OF MIXED RICE-FISH FARMING IN SOUTH-WEST REGION OF BANGLADESH

Mixed farming, Production efficiency, Scale to sale The present research aims at investigating the economic performance of mixed ricefish farming in south-west region of Bangladesh. In order to carry out the research objective descriptive statistics, profit function, Cobb-Douglas form of multiple linear regression model and t-test approaches have been applied. The study area has been selected using multi-stage sampling technique and convenient sampling method has been utilized to select the sample. In-depth interview technique has been employed to collect primary data by using pretested questionnaire from the samples. Results from descriptive statistics show that the average annual return on mixed rice-fish farming is BDT 56326.45 more than mono rice farming as well as production efficiency of mixed rice-fish farming is also found higher than mono rice farming. Besides, mixed rice-fish farming experiences increasing return to scale, whereas, mono rice farming undergoes decreasing return to scale. Moreover, test of hypothesis provides statistical evidence that mixed rice-fish farming is more profitable than mono rice farming in the study area.


INTRODUCTION
Bangladesh is a country of agriculture situated in south-east Asia.It is one of the highly populated least developed countries in the world with an area of 144000 square kilometers (Ahmed et al., 2011).Agriculture sector is the driving forces of economic growth of Bangladesh contributing 19.41 percent to the gross domestic product (GDP) in fiscal year 2011-2012 and providing a space of occupation for 47.5 percent of the total labor force in the same year (BBS, 2013).Rice and fish are two important products of agriculture becoming the part and parcel of the people of the country.As the main crop of Bangladesh, rice is grown around 29 million tons per year as annual production, whereas, the annual production of fish is around 2.70 million tons (Ahmed and Garnett, 2011).With a population of 16.4 crore, the demand of both rice and fish is increasing continuously in the country every year (Ahmed et al., 2011).The average size of cultivable land holding is decreasing in Bangladesh forcing the farmers to bear high risk and cost to maximize crop production within a shortest possible time due to high pressure from increasing population.In this backdrop, practice of mixed rice-fish farming can be a way to tackle the growing demand for food by utilizing farm resources with least time and cost of production.The exercise of rice-fish cultivation can also be source of decent supply of carbohydrate and animal protein (Ahmed and Garnett, 2011;Mamun et al., 2012;Roy et al., 2013).
The total amount of rice cultivated area in Bangladesh is near about 10 million hectare which covers about 75 percent of the total cropped areas (Ahmed and Garnett, 2011;Roy et al., 2013).The country's total production of rice is 344.30lakh metric tons and in case of fish, it remains 32.22 lakh metric tons in fiscal year 2011-2012.The available capacities for producing rice and fish are not being utilized fully although there exists potential to use them.Each year, a lot of food stuffs are imported to meet up the excess demand.In fiscal year 2011-2012, around 22.36 lakh metric tons food stuffs have been imported by both government and non-government sector (BBS, 2013).In this situation, producing rice-fish simultaneously in the same field paves the way for optimum resource utilization for maximum food production.In terms of cost and benefit, mixed rice-fish farming is also being preferred by farmers than mono rice farming in the recent times.One of the reasons may be that integration of fish and rice leads to lower production costs, where sails, insects, pests, and other harmful flies are caught and eaten by the fish.

Justification of the research
Extensive literature on mixed farming is still very few.In addition, studies on economic analysis on mixed rice-fish farming over mono rice production practice is also scarce.A little number of studies have been found on mixed farming, specifically on rice-fish mixed farming in Bangladesh perspective.Among them, a study conducted by Hasanuzzaman and the others investigate practice and economics of freshwater prawn farming in seasonally saline rice field at Shyamnagore Upazilla in Satkhira District of Bangladesh indicating that the culture of prawn in seasonally saline paddy field is economically viable (Hasanuzzaman et al., 2011).Another study explores the fish-paddy crop rotation practice at the farmer's pond at Sadar Upazilla of Bagerhat District, in south-west coastal region of Bangladesh describing that fish-paddy crop rotation system enhances the fertility of the ponds (Roy et al., 2013).However, another study conducted in Mymensingh District of northcentral Bangladesh provides evidence that integrated rice-fish farming can help Bangladesh keep pace with the current demand for food through rice and fish production (Ahmed and Garnett, 2011).On the other hand, Integrated Farming System (IFS) can eradicate high degree of risk and uncertainty because of seasonal, irregular and uncertain income and employment to the farmers by not only solving most of the existing economic and even ecological problems, but also provide other household needs like fuel, fertilizer and feed, along with increasing productivity of the farm manifold (Mamun et al., 2012).The objective of the research is to assess the economic performance of mixed rice-fish farming over mono rice farming at Dighalia Upazila in Khulna District of Bangladesh.Besides, the study also intends to estimate the production function of the two group of farmers to measure the production efficiency and returns to scale.

