Journal of Economics and Sustainable Development                                                www.iiste.orgISSN 2222-170...
Journal of Economics and Sustainable Development                                                  www.iiste.orgISSN 2222-1...
Journal of Economics and Sustainable Development                                                   www.iiste.orgISSN 2222-...
Journal of Economics and Sustainable Development                                                www.iiste.orgISSN 2222-170...
Journal of Economics and Sustainable Development                                                   www.iiste.orgISSN 2222-...
Journal of Economics and Sustainable Development                                                       www.iiste.orgISSN 2...
Journal of Economics and Sustainable Development                                               www.iiste.orgISSN 2222-1700...
Journal of Economics and Sustainable Development                                              www.iiste.orgISSN 2222-1700 ...
Journal of Economics and Sustainable Development                                                www.iiste.orgISSN 2222-170...
Journal of Economics and Sustainable Development                                                 www.iiste.orgISSN 2222-17...
Journal of Economics and Sustainable Development                                                  www.iiste.orgISSN 2222-1...
Journal of Economics and Sustainable Development                                                 www.iiste.orgISSN 2222-17...
Journal of Economics and Sustainable Development                                                www.iiste.orgISSN 2222-170...
Journal of Economics and Sustainable Development                                         www.iiste.orgISSN 2222-1700 (Pape...
Journal of Economics and Sustainable Development                                                www.iiste.orgISSN 2222-170...
Journal of Economics and Sustainable Development                                               www.iiste.orgISSN 2222-1700...
Journal of Economics and Sustainable Development                                                www.iiste.orgISSN 2222-170...
Journal of Economics and Sustainable Development                                              www.iiste.orgISSN 2222-1700 ...
Journal of Economics and Sustainable Development                                                 www.iiste.orgISSN 2222-17...
Journal of Economics and Sustainable Development                                                    www.iiste.orgISSN 2222...
Journal of Economics and Sustainable Development                                                 www.iiste.orgISSN 2222-17...
Journal of Economics and Sustainable Development                                               www.iiste.orgISSN 2222-1700...
Journal of Economics and Sustainable Development                                                    www.iiste.orgISSN 2222...
Journal of Economics and Sustainable Development                                            www.iiste.orgISSN 2222-1700 (P...
Journal of Economics and Sustainable Development            www.iiste.orgISSN 2222-1700 (Paper) ISSN 2222-2855 (Online)Vol...
Journal of Economics and Sustainable Development   www.iiste.orgISSN 2222-1700 (Paper) ISSN 2222-2855 (Online)Vol.2, No.6,...
Journal of Economics and Sustainable Development                     www.iiste.orgISSN 2222-1700 (Paper) ISSN 2222-2855 (O...
Journal of Economics and Sustainable Development                                             www.iiste.orgISSN 2222-1700 (...
Journal of Economics and Sustainable Development                                            www.iiste.orgISSN 2222-1700 (P...
3645754  content
3645754  content
3645754  content
3645754  content
3645754  content
3645754  content
3645754  content
3645754  content
3645754  content
3645754  content
3645754  content
3645754  content
3645754  content
3645754  content
3645754  content
3645754  content
3645754  content
3645754  content
3645754  content
3645754  content
3645754  content
3645754  content
3645754  content
3645754  content
3645754  content
3645754  content
3645754  content
3645754  content
3645754  content
3645754  content
3645754  content
3645754  content
3645754  content
3645754  content
3645754  content
3645754  content
3645754  content
3645754  content
3645754  content
3645754  content
3645754  content
3645754  content
3645754  content
3645754  content
3645754  content
3645754  content
3645754  content
3645754  content
3645754  content
3645754  content
3645754  content
3645754  content
3645754  content
3645754  content
3645754  content
3645754  content
3645754  content
3645754  content
3645754  content
3645754  content
3645754  content
3645754  content
3645754  content
3645754  content
3645754  content
Upcoming SlideShare
Loading in …5
×

3645754 content

923 views

Published on

The International Institute for Science, Technology and Education (IISTE) , International Journals Call for papaers: http://www.iiste.org/Journals

Published in: Business, Technology
0 Comments
0 Likes
Statistics
Notes
  • Be the first to comment

  • Be the first to like this

No Downloads
Views
Total views
923
On SlideShare
0
From Embeds
0
Number of Embeds
3
Actions
Shares
0
Downloads
3
Comments
0
Likes
0
Embeds 0
No embeds

No notes for slide

3645754 content

  1. 1. Journal of Economics and Sustainable Development www.iiste.orgISSN 2222-1700 (Paper) ISSN 2222-2855 (Online)Vol.2, No.6, 2011Supply Response of Rice in Ghana: A Co-integration Analysis John K.M. Kuwornu (Corresponding author) Department of Agricultural Economics and Agribusiness, P. O. Box LG 68, University of Ghana, Legon-Accra, Ghana Tel: +233 245 131 807 E-mail: jkuwornu@ug.edu.gh / jkuwornu@gmail.com Maanikuu P. M. Izideen Department of Agricultural Economics and Agribusiness, P. O. Box LG 68, University of Ghana, Legon-Accra, Ghana Tel: +233 241 913 985 E-mail: maanikuu@yahoo.com Yaw B. Osei-Asare Department of Agricultural Economics and Agribusiness, P. O. Box LG 68, University of Ghana, Legon-Accra, Ghana Tel: +233 209 543 766 E-mail: yosei@ug.edu.ghReceived: October 1st, 2011Accepted: October 11th, 2011Published: October 30th, 2011AbstractThis study presents an analysis of the responsiveness of rice production in Ghana over the period1970-2008. Annual time series data of aggregate output, total land area cultivated, yield, real prices of riceand maize, and rainfall were used for the analysis. The Augmented-Dickey Fuller test was used to test thestationarity of the individual series, and Johansen maximum likelihood criterion was used to estimate theshort-run and long-run elasticities. The land area cultivated of rice was significantly dependent on output,rainfall, real price of maize and real price of rice. The elasticity of lagged output (12.8) in the short run wassignificant at 1%, but the long run elasticity (4.6) was not significant. Rainfall had an elasticity of 0.004 andsignificant at 10%. Real price of maize had negative coefficient of -0.011 and significant at 10%significance level. This is consistent with theory since a rise in maize price will pull resources away fromrice production into maize production. The real price of rice had an elasticity of 2.01 and significant at 5%in the short run and an elasticity of 3.11 in the long run. The error correction term had the expected negativecoefficient of -0.434 which is significant at 1%. It was found that in the long run only real prices of maizeand rice were significant with elasticities of -0.46 and 3.11 respectively. The empirical results also revealedthat the aggregate output of rice in the short run was found to be dependent on the acreage cultivated, thereal prices of rice, rainfall and previous output with elasticities of 0.018, 0.01, 0.003 and 0.52 respectively.Real price of rice and area cultivated are significant 10% level of significance while rainfall and laggedoutput are significant 5%. In the long run aggregate output was found to be dependent on acreage cultivated,real price of rice, and real price maize with elasticities of 0.218, 0.242 and -0.01 respectively at the 1%significance level. The analysis showed that short-run responses in rice production are lower than long-runresponse as indicated by the higher long-run elasticities. These results have Agricultural policy implicationsfor Ghana.Key Words: Supply response, Rice, Error Correction Model, Co-integration Analysis, Ghana1. IntroductionOne of the most influential policy prescriptions for low-income countries ever given by developmenteconomists has been to foster industrialization by withdrawing resources from agriculture (e.g., Lewis,1954). There is robust evidence that the majority of policy makers followed this prescription at least untilthe mid-1980s. The results of a comprehensive World Bank study (Krueger et al., 1992), for example, showfor the period 1960 –1985 that in most countries examined, agriculture was taxed both directly via 1
  2. 2. Journal of Economics and Sustainable Development www.iiste.orgISSN 2222-1700 (Paper) ISSN 2222-2855 (Online)Vol.2, No.6, 2011interventions in agricultural markets and indirectly via overvalued exchange rates and import substitutionpolicies. It is not obvious whether the disincentives for agricultural production have continued to exist sincethe mid-1980s. On the one hand, most developing countries have adopted structural adjustment programswhich explicitly aim at a removal of the direct and indirect discrimination against agriculture. But, on theother hand, it is known that many of these programs were not fully implemented, especially in Sub-SaharanAfrica (Kherallah et al., 2000; Thiele and Wiebelt 2000; World Bank, 1997). Hence it can be concluded thata certain degree of discrimination still prevails.One of the most important issues in agricultural development economics is supply response of crops. Thisis because the responsiveness of farmers to economic incentives determines agricultures’ contribution to theeconomy especially where the sector is the largest employer of the labour force. This is often the case inthird world low income countries. Agricultural pricing policy plays a key role in increasing farm production.Supply response is fundamental to an understanding of this price mechanism (Nerlove and Bachman,1960).There is a notion that farmers in less developed countries respond slowly to economic incentives such asprice and income. Reasons cited for poor response vary from factors such as constraints on irrigation andinfrastructure to a lack of complementary agricultural policies. The importance of non-price factors drewadequate attention in the literature: rainfall, irrigation, market access for both inputs and output, and literacy.The reason cited for a low response to prices in less developed economies is the limited access to input andproduct markets or high transaction costs associated with their use. Limited market access may be eitherdue to physical constraints such as absence of proper road links or the distances involved between the roadsand the markets, or institutional constraints like the presence of intermediaries (Mythili, 2008).In Ghana, the goals of agricultural price policy are among others fair incomes for farmers, low food pricesfor urban consumers, cheap raw materials for manufacturing (Ministry of Food and Agriculture (MOFA),2005). Price support was one policy that was used by government in targeting these goals. Price support hasbeen used in many countries across the world. The prices of major commodities have been set below worldprices using subsidies and trade barriers, guaranteed prices (act as floors) and domestic market forcesdetermining actual prices (Food and Agriculture Organization (FAO), 1996). The effect of liberalization onthe growth of agriculture crucially depends on how the farmers respond to various price incentives.For many