ
Battese and Corra (1977) proposed γ =𝜎𝑢2 / (𝜎𝑣2 +𝜎𝑢2 ), The economically efficient input vector for the ith firm, Xie
instead of λ. However, there is an association between γ derived by applying Shepard’s Lemma and substituting the
and λ. According to Bravo and Pinheiro (1997) gamma can firms input price and adjusted output level into the resulting
be formulated as γ = [λ2 / (1+ λ2)].The reason is that λ could system of input demand equations.
be any non-negative value while γ ranges from 0 to 1,such ∂Ci
that the value of zero is associated with the traditional = Xie = (ɷ i, Yi ∗; α)
∂ɷn
response for which ui, is absent from the model, i.e. perfect Where n = 1, 2, 3...n are inputs used. The observed,
efficiency in production. The value one indicating that all technically and economically efficient costs of production
the deviations from the frontier are due to entirely technical of the ith firm are then equal to ɷ′i Xi , ɷ′i Xit ,ɷ′i Xie
inefficiency i.e. the random error on production is zero. A respectively. Those cost measures are used to compute
Stochastic Frontier model requires a prior specification of TE, AE and EE induces for the ith firm as follows:
the functional form, for this study, Cobb-Douglas ɷ′i Xit
production function will be used. Thus, to estimate a Cobb- TEi = ′ = Yi/Yi ∗
ɷi X i
Douglas production function, we must log all of input and
output data before analyzing the data (Coelli 1995). The
Where, Y* = frontier output and Y= actual yield
linear form is specified as:
ln(output) = β0 + β1 ln(land) + β2 ln(labor) + β3 ln(seed)
EEi =ɷ′i Xie /ɷ′i Xi = C*/C
+ β4 ln(Urea) + β5 ln(oxen) + β6 ln(Dap) + εi Where, C* = minimum (efficient) cost
C= actual cost
Technical efficiency (TE) of an individual firm is estimated AEi = ɷ′i Xit /ɷ′i Xi
as the ratio of the observed output to the corresponding 𝐴𝐸 = 𝐸𝐸𝑖/𝑇𝐸𝑖
frontier output. The measure is given as: Each of these efficiency measures takes a value between
Yi f(X i ; β)exp(vi− ui ) zero and one inclusive, with a value of one, indicating full
TEi = ∗ = = exp(−u) efficiency.
Y f(Xi ; β)exp(vi )
Determinants of efficiency
Following Bravo-Ureta and Rieger (1991) adjusted output
Y* is used to derive the technically efficient input vector,
To analyze the effect of demographic, socioeconomic,
X*. The technically efficient input vector for ith firm, is
farm attributes, institutional variables on efficiencies, a
derived by simultaneously solving equation and the
X second step procedure will be used where the estimated
observed input ratio 1 =k (i>1) where K i is equal to the efficiencies scores are regressed on selected explanatory
Xi
observed ratio of the two inputs in the production variables using censored Tobit model. This model is best
ofY*.Sharma et al. (1999) suggests that the corresponding suited for such analysis because of the nature of the
parameter of the dual cost frontier can be derived dependent variable (efficiency scores), which takes values
algebraically and written in a general form as: between 0 and 1 and yield the consistent estimates for
Ci=C(ɷ i,Yi*;α) unknown parameter vector (Maddala, 1999). Estimation

Economic Efficiency Analysis of Smallholder Sorghum Producers in West Hararghe Zone, Ethiopia
Lemma et al. 374

