




































 

 

 
8 

† Corresponding author 
© 2015 Conscientia Beam. All Rights Reserved. 

 

PRODUCTION EFFICIENCY OF SESAME IN SELAMAGO DISTRICT OF 

SOUTH OMO ZONE, SOUTHERN ETHIOPIA 

 

Ermiyas Mekonnen1† --- Endrias Geta2 --- Belaineh Legesse3 

1Jinka Agricultural Research Center, Jinka, Ethiopia 

2Harmaya University, Ethiopia  

3Haramaya University, Ethiopia 

 

ABSTRACT 

The aim of the present study was to measure the levels of technical, allocative and economic efficiencies of 

sesame producer and identify factors affecting them in selamago district of south Omo zone, Southern 

Ethiopia.  The study was based on the cross – sectional data collected in 2011/12 production season from 

120 randomly selected farm households. Stochastic production frontier model was used to estimate technical, 

allocative and economic efficiency levels, whereas Tobit model was used to identify factors affecting 

efficiency levels. The results indicated that there was substantial amount of inefficiency in sesame production 

in the study area.  Accordingly, the mean technical, allocative and economic efficiencies of sample households 

were 67.1 per cent, 67.25 per cent and 45.14 percent respectively. Labor and seed were the variables that 

positively affected the production of sesame. Results of the Tobit model revealed that soil fertility, non farm 

income and credit access positively and significantly affected TE. Soil fertility had positive and significant 

effect on AE. On the other hand experience in sesame production, distance of sesame farm form residence, 

non farm income and extension contact affected AE negatively and significantly. Soil fertility, non farm 

income and credit access had positive and significant impact on EE. However, extension contact affected 

EE negatively and significantly. These indicate that there is a room to increase the efficiency in sesame 

production of the study area. Therefore, government authorities and other concerned bodies should take into 

consideration the above mentioned socio economic and institutional factors to improve productivity of sesame 

in the study area. 

Keywords: Technical efficiency, Allocative efficiency, Stochastic production frontier, Tobit, Selamago, Sesame. 

 

 

 

 

Current Research in Agricultural Sciences 
2015 Vol. 2, No. 1, pp. 8-21 
ISSN(e): 2312-6418 
ISSN(p): 2313-3716 
DOI: 10.18488/journal.68/2015.2.1/68.1.8.21 
© 2015 Conscientia Beam. All Rights Reserved. 

 
 
 
 
 

http://crossmark.crossref.org/dialog/?doi=10.18488/journal.68/2015.2.1/68.1.8.21


Current Research in Agricultural Sciences, 2015, 2(1): 8-21 

 

 
9 

© 2015 Conscientia Beam. All Rights Reserved. 

Contribution/ Originality 

In this paper the production efficiency of farmers in the study area and the factors that affect 

their efficiency in sesame production are identified. This study can be used as a reference for 

fellow researchers who are interested in the area of production economics specifically that deals 

with production efficiency.  

 

1. INTRODUCTION 

Ethiopia is mainly an agrarian country. The agricultural sector accounts for roughly 43 

percent of GDP, and 90 percent of exports. Nevertheless, food security remains a critical issue for 

many households, and for the country as a whole. Moreover, expansion of the cropped area to 

more marginal lands has led to severe land degradation in some areas. With a total area of about 

1.13 million km2 and about 51.3 million hectares of arable land, Ethiopia has tremendous potential 

for agricultural development. Only about 11.7 million hectares of land, however, are currently 

being cultivated; just around 20 percent of the total arable area. Nearly 55 percent of all 

smallholder farmers operate on one hectare or less [1].   

Sesame seed has become one of the most important oilseeds for Ethiopia’s export earnings 

and for increasing the potential of generating income for the local population. In the last few 

years, sesame production and marketing has demonstrated highly significant growth. In 1997, the 

total area under sesame production was about 64,000 ha [2]. In nearly fourteen years time (up to 

2011), the total area of sesame production has increased by more than 500% to about 384,683 ha. 

The practice of sesame production has also expanded from the traditionally known regions 

(Northwest Humera, Wellega and North Gonder) to many new areas, including Benishangul, 

Illubabor and many other places. Similarly, the quantity of sesame produced during the same 

period, which is mainly intended for export, has also increased from 42,000 tones  to about 

327,740.9 tones [3] which is again an increment of over 650%. The potential to increase the area, 

production and productivity of sesame is still large [4]. 

In Ethiopia, the existing production system of sesame suffers from traditional farming 

practices, unimproved seed, low fertilizer use, etc. This situation has caused productivity of the 

crop to be far below the estimated FAO potential, which is about 16 quintals/ha (Wijnands et al., 

2007 as cited in Sorsa [4]). The Ethiopian Statistical Agency report of 2011 indicated that the 

crop’s productivity level was 8.52 quintals per hectare countrywide.  However, it is understood 

that the current productivity level of sesame in Ethiopia in general and in the study area in 

particular is far below the average level. Moreover, there was no study regarding the efficiency of 

sesame production in the study area. 