low-income countries, the impact of structural reforms on economic growth and povertyalleviation crucially depends on the response of aggregate agricultural supply to changing incentives.Despite its policy relevance, the size of this parameter is still largely unknown (Thiele, 2000).Rice is a very important crop especially for those areas where it is produced. Rice is gradually taking theposition as the main staple food for the majority of families in Ghana especially in the urban centres. Thegrowth in consumption of rice has been not been matched with a corresponding growth in local productionof the crop. Per capita consumption of rice has steadily increased over the years since the 1980s from about12.4kg/person /year in 1984 to about 20kg/person /year (Statistical Research and Information Division(SRID), 2005). It continued to rise to 38kg/person/year in 2008 (National Rice Development Policy(NRDP), 2009). Over the last decade rice per capita consumption has increased by more than 35%. This isattributed mainly to the rapid urbanization and changes in food consumption patterns. It is estimated basedon population and demand growth rates that per capita consumption will reach 41.1kg by 2010 and 63.0kgby 2015 giving an aggregate demand of 1,680,000tons/year (Statistical Research and Information Division(SRID), 2005). Changes in population dynamics and the taste or preference for foreign products iscontributing to this trend. These changes are significant considering the rate of population growth. Ghana’spopulation has grown by about 70% from 12.3 million in 1984 to about 21 million in 2004. The populationis now 24 million (Ghana Statistical Service (GSS), 2011).The growing quantum of rice imports into the country has also been triggered partly by the fall in worldmarket price. This is largely due to increased output levels in India and South Asia. Massive subsidies inrice exporting countries have also contributed to this phenomenon. Despite all these developments thereseem to be a lack of coherent and comprehensive national policy for rice in Ghana. From the period ofeconomic recovery and structural adjustment, the country has had to embark on trade liberalization and theremoval of all forms of subsidies on agriculture. These policies no doubt have adversely affected riceproduction in the country. The smallholder rural farmer is faced with unfair competition from abroad. The 2
  3. 3. Journal of Economics and Sustainable Development www.iiste.orgISSN 2222-1700 (Paper) ISSN 2222-2855 (Online)Vol.2, No.6, 2011role of incentives to farmers has generally been sidelined or ignored in most developing countries due toconditions of trade liberalization and global trade integration. However, a number of empirical studies inother developing economies addressed the question of farmers’ response to economic incentives andefficient allocation of resources (e.g. Chinyere, 2009). The agricultural sector in Ghana has undergonevarious policy regimes which has affected both the factor and product market resulting in changes in thestructure of market incentives (prices) faced by farmers. Most of these policies have, however, not beencrop specific and therefore has wide variations in the quantum of changes in the incentives. This studytherefore follows the supply response framework of analysis to examine the dynamics of the supply of ricein Ghana. Effort in this direction will have to be preceded by a thorough analysis of the factors that affectsthe supply of rice. These teething problems lead to the following research questions;How responsive is rice production to price and non-price factors? What are the long run and short runelasticities of rice production? What are the trends in area cultivated, output and real prices of output?Therefore, objective of the study is threefold: (a) Examine the acreage and output response of riceproduction in Ghana. (b) Estimate and compare the long run and short run elasticities of rice production.(c) Analyze the trends in output, area cultivated and real prices.The rest of the paper is structured as follows. Section 2 outlines the methodology. In section 3 we presentthe empirical applications and the results. Section 4 provides the conclusions.2.0 MethodologyThis section presents the methodology of the study.2.1 Linear RegressionLinear regression was conducted to determine the growth rates of the variables over the study period. Timein years was regressed on acreage cultivated, aggregate output, real price of rice separately.2.2 Time series analysisIn empirical analysis using time series data it is important that the presence or absence of unit root isestablished. This is because contemporary econometrics has indicated that regression analysis using timeseries data with unit root produce spurious or invalid regression results (e.g. Townsend, 2001). Most timeseries are trended over time and regressions between trended series may produce significant parameterswith high R2s, but may be spurious or meaningless (Granger and Newbold, 1974). When using the classicalstatistical inference to analyze time series data, the results are only stationary when the series are stationary.The solution to this problem was initially provided by Box and Jenkins (1976), by formulating regressionsin which the variables were expressed in first difference. Their approach simply assumed thatnon-stationary data can be made stationary by repeated differencing until stationarity is achieved and thento perform the regression using these differenced variables. However, according to Davidson et al., (1978),this process of repeated differencing even though leads to stationarity of the series; it is achieved at theexpense of losing valuable long run information. This posed a new challenge to time series econometrics.The concept of cointegration was introduced to solve these problems (Granger, 1981; Engle and Granger,1987). By using the method of cointegration an equation can be specified in which all terms are stationaryand so allow the use of classical statistical inference. It also retains information about the long runrelationship between the levels of variables, which is captured in the stationary co-integrating vector. Thisvector will comprise the parameters of the long run equilibrium and corresponds to the parameters of theerror correction term in the second stage regression (Mohammed, 2005). The cointegration approach takesinto consideration the long-run information such that spurious results are avoided.2.3 Stationarity TestsA data series is said to be stationary if it has a constant mean and variance. That is the series fluctuates aroundits mean value within a finite range and does not show any distinct trend over time. In a stationary seriesdisplacement over time does not alter the characteristics of a series in the sense that the probabilitydistribution remains constant over time. A stationary series is thus a series in which the mean, variance andcovariance remain constant over time or in other words do not change or fluctuate over time. In a stationary 3
  4. 4. Journal of Economics and Sustainable Development www.iiste.orgISSN 2222-1700 (Paper) ISSN 2222-2855 (Online)Vol.2, No.6, 2011series the mean always has the tendency to return to its mean value and to fluctuate around it in a more or lessconstant range, while a non-stationary series has a changing mean at different points in time and its variancechange with the sample size (Mohammed, 2005). The conditions of stationarity can be illustrated by thefollowing:Yt = ѳYt-1 + µ t t=1………T (1)Where µ t is a random walk with mean zero and constant variance. If ѳ < 1, the series Yt stationary and ifѳ = 1 then the series Yt is non-stationary and is known as random walk. In other words the mean, varianceand covariance of the series Yt changes with time or have an infinite range. However Yt can be madestationary by differencing. Differencing can be done multiple times on a series depending on the number ofunit roots a series has. If a series becomes stationary after differencing d times, then the series contains dunit roots and hence integrated of order d denoted as I (d). Thus, in equation (1) where ѳ = 1, Yt has a unitroot. A stationary series could also exhibit other properties such as when there are different kinds of timetrends in the variable.The DF (Dickey-Fuller)-statistic used in testing for unit root is based on the assumption that µ t is whitenoise. If this assumption does not hold, it leads to autocorrelation in the residuals of the OLS regressionsand this can make invalid the use of the DF-statistic for testing unit root. There are two approaches to solvethis problem (Towsend, 2001). In the first instance the equations to be tested can be generalized. Secondlythe DF-statistics can be adjusted. The most commonly used is the first approach which is the AugmentedDickey-Fuller (ADF) test. µ t is made white noise by adding lagged values of the dependent variable to theequations being tested, thus:∆Yt = (ѳ1 – 1) Yt-1 + IYt-I + µ t (2)∆Yt = α2 + (ѳ2 – 1) Yt-1 + I ∆Yt-I + µ t (3)∆Yt= α3+β3t(ѳ3–1)Yt-1+ I ∆Yt-I + µ t (4)The ADF test uses the same critical values with DF. The results of the ADF test for unit roots for each ofthe data series used in this study are presented in in the next section using equation (4) where Yt is theseries under investigation, t is the time trend, α3 is the constant term and µ t are white noise residuals.Eviews was used in the analysis, all the data series was tested for stationarity and the results are presentedin section three.2.4 CointegrationCointegration is founded on the principle of identifying equilibrium or long run relationships betweenvariables. If two data series have a long run equilibrium relationship it implies their divergence from theequilibrium are bounded, that is they move together and are cointegrated. Generally for two or more seriesto be co-integrated two conditions have to be met. One is that the series must all be integrated to the sameorder and secondly a linear combination of the variables exist which is integrated to an order lower thanthat of the individual series. If in a regression equation the variables become stationary after firstdifferencing, that is I (1), then the error term from the cointegration regression is stationary, I (0)(Hansen and Juselius, 1995). If the cointegration regression is presented as:Yt = α + βXt + µ t (5)where Yt and Xt are both I (1) and the error term is I (0), then the series are co-integrated of order I (1,0 )and β measures the equilibrium relationship between the series Yt and Xt and µ t is the deviation from thelong-run equilibrium path. An equilibrium relationship between the variables implies that even though Ytand Xt series may have trends, or cyclical seasonal variations, the movement in one are matched bymovements in the other. The concept of cointegration has implications for economists. The economicinterpretation that is accepted is that if in the long-run two or more series Yt and Xt themselves arenon-stationary, they will move together closely over time and the difference between them is constant(stationary) (Mohammed 2005).2.4.1 Testing for CointegrationThere are two most commonly used methods for testing cointegration. The Augmented Dickey-Fullerresidual based test by Engle and Granger (1987), and the Johansen Full Information Maximum Likelihood(FIML) test (Johansen and Juselius, 1990). For the purpose of this study the Johansen Full InformationMaximum Likelihood test is used due to its advantages. The major disadvantage of the residual based test is 4