with OLS regression of the efficiency score would lead to Production function variables
a biased parameter estimate since OLS regression
assumes normal and homoscedastic distribution of the The relative contribution of both usual noises and the
disturbance and the dependent variable Greene (2003). inefficiency component on total variability should be
Following Tobin (1958) the model can be specified as: determined. The ratio of the standard error of u (σu) to the
n
standard error of v (σv), known as lambda (λ), is 2.146.
Ei∗ = ∑ βj Xj + Vi Based on λ, gamma (γ) which measures the effect of
i=1 technical inefficiency in the variation of observed output
1if Ei∗ ≥ 1 can be derived (i.e. γ= λ2/ [1+λ2]). In this case, the value of
Ei = {Ei∗ if E𝑖∗ < 1 this discrepancy ratio (γ) calculated from the maximum
0if Ei∗ ≤ 0 likelihood estimation of the full frontier model was 0.822
Where Ei is an efficiency score, representing technical, with standard error of 0.016 and it is much higher than its
allocative and economic efficiencies; and 𝑉 ∼ 𝑁(0 , 𝛿 2 )and standard error.
βj are the vector of parameters to be estimated. The Xj
represents various farm specific variables and Ei∗ is the Table 1. Parameter Estimates of the Stochastic
latent variable, with 𝐸 [ Ei∗ ⁄Xi ] equals Xi β. Production Frontier Model
Variables Coefficient Standard Error
Labour 0.486*** 0.171
RESULTS OF ECONOMETRIC ANALYSIS DAP 0.121* 0.059
UREA 0.101 0.187
Hypothesis Test Seed 0.036* 0.019
Area 0.924*** 0.167
Before proceeding to the estimation of the parameters
Oxen 0.376*** 0.182
from which individual level of efficiencies are estimated, it
Constant 1.095*** 0.391
is essential to examine various assumptions related to the
model specification. To do this, two hypotheses were Gamma 0.822*** 0.017
tested the first one is testing for the existence of the Lamda 2.146*** 0.007
inefficiency component of the total error term of the Sigma square 0.397*** 0.052
stochastic production function. The result reveals the *and *** Significant at 10% and 1%, significance level
existence of inefficiency or one-sided error component in respectively
the model. Hence, the hypothesis that sorghum producers Source: Model output based on survey data, 2019
in the area are efficient is strongly rejected. As a result, the
production behavior of sorghum producers of the study The coefficient for the parameter γ can be interpreted in
area can better be represented by the stochastic such a way that about 82.2% of the variability in sorghum
production function than the average response function. output in the study area in the year 2019 was attributable
The second related hypothesis to be tested was that all to technical inefficiency effect, while the remaining 17.8%
coefficients of the inefficiency effect model are variation in output was due to the effect of random noise.
simultaneously equal to zero (i.e. Ho: = δ0 = δ1 = δ2 … = This indicates that there is a room for improving output of
δ14 = 0) against the null hypothesis, which states that all sorghum by first identifying those institutional,
parameter coefficients of the inefficiency model are socioeconomic and farm specific factors causing this
different from zero. It is to mean that the explanatory variation.
variables in the inefficiency effect model do not contribute
significantly to the explanation of the economic inefficiency The dual frontier cost function derived analytically from the
variation for the sorghum-growing households. So based stochastic production is given as:
on the test result, the null hypothesis is rejected in favor of 𝑙𝑛𝐶𝑖 = 5.450 + 0.894𝑙𝑛𝑌𝑖 ∗ +0.35 𝑙𝑛ɷ𝑙𝑎𝑏𝑜𝑟 + 0.346 𝑙𝑛ɷ𝑜𝑥𝑒𝑛
+ 0.1010 𝑙𝑛ɷ𝑈𝑅𝐸𝐴 + 0.1094 𝑙𝑛ɷ𝐷𝐴𝑃
the alternative hypothesis that explanatory variables
+ 0.4873 𝑙𝑛ɷ 𝐴𝑟𝑒𝑎 + 0.0305 𝑙𝑛ɷ 𝑠𝑒𝑒𝑑
associated with inefficiency effects model are
Where C is the minimum cost of production of the ith
simultaneously different from zero.
farmer, Y* refers to the index of output adjusted for any
Table 1. Generalized Likelihood Ratio test of hypotheses statistical noise and scale effects and ɷ stands for input
for parameters of SPF prices.
Null λ Degree Critical Decision
hypothesis of value Technical, allocative and economic efficiency scores
freedom (X2,
95%) The model output indicates that farmers in the study area
H0: γ = 0 17.32 1 6.34 Rejected were relatively good in technical efficiency than allocative
efficiency or economic efficiency. The mean level of
H0: = δ0 = δ1 = 41.23 14 22.72 Rejected
δ14 = 0 technical efficiency of sorghum growing sample
households was about 78.9%. This means in the short run
Source: Model output based on survey data, 2019