Agricultural output can be increased either through introduction of modern technologies or 

by improving the efficiency of inputs such as labour and management at the existing technology. 

In other words, productivity can be increased through dissemination of improved technologies 

such as fertilizer and high yielding varieties (HYV) and/or by improving the productive capacity 

of saying the manager (the farmer). These two are not exclusive because the introduction of 

modern technology could not bring the expected shift of production frontier, if the existing level 



Current Research in Agricultural Sciences, 2015, 2(1): 8-21 

 

 
10 

© 2015 Conscientia Beam. All Rights Reserved. 

of efficiency is low. This implies the need for the integration of modern technologies with 

improved level of efficiency [5]. If farmers are sufficiently efficient then increases in productivity 

require new inputs and technology to shift the production possibility frontier upward. But, on the 

other hand, if there are significant opportunities to increase productivity through more efficient 

use of farmers' resources and inputs with current technology, a stronger case could be made for 

productivity improvement through ameliorating the factors or determinants of inefficiency. 

So it is relevant to ask, what are the existing levels of technical, allocative and economic 

efficiencies in the production of sesame in the study area? Is there any room for improvement in 

the level of efficiency? What are the main causes for the existing levels of inefficiencies in the area 

(if any)? What are the main possible solutions to reduce the existing levels of inefficiencies? By 

what level can inputs be reduced given the existing level of output? By what level can cost be 

reduced given the existing level of output? These are important research questions that have to 

be answered first if an intervention aiming at improving production and productivity of farmers is 

to be made. And this study has attempted to answer these questions. The general objective of this 

study was to assess economic efficiency of smallholder farmers in the production of sesame in 

Selamago district of South Omo zone. The specific objectives were:to estimate the level of 

technical, allocative and economic efficiencies of sesame production in the study area; and to 

identify the sources of differences in technical, allocative and economic efficiencies in the study 

area. 

 

2. METHODOLOGIES 

Selamago district is one of the eight districts found in South Omo zone of SNNPR, Ethiopia 

with an area of 4450.1 km2 and is located at 5.32-6.460 N and 35.81-36.450 E. It has a human 

population of 32135 with a population density of 7 persons per km2. It is bordering with Keffa 

zone and Basketo special district in north, Dasenech, Gnangatom and Benatsemay districts in 

south, South Ari district in east and Bench Maji zone in west. There are 20 kebeles in the district. 

The altitude of the district ranges between 485 and 2500 m, a.s.l. The traditional agro ecologies 

woinadega, kola and kefilbereha cover 32, 66 and 2 percent respectively. The mean annual rainfall 

ranges between 400 mm and 1600 mm whereas the mean annual temperature ranges between 

17.60c. and greater than 27.50c.  

The study was conducted in Selamago district which is purposively selected due to its large 

extent of sesame production. Even though the district consists of 20 kebeles, only 5 kebeles were 

engaged in the production of sesame. Out of these, 2 kebeles were selected randomly and the 

number of sample households from each kebele was determined randomly based on the 

probability proportional to size of households in each kebele. The sample size was 120 which is 

determined on the basis of the following formula given by Yamane [6]. 

n=
 

       
                                                                                                                             (1) 

For this study, both primary and secondary data from different sources were used. The 

primary data on socio-economic variables such as demographic characteristics, extension visits, 



Current Research in Agricultural Sciences, 2015, 2(1): 8-21 

 

 
11 

© 2015 Conscientia Beam. All Rights Reserved. 

credit access, livestock holding, price data, wealth indicators, amount and cost of labour used,  the 

amount and cost of inputs used such as seed and fertilizer and the amount of outputs obtained 

were collected using structured questionnaire which was administered by trained enumerators 

from February 1-7, 2013. Before starting the actual data collection some preliminary information 

about the overall farming system of the district was assessed through informal survey. Pre-

testing of the questionnaire was also conducted and appropriate refinements and modifications 

were made in the questionnaire. Secondary data were collected by reviewing relevant sources 

such as documents of the office of agriculture of the district and other relevant organizations. 

To address the objectives of the study, both descriptive statistics and econometric models 

were employed. Accordingly, in the descriptive part, simple measures of central tendency, 

frequency and percentages were used; and in the econometric analyses, a stochastic frontier 

approach and a Tobit model were used to estimate the level of technical, allocative and economic 

efficiencies and the relation between farm level socio-economic and institutional variables and 

inefficiencies, respectively. 

The stochastic frontier production function was employed to assess the technical, allocative 

and economic efficiencies of sesame producers. The function was autonomously developed by 

Aigner, et al. [7] and Meeusen and Van Den Broeck [8]. The approach offers some sensible 

advantages over the other methods that are usually used in efficiency analysis. In the first place, it 

is easy to implement and interpret. Most importantly, the model allows segregating the effect of 

statistical noises from systematic sources of inefficiency. Besides, the technique is consistent with 

most of the agricultural production efficiency studies [9-11]. 