  5. 5. Journal of Economics and Sustainable Development www.iiste.orgISSN 2222-1700 (Paper) ISSN 2222-2855 (Online)Vol.2, No.6, 2011that it assumes a single co-integrating vector. But if the regression has more than one co-integrating vectorthis method becomes inappropriate (Johansen and Juselius, 1990). The Johansen method allows for allpossible co-integrating relationships and allows the number of co-integrating vectors to be determinedempirically.2.4.2 Johansen Full Information Maximum Likelihood ApproachThe Johansen approach is based on the following Vector AutoregressionZt = AtZt-1 + … + AkZt-k + µ t (6)Where Zt is an (n×1) vector of I(1) variables (containing both endogenous and exogenous variables), At is(n×n) matrix of parameters and µ t is (n×1) vector of white noise errors. Zt is assumed to be nonstationaryhence equation (6) can be rewritten in first difference or error correction form as;∆Zt = Γ1∆Zt-1 + … + Γk-1∆Zt-k+1 + πZt-k + µ t (7)where Γ1 = - ( 1- A1 - A2 - … -Ai), (i = 1, …, k-1) and π = - (1- A1-A2- …-Ak).Γ1 gives the short run estimates while π gives the long run estimates. Information on the number ofco-integrating relationships among variables in Zt is given by the rank of the matrix π. If the rank of πmatrix r, is 0 < r > n, there are r linear combinations of the variables in Zt that are stationary. Thus π can bedecomposed into two matrices α and β where α is the error correction term and measures the speed ofadjustment in ∆Zt and β contains r co-integrating vectors, that is the cointegration relationship betweennon-stationary variables. If there are variables which are I(0) and are significant in the long runco-integrating space but affect the short run model then equation (7) can be rewritten as:∆Zt = Γ1∆Zt-1 + πZt-k + vDt + µ t (8)where Dt represents the I(0) variables.To test for co-integrating vector two likelihood ratio (LR) tests are used. The first is the trace test statistic;Λtrace = -2lnQ = -T i) (9)Which test the null hypothesis of r co-integrating vectors against the alternative that it is greater than r. Thesecond test is known as the maximal-eigen value test:Λmax = -2 ln(Q: r 1 r + 1 = -T ln(1-λr+1 ) (10)which test the null hypothesis of r co-integrating vectors against the alternative of r+1 co-integratingvectors. The trace test shows more robustness to both skewness and excess kurtosis in the residuals than themaximal eigen value test (Harris, 1995). The error correction formulation in (7) includes both the differenceand level of the series hence there is no loss of long run relationship between variables which is acharacteristic feature of error correction modeling.It should be noted that in using this method, the endogenous variables included in the VectorAutoregression (VAR) are all I(1), also the additional exogenous variables which explain the short runeffect are I(0). The choice of lag length is also important and the Akaike Information Criterion (AIC), theScharz Bayesian Criterion (SBC) and the Hannan-Quin Information Criterion (HQ) are used for theselection.According to Hall (1991) since the process might be sensitive to lag length, different lag orders should beused starting from an arbitrary high order. The correct order is where a restriction on the lag length isrejected and the results are consistent with theory.2.5 Error Correction Models (ECMs)The idea behind the mechanism of error correction is that a proportion of disequilibrium from one period iscorrected in the next period in an economic system (Engle and Granger, 1987). The process of making adata series stationary is either done by differencing or inclusion of a trend. A series that is made stationaryby including a trend is trend stationary and a series that is made stationary by differencing is differencestationary. The process of transforming a data series into stationary series leads to loss of valuable long runinformation (Engle and Granger, 1987). Error correction models helps to solve this problem.The Granger representation theorem is the basis for the error correction model which indicates that if thevariables are cointegrated, there is a long-run relationship between them and can be described by the errorcorrection model. The following equation shows an ECM of agricultural supply response involving thevariables Y and X in its simplest form: 5
  6. 6. Journal of Economics and Sustainable Development www.iiste.orgISSN 2222-1700 (Paper) ISSN 2222-2855 (Online)Vol.2, No.6, 2011∆Yt = α∆Xt – ѳ(Yt-1 – γXt-1)+µ t (11)Where µ t is the disturbance term with zero mean, constant variance and zero covariance. Parameter α takesinto account the short run effect on Y of the changes in X, while γ measures the long-run equilibriumrelationship between Y and X that is:Yt = γXt + µ t (12)Where Yt-1 – γYt-1 + µ t-1 measures the divergence (errors) from long-run equilibrium. Also ѳ measures theextent of error correction by adjustment in Y and its negative sign indicates that the adjustment is in thedirection which restores the long-run relationship (Hallam and Zanoli, 1993). In order to estimate equation(5), Engle and Granger (1987) proposed a two stage process. Firstly the static long run cointegrationregression (6) is estimated to test cointegration between the two variables. If cointegration exists the laggedresiduals from equation (5) are used as error correction term in the Error Correction Model (in equation 6)to estimate the short run equilibrium relationship between the variables in the second stage. The validity ofthe Error Correction Models (ECMs) depends upon the existence of a long-run or equilibrium relationshipamong the variables (Mohammed, 2005).The Error Correction Model (ECM) has several advantages. It contains a well-behaved error term andavoids the problem of autocorrelation. It allows consistent estimation of the parameters by incorporatingboth short-run and long-run effects. Most importantly all terms in the ECM are stationary. It ensures that noinformation on the levels of the variables is lost or ignored by the inclusion of the disequilibrium terms(Mohammed, 2005). ECM solves the problems of spurious correlation because ECMs are formulated interms of first difference which eliminates trends from the variables (Ganger and Newbold, 1974). It avoidsthe unrealistic assumption of fixed supply based on stationary expectations in the partial adjustment model.2.6 Supply Response ModelsThis study estimated the total area cultivated and aggregate output of rice in Ghana using doublelogarithmic regression models.Area cultivated of rice (lgarea) is a function of own price (Lrp), rainfall (lgrain), aggregate output (lgoutput)and price of maize (lmp). The equation used for the regression is:Lgarea = f1 (Lgoutput, Lrp, Lgrain, Lmp) (13)Aggregate output of rice (lgoutput) is a function of the area cultivated (lgarea), the price of rice (Lrp), priceof maize (Lmp), rainfall (lgrain). The estimating equation is:Lgoutput = f2 (lgrain, Lrp, Lmp, Lgarea)1. (14)3. Empirical Application and Results3.1 Results for linear regressionTable 1: The results of the regression analysis for the trend of the variables against timeVariable Coefficient Std error t-statistic R2 F-statistic ProbLgarea 1.334274 1.149167 1.61079 0.035154 0.253046 0.2530Lgoutput -13.86625 68.09380 -0.20364 0.005889 0.844432 0.8444lgyield -25.99348 35.51112 -0.73198 0.071100 0.487957 0.4880Lrp 2.588941 5.221601 4.956136 0.399186 24.58311 0.0000As can be seen from table 1 the results indicated that the regression for area, output and yield turned outinsignificant. Only the trend of real price of rice was significant at 1% significance level. Real rice priceyielded a positive coefficient of 2.589 which implies that for each year the real price of rice grew by 2.589units.3.2 Unit Root Test ResultsAs a requirement for cointegration analysis the data was tested for series stationarity and to determine theorder of integration of the individual variables. For cointegration analysis to be valid all series must beintegrated of the same order usually of order one (Towsend, 2001). Eviews was used to perform these tests.1 Price of maize is included because we assume that the same resources (land type, fertilizer etc.) can be used toproduce both maize and rice. Hence, a rise in the price of maize will pull resources away from rice production to maizeproduction. 6
  7. 7. Journal of Economics and Sustainable Development www.iiste.orgISSN 2222-1700 (Paper) ISSN 2222-2855 (Online)Vol.2, No.6, 2011The data series on annual acreage cultivated (Lgarea), aggregate output (Lgoutput), aggregate yield(Lgyield), real price of rice (Lrp), real price of maize (Lmp), rainfall (Lgrain) was tested for unit root forthe study period 1970 - 2008. The Augmented Dickey-Fuller test was used for this test. The results arepresented below.Table 2: Results of unit root test at levelsSeries ADF test Mackinnon Lag-length Prob Conclusion statistic critical valueLgarea 2.221125 3.615588 2 0.2024 Non-stationaryLgoutput 1.519278 5.119808 2 0.4582 Non-stationaryLgyield 0.103318 4.580648 2 0.9419 Non-stationaryLmp 0.378290 3.621023 2 0.9026 Non-stationaryLrp 1.429037 3.615588 2 0.5579 Non-stationaryLgrain 3.006033 2.963972 7 0.0457 StationaryThe results of the unit root test after first differencing are presented in table 3.Table 3: Results of unit root test at first differencesSeries ADF test Mackinnon Lag-length Prob Conclusion statistic critical valueLgarea 7.517047 3.621023 2 0.0000 I(1)Lgoutput 9.577098 3.621023 2 0.0000 I(1)Lgyield 7.517047 3.621023 2 0.0000 I(1)Lmp 10.02189 3.621023 2 0.0000 I(1)Lrp 6.346392 3.626784 2 0.0000 I(1)Note; All variables are in log form. The ADF method test the hypothesis that H0 : X ~ I(1), that is, has unitroot (non-stationary) against H1 : X ~ I(0), that is, no unit root (stationary). The Mackinnon critical valuesfor the rejection of the null hypothesis of unit root are all significant at 1%. Lgarea denotes log of areacultivated, Lgoutput denotes log total output, Lgyield denotes log of yield, Lmp denotes log of maize price,Lrp denotes log of rice price and Lgrain denotes log of rainfall.The results of the unit root tests showed that all the series are non-stationary at levels except for rainfallwhich is stationary at levels as shown in table 2 above. However as expected all the non-stationary seriesbecame stationary after first differencing. From table 2 the null hypothesis of unit root could not be rejectedat levels since none except rainfall of the ADF test statistics was greater than the relevant MackinnonCritical values. Hence the null of the presence of unit root is accepted.However the hypothesis of unit root in all series was rejected at 1% level of significance for all series afterfirst difference since the ADF test statistics are greater than the respective Mackinnon critical values asshown in table 3 above.3.3 Cointegration ResultsWhen the order of integration of the data series have been established, the next step in the process ofanalysis is to determine the existence or otherwise of cointegration in the series. This is to establish theexistence of valid long-run relationships between variables. Basically there are two most commonly usedmethods to test for cointegration. These were suggested by Engle and Granger (1987) and Johansen andJuselius (1990). This study applies the Johansen approach which provides likelihood ratio tests for thepresence of number of co-integrating vectors among the series and produces long-run elasticities. The Errorcorrection model was then used to estimate short-run elasticities. 7