Economic Efficiency Analysis of Smallholder Sorghum Producers in West Hararghe Zone, Ethiopia
 Int. J. Agric. Educ. Ext. 375

there are opportunities for reducing sorghum production Table 4. Frequency distribution of efficiency estimates of
inputs by 21.1% through adopting technologies used by sample farmers
the best practice of sorghum farmers. The mean allocative TE AE EE
efficiency of farmers in the study area was 38.6% Efficiency N Percent N Percent N Percent
indicating there is a need to improve the present level of level
allocative efficiency. The estimates depicted that the 00-09.999 0 0.00 0 0.00 0 0.00
farmers have ample opportunities to increase their 10-19.999 0 0.00 2 1.00 16 0.00
allocative efficiency. For instance, farmer with average 20-29.999 0 0.00 15 7.50 77 38.50
level of allocative efficiency would enjoy a cost saving of 30-39.999 0 0.00 105 52.50 98 49.00
about 32.6% derived from (1 – 0.386/0.573)*100 to attain 40-49.999 0 0.00 64 32.00 7 3.500
the level of the most efficient farmer. The most allocative 50-59.999 16 8.00 14 7.00 2 1.00
inefficient farmer would have an efficiency gain of about 60-60.999 25 12.50 0 0.00 0 0.00
65.6% derived from (1-0.197/0.573)*100 to attain the level 70-79.999 30 15.00 0 0.00 0 0.00
of the most technically efficient farmer.
80-89.999 87 43.50 0 0.00 0 0.00
90-99.999 42 21.00 0 0.00 0 0.00
The mean economic efficiency showed that there was a
Source: Model output based on survey data, 2019
significant level of inefficiency in the production process.
That is the producer with an average economic efficiency Determinants of efficiency differentials among
level could reduce current average cost of production by farmers
66.4% to achieve the potential minimum cost level without
reducing output levels. It can be inferred that if farmers in After measuring levels of farmers’ efficiency and
the study area were to achieve 100% economic efficiency, determining the presence of efficiency differences among
they would experience substantial production cost saving farmers, finding out factors causing efficiency disparity
of 66.4%. This implies that reduction in cost of production among farmers was the next most important step of this
through eliminating resource use inefficiency could add study. The result of Tobit model showed that; frequency of
about 66.4% of the minimum annual income. The result extension contact, cultivated land and educational level of
also indicated that the farmer with average level of household head were important factors influencing
economic efficiency would enjoy a cost saving of about economic efficiency of farmers in the study area. The
34.6% derived from (1-0.3360/0.514)*100 to attain the model also revealed that three variables were found to
level of the most efficient farmer. The most economically significantly influence allocative efficiency of sorghum
inefficient farmer would have an efficiency gain of 66.5% producers. These variables were frequency of extension
derived from (1-0.173/0.516)*100 to attain the level of the contact, experience of house hold head on sorghum
most efficient farmer production and educational level of the household.
However, the sign for extension contact in allocative and
Table 3. Summary of descriptive statistics of efficiency economic efficiencies was not as anticipated. The result
measure also show that frequency of extension contact, soil fertility,
Types of Minimum Maximum Mean Standard experience, slope and cultivated land were found to be
Efficiency deviations statistically significant to affect the level of technical
Technical 0.518 0.943 0.789 0.072 efficiency of farmers.
Efficiency Frequency of extension contact: Frequency of extension
Allocative 0.197 0.573 0.386 0.057 contact had statistically significant positive relationship
Efficiency with technical efficiency at 1% significance level as it was
Economic 0.173 0.516 0.336 0.045 expected. This implies that a frequent contact facilitates
Efficiency the flow of new ideas between the extension agent and the
Source: Model output based on survey data, 2019 farmer thereby giving a room for improvement in farm
efficiency. Advisory service rendered to the farmers in
Frequency distribution of efficiency estimates of sample general can help farmers to improve their average
farmers is presented in Table 20, Most of households had performance in the overall farming operation as the service
a higher technical efficiency levels. About 21% of sorghum widens the household’s knowledge with regard to the use
farmers in the study area were operating above the of improved agricultural inputs and agricultural
efficiency level of 90% and 43.50% of them were operating technologies. This result is also similar to those obtained
in the range of 80-90% of technical efficiency levels. On by Jema, (2008) and Tolosa et al. (2019). However, the
the other hand, none of the farmers was operating below negative coefficient of extension contact which is
50% of technical efficiency level. The result indicated that significant at one percent in allocative efficiency and
the potential to improve sorghum productivity for individual economic efficiency indicates that efficiencies in resource
farmers through improvement in the level of TE is the allocation are deteriorating as the frequency of extension
smallest as compared to that of the AE and EE. contact increases. This may be due to the fact that
extension workers are only interested in maximizing output
at any cost.