Following Aigner, et al. [7] and Meeusen and Van Den Broeck [8] the SPF model is defined 

as: 

           ∑                               (2) 

          

Here ln denotes the natural logarithm; j represents the number of inputs used; i represents 

the ith farm in the sample; Yi represents the observed sesame production of the ith farmer; Xij 

denotes jth farm input variables used in sesame production of the ith farmer; ß stands for the vector 

of unknown parameters to be estimated; εi is a composed disturbance term made up of two 

elements (vi and ui). The random error (vi) accounts for the stochastic effects beyond the farmer’s 

control, measurement errors as well as other statistical noises and ui captures the technical 

inefficiency. 

Stochastic frontier functional approach requires a priori specification of the production 

function to estimate the level of efficiency. Cobb-Douglas production function met the 

requirement of being self dual and has been employed in many researches dealing with efficiency 

[9, 10, 12]. Therefore, it was adopted for this study. 

The production function could also be estimated through an alternative form, called dual, 

such as cost or profit function. Sharma, et al. [13] suggests that the corresponding dual cost 

frontier of the Cobb Douglas production functional form in equation (1) can be rewritten as: 

Ci = C (Wi, Yi
*; α)                                                                                                         (3) 



Current Research in Agricultural Sciences, 2015, 2(1): 8-21 

 

 
12 

© 2015 Conscientia Beam. All Rights Reserved. 

Where i refers to the ith sample household; Ci is the minimum cost of production; Wi denotes 

input prices; Yi* refers to farm output which is adjusted for noise vi and α's are parameters to be 

estimated. The economically efficient input vector of the ithhousehold Xie is derived by applying 

Shepards’ lemma [14] and substituting the firms input prices and adjusted output level, a system 

of minimum cost input demand equation can be expressed as: 

Ci/ Wn= Xie (Wi, Yi
*; α)                                                                                               (4) 

Where n is the number of inputs used. The observed, technically and economically efficient 

costs of production of the ith farm are then equal to W’Xi, W’Xit and W’ie; respectively. 

The minimum cost is derived analytically from the production function, using the 

methodology used in Arega and Rashid [14]. Given input oriented function, the efficient cost 

function can be specified as follows: 

    ∑   ∑    

 

    

 

Subject to            
    ̂∏  

 ́                              (5) 

Where  ̂       ̂   

The solution for the problem in the above equation is the basis for driving dual cost frontier. 

Substituting the input demand equations derived using shepherd`s lemma (Eq. 3) and Yield 

adjusted for stochastic noise (predicted value of yield) in the minimization problem above, the dual 

cost function can be written as follows: 

    
        

  ∏   
  

                                                                                                     (6) 

where;       ̂        ∑  ̂  -1  and     
 

 
  ̂∏  ̂  

 
 -µ 

All the Parameters are known; hence we can calculate the minimum (efficient) cost of 

production.According to Sharma, et al. [13] the above cost measures are used to estimate the 

technical, allocative and economic efficiencies respectively. 

We can define the farm–specific technical efficiency in terms of observed output (Yi ) to the 

corresponding frontier output (Yi*) using the existing technology. 

TEi = Yi/Yi*                                                                                                                (7) 

The farm specific economic efficiency is defined as the ratio of minimum observed total 

production cost (C*) to actual total production cost (C). 

EE = C*/C                                                                                                                   (8) 

Following Farrell [15] the AE index can be derived from Equations (6) and (7) as follows: 

AE = EE/TE                                                                                                                  (9) 

After estimating the level of efficiency, tobit model was estimated to identify factors affecting TE, 

AE and EE.Following Gujarati [16] the tobit model was estimated as follows: 

E* = δ0 + δkZi + v, v/z ≈ Normal (0, δ2)         

                                                                                                                          (10) 

E = max (0, E*) 



Current Research in Agricultural Sciences, 2015, 2(1): 8-21 

 

 
13 

© 2015 Conscientia Beam. All Rights Reserved. 

Where I represents thei
th 

farm in the sample; k is the number of factors affecting efficiency; 

Zi represents farm specific factors affecting efficiency; δ is parameter to be estimated; E is 

efficiency(TE, AE and EE) measure. 

Equation (10) implies that the above observed variable, E, equals E* when E*> 0, but E = 0 

when E*≤ 0. 

In SPF hypothesis tests can be made using ML ratio test that are not possible in non-

parametric models. A number of tests of hypotheses were made in this study using the usual 

Likelihood Ratio (LR) test given by Equation (10). 

LR=       [          ⁄ ] 

        λ = -2[ln L(Ho) – ln L(H1)]                                                                                      (11) 

Where, λ is the likelihood ratio (LR), 

L (Ho)= the log likelihood value of the null-hypothesis, 

            L (H1) = log likelihood value of the alternative hypothesis, and 

            ln is the natural logarithms. 