  8. 8. Journal of Economics and Sustainable Development www.iiste.orgISSN 2222-1700 (Paper) ISSN 2222-2855 (Online)Vol.2, No.6, 20113.3.1 Supply Response of Rice OutputFirstly, the Vector Error Correction Modeling (VECM) procedure using the Johansen method involvesdefining an unrestricted Vector Autoregression (VAR) using the following equation.Zt=AtZt-1+…+AkZt-1+µ t (15)Likelihood Ratio (LR) tests were conducted with maximum of three lags due to the short time series. Theresults are shown in table 4.Table 4: Results for lag selection output model VAR Lag Order Selection Criteria Endogenous variables: OUTPUT AREA MP RP Exogenous variables: C RAIN Lag LogL LR FPE AIC SC HQ0 -1208.968 NA 6.64e+32 86.92626 87.30689* 87.042621 -1184.928 37.77692 3.84e+32 86.35198 87.49387 86.701072 -1174.563 13.32579 6.37e+32 86.75452 88.65766 87.336333 -1138.068 36.49514* 1.92e+32* 85.29058* 87.95499 86.10511* * indicates lag order selected by the criterion LR: sequential modified LR test statistic (each test at 5% level) FPE: Final prediction error AIC: Akaike information criterion SC: Schwarz information criterion HQ: Hannan-Quinn information criterion LogL: Log-likelihoodThe results indicate that the Akaike Information Criterion (AIC) and the Hannan-Quin information criterion(HQ) selected lag order three while the Scharz Bayesian Criterion (SBC) selected the lag order zero. Thus,this study selects the lag order three as the order for the VAR models. The next step in the Johansen methodis to test for the number of co-integrating vectors among the series in the model.The results for the cointegration test imply that both the trace test and the maximum eigen value test selectsthe presence of one co-integrating vector, thus it can be concluded that the variables in the model areco-integrated. The Johansen model is a form of Error Correction Model. When only one co-integratingvector is established its parameters can be interpreted as estimates of long run co-integrating relationshipbetween the variables (Hallam and Zanoli, 1993). This implies that the estimated parameter values fromthis equation when normalized on output are the long run elasticities for the model. Eviews automaticallyproduces the normalized estimates. These coefficients represent estimates of long-run elasticities withrespect to area cultivated, real price of rice and real price of maize. The normalized cointegration equationfor rice output is given below;Output=0.216748area+0.241522rp-0.009975mp-12841.69 (16)Since cointegration has been established among the variables, then the dynamic ECM can be used forsupply response analysis since it provides information about the speed of adjustment to long runequilibrium and avoids the spurious regression problem between the variables (Engle and Granger, 1987).The ECM for rice output is presented as;∆output = α0+ 1i∆outputt-i + 2i∆areat-i + 3i∆rpt-I + 4i∆mpt-I + 5rain – θECt-I (17)Where θECt-I = α (β1outputt-I – β2areat-I – β3rpt-I – β4mpt-I) 8
  9. 9. Journal of Economics and Sustainable Development www.iiste.orgISSN 2222-1700 (Paper) ISSN 2222-2855 (Online)Vol.2, No.6, 2011In the ECM model above, the α’s explain the short run effect of changes in the explanatory variables on thedependent variable whereas β’s represent the long run equilibrium effect. θECt-I is the error correction termand correspond to the residuals of the long run cointegration relationship, that is the normalized equation(16). The negative sign on the error correction term indicates that adjustments are made towards restoringlong run equilibrium. This representation ensures that short run adjustments are guided by and consistentwith the long run equilibrium relationship. This method provides estimates for the short run elasticities, thatis the coefficients of the difference terms and, whereas the parameters from the Johansen cointegrationregression are estimates of the long run elasticities (Townsend and Thirtle, 1994). In selecting the best ECMestimates, models with lag lengths are estimated and those with insignificant parameters are eliminated.Note that the estimates presented above represent the short run effect of the explanatory variables on thedependent variable. The long run effect is captured by the estimates of the normalised Johansen regressionresults presented in equation (16). The diagnostic tests are the t-ratio test of the coefficients, LM test forautocorrelation and the Jarque Berra test for normality of residuals.Table 5: ECM results for rice outputVariable Coefficient t-statistic Prob∆lgarea 0.017611 0.668907 0.0510∆lgoutput(-1) 0.523393 2.283959 0.0319lgRain 0.003740 0.997918 0.0328∆lgmp -0.001107 -0.298174 0.7683∆lgrp 0.009922 0.307758 0.0761Residual -0.174253 -1.321005 0.0043R2 (0.754242) F-statistic (2.523411) Prob(F-stat) 0.058261It indicates that aggregate rice output is dependent on area cultivated, previous year’s output, and previousyear’s price of rice. The coefficient of maize price was not significant.An R2 of 0.754 indicates that 75% of the variation in the dependent variable can be accounted for byvariation in the explanatory variables. The results indicate that area cultivated was significant at 10%, withelasticity of 0.0176, which implies that a one percent increase in area cultivated of rice will lead to a0.018 % increase in output in the short run. This is rather on the low side. This can be attributed to the factthat an increase in the land cultivated without a necessary increase in the resources of the farmer willincrease adverse effect on the minimal resources available to the farmer hence resulting in this marginalincrease in output. This result also indicates that the productivity of the farmer reduces with increase infarm size. The long run elasticity of area cultivated is 0.2167 (in equation (16)) is higher than the 0.0176 forthe short run. This indicates that over the long run farmers adjust their farm sizes more to output than in theshort run. That is, a 1% increase in area cultivated will lead to 0.217% increase in output in the long run.Lagged output has elasticity of 0.523393 in the short run and is significant at the 5% significance levelimplying that a one percent increase in output in a year will lead to a 0.52% increase in output thesubsequent year. This simply means that an increase in output will make capital available for the farmer toinvest in the subsequent year’s rice production activities. This is only true if the additional resource isinvested into the following year’s rice production activities.Rainfall is significant at 5%. An elasticity of 0.003740 for rainfall indicates that a 1% increase in rainfallwill result in a 0.004% rise in output. This low impact of rainfall can be explained by the fact that a largeproportion of total rice output is produced under the irrigation projects across the country. This makes theresponse of output to rainfall highly inelastic.A coefficient of 0.009922 which is significant at 10% for real price of rice indicates that a 1% rise in real 9
  10. 10. Journal of Economics and Sustainable Development www.iiste.orgISSN 2222-1700 (Paper) ISSN 2222-2855 (Online)Vol.2, No.6, 2011prices will lead to 0.01% increase in output the subsequent year in the short run. This makes prices alsoinelastic. This low rice price elasticity suggests that farmers do not always necessarily benefit fromincreasing prices due to the structure of the marketing system. Middlemen and other marketing channelmembers purchase the rice from farmers at the farm gate and thereafter transport it to market centres to besold. Hence when there is a rise in prices a little fraction of it is transmitted to the famer. The low priceelasticity could also be attributed to the fact that farmers are hindered by an array of constraints such asland tenure issues, rainfall variability, lack of capital resources and credit facilities which limits theircapacity to respond to price incentives. The long run coefficient of 0.2415 indicates that 1% percentincrease in real prices will lead to a 0.24% increase in output. The long run elasticity far exceeds the shortrun. This is because over the long run when prices show a continual rise, farmers are able to accumulatecapital enough to enhance production.The residual which is the error correction term is significant at 1% and has the expected negative sign. Itmeasures the adjustment to equilibrium. Its coefficient of -0.174 indicates that the 17.4% deviation of riceoutput from long run equilibrium is corrected for in the current period. This slow adjustment can beattributed to the fact that famers in the short run are constrained by technical factors as mentioned earlierwhich limits their ability to adjust immediately to price incentives.In the long run maize price is significant at 1%. With a negative coefficient of 0.009975 it implies that a 1%increase in the price of maize will lead to 0.01% reduction in the output of rice. This is to say that resourceswill be diverted to maize production relative to rice production leading to the fall in rice output. However,Maize price was not significant in the short run. The results of the LM test of serial correlation for up tofifth order show that there is no serial correlation in the data set.3.3.2 Supply Response of Area CultivatedTesting for the selection of lag length yielded the same results for both the acreage and output models.Hence the same lag order three is therefore used for the Vector Error Correction model (VEC).The trace test selects one co-integrating vector while the eigen value test selects two co-integrating vectors.But since the trace test is the more powerful test (Mohammed, 2005), the result from the trace test is usedhere. Thus the acreage model has one co-integrating vector. The normalised cointegration equation for areacultivated is given as;lgarea =4.613649lgoutput+3.11429lrp–0.46020lmp–59247 (18)The ECM for area cultivated is presented as;∆lgarea = α0 + 1i ∆lgoutputt-i + 2i ∆lgareat-i + 3i ∆lgrpt-I + 4i ∆lgmpt-I + 5l lgrain –θECt-I (19)Where θECt-I = α (β1areat-I – β2outputt-I – β3rpt-I – β4mpt-I)The result of the estimated ECM is given in table 6 below.Table 6: ECM results for area cultivatedVariable Coefficient t-statistic Prob∆lgarea(-1) 0.169747 1.343021 0.1192∆lgoutput(-1) 12.80810 1.816372 0.0824lgRain 0.003984 0.138622 0.0891∆lmp(-1) -0.011350 -0.401819 0.0691∆lrp(-1) 2.016747 1.440683 0.0265Residual -0.435411 -1.043641 0.0047R2 (0.770669) F-statistic Prob(F-stat) 0.017301 (1.707148)An R2 of 0.77067 indicates that 77% 0f the variation in the dependent variable is accounted for by variationin the explanatory variables. Output is significant at 10% with an elasticity of 12.80 indicating that a 1%increase in output this year will increase land cultivated in the subsequent year by 12.8% in the short run.Thus acreage cultivated is highly elastic with respect to output. An increase output level gives the farmersan opportunity to acquire necessary resources and equipment to put more land under cultivation. Output 10
