ARTICLE # 13884-RJBS
Stochastic Frontier Model for Climatic Determinants on Rice Productive Efficiency
Ismail Hossain1Md. Azizul Baten1,2
1Department of Statistics, School of Physical Sciences, Shahjalal University of Science and Technology Sylhet-3114, Bangladesh. Email: . Phone Number: +8801613938989, +8801713938989
2Department of Decision Science, School of Quantitative Sciences, Universiti Utara Malaysia, 01610 UUM Sintok, Kedah, Malaysia.Email, . Phone Number: +6049286886
Abstract
Identifying climatic determinants of efficiency is a major task in efficiency analysis. An attempt has been made to investigate the climatic effects on efficiency levels of rice production in Bangladesh. Three types of rice crops (Boro, Aus and Aman) in Bangladesh are considered and the model proposed by Battese and Coelli (1995) are revisited in this study to see the effects on rice productive efficiency in Bangladesh. We used the collected data from agricultural sector consisting three main rice crops in Bangladesh for the period of 1980 to 2008.The results showed that temperature has a significant effect on rice production. The negative coefficients of rainfall, fertilizer are indicated that the technical efficiency level tended to decrease by 37 and 16 percent respectively over the time period. It was observed that humidity and wage rate of labor were found decreasing the level of inefficiency. The coefficient of wage rate of labor was found negative but significant which indicated that the labors involved in rice production are not skilled enough. The most efficient rice production system in Bangladesh is Boro with the technical efficiency (0.86) than other two crops.
Key Words: Stochastic frontier model, Climatic change, Technical inefficiency; Bangladesh rice production; Panel data.
Corresponding author name:
Visiting Associate Professor Dr. Md. AzizulBaten,
Department of Decision Science, School of Quantitative Sciences, Universiti Utara Malaysia, 01610 UUM, Sintok, Kedah, Malaysia
Email:
1 Introduction
Climate is one of the major controlling factors for well-being of the residents of the world. But Climate variability and change, its impacts and vulnerabilities are growing concern worldwide.Climate change and agriculture are interrelated processes, both of which take place on a global scale (Fraser 2008). The climate of Bangladesh is changing and it is becoming more unpredictable every year. Global warming induced changes in temperature and rainfall are already evident in many parts of the world, as well as in Bangladesh (Ahmedand M.Alam, 1999). The overall effect of changing climatic condition on agriculture will depend on the balance of these effects. Assessment of the effects of global climate condition change on agriculture might help to properly anticipate and adapt farming to maximize agricultural production. Rice is one of the most important foodsin the world, being the staple food of over 50% of the world population, particularly in India, China, and a number of other countries in Africa and Asia(Olorunfemi and Victor 2006). Riceis a major source of livelihood of rural population in most Asian countries as like as in Bangladesh. There are about 4 billion people who consume over 90 percent of the world’s rice production. Rice was selected in this study because of its prominent position in the national economy of Bangladesh. The share of agriculture to GDP in Bangladesh is about 18.64 percent (BER, 2008-09). Agriculture accounts for 21% of the gross domestic product (GDP) (different information about GDP) and 50% of overall employment (Bangladesh Agricultural Census, 2008).Therefore we have to be given the importance of changing climatic conditions on rice production because of adaptation problem of producing rice. It is observed that there have been a few studies on the efficiency of rice production measurement on the basis of climatic condition in Bangladesh. Have farmers grown their product efficiently with the available technologies? How have the policies undertaken to be aware of farmers regarding the effect of climatic condition on rice production in Bangladesh. These are some of the questions the study partly sought to answer.