Economic Efficiency Analysis of Smallholder Sorghum Producers in West Hararghe Zone, Ethiopia
Lemma et al. 376

Table 5. Maximum likelihood estimates of the tobit model

Variables TE AE EE
Coef. Std. Err Coef. Std. Err Coef. Std. Err
Constant 0.5286*** 0.0437 0.5682*** 0.0325 0.3083*** 0.0087
EXTNCON 0.0040*** 0.0009 -0.0026*** 0.0006 -0.0025 *** 0.0012
Soil fertility 0.0782*** 0.0197 0.02416 0.0145 0.0004 0.0039
Experience 0.0027*** 0.0009 0.0018** 0.0007 0.0001 0.0002
House hold size 0.0002 0.0038 -0.0007 0.0028 -0.0008 0.0008
slope -0.0398* 0.0216 -0.0167 0.0160 -0.0012 0.0090
Cultivated land 0.0358* 0.0195 -0.0110 0.0145 0.0108*** 0.0039
Fragmentation 0.0078 0.0170 -0.0098 0.0126 -0.0012 0.0034
OFNFA 0.0050 0.0176 -0.0104 0.0131 -0.0016 0.0035
Livestock -0.0000 0.0014 -0.0000 0.0010 -0.0002 0.0002
Credit 0.0221 0.0221 -0.0060 0.0164 0.0031 0.0044
Education 0.0108 0.0187 0.0126* 0.0077 0.0082** 0.0037
Weeding 0.0151 0.0221 0.0034 0.0138 -0.0010 0.0020
Proximity -0.0007 0.0054 0.0013 0.0040 0.0002 0.0011
Inter cropping 0.02416 0.0145 -0.0041 0.0097 0.0046 0.0079
*, **and***Significant at 10%, 5% and 1%, significance level respectively
Source: Model output based on survey data, 2019
Perceived fertility status of soil: It was hypothesized that obtained by Azeb et al (2019), which slope significantly
the fertility status of the soil would have significant impact and negatively determines efficiency of sorghum producer.
on the output of sorghum. The result indicates that the Cultivated land: Total cultivated or farm land was found to
coefficient of fertility of soil is positive and statistically have significant and positive impact on technical efficiency
significant at one percent level of significance implying that and economic efficiency at ten and one percent
fertility of soil is an important factor in influencing the level respectively, which is in line with the hypothesis made.
of efficiency in the production of sorghum. Therefore, This might be because of Farmers with larger area of
development programs in improving and maintaining the cultivated land have the capacity to use compatible
fertility of land will have positive impact in raising efficiency. technologies that could increase the efficiency of the
The study result is similar with Ermiyas et al. (2015) that farmer, enjoy economies of scale. This result is in line with
showed that soil fertility had positive relationship with the argument of Mustefa (2014). Therefore, larger farms
efficiency. are relatively better efficient than small size farms.

Experience: The estimated coefficients of experience for Education: The coefficient for educational level was
technical efficiency was positive signs and significant at significant and positively related to allocative and
one percent. Dawang et al. (2011) and Azeb et al. (2019) economic efficiencies at ten and five percent respectively.
which is because of the accumulated experiences that The result indicates that, allocative efficiency and
have been gathered over time they become skillful as they economic efficiency require better knowledge and
get older and may have an interest in the use of new managerial skill. The positive sign indicates that increase
methods of production. The estimated coefficients of in human capital enhances the efficiency of farmers. The
experience for allocative efficiencies were also positive sign was as expected because the more educated the
and significant at five percent significance level. This may farmers are the more will be the likelihood of being efficient
be because allocative efficiency requires greater in resource allocation. This result is consistent with that of
knowledge and skill gathered over time, which increases Musa (2013) and Tolosa et al. (2019). Because of their
the capacity of farmers for optimal allocation of resources better skills, access to information and good farm planning;
and technology. Therefore, experienced the farmers are literate farmers are better to manage their farm resources
the more technically and allocatively efficient they are. and agricultural activities and willing to adopt improved
production technologies.
Slope: The slope of land was hypothesized to determine
technical efficiency negatively. The result shows that slope
determined efficiency negatively and it is statistically CONCLUSION AND RECOMMENDATIONS
significant at ten percent level of significance. This is
because as slope increase soil erosion also increase and Thus, the results of the study give information to policy
the nutrients (fertilizers) applied to the soil also lost through makers and extension workers on how to better aim efforts
erosion and this can reduce the availability of nutrients to to improve farm efficiency as the level and specific
the crop and consequently minimize the yield obtained and determinant for specific efficiency types are identified. This
efficiency of the farmers. The result is similar to that could contribute to compensation of high production cost,