All the tests were carried out using generalized likelihood ratio statistics. The test statistics 

is defined by χ2 = –2 [L(H0) – L(H1)], where L(H0) and L(H1) are the values of the likelihood 

function for the model under the null hypothesis, H0, and the alternative hypothesis, H1, that are 

involved. 

 

3. RESULTS AND DISCUSSION 

3.1. Summary of Variables Used in the Model 

Before embarking on estimation of the model, it is preferable to see the characteristics of 

variables considered in the model. The production function for this study was estimated using five 

input variables. To draw some picture about the distribution and level of inputs, the mean and 

range of input variables is discussed as follows: 

 

Table-1. Summary statistics of variables used to estimate the production function 

Variable description Minimum  Maximum  Mean Std. deviation 

Output (Qt) 1 24 4.83 3.65 

Land (ha) 0.125 2 0.775 0.44 
Seed (Kg) 1 40 11.4 6.36 
Labor (MDs) 12.5 122.5 44.26 22.1 
Oxen (Oxen days) 1.88 32.5 11 6.68 
DAP (Kg) 0 200 27.62 37 

       Source: Own survey (2013) 

 

On average farmers produced 4.83 quintal of sesame, which is dependent variable in the 

production function. The land allocated for sesame production, by the sample households during 

the survey period, ranges from 0.125 to 2 ha with average of 0.775 ha. The other very important 

variable is seed. The average amount of seed that sample households’ used was 11.4 Kg. Like 

other inputs human and animal labor inputs were also decisive, given a traditional farming system 



Current Research in Agricultural Sciences, 2015, 2(1): 8-21 

 

 
14 

© 2015 Conscientia Beam. All Rights Reserved. 

in the study area. Sample households, on average, use 44.26 man equivalent labor and 11 oxen 

days for the production of sesame during 2012 production season. In the study area farmers use 

only DAP for sesame production and on average farmers used 27.62 Kg of DAP.Similar to the 

production function, the mean and standard deviation of each of the variables used in the cost 

function are depicted as follows: 

 

Table-2. Summary statistics of variables used to estimate the cost function 

Variable description Minimum Maximum Mean Std. deviation 

Output (Qt) 1 24 4.83 3.65 

Total cost of production (Birr) 732.88 10290 3561.9 1930.14 
Cost of land (Birr) 46.9 1000 318.62 175.05 
Cost of seed (Birr) 12 480 136.7 76.38 
Cost of human labor (Birr) 500 4900 1770.5 884.18 
Cost of oxen labor (Birr) 150 2600 878.75 534.47 

Cost of DAP (Birr) 0 3312 457.33 612.86 

Variable Source: Own survey (2013) 

 

On average, the total cost of 3561.9 Birr was required to produce 4.83 quintal of sesame. 

Among the various factors of production, the cost of human labor accounted for the highest share 

(1770.5 Birr). Following the cost of labor, cost of oxen labor takes major share out of total cost of 

production which is 878.75 Birr. Among other inputs, cost of seed takes the smallest (136.7 Birr) 

share out of the total cost of sesame production. A total of 11 variables were hypothesized to 

affect efficiency of sesame producers, out of which five of them were dummy variables. Table 3 

portrays summary of these variables. 

 

Table-3. Summary of efficiency model variables 

Variables  Mean  Std. dev. 
Percentage of the 
mean with dummy = 1 

Percentage of the 
mean with dummy = 0 

Age  36.11 11.14 - - 
Education  2.53 2.8 - - 

Family size 5.76 2.58 - - 
Sesame prodn. 
Experience 5.36 1.87 - - 
Proximity to sesame 
farm 36.92 25.29 - - 
Number of oxen 1.97 1.39 - - 

Soil fertility - - 35.8 64.2 
Non-farm income - - 53.3 46.7 
Credit access - - 75 25 
Extension contact - - 98.3 1.7 
Training - - 72.5 27.5 

Source: Own survey (2013) 

 



Current Research in Agricultural Sciences, 2015, 2(1): 8-21 

 

 
15 

© 2015 Conscientia Beam. All Rights Reserved. 

4. ECONOMETRIC RESULT 

4.1. Hypothesis Testing 

The first important hypothesis test carried out was checking for the existence of the 

inefficiency component of the total error term of the stochastic production function. Using Cobb-

Douglas functional form and assuming a half-normal distribution of one-sided error term (ui), the 

hypothesis to be tested was whether technical inefficiency is absent in the model and hence the 

conventional (average) production function is appropriate or not.Tests of hypotheses for the 

parameters of the frontier model are conducted using the generalized likelihood ratio statistics, λ, 

defined by Equation 10. As explained in Table 4, one-sided generalized LR test of γ =0 provide a 

statistic of 9.36 for sesame production; which is significantly higher than the critical value of χ2 

for the upper 5% at one degree of freedom (3.84). Rejecting the null hypothesis implies that the 

average response function estimated by OLS, which assumes all farmers are technically efficient is 

an inadequate representation of the data, given the stochastic frontier and the inefficiency effects 

model. Consequently, the null hypothesis that sesame producers in the area are fully efficient is 

rejected.  