Efficiency measures are important because of its vital role on productivity growth. The efficiency of rice production has been of longstanding interest to the economists and policymakers in Asia because of the strong relationship between rice production and food security in the region (Richard et al. 2007). A number of studies examined productivity efficiency of crops in Asia and other country (Kalirajan1981; Erwidodo 1990;Kalirajan and Shand1989; Ekanayake, S.A.B. (1987); Hanley, N. and Spash, C. L. 1993 ; Mythili and Shanmugan2000; ‘Shanmugam2000;Squires and Tabor1991;Ahad, Rafiq and Ali 1999;Ajibefun, Battese and Daramola2002; Ali and Flinn 1989;Battese 1992; Battese and Coelli 1992;Battese and Coelli 1995;Travers and Ma 1994; Fan et al. 1994; Wang et al. 1996a, 1996b; Xu and Jeffrey 1998; Fan 1999; and Tian and Wan 2000; Wadud and White 2000). The impacts of using advanced techniques on the rice production efficiency in developing countries have been conducted by Bordey 2004; Chengappa et al. 2003; and Khuda 2005.In this context,Stochastic Frontier Approach has found wide acceptance within the agricultural economics (Battese and Coelli 1992; Coelli and Battese 1995). There exist few literatures in estimating stochastic frontier production and consequently dealing with technical inefficiency in rice production ofBangladesh (Rahman et al. 1999; Deb and Hossain 1995; Banik 1994; Rahman 2002). Khan et al. (2010) examined the influenced of farmers’ age, education and experience ontechnical efficiency of boro and aman production using Cobb-Douglas production function and found that education was positively related to technical efficiency. However, none of these studies focused on the potential influenceof climatic conditions to enhancing technical efficiency of rice production. Given this back drop, the present study assesses the effects of climatic conditions (Rainfall, Humidity and Temperature) to Rice production in Bangladesh on technical efficiency by applying stochastic frontier model (Battese and Coelli1995). This studyalso attempts to investigate the effect of other agricultural inputs such as area, seed and variety of fertilizer in rice production in Bangladesh. An improved understanding of theserelationships can help the farmers to allocate scarce resources more efficiently and may assist policymakers to design and formulate agricultural policy to increase agricultural production in Bangladesh.
2 Materials and Methods
2.1Data Sources and Variables Construction
2.1.1Data Set
The data on rice production in Bangladesh is obtained from the year book agricultural statistics of Bangladesh, conducting by the Bangladesh Bureau of Statistics (BBS) every year. The dependent variable, rice production and other independent variables such as area, the amount of seed, the amount of fertilizer, rainfall, wage rate per labor without food and wage rate of bullock pair for each crop are collected from yearly book of agriculture statistics of Bangladesh. For this study we consider 29 time periods from 1980-1981 to 2008-2009.The meteorological data such as rainfall, temperature and humidity are collected from the meteorological department in Bangladesh. The yearly distributions of fertilizer data for each rice crop (Boro, Aus and Aman) are collected from Bangladesh Agriculture Development Corporation.
2.1.2 Descriptions of the Variables
Dependent Variable
Production (Y): Total Boro (Local, HYV and Hybrid Boro), Aus (Local and HYV Aus) and Aman(broadcast, local transplant and HYV Aman) have been estimated at thousand metric tons.
Independent Variables
Area: The total areas of Boro, Aus and Aman rice have been estimated at hectares. Here we considered the total area of cultivated land where specifically Boro are cultivated crop. In this study we considered all the varieties of each crop.
Seed: Seed is the very important input to increase rice production. Therefore it is required that the farmers must use pure, healthy seeds as per the minimum certification standards which have standard percentage. In fact the seeds are foundation of farming; the high good quality seeds are those which have genetic purity, physical purity, health standards and moisture percentage in accordance with the minimum seed certification standards. For this study the amount of seed is considered of each crop and it is measured in thousand metric tons.
Fertilizer in Urea: Fertilizer in Urea is a kingpin in enhancing crop production. No country has been able to increase agricultural productivity without expanding the use of chemical industry. The total amount of fertilizer in urea is used in each crop separately and the unit of fertilizer in urea is in metric tons.
Fertilizer in TSP: Triple supper phosphate (TSP) is the major fertilizer which is applied in agricultural land in various proportions for rice production system in Bangladesh. For this study, the unit offertilizer in TSP is in metric tons.
Rainfall: The primary source for agriculture production for most of the world is rainfall. For this study the total rainfall is considered for Boro rice in millimeter during February-July session. Similarly the total rainfall is considered for Aus and Aman rice during March-June and July-December respectively.
Explanatory Variable
Temperature: Rice crop is influenced by seasonal characteristics and different variables of climate such as temperature. Extreme temperature due to climate change usually affects livestock. High temperature affects livestock in a number of ways: causes great discomfort as in the case of human, decreases feed intake and alters nutrient metabolism leading to high loss of energy and the combined effects of discomfort and nutrient metabolism reduces their productivity, resulting in financial loss of the farmers. Here we took the average monthly temperature (in Celsius) in each rice crop session.