Study area
Present study has been conducted in the southeast region of Bangladesh.Multistage sampling technique has used to select the study area.In the first stage, Khulna District has been chosen for purpose of the study from total of 64 districts in Bangladesh; because, this district is situated in the south-west region of Bangladesh.In the second stage, Dighalia Upazila, shown in Figure 1, has been selected from Khulna Districts because most of the farmers in Dighalia Upazila are engaged in rice cultivation and fish cultivation.At the third step, Gazirhat Union has been selected purposively as the farmers in this area are engaged in both rice-fish farming and mono rice production under double cropping system all the year round.At the last stage, a total of 8 villages of the union have been selected randomly.

Sampling and data collection
For study purpose, farmers who produce only rice and farmers who produce rice-fish have been considered as population.Existing literature survey and a pilot field survey confirm that double cropping system is in operational for rice cultivation in the study area.Generally, cropping season for Boro rice ranges from November to April and duration of Aman rice is June to September (Ahmed and Garnett, 2011; Ahmed et al., 2011;Hasanuzzaman et al., 2011).On the other hand, two widely practiced rice-fish cultivation systems are concurrent culture (integrated)growing the fish together with the rice in the same areaand rotational culture (alternate)where the rice and fish are grown at different times (Ahmed et al., 2011;Hasanuzzaman et al., 2011;Roy et al., 2013).Time frame for integrated rice-fish farming is July to November and December to April for Boro rice production.On the other hand, alternate rice-fish farming involves producing of fish in the monsoon from June to December and cultivation of Boro rice from November to April.
In case of study area, quick field survey reveals that mono rice farmers produce Boro and Aman in two cropping season, on the other hand, most of the rice-fish farmers are practicing alternate culture due to the lack of enough water and fish prawn in the dry season.So, the samples of the present study include 30 mono rice farmers and 30 alternate rice-fish producing farmers from the time period of June 2015 to April 2016.The sample has been selected purposively for the study.
For accomplishing the study, data have been collected from both primary and secondary sources.In order to collect primary data, well-structured questionnaire, pretested and revised on the basis of findings and experiences originated at the time of pilot survey has been used.Primary cross-section data have been collected through questionnaire survey at the time of final field survey.Besides, necessary secondary data have been collected from published book, journals, working papers, internet etc.

Data analysis and model specification
In order to analysis data, some mathematical tools have been applied in the study.Profit function has been used to identify the net profit of the farmers from mono rice and rice-fish cultivation, and a Cobb-Douglas production function has been applied to estimate the production efficiency of rice-fish farming and mono rice farming in association with measuring the returns to scale from both production function.In the Table 1, description of the variables used has been mentioned with measurement unit.Source: Author's compilation, 2016

Estimation of profit function
For calculating profit level of the farmers, the following general profit functions have been used.

Estimation of production efficiency model
Production efficiency of rice-fish production and mono rice production process has been estimated to observe the efficiency level of the farmers.Econometric model of Cobb-Douglas production function is widely used to estimate production efficiency in agriculture.Since, alternative mixed rice-fish technique is prevalent in the study area, separate equation has been applied to estimate the efficiency level of alternate mixed rice-fish and mono rice production.
Where, r = Returns to scale ∑βi = Sum of all the production coefficients r < 1 = Decreasing returns to scale r = 1 = Constant returns to scale r > 1 = Increasing returns to scale

Hypothesis testing
This study intends to find out the economic difference between mixed rice-fish and mono rice farming in the study area.For this reason, a research hypothesis has been articulated as given below: Null Hypothesis: H0 = There is no difference between mixed rice-fish and mono rice farming in terms of economic returns.Alternative Hypothesis: H1 = There is economic difference between mixed rice-fish and mono rice farming in terms of economic returns.
For data analysis, a number of computer software have been used.At first, collected primary data have been organized and stored by MS Excel 2016.In the second step, mathematical and statistical calculations have been compiled through using STATA 12 and IBM SPSS Statistics 20.

Social profile of the respondents
Social profile of the respondents helps to get general idea of the samples in the study area.Descriptive analysis presented in the Table 2 shows that among the samples average ages of the mixed faming farmers and mono rice farmers are 40 and 42 respectively.Among the mixed farmers, mean education level is 7 years and it is 5 years for the mono rice produces.In addition, there is little difference between the two type of respondents in terms of household size.However, the average farming experience of the former group of farmers is 16 years, whereas, the latter group of farmers have on an average 20 years of farming experience.