Table-4. Generalized LR test of hypotheses for parameters of SPF 

Null hypothesis 
Critical value (χ2 ,  
0.95) LR Decision 

H0: γ = 0 3.84 9.36 reject H0 

H0: ui = δ0 = δ1 = …= δ11 = 
0 19.68 46.42 reject H0 

              Source: Own computation (2013) 

 

The second hypothesis tested was that all coefficients of the inefficiency effect model are 

simultaneously equal to zero (i.e. H0: δ0 = δ1 = δ2 … = δ11 = 0) against the alternative hypothesis, 

which states that all parameter coefficients of the inefficiency model are different from zero. It is 

to mean that the explanatory variables in the inefficiency effect model do not contribute 

significantly to the explanation of the inefficiency variation for the sesame producing farmers. 

Using the formula in Equation (10), the LR value obtained was 46.42, which is higher than the 

critical χ2 value (19.68) at the degree of freedom equal to the number of restrictions to be zero (in 

this case the number of coefficients of the inefficiency effect model which are 11). As a result, the 

null hypothesis is rejected in favor of the alternative hypothesis that explanatory variables 

associated with inefficiency effects model are simultaneously different from zero. 

 

4.2. Estimation of Production and Cost Functions 

The ML estimates of the parameters, of the SPF specified in equation (1), were obtained 

using the STATA 11 computer program. These results together with the standard OLS estimates 

of the average production function are presented in Table 5. 

 

 

 



Current Research in Agricultural Sciences, 2015, 2(1): 8-21 

 

 
16 

© 2015 Conscientia Beam. All Rights Reserved. 

Table-5. OLS and MLE results of the production frontier for the sample households 

  OLS MLE 

Variables Coefficients Std. Err Coefficients  Std. Err 

Constant  1.8099** 0.7725 0.7514***  0.6278 
Land 0.03668 0.2074 0.1002  0.1700 
Seed 0.3122*** 0.0941 0.2134***  0.0820 
DAP -0.0053 0.0055 0.0004 0.0044 
Labor 0.9264*** 0.1548 0.8284*** 0.1218 

Oxen 0.0082 0.1611 0.0681 0.1328 
R2 0.7337       
F statistics 62.81***       

ζ2 = ζv2 + ζu2 - 4.1054***     

λ = ζu/ ζv - 0.3297***     
Log likelihood -   -41.223   

          Note: ** and *** refers to 5% and 1% significance level, respectively.   

           Source: Own computation (2013) 

 

From the total of five variables considered in the production function, two (labor and seed) 

had a significant effect in explaining the variation in sesame yield among farmers. The coefficients 

of the production function are interpreted as elasticity. Hence, high elasticity of output to labor 

(0.8284) suggests that sesame production was highly sensitive to labor. As a result, 1% increase in 

amount of labor will result in 0.8284% increase in sesame production, keeping other factors 

constant. Alternatively, this indicates sesame production was responsive to labor and seed. 

The diagnostic statistics of inefficiency component reveals that sigma squared (ζ2) was 

statistically significant at 1 percent (Table 5). This indicates goodness of fit, and the correctness 

of the distributional form assumed for the composite error term. 

The returns to scale analysis can serve as a measure of total factor productivity [17]. The 

scale coefficient was calculated to be 1.21, indicating increasing returns to scale (Table 6). This 

implies that there is potential for sesame producers to continue to expand their production 

because they are in the stage I of the production surface, where resource use and production is 

believed to be inefficient. In other words, a percent increase in all inputs proportionally will 

increase the total production by 1.21%. This result is consistent with Ajibefun [17], Fekadu [5] 

and Amos [18].  

 

Table-6. Elasticity and returns to scale of the parameters in the production function 

Variables Elasticity’s 

Land 0.1002 
Seed 0.2134 
DAP 0.0004 
Labor 0.8284 
Oxen 0.0681 
Return to scale 1.2105 

                                           Source: Own computation (2013) 

 



Current Research in Agricultural Sciences, 2015, 2(1): 8-21 

 

 
17 

© 2015 Conscientia Beam. All Rights Reserved. 

The dual cost function which is specified in equation (2) and derived analytically from the 

stochastic production function is given as follows: 

lnCsi=2.27 + 0.083lnW1+ 0.176lnW2 + 0.0004lnW3+ 0.056lnW4 + 0.684lnW5 + 0.826lnYi* 

Where Cs is cost of producing sesame; W1 refers to the price of land, W2 is cost of seed; W3 

is cost of DAP; W4 is cost of oxen; W5 is cost of labor; Y is output; i refers to the ith sample 

household. 