Humidity: Humidity is a measure of the amount of water vapor in a body of air at a certain temperature. Humidity is an important measure in agriculture because it has an effect on evaporation rates. Where there is high humidity, animals may get stressed, and moulds, mildews and fungus may grow which is indirectly effect in agriculture especially in rice production. For this study we considered average humidity in percentage in each crop.
Daily average wage rate of labour without food: The wage rate is important factor which has an indirect impact in the production of rice. In this study, the average daily wage rate per man without food is considered and it is measured in taka.
Wage rate of Bullock pair in a day: Most of the cultivation areas in Bangladesh are in rural where the improved agriculture technology is not available. The farmer is depended on the traditional method to prepare their land for cultivation. Wage rate of bullock pair is considered in a day.
2.2.1Analytical Framework
In this framework, the output (rice production) is treated as a stochastic production process and is defined as the specification of Battese and Coelli (1995) model specification may be expressed as:
Where represent the production at the t-th observation (t = 1,2,..., T) for the i-th firm (i = 1,2,..., N); a (1 X k) vector of values of known functions of inputs of production in rice and other explanatory variables associated with the i-th firm at the t-th observation , is a vector of unknown parameters for the stochastic frontier. assumed to be iid N (0, ) random errors, independently distributed of the , represents non-negative random variables, associated with technical inefficiency of production, which are assumed to be independently distributed, such that is obtained by truncation (at zero) of the normal distribution with mean, and varianc ;is a (1 X m) vector of explanatory variables associated with technical inefficiency of production of overtime and
represents (m x 1) vector of unknown coefficient to be estimated. Since Rice producing always operates under uncertainty, again the present study employs a stochastic production frontier approach introduced. In this framework we add the following assumption that ’s are non-negative random variables which are assumed to account for technical inefficiency in production and to be independently distributed as truncations at zero of the distribution; where where is a vector of explanatory variables which may influence the inefficiency of rice producing in Bangladesh and is avector of parameters to be estimated ……………………(2)
Where, the random variable follows truncated normal distribution with mean zero and variance, such that the point of truncation is .Parameters of the stochastic frontier given by equation (1) and inefficiency model given by equation (2) are simultaneously estimated by using maximum likelihood estimation. After obtaining the estimates of the technical efficiency of the t-th observation (t = 1, 2... T) inthe i-th rice producing or farmers is given by:
2.2.2Empirical Stochastic Frontier Model
There are several functional forms for estimating the relationship between inputs and output. Since the Cobb-Douglas functional form is preferable to other forms if there are three or more independent variables in the model (Hanley and Spash 1993), the Cobb-Douglas production function with five independent variables was applied in this study. Following Battese and Coelli (1995) the model (1) can be expressed into the Cobb-Douglas stochastic frontier production functional form with logarithm. The empirical Stochastic Cobb-Douglas frontier production model for technical inefficiency effect on Boro rice production can be expressed equation (1) as follows
+++
i=1, 2, 3; t=1, 2, 3……..29 (4)
Where
= production in the i-thrice (Boro, Aus and Aman) with t-th period.
= Area in the i-th rice with t-th period
= The quantityof seed of the -th rice in the i-th rice with t-th period
=The amount of Urea is used in the i-th rice with t-th period
=the amount of TSP is used in the i-th rice with t-th period
=the Amount of rainfall in the i-th rice with t-th period
=unknown parameter to be estimated
Refers to the natural logarithm
i=the number of rice (Boro, Aus and Aman)
t = time period
2.2.3 Technical Inefficiency Effect Model
The technical inefficiency effects model (2), are defined as
Where’s are non-negative random variables, assumed to be independently distributed such that the technical inefficiency effect for the ith farmer, , were obtained by truncation of normal distribution with mean zero and variance, such that
= the average temperature in the i-th rice with t-th period
= the average humidity in the i-th rice with t-th period
= in the i-th rice with t-th period the average wage rate of human labour without food.