  11. 11. Journal of Economics and Sustainable Development www.iiste.orgISSN 2222-1700 (Paper) ISSN 2222-2855 (Online)Vol.2, No.6, 2011elasticity is even higher than that for real prices. This can be attributed to the fact that a large proportion ofoutput is for household consumption and thus is independent of prices. Increased output alone is enoughmotivation to increase acreage under cultivation. In the long run output is insignificant. This because in thelong run land will get exhausted and output will be dependent on productivity (and not the area of landcultivated).Rainfall is significant at 10% with a coefficient of 0.003984; this implies a 1% increase in rainfall leads to0.004% increase in area cultivated. This low response to rain can be attributed to the fact that large areasunder rice cultivation are in irrigated fields which depends less on rainfall for cultivation. Rainfall is anexogenous variable (I(0)), hence it measures short run effect.Maize price is significant at 10% with a negative coefficient of -0.011350 in the short run and -0.460 in thelong run. This means over the long run if maize prices are continually increasing farmers will commit moreresources into maize farming relative to rice farming. Thus as the price of maize rises area under ricecultivation reduces both in the short and long run. In the short run a 1% increase in maize price reduces areacultivated of rice by 0.0114% and by 0.46% in the long run.Real price of rice was significant at 5% with elasticities of 2.017 and 3.11 in the short run and long runrespectively, thus price is elastic. This implies a 1% percent increase in real price of rice will result 2.017%and 3.11% increase in acreage cultivated in the short run and long run respectively. This is expected sincein the long run farmers are able to adjust to overcome some the major challenges of production and henceare able to adjust more to price incentives.The error correction term has the expected negative sign and with a coefficient of 0.4354 indicating that43.54% of the deviation from long run equilibrium is corrected for in the current period.The results of the test for serial correlation show that is no autocorrelation in the series. The Jarque-Berratest statistic of 10.18 with probability value of 0.25 implies that the residuals are normally distributed.4. ConclusionsEconomic theory in the past had been based on the assumption that time series data is stationary and hencestandard statistical techniques (Ordinary Least Squares regression (OLS)) designed for stationary series wasused. However, it is now known that many time series data are non-stationary and therefore usingtraditional OLS methods will lead to invalid or spurious results. To address this problem, the method ofdifferencing was introduced. Differencing according to Granger however leads to loss of valuable long-runinformation. Granger and Newbold (1974) introduced the technique of cointegration which takes intoaccount long-run information therefore avoiding spurious results while maintaining long-run information.This study presents an analysis of the responsiveness of rice production in Ghana over the period1970-2008. Annual time series data of aggregate output, total land area cultivated, yield, real prices of riceand maize, and rainfall were used for the analysis.The Augmented-Dickey Fuller test was used to test the stationarity of the individual series. The Johansenmaximum likelihood criterion is used to estimate the short-run and long-run elasticities.The trend analysis for rice output, rice price, acreage cultivated, and yield revealed that only rice price issignificant at 1% significance level. The results imply that for each year, the price of rice will increase by2.589 units.All the time series data that was used were tested for unit root. They were found to be non-stationary atlevels but stationary after first differencing at the one percent significance level except for rainfall whichwas stationary at levels at the 5% significance level. The Likelihood Ratio tests selection of lag orderselected order three by the AIC and HQ criteria for both models. The Johansen cointegration test selectedone cointegration vector (in both rice output and rice price models) indicating that the variables areco-integrated. The diagnostic tests of serial correlation and normality test was done using the LagrangeMultiplier test for autocorrelation and the Jarque-Berra test for normality of residuals. The results indicateno serial correlation and normally distributed residuals.The land area cultivated of rice was significantly dependent on output, rainfall, real price of maize and realprice of rice. The elasticity of lagged output was 12.8 in the short run and was significant at 1%. However,this elasticity was not significant in the long run.Rainfall had an elasticity of 0.004 and significant at 10%. Also real price of maize had negative coefficientof -0.011 which was significant at 10%. This is consistent with theory since a rise in maize price will pull 11
  12. 12. Journal of Economics and Sustainable Development www.iiste.orgISSN 2222-1700 (Paper) ISSN 2222-2855 (Online)Vol.2, No.6, 2011resources away from rice production into maize production. The coefficient of the real price of maizeestimated by Chinere (2009) was -0.066 for rice farming in Nigeria. This is higher than the -0.011 estimatedfor Ghana (in this study).The real price of rice had an elasticity of 2.01 and significant at 5% in the short run and an elasticity of 3.11in the long run. The error correction term had the expected negative coefficient of -0.434 which issignificant at 1%. It was found that in the long run only real prices of maize and rice were significant withelasticities of -0.46 and 3.11 respectively. The error correction estimated by Chinere (2009) was -0.575.This indicates that adjustment to long run equilibrium is faster in Nigeria. This could be attributed to betteragricultural infrastructure in Nigeria compared to Ghana. The error correction for Indian rice in the ricezone as estimated by Mohammed (2005) was -0.415. This could be attributed to the fact that Indianagriculture was highly constrained by land problems hence leaving room for little adjustments in terms ofincreasing acreage cultivated.The aggregate output of rice in the short run was found to be dependent on the acreage cultivated, the realprices of rice, rainfall and previous output with elasticities of 0.018, 0.01, 0.004 and 0.52 respectively. Realprice of rice and area cultivated are significant 10% level of significance while rainfall and lagged outputare significant 5%. In the long run aggregate output was found to be dependent on acreage cultivated andthe real price of rice and real price maize with elasticities of 0.218, 0.242 and -0.01 respectively at the 1%significance level.Mythili (2008) used panel data to estimate the supply response of Indian farmers. His findings also supportthe view that farmers’ response to price is low in the short-run and their adjustment to reaching desiredlevels is low for food grains.The analysis showed that short-run response in rice production is lower than long-run response as indicatedby the higher long-run elasticities. This is because in the short run the farmers are constrained by the lack ofresources needed to respond appropriately to incentives. In the short-run inputs such as land, labour, andcapital are fixed. To address these concerns government should devise policies to make land available tofarmers so that prospective farmers could increase acreage cultivated. More irrigation facilities should beconstructed to put more land under cultivation.It was also observed that the acreage model had higher elasticities than the output model. Thus farmers tendto increase acreage cultivated in response to incentives. This implies that farmers have more control overland than the other factors that influence output.Efforts should be put in place to make the acquisition of inputs such as tractors and fertilizer moreaccessible and affordable to farmers, and to improve the road network linking farming communities and theurban centres.Price control policy should be introduced and enforced to address the problem of frequent price fluctuationwhich is the main reason for the low response to prices. Since farmers are aware of these price fluctuationsthey are reluctant to immediately respond positively to price rises.Though there is a market for rice in Ghana, recent developments have shown that consumers prefer foreignpolished rice; therefore, government should put in place the needed infrastructure to process the locallyproduced rice to ensure the sustainability of local rice production.ReferencesBox G. E. P. and Jenkins G. M. (1976). ‘Time Series Analysis: Forecasting and Control’, 2nd ed, SanFrancisco, Holden-day. Cambridge University Press.Chinyere G. O. (2009). ‘Rice output supply response to changes in real prices in Nigeria; An AutoregressiveDistributed Lag Model approach’. Journal of Sustainable Development in Africa, 11(4), 83 -100. ClarionUniversity of Pennsylvania, Clarion, Pennsylvania.Davidson, J. E. H., Handry, D. F., Srba F., and Yeo S. (1978). ‘Econometric modeling of aggregate timeseries relationships between consumer’s expenditures and income in the UK’, Economic Journal. 88,661-692. 12
  13. 13. Journal of Economics and Sustainable Development www.iiste.orgISSN 2222-1700 (Paper) ISSN 2222-2855 (Online)Vol.2, No.6, 2011Dickey D. A. and Fuller W. A, (1981). ‘Likelihood ratio statistics for autoregressive time series with a unitroot’. Econometrica, 49, 1057-1072.Engle R. F, and Granger C. W. J. (1987). ‘Cointegration and error correction: Representation, estimationand testing’, Econometrica, 55, 251 -276.FAO (1996). ‘Report of expert consultation on the technological evolution and impact for sustainable riceproduction in Asia and Pacific’. FAO-RAP Publication No. 1997/23, 206 pp. Regional Office for the Asiaand the Pacific.Granger C. W. J. (1981). ‘Some properties of time series data and their use in econometric modelspecification.’ Journal of Econometrics, 16 (1), 121-130.Granger C. W. J. and Newbold P. (1974). ‘Spurious Regression in Econometrics’, Journal of Econometrics2, 111-120.Ghana Statistical Service (2011). ‘Ghana Statistical Service release’.Hall S. G. (1991). ‘The effect of varying length VAR on the maximum likelihood estimates ofco-integrating vectors.’ Scottish Journal of Political Economy, 38, 317- 323.Hallam D. and Zanoli R. (1993). ‘Error Correction Models and Agricultural Supply Response’. EuropeanReview of Agricultural Economics, 20, 151-166.Hansen S. and Juselius K. (1995). ‘Cats in rats: Cointegration analysis of time series, Evanston, Illinois.Harris R, (1995). Using cointgration analysis in econometric modeling. Prentice Hall, HarvesterWheatsheaf, England.Johansen S. and Juselius K. (1990). ‘Maximum likelihood estimation and inference on cointegration- withapplication to the demand for money’. Oxford bulletin of Economics and statistics, 52, 170-209.Kherallah M C., Delgado E., Gabri-Madhin N. M., and Johnson M., (2000). ‘The Road Half Travelled:Agricultural Market Reform in Sub-Saharan Africa’. International Food Policy Research Institute, FoodPolicy Report 10. Washington D.C. Kiel Working Paper No. 1016.Krueger A. O., Schiff M, and Valdés A. (1992). ‘The Political Economy of Agricultural Pricing Policies’. AWorld Bank Comparative Study, Baltimore.Lewis W. A, (1954). ‘Economic Development with Unlimited Supplies of Labour’. Manchester School 22:139–191.Ministry of Food and Agriculture, Ghana (2005), Ministry of Food and Agriculture report.Mohammed S. (2005). ‘Supply response analysis of major crops in different agro-ecological zones inPunjab.’ Pakistan Journal of Agricultural Research, 2007, 124 (54).Mythili G. (2008). ‘Acreage and yield response for major crops in the pre – and post reform periods in India:A dynamic panel data approach’, Report prepared for IGIDR-ERS/USDA project: Agricultural markets andpolicy, Indira Gandhi Institute of Development Research, Mumbai. 13
  14. 14. Journal of Economics and Sustainable Development www.iiste.orgISSN 2222-1700 (Paper) ISSN 2222-2855 (Online)Vol.2, No.6, 2011Nerlove M. and Bachman K L., (1960). ‘Analysis of the changes in agricultural supply, Problems andApproaches’. Journal of farm economics. 3(XLII), 531-554.National Rice Policy Development (2009). National Rice Policy Development Report.Statistical Research and Information Division (SRID). (2005). Statistics Research and InformationDepartment Release.Thiele R., (2000). ‘Estimating the aggregate supply response, a survey of techniques and results fordeveloping countries.’ Kiel Institute of world Economics Duuesternbrooker Weg 120 241015 Kiel, Germany.Kiel working paper No 1016 Rainer Thiele.Thiele R, and Wiebelt M, (2000). ‘Adjustment lending in Sub Saharan Africa Kiel Institute of WorldEconomics.’ Kiel Discussion Papers 357, Kiel. Tobacco In Zimbabwe Contributed Paper Presented at the41st Annual.Townsend R and Thirtle C. (1994). ‘Dynamic acreage response. An error correction model for maize andtobacco in Zimbabwe’. Occasional paper number 7 of the International Association AgriculturalEconomics.Townsend T. P. (2001). "World Cotton Market Conditions". Beltwide Cotton Conferences, Proceedings,Cotton Economics and Marketing Conference, National Cotton Council, Memphis, TN, pp. 401-405.World Bank. (1997). "Cotton and Textile Industries: Reforming to Compete." Vol. I, II, Rural DevelopmentSector Unit; South Asian Region, World Bank. Report No. 16347- IN, Washington, D.C, 1997. 14