Economic Efficiency Analysis of Smallholder Sorghum Producers in West Hararghe Zone, Ethiopia
 Int. J. Agric. Educ. Ext. 377

hence improve farm revenue, welfare and generally help with increased cultivated area, technical and allocative
agricultural as well as economic development. These efficiencies of sorghum production would increase. This
findings stress the need for appropriate policy formulation might be because of farmers with larger area of cultivated
and implementation to enable farmers reduce their land have the capacity to use compatible technologies that
inefficiency in production as this is expected to have could increase the efficiency of the farmer, enjoy
multiplier effects ranging from farm productivity growth to economies of scale. Thus, provision of technologies that
economic growth and poverty reduction at macro level. would help to carry out such operations more efficiently
would improve the technical and allocative efficiencies
In the study area, levels of different types of efficiencies level of the farmers and complementary inputs including
and their determinants were found to be different and employment of labor.
allocative efficiencies and economic efficiencies were
found to be low. Therefore, an intervention aiming to Education was very important determining factor that has
improve efficiency of farmers in the study area has to give positive and significant impact to both allocative efficiency
due attention for resource allocation in line with output and economic efficiency in the study area. It is central to
maximization as there is big opportunities to increase adopt and use modern agricultural technologies and
output without additional investment. practices, agricultural information and institutional
accessibilities which in turn increase and improve farm
The result indicated that extension contact has positive household’s efficiencies. Thus, government has to give
and significant contribution to technical efficiency. Since due attention for training farmers through strengthening
extension services are the main instrument used in the and establishing both formal and informal type of framers'
promotion of demand for modern technologies, education, farmers' training centers, technical and
appropriate and adequate extension services should be vocational schools as farmer education would reduce both
provided. This could be done by designing appropriate allocative and economic inefficiencies.
capacity building program to train additional development
agents to reduce the existing higher ratio of farmers to Finally, it is interesting to note that most efficiency studies
development agents as well as to provide refreshment in the developing countries have focused mainly on the
training for development agents. The study also indicated measurement of technical efficiency, even though it is by
that extension contact has negative and significant improving the overall economic efficiency that major gains
contribution to allocative and economic efficiencies. in production could be achieved. This means, additional
Therefore, extension agents have to give due attention for efforts should be devoted to examining the impact of both
appropriate input allocation and cost minimization in allocative and economic efficiencies on performance for
addition to their acknowledgeable effort to increase different types of crops and areas at various points in time.
production. This calls for the need to more effective policy
support for extension services and additional efforts need
to be devoted to upgrade the skills and knowledge of the REFERENCES
extension agents.
The result of the study shows that soil fertility is a crucial Agricultural Growth Program (2012). Global Agriculture
factor in determining technical efficiency of households. and Food Security Programme Request for Funding
Therefore, households have to work to improve the fertility Public Sector Window. Summary on Agriculture and
status of the soil though it is difficult to achieve this in the Food Security Strategy and Post Compact Policy and
short run. Households can do this by applying fertilizers Investment Framework.
that are suitable for the farm and practicing soil Azeb L. Belaineh L. Mengistu K. (2019). Technical
conservation practices. Strengthening soil fertility Efficiency of Smallholder Sorghum Producers in West
maintenance program is required and extension workers Hararghe Zone, Oromia Region, Ethiopia. Journal of
can play a great role in improving the status of the soil by Agricultural Economics and Rural Development. 6(1):
working closely with the farmers in this regard. 718-725.
Battese G.E. and Broca, S.S (1996). Functional forms of
Slope significantly reduced the technical efficiency of stochastic production functions & model for technical
sorghum producers when lands are vulnerable to erosion inefficiency effects: A comparable study of wheat
damages and their fertility is likely to be poor due to high farmers in Pakistan. CEPA Working Papers, No. 4
run-off. If soil conservation measures such as check dam Department of Econometrics, University of New
and water way measures are not practiced, it reduces England, Armidale.
efficiency thereby reducing sorghum production. So, Bekele, A (2013). Technical efficiency variation for
development agents should encourage households to smallholder irrigated maize producers: stochastic
strengthen the soil conservation measures such as check Frontier Approach. The case of Tibila surface water
dam and water way to reduce soil erosion. irrigation scheme. MSc Thesis, Mekelle University,
Mekelle, Ethiopia.
Total cultivated land was found to be related to technical Bravo-Ureta B.E. and A. Pinheiro. (1997). Technical,
and allocative efficiencies level positively. This shows that economic and allocative efficiency in peasant farming:

Economic Efficiency Analysis of Smallholder Sorghum Producers in West Hararghe Zone, Ethiopia
Lemma et al. 378

evidence from Dominica Republic. Journal of Mustefa, B (2014). Economic efficiency of barely
Developing Economics, 38(1): 48-67. production: the case of chole district, in East Arsi Zone.
Coelli, T.J (1998). An Introduction to Efficiency and M.S.c. Thesis, Haramaya University, Haramaya,
Productivity Analysis. Kluwer Academic Publishers, Ethiopia.
Boston. Peter Schmidt (2000). One-Step and two-step estimation
Ermiyas, M. Endrias, G. and Belaineh, L (2015). of the effects of exogenous v
Production efficiency of sesame in Selamago district of Shumet, A (2012). Technicall efficiency in crop production
South Omo Zone, Southern Ethiopia. Current Research in Tigray region, Ethiopia stochastic frontier approach.
in Agricultural Sciences, 2(1): 8-2. Global Advanced Research Journal of Agricultural
Forsund, F. R., Lovell, C.K. and Schmidt, P (1986). Survey Science, 2(3): 9-20.
of frontier production functions and their relationships Tolosa T., Temesgen K. and Zekarias Shu (2019).
to efficiency measurement. Econometrics Journal Economic efficiency of small holder maize producers in
13:5–25. Gudeya Bila district of oromia region, Ethiopia.. Journal
Jema, H (2008). Economic efficiency and marketing of Applied Agricultural Economics and Policy Analysis.
performance of vegetable production in Eastern and 2(1):1-7.
central parts of Ethiopia, Doctoral Thesis, Acta
University Agriculture Sueciae (SLU) Uppsala,
Sweden.
Jondrow, J., C.K. Lovell, I.S. Materov and P. Schmidt
(1982). on Estimation of Technical Inefficiency in the
Stochastic Frontier Production Function Model. Journal Accepted 15 June 2020
of Econometrics, 19: 233–38
Kopp, R.J. and V.K. Smith (1980). Frontier Production Citation: Lemma A, Diyana T, TekleHana G (2020).
Function Estimates for Steam Electric Generation. A Economic Efficiency Analysis of Smallholder Sorghum
Comparative Analysis. Southern Economic Journal, Producers in West Hararghe Zone, Ethiopia. International
47:. 1049–1059. Journal of Agricultural Education and Extension, 6(2): 371-
Lovell, C Fried, H. and Schmidt, S (1993). Note on theory 378.
of productive efficiency and stochastic frontier models.
European Research Studies, 13 (4): 109-117.
Mohammed, A (2011). Analysis of technical efficiency
between extension participant and non-participant farm
households evidence from Tigray Northern Ethiopia Copyright: © 2020 Lemma et al. This is an open-access
Thesis, Haramaya University, Haramaya, Ethiopia. article distributed under the terms of the Creative
Musa, H (2013). Economic Efficiency of Smallholder Commons Attribution License, which permits unrestricted
Farmers Maize Production: The Case of Arsi Negelle use, distribution, and reproduction in any medium,
District of West Arsi Zone. MSc Thesis, Haramaya provided the original author and source are cited.
University, Haramaya, Ethiopia.

Economic Efficiency Analysis of Smallholder Sorghum Producers in West Hararghe Zone, Ethiopia