= The wage rate of bullock pair in daily in the i-th rice with t-th period,
are parameters to be estimated
were unobservable random variables or classical disturbance term, which are assumed to be independently distributed, obtained by truncation of the normal distribution with mean zero and unknown variance, , such that is non- negative. The β, and δ coefficients are unknown parameters to be estimated, together with the variance parameters which are expressed in terms of , , γ is the ratio of variance of farm specific technical efficiency to the total variance of output and has a value between zero and one. The estimates for all parameters of the inefficiency model (5) were estimated by using the Maximum Likelihood (ML) method. The econometric computer software package FRONTIER 4.1 (Coelli, 1996) was applied to estimate the parameters of stochastic frontier models using the ML method.
3 Results and Discussions
3.1.1Maximum-Likelihood Estimates ofStochastic Cobb-Douglas Frontier Production Model
The estimated parameters of technical efficiency (TE) model were reported in the Table 1. In the TE model, a positive coefficient indicatesthe improvement of technical efficiency and vice-versa. In the result, seed and area are found significant with positive coefficient. That is these variables have direct influence to increase the level of efficiency. Again it was observed that coefficient of seed, fertilizer (urea) and area are significant these indicated that these inputs are directly related with the TE. The negative coefficient of rainfall, fertilizer (urea) and fertilizer (TSP) are indicated that over rainfall and more fertilizer will be cause of decreasingTE.
The value of γ (0.8430)is positive and significant at 1% level of significance. It can be interpreted as follows: 84 percent of random variation around in rice production due to inefficiency and 1 percent due to stochastic random error.the estimated value of σ is (0.0019) was significantly different from Zero, indicate a good fit.
In inefficiency effects model, a positive coefficient indicates the decrease of technical efficiency and vice-versa. The coefficient of temperatureis positive and significant at 1 percent level of significance which is indicated that is temperature has direct influence on rice production. Hence from the result, it was observed that humidity and wage rate of labor were found decreasing the level of inefficiency. Hence the variable humidity does not play any contributors role in producing rice production. The coefficient of wage rate of labor was found negative but significant which indicated that farmers are unable to pay high rate of wage.
3.1.2 Year wise Efficiencies ofRice inBangladesh
The year wise efficiency of three rice crop (Boro, Aus, Aman) in Bangladesh was displayed in Table 2: and Figure 1. From this investigation we observed that the highest efficiency was in 2008 for Boro rice and the lowest efficiency was in 1980 for each crop. The mean efficiency of three rice crops was found 0.86, .0.79 and 0.82 for Boro, Aus and Aman respectively. Hence the mean efficiency of Boro rice is better than others. Time had an important effect in increasing efficiency.It was also revealed that the technical efficiency of the Aman rice during the period 1980to 2008 was found to be 0.82. This implied that 82 percent of potential output was being realized by the rice production that the technical efficiency had increased over time. From the figure, the overall situation of three types of rice crops is to be clearly understood. Borois most efficient (86 percent) followed by the Aman rice was second most efficient (82 percent). These findings are in line with the argument that included in Boro rice is superior as it is regular. Amanand Aus rice were relatively less efficient than Boro rice. So it can be concluded that technical efficiency of Boro rice was improved more than Aus and Amanover time. The result can interpret as government may give more attention to support in rice production care in rice production providing technical support in the context of agricultural sector. From the figure it was clear that but sometimes the Te are ups and down.Last five year from 2004 to 2008, the technical efficiency was found higher and maintained almost same.
Conclusion
The results of this study show that the rice related productive factor included in the inefficiency effects model have had significant influences upon the technical inefficiency. Theimpact of climate condition on rice production will depend on actual patterns of climatic factors change in rice growing regions. From the study we conclude that the agricultural input factor such as area, seed and fertilizer are responsible factor in rice production system. The inefficiency effects model we observed that the climatic conditions are playing an importantrole in our rice production i.e. the effects of maximum temperature would drastically reduce total rice production and humidity indicates the decreasing the level of inefficiency. So, the effects of changing climatic conditions on rice production give the higher technical inefficiency rate and we lost a significant amount of rice production may hamper for only effects of those climatic conditions.In this regards proper management practices (should be proper distribution and application of seed and fertilizer on the basis of area) would also help to meet the food demandand achieve our desire rice production target under changing climatic condition in future. Public awareness of the impact of climate conditions on the agricultural production systems deserves priority consideration, and mitigating technologies must be developed, which will require increased public and private investment. .