  15. 15. Journal of Economics and Sustainable Development www.iiste.orgISSN 2222-1700 (Paper) ISSN 2222-2855 (Online)Vol.2, No.6, 2011Sustainable Use of Water Resources in the Form of Pisciculture to Generate Income in West Bengal -A study report. Bairagya Ramsundar Department of Economics, SambhuNath College, Labpur, Birbhum, West Bengal, India, Pin-731303 Mobile No. +919474308362 Fax: +913463 266255, E-mail: ramsundarbairagya@gmail.comReceived: October 1st, 2011Accepted: October 11th, 2011Published: October 30th, 2011AbstractThough the two-third of earth’s volume is comprised of water the world is facing the problem of scarcity offresh- water. Because most of the water is sea water which is salt in nature. It is the modern day tragedy thatdue to the scarcity of these valuable lives saving natural resources life will exist no more. So the sustainableuse of this resource is very much important today. India was blessed with vast inland natural waterresources. But Indian Economy faces the problem of proper utilization of these huge water resourcesspread over its vast stretches of land. A proper policy for utilization of these resources would become agoverning direction of economic growth. The role of fisheries in the country’s economic development isamply evident. It generates employment, reduces poverty, generates income, increases food supply andmaintains ecological balance between flora and fauna. Scientists have shown that 3 bighas forest areas areequivalent to 1 bigha plank origin in water bodies which create more O2. This paper intends to develop ascientific plan use of water with a view to sustainable management of water resources to generate incomefrom fishery in the field of pisciculture by using the scarce water resources in a sustainable eco-friendlymanner.Keywords: Ecology, Growth, Perishable, Pisciculture, Pollution, Project, Sustainable development,Composite fish culture1. Introduction According to 2001 Census India is the second (next to China) largest populous country in the worldabout 17.5% of world populations live in India which covers only 2.4% (Dutt R. and Sundharam K.P.M.2009) of geographical land area. The density of population is 324 persons per sq. km. If the current growthrate of population (i.e. 2.11% per annum) continues within 2030 India will be the most populous country inthe world creating a food scarcity and a huge amount of unemployment in a massive scale. About 68% oftotal population spends their livelihood from agriculture and the per capita income is very low. Waterscarcity is too much related to food scarcity. Hence it is very crucial to develop a scientific plan of wateruse with a view to sustainable management of water resource. Management of shortage of water andmanagement of water pollution are complex task. These issues have drawn the attention of the developedand developing countries as well as the various national and international organizations including theJohannesburg Summit 2002, Year 2003 has been declared as year of Fresh -Water. But in India there arehuge prospect to utilize the unused fresh water resources for pisciculture which can play an important rolein this aspect. Realizing its importance during the 5th - five year plan the Government of India introduced 15
  16. 16. Journal of Economics and Sustainable Development www.iiste.orgISSN 2222-1700 (Paper) ISSN 2222-2855 (Online)Vol.2, No.6, 2011beneficiary-oriented programme in the form of a pilot project entitled ‘Fish Farmers Development Agency’(FFDA) to provide self employment, financial, technical and extension support to fish farming in ruralareas. In 1974-75 this Programme was further extended under World Bank-assisted, Inland fisheries projectto cover about 200 districts of various states in India. 2. Sustainable development and Pisciculture Sustainable development is a pattern of resource use that aims to meet human needs while preservingthe environment so that these needs can be met not only in the present, but also for future generations. Theterm was used by the Brundtland Commission which coined what has become the most often-quoteddefinition of sustainable development as development that “meets the needs of the present withoutcompromising the ability of future generations to meet their own needs”. It is usually noted that thisrequires the reconciliation of environmental, social and economic demands - the “three pillars” ofsustainability. This view has been expressed as an illustration using three overlapping ellipses indicatingthat the three pillars of sustainability are not mutually exclusive and can be mutuallyreinforcing. Sustainable development ties together concern for the carrying capacity of natural systems withthe social challenges facing humanity. As early as the 1970s “sustainability” was employed to describe aneconomy in equilibrium with basic ecological support systems [Wikipedia]. A primary goal of sustainabledevelopment is to achieve a reasonable and equitable distributed level of economic well being that can beperpetuated continually for next generation. Thus the field of sustainable development can be broken intothree constituent parts i.e. environmental, economic and social sustainability. It is proved that socio-economic sustainability is depended on environmental sustainability because the socio- economic aspects,like agriculture, transport, settlement, and other demographic factors are born and raised up in theenvironmental system. All the environmental set up is depended on a piece of land where it exists. Water is a renewable natural resource and is a free gift of nature. In the early days the supplyof water was plenty in relation to its demand and the price of water was very low or even zero but in courseof time the scenario has totally changed (Bairagya R. and Bairagya H. 2011). Water is a prime need forhuman survival and also an essential input for the development of the nation. For sustainable developmentmanagement of water resources is very important today. Due to rapid growth of population, expansion ofindustries, rapid urbanization etc. the demand of water rises many-fold which is unbearable in relation to itsresources in the earth. All the environmental set up is depended on a piece of land where it exists. So, to getsustainable environmental management, sustainable land management is necessary. As a result the use ofwater resources in the form pisciculture in a sustainable manner may gain priority. 4. Importance of the study Social scientists have always identified the rural areas for investigation. In the case of India to a largenumber of studies have been carried out in rural situations including panchayats and co-operative societies.Though many research works have been done in the biological and marine sciences, the economicinvestigations of pisciculture have not yet been done so far. In this respect the present study has a cleareconomic importance for the upliftment of the rural economy at the grass root level. Thus pisciculture has apositive effect on ecology and hence to maintain ecological balance a meaningful use of unused waterresources has an important role. Scarcity of water is a modern day tragedy human society will exist no more 16
  17. 17. Journal of Economics and Sustainable Development www.iiste.orgISSN 2222-1700 (Paper) ISSN 2222-2855 (Online)Vol.2, No.6, 2011due to these scarce resources. So a meaningful use of this resources in the form of pisciculture to generateincome of the rural poor gain topmost priority.The state of West Bengal plays an important role for the implementation of the programme. Though the twodistricts Burdwan and Birbhum of W.B. are primarily agricultural districts, there is huge scope forpisciculture. A few studies have been undertaken by several experts, notably, D. Prasad (1968), R. Charan(1981), A.V. Natarajan (1985), K. M. B. Rahim (1992,93), A Chakravorty (1996), I Guha and R. Neogy(1996), P.K. Ghosh (1998) and others on the economic evaluation of pisciculture. Though these are usefulguides to researchers, yet there is ample scope for further works relating to pisciculture in the rural areas ofW.B. Besides, there is the necessity of developing studies concerning the impact of these programmes onthe rural economy. The present study is a modest attempt in remedying this inadequacy.4.1 ObjectivesThis pilot study was conducted on the following objectives i) To generate employment in rural India. ii) Torise in food supply and reduce mal-nutrition. iii) For proper utilisation of unused water resources. iv) Tomaintain ecological balance in a sustainable manner. V) Identification of fish farmers suitable and willingto develop fish farming in ponds.vi) Arranging training in organized manner and ensuring extensionservices to fish farmers.4.2 Methodology of the study4.2.1 Selection of study areaThe present study was confined to survey the rural areas of Burdwan and Birbhum districts of W. B. inrespect of implementation of the programme of Fish Farmers Development Agency (Mishra S. 1987). Inthe selection of districts following considerations weighed most: i) Both the districts are covered with waterareas constituting half of the total inland water resources. ii) There is a heavy concentration of tanks andponds in both the districts. iii) Both the districts are considered as having one homogeneous agro-climaticzone in view of the broad similarities of soils, climate and other features. Since they are also neighboringdistricts a suitable comparison can easily be made. iv) In both the districts, the FFDA programmes are beingimplemented in full fledged form by the Government authority. v) Data from both the districts can beobtained because of my personal knowledge about the two districts.4.2.2 Selection of sampleKeeping in view the time factor, limited fund and limited ability it was not possible to collect data from allthe recorded fish farmers of the two districts. At first 5 blocks from each district have been purposefullyselected. Then 10 recorded fish farmers from each block have been selected purposefully. Thus 50 fish 17