 

4.3. Efficiency Scores 

The results of the efficiency scores indicate that there were wide ranges of differences in TE, 

AE and EE among sesame producer farmers. The mean TE of sample households during the 

survey period was 67.11%. The TE among the households ranges from 23.45 to 95.81%, with 

standard deviation of 0.1805. Similarly, the mean AE and EE of sample households were 67.25 

and 45.14%, respectively. Unlike TE and EE there was high average AE score. Generally there is 

a considerable amount of efficiency variation among sesame producer farmers in all measures of 

efficiency. This result is consistent with study of Jema [9]. 

 

Table-7. Descriptive statistics of efficiency measures 

Efficiency parameter Minimum Maximum Mean Std.dev 

TE 0.2345 0.9581 0.6711 0.1805 

AE 0.4383 0.8454 0.6725 0.0977 
EE 0.1332 0.7305 0.4514 0.1362 

                       Source: Own computation (2013) 

 

The distribution of the TE scores showed that the majority (more than 61%) of the sample 

households had TE score of greater than 60% (Table 8). But there were also some households 

whose TE levels were limited to the range 20 to 40%. Households in this group have a room to 

enhance their sesame production at least by 60%, on average. Out of the total sample households, 

only 7.5% had TE greater than 90%. This implies that about 92.5% of the households can increase 

their production at least by 10%. Moreover, 50% of the sample households can increase their 

production at least by 32.9%.  

 

Table-8. Distribution of technical, allocative and economic efficiency scores 

 
Efficiency range 

                                 Number of households 

TE AE EE 

0 - 9.99 0 0 0 
10 – 19.99 0 0 3 
20 – 29.99 4 0 16 
30 – 39.99 6 0 21 
40 - 49.99 10 4 35 
50 – 59.99 26 31 24 
60 – 69.99 20 30 17 
70 – 79.99 17 40 4 
80 – 89.99 28 15 0 
90 – 99.99 9 0 0 

Source: Own computation (2013) 



Current Research in Agricultural Sciences, 2015, 2(1): 8-21 

 

 
18 

© 2015 Conscientia Beam. All Rights Reserved. 

According to Table 8, the AE distribution scores indicate that the largest efficiency group of 

sesame producers (58.33%) operated between 60% and 79.99%. Households in this group can save 

at least 20% of their current cost of inputs by behaving in a cost minimizing way. Only 12.5% of 

the total sample households had an AE score that ranged between 80 and 89.99%. This shows 

that all sesame producing farmers (100%) can at least save 10% of their current input cost by 

reallocation of resources in cost minimizing way. 

The distribution of EE scores (Table 8) implies that the majority of the farmers were 

performing under the average efficiency level. The low average level of EE was the total effect of 

both technical and allocative inefficiencies. This also indicates the existence of substantial 

economic inefficiency in the production of sesame during the study period. 

 

4.4. Determinants of Efficiency Differentials Among Farmers 

The results obtained from the first stage estimations indicated that the average efficiency 

scores were low and there existed efficiency variations among farmers. The technical, allocative 

and economic efficiency estimates derived from the model were regressed on socioeconomic and 

institutional variables that explain variations in efficiency across farm households using Tobit 

regression model (Table 9).  

 

Table-9. Tobit model estimates for different efficiency measures 

  TE   AE   EE   

Variable 
Marginal 
Effect Std.Err Marginal Effect Std.Err. 

Marginal 
Effect Std.Err. 

AGEHH 0.0012 0.0014 0.0004 0.0009 0.0009 0.0011 
EDULVL 0.002 0.005 0.004 0.0031 0.0037 0.0039 
FAMSIZ -0.005 0.0065 -0.0049 0.004 -0.0059 0.005 
SESPROEXP 0.0071 0.0074 -0.0099** 0.0046 -0.0014 0.0057 
PROX 0.0006 0.0006 -0.0008** 0.0003 -0.0002 0.0004 
SOILFERT 0.1767*** 0.0281 0.0517*** 0.0173 0.1560*** 0.0214 
NOFOXEN 0.0116 0.0116 -0.0019 0.0071 0.0036 0.0088 
NONFRMINC 0.0997*** 0.028 -0.0316* 0.0172 0.0431** 0.0213 
EXTN -0.0648 0.1098 -0.2239*** 0.0677 -0.1951** 0.0836 

CRDT 0.0859*** 0.0306 0.0106 0.0188 0.0637*** 0.0233 
TRNG -0.0304 0.1303 0.0219 0.0186 -0.0089 0.023 
LOG L 65.55 123.7 98.36 

       Note: *, ** and *** refers to 10%, 5% and 1% significance level, respectively. 