Household farm production related information
Farm related information of the household provides significant insight of the production process of the respondents in the study area.To make it clear, cropping season-wise summary statistics of the respondents has been presented in the Table 3.It has been seen from the Table 3 that, Boro production under alternate mixed rice-fish farming system has an average rice production of 4184 kg per acre.In terms of using inputs per acre, on an average, 57 labor man day, 10 kg of rice seed and 308 kg of fertilizer are applied per acre of land.On the other hand, fish production under the same farming system yields on an average 1233 kg of white fish per acre of land in the same area.In order to produce that level of output, it takes on an average 15 labor man day, 195 kg of fish feed, 68 kg of fish prawn and 60 kg of fertilizer.There is variation in the output and inputs employed from respective farmer to farmer as indicated by standard deviation.
On the other hand, Aman production under mono rice production represents different results apart from mixed farming in the same area.It has been found that, average production of Aman per acre of land is 1836 kg.It takes on an average, 12 labor man day, 13 kg rice seed and 51 kg of fertilizer to produce that level of output.However, the production volume of Boro rice under mono rice farming shows different picture.Summary statistics presented in the Table 3 articulates that on an average 3733 kg of rice is produced per acre.To come up with this amount of output, it takes on average, 28 labor man day, 8 kg of rice seed, and 111 kg of fertilizer.There is variation in the output and input level from producer to producer which is shown by standard deviation in the Table 3.   Estimation of profit level of mixed rice-fish and mono rice production Integrated rice-fish farming provides benefits such as economic, optimum and double utilization of paddy field, where it is possible to produce fish and rice simultaneously without supplying any excess fertilizer, food and water and rice yield also will increase (Noorhosseini-Niyaki and Bagherzadeh-Lakani, 2011).One of the prime motives of any production activity is to earn revenue.In this case, attempts have been taken to figure out the level of revenue earned by the farmers who practice rice-fish mixed farming and mono rice farming in the study area.Net return from both farming type has been calculated separately based on cropping season and combinedly as presented in the Table 4.It has been seen from the Table 4 that, farmers have to spend money on labor, seed, land preparation, fertilizer, fish stock, fish fee etc.On the other hand, they get return from rice production, rice straw etc.Estimated results shown in the Table 4 indicates that on an average, Boro rice production under alternate mixed rice-fish farming incurs a total of BDT 20238 as input cost.On the other hand, it gets return amounted to BDT 91426.In the end, the amount of net return reaches to near about BDT 71188 to the producer.In the same manner, on an average, a total of BDT 124955 is left as net return in fish farming under alternate mixed rice-fish farming in the same area.In that case, the total cost is BDT 23004 and

Production efficiency and reruns to scale of mixed rice-fish and mono rice farming
Production efficiency of the respective farming system has been estimated using econometric regression model of Cobb-Douglas production function.Results from alternative mixed rice-fish farming show that coefficient of multiple determination (R 2 ) ranges from 0.91 to 0.32.It implies that 32 to 91 percent of total variation in the production of mixed rice-fish farms can be explained by the explanatory variables used in the models.In addition, it omits variables that can explain 9 to 68 percent of the model.The estimated coefficients of farm size and fertilizer are 0.77 and 0.06 respectively.This indicates that 1 percent increase in farm size and fertilizer, keeping other thing constant, would increase the production by 0.77 percent and 0.06 percent respectively and it is statistically significant at 1 percent.In case of fish production under the same farming system, estimated coefficient value of fish feed is 0.47 which is significant at 5 percent level.It states that 1 percent increase in input fish feed results in 0.47 percent increase in the fish production.The other variables are found statistically insignificant.On the other hand, the estimated coefficients of R 2 of Aman and Boro rice production under mono rice farming system are 0.21 and 0.78.It refers that 21 to 78 percent of total variation in the production of mono rice farms can be explained by the explanatory variables used in the models, the other thing remaining constant.In addition, it excludes explanatory variables that could explain 22 to 79 percent of the model.However, estimated coefficient of input rice seed is 0.67 indicating 1 percent increase in rice seed would increase Aman rice production by 0.67 percent and it is statistically significant at 10 percent level.Besides, estimated coefficient of farm size of Boro rice production is 1.05 meaning 1 percent increase in farm size would increase Boro rice production by 1.05 percent, which is statistically significant at 1 percent level.The other variables are found statistically insignificant.
The summation of all the production coefficients (Σbi) in the alternate rice-fish mixed farming system is 1.36.This result is closer to the findings of (Ahmed and Garnett, 2011), who have revealed that return to scale of alternate rice-fish mixed farming is equivalent to 1.41.Since it is greater than 1, the returns are supposed to increase with the increase in inputs.For instance, if all the inputs mentioned in the model are increased by 1 percent, farm production will increase by 1.36 percent in the alternate rice-fish mixed farming system.This results suggests that there is still scope to increase production under the alternate mixed farming system in the study area by increasing the amount of inputs.On the other hand, the summation of all the production coefficients (Σbi) in the mono rice farming system is 0.763.Since it is smaller than 1, the returns are supposed to decease with the increase in inputs in the study area.

Hypothesis testing
The result of hypothesis testing has been mentioned in the Table 6.It is seen from Table 6 that, there is statistically significant difference between return of mixed rice-fish farming and mono rice farming in the study area.The mean difference of gross return between the two-farming system is found equivalent to BDT 129421.70.The probability of mean difference to be equal to zero or greater than zero is significant at 1 percent level.So, the null hypothesis implying there is no difference of economic return from mixed rice-fish

Figure 1 .
The study area Source: Asiatic Society of Bangladesh(2006)

Table 2 .
Summary of social profile of the respondents [Note: Obs.= Observation, Std.Dev.= Standard Deviation, Min = Minimum, Max = Maximum] Source: Author's compilation based on field survey, 2016

Table 3 .
Summary of farm production information of the respondents [Note: Obs.= Observation, Std.Dev.= Standard Deviation, Min = Minimum, Max = Maximum] Source: Author's compilation based on field survey, 2016

Table 4 .
Profit level of mixed rice-fish and mono rice production