  18. 18. Journal of Economics and Sustainable Development www.iiste.orgISSN 2222-1700 (Paper) ISSN 2222-2855 (Online)Vol.2, No.6, 2011farmers from each district have been selected for interview. For comparative analysis of data each farmerwas taken as the unit. For this study data have been collected on different aspects of the programme such asfarmers’ income, water area, finance, total production, product price, cost of production, profit, duration oftraining period etc. The data have been collected by personal interview method through a questionnaire.Thus the collected data are entirely primary in nature.4.2.3 Location map of the study areaThe state of West Bengal lies between latitude 21038/ to 27010/ North and longitude 85038/ to 89050/ East.The district Burdwan lies between 22056/ and 25053/ North latitude and 86048/ and 88025/ East longitude.The district Birbhum lies between 23032/30// and 24035/00// North Latitudes and 88001/40// and 87005/25//East Longitudes shown in figure-1.5. Fisheries in IndiaIndian fisheries can broadly be divided into two categories: i) Inland fisheries & ii) Marine fisheries.Further Inland fisheries can be classified into two types: i) Capture & ii) Culture. Capture fisheries consistof rivers, lakes canals etc. where farmers do not cultivate fishes. Natural breeding process is the commonphenomenon there. On the other hand, culture fisheries consist of ponds, tanks, swamps, marshes etc. Inthis case, farmers have to sow fish seeds, nurse it and send it to proper size before harvesting. India is thethird largest producer of fish in the world and the second in inland fish production. Indian fishing resourcescomprising of 2 million sq.km.of EEZ for deep sea fishing, 7,520 km. of coast line, 29,000 km. of Rivers,1.7 million Ha. of reservoirs, 1 million Ha. of brackish water and .8 million Ha. of ponds, lakes and tanksfor inland and marine fish production (Giriappa S. (ed) 1994) . About 14 million fishermen draw theirlivelihood from fishery. During the period 1981 to 2002 the contribution of fishery to GDP has increasedfrom Rs.1230 crores to Rs.32060 crores. The fish production increased from 0.7 million tons in1951 to 6.8million tons in 2006.5.1 Fisheries in West BengalWest Bengal has a vast water resource potentiality. By utilizing these water resources there is a hugeprospect of pisciculture (Chakraborty S. K.1991). These resources can be divided into two categories: i)Inland water resources and ii) Marine water resources. Inland resources constitute ponds, rivers, marshylands, canals, reservoirs etc. Water Resources of West Bengal shown in figure-2.It should be noted that tanks/ponds occupy the major share i.e. 46.70% of total inland water resources. Butout of 2, 76,202 Ha. area under ponds and tanks only 2, 20,000 Ha. i.e. 79.65% are presently used forpisciculture which means 20.35% remains unused. Moreover, out of 5, 91,476.71 Ha. total inland waterresource only 2.87000 Ha. water area is brought under pisciculture i.e. 48.56% are presently used and51.44% remains unused. These unused water resources can be brought under pisciculture through properutilization. 18
  19. 19. Journal of Economics and Sustainable Development www.iiste.orgISSN 2222-1700 (Paper) ISSN 2222-2855 (Online)Vol.2, No.6, 2011In West Bengal marine fishery has a substantial share amounting to a coast line of 158 km. inshore area upto 10 fathoms depth is 770 sq. km. (Mamoria C.B. 1979), offshore area (10-40) fathoms depth is 1813 sq.km. and a continental shelf up to 100 Fathom is 17,049 sq.km. Out of 19 districts of West Bengal only twodistricts East- Midnapur and South 24-Parganas are coastal. West Bengal is on the top of the list in fishproduction in the country. With the passage of time, more and more people are getting themselves involvedin fishery. As fish constitutes the staple food of the people efforts are being made to augment fishproduction.From the period 1986 to 2000, the total fish production increased from 424000 tons to 1045,000 tons (i.e.2.46 times). At the same period the inland fish production increased 2.25 times and marine fish productionincreased 4.50 times. In the year 1986, the share of inland and marine fisheries to total fish production were91% and 9% respectively. But in the year 2000, the share of inland and marine fisheries to total fishproduction are 83% and 17% respectively. Thus we see that during the period 1986 to 2000, the share ofinland fish to total production declined (from 91% in 1986 to 83% in 2000) and the share of marinefisheries increased (from 9% in 1986 to 17% in 2000).Not only in fish production but also in the demand for fish West Bengal is the highest in the country. Thedomestic demand for fish in West Bengal is high because almost all the people of West Bengal arefish-eating. But the state has a higher demand for fish than its production of fish i.e. this state has a deficitin fish supply. To meet this gap the state West Bengal has to import fish from other states like AndhraPradesh, Tamil Nadu etc. At the same time various efforts have been made to augment fish production tobridge this gap.5.2 Fisheries in Burdwan districtThe district of Burdwan is mainly an agricultural district though it is also well advanced in industrialproduction. It is called the ‘granary’ of West Bengal. The district is filled with well fertile productive land.Not only that, the district is also well endowed with a large number of water bodies in the form of rivers,ditches, canals, marshy lands, ponds, tanks etc. The principal rivers are Ajoy, Damodar and Kunur etc.There is a good prospect of pisciculture by utilizing these water resources. The distribution of total waterresources in Burdwan district (i.e. 66480.82 Ha.) is shown in figure-3.Though the district has huge water resources, during the period 1982-1998 only 50% of this water area wasbrought under scientific pisciculture. In the year 1999, the district’s total fish production was 55,500 metrictons while demand in that year was 65,000 metric tons. Thus there was a deficit of 9,500 metric tons. Tomeet this demand the district had to import fish from neighboring states. The district will be self-sufficientin fish production if the unused 50% water resources are brought under pisciculture.5.3 Fisheries in Birbhum district 19
  20. 20. Journal of Economics and Sustainable Development www.iiste.orgISSN 2222-1700 (Paper) ISSN 2222-2855 (Online)Vol.2, No.6, 2011The district of Birbhum is a part of Rarh area. The district is well-drained by a number of rivers and rivulets.The principal rivers of this district are: Ajoy, Brahmani, Dwarka, Kopai, Mayurakshi etc. But most of therivers remain dried up during a greater part of the year. Due to this adverse natural factor pisciculture has notmade any significant progress in this district. Agriculture is the main occupation of the common people ofthis district. The main water areas for pisciculture of this district are rivers, tanks, ponds, khals, Bills, baors,reservoirs etc. The distribution of total water resources in Birbhum district (i.e. 45215.81 Ha.) is shown infigure-4.The figure-4 shows that this district has a large number of water resources. But during the period 1980-1999,only 46% of this water area was brought under pisciculture through FFDA. So there is a huge prospect ofpisciculture by utilizing these unused water resources. In the year 1999 total number of fish farmers in thisdistrict was 2, 00747 out of which 56.5% were males and 43.5% were females. The total number of fishermanfamily in this district was 4,975. The average fish production of this district was 21,000 metric tons and theannual demand was 24,000 metric tons. Thus the district had a deficit (3000 mt) in fish production. To meetits own demand the district has also to import fish from the neighbouring states.6. Findings6.1 Income distribution of sample fish farmers before and after assistance from FFDAIt was observed that 26 farmers in Burdwan district & 32 farmers in Birbhum district have pisciculture asthe main occupation while others have pisciculture as a subsidiary occupation. Their basic occupations areagriculture, business and service and have various types of incomes from those occupations. But as incomegeneration scheme only income from fishery was taken into account i.e. income means income earned fromselling fishes. From sample survey we have collected two sets of income data (one showing income beforeand other showing income after the assistance of FFDA) for each fish farmer. Thus the amount of incomereveals the income of the fish farmer before the receipt of assistance of FFDA and the income after theassistance of FFDA. In this regard the income comparison has done in two areas (namely within thedistrict, between the district and the overall change of income) and‘t’ (Gujarati D.N. 2009) test was used forstatistical analysis. Now the various types of incomes distribution of fish farmers of the two districts areshown in terms of four tables.From table- 1 it is clear in both the districts the average income of most of the fish farmers (i.e. 89 out of100 fish farmers of the two districts taken together) has monthly income before assistance below Rs. 4000i.e. they are basically come from poor family income group i) From the field survey we get that the per capita availability of fish in Burdwan district is 24 kg. per headper year while in Birbhum district it is 23.5 kg. per head per year. Now the world’s per capita availability offish is 11.5 kg. per head per year and for developed countries it is 25 kg. per head per year. But India has alow per capita availability of 3.5 kg. per head per year only. Thus we see that the fishermen have a standardlevel of per capita fish consumption. The point is very important to reduce mal-nutrition (which is acommon phenomenon for the country) by means of intensive pisciculture. But India has a poor per capita 20
  21. 21. Journal of Economics and Sustainable Development www.iiste.orgISSN 2222-1700 (Paper) ISSN 2222-2855 (Online)Vol.2, No.6, 2011availability of fish of only 3.5 kg per head per year. Thus we see that the per capita availability of fish forthese 100 farmers’ families of these two districts are much higher than that of India and world’s averageand even equal to the standard of the developed countries.ii) It was found that the average monthly income of fish farmers before the assistance of FFDA in Burdwandistrict was Rs. 2680 and in Birbhum district it was Rs. 2045. But after the assistance of FFDA the averagemonthly incomes of the fish farmers of the two districts increased to Rs.3935 and Rs.3721 respectively.Thus we see that the average monthly income of fish farmers of Burdwan district has increased 1.47 timesand in case of Birbhum district it has increased 1.8 times. The result indicates that the average income offish farmers of Burdwan district was higher than the average income of those of Birbhum district in respectof both before and after the assistance of FFDA. However, the rate of increment was higher in Birbhumthan that of Burdwan district. Thus we can say that the incomes of the fish farmers have improved due topisciculture. iii) Statistical methods were adopted is to judge whether the change of income of the sample fish farmersbefore and after the assistance of FFDA was significant or not. This has been done in two areas: - a) Changeof income within the district and b) Overall change of income of the two districts.6.2 Change of income within the districtBurdwan district- It is seen that the value of‘t’ = 3.31 is significant at .01 level, meaning thereby that thechange income of sample fish farmers of Burdwan district is significant. The gain is in favor of theassistance of FFDA. It may be concluded from the result that there is a positive impact of assistance on fishfarming (i.e. the production of fish). It may also be concluded that production has increased significantly inBurdwan district.Birbhum district-The result indicates that “t” is significant at .01 level. Hence it may be said that inBirbhum district the change of income of fish farmers is significant. 6.3 Overall change of income taking the two districts togetherIt is seen from the above table that the value of‘t’ is significant at .01 level which indicates that the incomeof fish farmers before and after assistance differ significantly. The result reveals that the mean income after assistance is significantly greater than that before assistance. It may be concluded from the overall results considering both the districts that the impact of assistance on fish production is significant.The finding of these results was that the assistance of FFDA programme has raised the fish farmer’s incomeundoubtedly in both the districts. Not only that the change of income was statistically significant but alsothe result revealed that there was a positive impact of FFDA assistance on fish production (i.e. fish 21