        Source: Own computation (2013) 

 

As can be seen in Table 9, perceived soil fertility had a significant and positive impact on 

technical, allocative and economic efficiencies, as expected. This implies that fertility of land is an 

important factor in influencing the level of efficiency in the production of sesame. In other words, 

farmers with fertile farm were more efficient than farmers with less fertile farm. The result is 

consistent with that of Fekadu [5]. The positive and significant coefficient of the non-farm 

income in technical and economic efficiencies suggests that the income obtained from such non-

farm activities could be used for the purchase of agricultural inputs and augment financing 

household expenditures which would otherwise put pressure on on-farm income. This could be 



Current Research in Agricultural Sciences, 2015, 2(1): 8-21 

 

 
19 

© 2015 Conscientia Beam. All Rights Reserved. 

due to the fact that most of the non-farm activities (butchery, grinding mills, handicraft, and 

selling of local drinks) performed by the sample households do not compete with time allocated 

for farm activities. The result is consistent with Jema [9] but inconsistent with Dolisca and 

Curtis [19]. Unexpectedly, extension contact was found to have a negative and significant 

relationship with allocative and economic efficiency of farmers. This might be due to the fact that 

the involvement of extension workers in many non-extension activities such as credit applications 

processing, input distributions, and collection of loans. Jema [9] also found a negative 

relationship between extension visit and technical efficiency. 

Credit has been found to be an important variable in explaining the variation of technical and 

economic efficiency among farmers. The positive and significant impact of credit in this study 

implies that credit availability enables farmers to make timely purchases of inputs that they 

cannot provide otherwise from their own resources. In other words, farmers who had access to 

credit were more technically and economically efficient than farmers who had no access to credit. 

This result is consistent with Kinde [20] and Dolisca and Curtis [19]. As hypothesized distance 

of sesame farm from farmers’ residence was found to be negatively and significantly related to 

allocative efficiency. This implies that as the distance of the farm from home increases the 

allocative efficiency decreases. Increased farming experience may lead to better assessment of 

importance and complexities of good farming decision, including efficient use of inputs. 

Unexpectedly, experience in sesame production was found to have a negative and significant 

relationship with allocative efficiency. Wilson, et al. [21] also found a negative relationship 

between experience and efficiency in potato production in UK, implying that farmers with fewer 

years of experience achieved higher levels of efficiency. Rahman [22] also reported similar results 

for Bangladesh rice farmers. The reason may be that those with little experience are likely to seek 

out for new technology, unlike those with experience or are better at managing their resources. 

 

5. CONCLUSION 

An important conclusion stemming from the analysis of the efficiency of sesame production is 

that, there exists a considerable room to enhance the level of technical, allocative and economic 

efficiency of sesame producer farmers. Result of the production function indicated that labor and 

seed were limiting constraints, with positive sign as expected. The positive coefficients of these 

variables indicate that, increased use of these inputs will increase the production level to greater 

extent. The average TE, AE and EE values of the sample households were 67.11, 67.25 and 

45.14%, respectively. This implies that farmers can increase their sesame production on average 

by 32.9% if they were technically efficient. Similarly, sesame producers can reduce current cost of 

inputs, on average, by 32.75% if they were allocatively efficient. The result also indicated that if 

these farmers operate at full efficiency levels, they on average could reduce their costs of 

production by 54.85% and still produce the same level of output. 

The factors that affect the level of efficiency were identified, to help different stakeholders to 

enhance the current level of efficiency in sesame production. Accordingly, soil fertility, non-farm 

income and credit access positively and significantly affected TE. This implies that farmers with 



Current Research in Agricultural Sciences, 2015, 2(1): 8-21 

 

 
20 

© 2015 Conscientia Beam. All Rights Reserved. 

fertile land, better access to credit and more non-farm income were technically efficient than 

their counterparts. Soil fertility had positive and significant impact on AE. This again implies 

that farmers with fertile land were allocatively efficient than others. But exper ience  in  

sesame production ,  d is tance  of  sesame farm,  non - farm income and extens ion 

contact  a ffected AE negatively and significantly. From this, we can conclude that households 

who had a distant farm, more years of sesame production experience, better non-farm income and 

extension contact were allocatively less efficient than others. Soi l  fert i l i ty ,  non-farm 

income and credit  access  had positive and significant impact on EE. Here ,  we can 

conc lude  that farmers  with  fer t i le  land ,  more non -farm income and better  

access  to  cred i t  were  economica l ly  e f f i c ient  than the  rest .  However, extension 

contact affected EE negatively and significantly in contrary to the expectation. 

 

6. RECOMMENDATIONS 

1. Even though non-farm income was related negatively to allocative efficiency, it had a 

positive impact on technical and overall efficiency. This indicates that there is a need to 

introduce activities that could enhance the non-farm/off farm income of households 

without affecting their farm time allocation so that the farmers would be in a position to 

invest the required amount of resources in sesame production. 

2. Access to credit affected both technical and economic efficiency of sesame producers 

positively. Hence policy makers should devote a great effort to create more access to 

credit services for farmers in the study area. 

3. Fertility of sesame farm affected technical, allocative and economic efficiency of farmers 

positively. Therefore, development programs should give due emphasis to improve and 

maintain the fertility of land through awareness creation and introduction of 

technologies that improve and maintains fertility so that the efficiency of the farmers 

increases.  