  22. 22. Journal of Economics and Sustainable Development www.iiste.orgISSN 2222-1700 (Paper) ISSN 2222-2855 (Online)Vol.2, No.6, 2011production has significantly increased). Moreover, it was also found that the gain of income of fish farmersof Birbhum district was higher than that of Burdwan district. Thus we can conclude that in case of theimplementation of the programme of FFDA the district Burdwan has more successful achievements thanthe district of Birbhum.7 Suggestions7.1 General suggestionsi) In pisciculture, fishermen are not only directly employed in fishing but also some other alternativeoccupations like net making, marketing of fish seed and fishery product , transport, boat making etc. manyrural people earn income. Since fish is a perishable commodity proper marketing channels should beestablished. Hence to reduce pressure from agriculture pisciculture may be alternative occupations forgenerating income and employment for a large number of poor people. ii) In Burdwan district there aresome open caste pits (OCP) in Ranigang coal belt and in Birbhum district also such pits are available atKhoirasole block, calamines at Md.Bazar block. Fish production may increase by utilizing these waterresources for pisciculture purposes.iii) The urban waste (i.e. garbage) may be recycled as fish feed (to thoseponds and water areas lying near the towns) to raise fish production and to prevent the environmentalpollution in those areas. v) To make financial support for the poor fishermen Bank should grant loan on along-term basis and at a low rate of interest in proper amount and in proper time.vi) the selection of actualbeneficiary is very much essential and it should be made on the basis of need and neutrally not in politically.vii) Since fish is a perishable commodity storage facilities should be provided such that the fishermen arenot forced to sell their product at a lower price. Prices of organic manures and fish seeds should be kept aslow as possible or the Government should give more subsidies in the case of chemical fertilizers so that thepoor fish farmers can buy them. viii) For welfare measures of the rural poor fishermen some Group andPersonal Insurance Schemes, Old- Age Pension Schemes should be taken and the Fishery Dept. shouldissue “Identity Card” to each fisherman. ix) Since both the districts are agricultural based we shouldinterlink agriculture with pisciculture shown in figure-5. Along with pisciculture in ponds other alliedculture can be inter-linked in composite farming. The concept of composite farming e.g. in the pond- thereare pisciculture and on one side of the pond mulberry trees can be cultivated for the development ofsericulture industry-from there silk industry can be grown. Thus the final products (i.e. silk yarn and silkcloth) come to the market and their waste materials are drained off to the pond and used as fish feed. In thesame pattern on another side of the pond animal husbandry can be practiced (e.g. poultry, duccary andpiggery). Their waste can be used as a valuable manure for fish feed and the residual can be utilized foragricultural production. The excreta of the animal husbandry can also be used in the bio-gas plant for fueland light. The products of animal husbandry e.g. milk, meat and egg come to the market directly. Onanother side of the pond some fruit plants such as papine, guava, mango etc. can be cultivated by using theexcess manure of animal husbandry and the products can be sold in the market.x) Composite fish culture: Stocking of various species should be in a certain proportion such that varioustypes of fishes live in various layers and eat the entire food organism (so called Polyculture or CompositeFish Culture). Stocking is an important factor to raise fish production. On an average the survibility ofstocking is 80% (Jhingran V.G. 1991). This means that 20% of stocking is lost for various reasons such asimproper handling, netting and also eaten by preratory animals like snakes, frogs etc. The main speciescultured in India are Rahu, Catla, Mrigel, Silver Carp (S.C.), Grass Carp (G.C.), Cyprinus Carpio (Cy, Ca.) 22
  23. 23. Journal of Economics and Sustainable Development www.iiste.orgISSN 2222-1700 (Paper) ISSN 2222-2855 (Online)Vol.2, No.6, 2011and Bata etc. In polyculture (where various fishes are cultured simultaneously) (Agarwal S.C. 1990) systemall these species are cultured in certain ratios. It is found that all of these species do not generally live in thesame layer. In mixed (or polyculture or composite culture) culture all these fishes are cultured in such a waythat all layer’s feed are used. If we divide the whole layer into 3 parts i.e. upper, middle and lower layers,then Cattla and S.C live in the upper layer, Rahu and G.C. live in the middle layer and Mrigel and Cy. Ca.live in the lower layer. It is also to be noted that these two species in the same layer are not competitors ofeach other. In their feed habit one eats waste weeds and the other eats green weeds. Hence in polyculture allthe pond’s feed are used scientifically and economically. The production of Big head and Big head Africangiant hybrid magur are totally banned because their food habits are too much and hamper the foods of othercommon species. Suppose 1000 fish seed is cultivated in the pond. Then the stocking ratios for variousfishes are Rahu: Catla: Mrigel: S.C.: G.C.: Cy.Car = 3 : 1 : 1.5 : 1.5 : 1 : 2 Thus the number of seeds are,Rahu = 300, Catla = 100, Mrigel = 150, S.C = 150, G.C. = 100, Cy. Ca. = 200. Now the survibility rate is80%.Thus the number of produced matured fishes are Rahu = 240, Catla = 80, Mrigel = 120, S.C. = 120,G.C. = 80, Cy.Car. = 160. Fishes are cultivated for 9-12 months. The culture period may extend up to 18months with partial harvesting between. Moreover, the general annual growth of these fishes are, Rahu =1kg, Catla 1.5 kg. Mrigel 0.75kg, SC. = 0.8 kg, G.C. = 1.5 kg, Cy. Car. = 0.8 kg. The fish farmers collectfish seeds from private traders or from Government fish farm.xi) Since training is a necessary ingredient to raise fish production the Government has initiated somespecial training programmes for fish farmers. Moreover to motivate the poor fish farmers some amount ofremuneration or stipend is paid to the participant during training. The amount of remuneration dependsupon the kind of training, place of training organization and duration of the training programmes conductedby the FFDA. To attend the district level training programmes for the farmers of distant villages, travelingallowances are also paid.xii) It is very essential to use mahua cake because it has dual roles in pisciculture. It firstly acts as apesticide but later it acts as organic manure and produces a desired quantity of fish food organism for thebaby fishes. The presence of ‘Saponin’ in mahua cake kills the pre-ratory animals / fishes of the pond. Thispoisonous effect continues about 10-15 days and after that it acts as manure. On the other hand, the use ofpesticides has a negative long-term effect to ecological balance. So the pesticides are not generally usedthough they are cheaper. The mahua cake should be applied at the rate of 333.33kg/ Yr. / bigha.7.2 Limitations of the studyThe following are the limitations of the study: i) Regarding the change in the level of income before andafter the assistance of FFDA, the statements of the fish farmers have been taken into consideration on goodfaith. ii) Due to limited time, ability and resource constraints, data have been collected from a small numberof fish farmers of the two districts and results are assumed to be the representative for the district as awhole. Iii) Due to difficulties in getting responses from the sample fish farmers sometimes we had to relyon the different FEOs’ opinions and also on different official records on good faith. iv) The observationsmade on the basis of collected data are obviously particularistic in nature in so far as these data relate tomicro-study like the present one (i.e. only 100 fish farmers from the two districts were taken intoconsideration). Micro-studies do not attempt to build general theories but the utility of this type of study isthat large number micro-studies may, in course of time, be helpful in constructing meaningfulgeneralizations. Moreover, it is an explorative study which seeks to explore the conditions of fish farmers 23
  24. 24. Journal of Economics and Sustainable Development www.iiste.orgISSN 2222-1700 (Paper) ISSN 2222-2855 (Online)Vol.2, No.6, 2011before and after the assistance of FFDA programme. A much larger study may be undertaken to vindicatethe results obtained from this explorative study.8. ConclusionThe study indicates that the introduction of the programme of FFDA had a clear positive impact on the ruraleconomy through employment and income generation and also raising the standard of living andsocio-economic performances of the rural community of the two districts. It is environmentally viable,sustainable and eco-friendly in nature. So for sustainable use of the scarce water resources it is veryessential in the present context. Unemployment is a curse for the society today and pressure of populationon agricultural sector is rising day by day. Thus fish farming may be an alternative occupation for thoseunemployed people and undoubtedly generate income in a viable method. Therefore it is recommended thatthe present programme should be further spread in the rural areas by means of proper planning, adequatesupervision, effective implementation and better monitoring in a sustainable manner.References:Agarwal S.C. (1990), Fishery Management, Ashis Publishing House, New-Delhi. Bairagya R. and Bairagya H. (2011), “Water Scarcity a Global Problem- An Economic Analysis”,Indian Journal of Landscape Systems and Ecological Studies, Vol.-34, Institute of Landscape, Ecology &Ekistics, Kolkata, 127-132 Bairagya R. (2004), “A Comparative Study of the Functioning of Fish Farmers DevelopmentAgency (FFDA) in the Districts of Burdwan and Birbhum (1985-1995)”, PhD Thesis, Department ofCommerce, the University of Burdwan. Bureau of Applied Economics, Burdwan (1998), Key Statistics of Burdwan, Govt. of W.B.Bureau of Applied Economics, Birbhum (1998), Key Statistics of Birbhum, Govt. of W.BChakraborty S. K. (1991), A Study in Bhery Fisheries, Agro-Economic Reaserch Centre, Visva- Bharati.Dutt R. and Sundharam K.P.M. (2009), Indian Economy, S. Chand and Co. New-Delhi, 101-103Jhingran V.G. (1991), Fish and Fisheries in India, Hindustan Publishing Cor.-Delhi.Giriappa S. (ed) (1994), Role of Fisheries in Rural Development, Daya Publishing House, Delhi. Gujarati D.N. (2009), Basic Econometrics, McGraw-Hill., 98-133Mamoria C.B. (1979), Economic & Commercial Geography of India, Shiblal Agarwala Company, Agra-3Mishra S. (1987), Fisheries in India, Ashis Publishing House, New-DelhiAcknowledgement: In preparation of this paper I deeply acknowledge my respected sir Prof. JaydebSarkhel, Department of Commerce, Burdwan University, West Bengal, India, e-mail: 24
  25. 25. Journal of Economics and Sustainable Development www.iiste.orgISSN 2222-1700 (Paper) ISSN 2222-2855 (Online)Vol.2, No.6, 2011jaydebsarkhel@gmail.com Tables and Figures: Figure-1 25
  26. 26. Journal of Economics and Sustainable Development www.iiste.orgISSN 2222-1700 (Paper) ISSN 2222-2855 (Online)Vol.2, No.6, 2011 26
  27. 27. Journal of Economics and Sustainable Development www.iiste.orgISSN 2222-1700 (Paper) ISSN 2222-2855 (Online)Vol.2, No.6, 2011 Figure-2Figure-3 Figure-4Table -1 Change of income distribution of Burdwan. 27
  28. 28. Journal of Economics and Sustainable Development www.iiste.orgISSN 2222-1700 (Paper) ISSN 2222-2855 (Online)Vol.2, No.6, 2011 Table -2 Change of income distribution of Birbhum districtTable--3 Classification of sample fish farmers according to their monthly incomes after the assistance 28
  29. 29. Journal of Economics and Sustainable Development www.iiste.orgISSN 2222-1700 (Paper) ISSN 2222-2855 (Online)Vol.2, No.6, 2011 Table- 4 Classification of sample fish farmers according to their monthly incomes after the assistance Source: All tables data sources are computed from field survey. 29

×