 

REFERENCES 

[1] MoARD (Ministry of Agriculture and Rural Development), "Ethiopia’s agricultural sector policy and 

investment framework 2010-2020," Draft Final Report, 15 September 20102010. 

[2] A. Kindie, "Sesame market chain analysis: The case of Metema Woreda, North Gondar zone, Amhara National 

Regional State," Thesis Presented to School of Graduate Studies of Haramaya University, 2007. 

[3] CSA (Central Statistical Authority), "Statistical report on area and crop production. Addis Ababa, Ethiopia," 

2011. 

[4] D. Sorsa, Sesame trade arrangements, costs and risks in Ethiopia: A baseline survey: Wageningen, 2009. 

[5] G. Fekadu, "Analysis of technical efficiency of wheat production: A study in machkel district," M.Sc. Thesis, 

Presented to School of Graduate Studies of Haramaya University, 2004. 

[6] T. I. Yamane, Statistics: An introductory analysis, 2nd ed. New York: Harper and Row, 1967. 

[7] D. J. Aigner, K. C. A. Lovell, and P. S. Schmidt, "Formulation and estimation of stochastic models," Journal of 

Econometrics, vol. 6, pp. 21-37, 1977. 



Current Research in Agricultural Sciences, 2015, 2(1): 8-21 

 

 
21 

© 2015 Conscientia Beam. All Rights Reserved. 

[8] W. Meeusen and J. Van Den Broeck, "Efficiency estimation from cobb-douglas production functions with 

composed error," International Economic Review, vol. 18, pp. 435-444, 1977. 

[9] H. Jema, "Economic efficiency and marketing performance of vegetable production in Eastern and central parts 

of Ethiopia," Doctorial Thesis, Acta Universitatis Agriculture Sueciae (SLU) Uppsala, Sweden, 2008. 

[10] R. O. Kareem, A. O. Dipeolu, A. B. Aromolaran, and A. Samson, "Analysis of technical, allocative and economic 

efficiency of different pond systems in Ogun State, Nigeria," African Journal of Agricultural Research, vol. 3, pp. 

246-25, 2008. 

[11] A. L. Kehinde and T. T. Awoyemi, "Analysis of economic efficiency in sawn wood production in South West 

Nigeria," Journal of Human Ecology, vol. 26, pp. 175-183, 2009. 

[12] B. Solomon, "Economic efficiency of wheat seed production: The case of smallholders in womberma Woreda of 

West Gojjam zone," M.Sc. Thesis, Presented to School of Graduate Studies of Haramaya University, 2012. 

[13] K. R. Sharma, P. Leung, and H. M. Zaleski, "Technical, allocative and economic efficiencies in swine production 

in Hawaii: A comparison of parametric and non-parametric approaches," Agricultural Economics, vol. 20, pp. 23-

35, 1999. 

[14] A. Arega and M. H. Rashid, "The efficiency of traditional and hybrid maize production in Eastern Ethiopia: An 

extended efficiency decomposition approach," Journal of African Economics, vol. 15, pp. 91-116, 2005. 

[15] M. J. Farrell, "The measurement of productive efficiency," Journal of Royal Statistical Society. Series A, vol. 120, 

pp. 253-290, 1957. 

[16] D. Gujarati, Basic econometrics. Tokyo: McGraw-Hill Companies, 2004. 

[17] I. A. Ajibefun, "An evaluation of parametric and non-parametric methods of technical efficiency measurement: 

Application to small scale food crop production in Nigeria," Journal of Agricultural and Social of Sciences, vol. 4, 

pp. 95-100, 2002. 

[18] T. T. Amos, "An analysis of productivity and technical efficiency of smallholder cocoa farmers in Nigeria," 

Journal of Social Science, vol. 15, pp. 127-133, 2007. 

[19] F. Dolisca and M. J. Curtis, "Technical efficiency of traditional and non-traditional crop production: A case 

study from Haiti," World Journal of Agricultural Sciences, vol. 4, pp. 416-426, 2008. 

[20] T. Kinde, "Technical efficiency of maize production: A case of small-holder farmers in Assosa district," Thesis 

Presented to School of Graduate Studies of Haramaya University, 2005. 

[21] P. Wilson, D. Hadley, S. Ramsden, and I. Kaltsas, "Measuring and explaining technical efficiency in UK potato 

production," Journal of Agricultural Economics, vol. 49, pp. 294-305, 1998. 

[22] S. Rahman, "Profit efficiency among Bangladesh rice farmers." University of Manchester, School of Economic 

Studies. Discussion Paper Series No. 0203, 2002. 

 

 

 

 

 

Views and opinions expressed in this article are the views and opinions of the author(s), Current Research in Agricultural Sciences 
shall not be responsible or answerable for any loss, damage or liability etc. caused in relation to/arising out of the use of the content. 

 


