nuhfil.lecture.ub.ac.idnuhfil.lecture.ub.ac.id/files/2012/12/FINALseptember2012...PUBLISHER Islamic...

Vol 2(3), September 2012

ISSN: 2159-5852 (Print)

ISSN:2159-5860 (Online)

PUBLISHER

Islamic Azad University, Rasht Branch, Iran.

Editor-in-Chief Dr. Mohammad Sadegh Allahyari, Ph.DDepartment of Agricultural Management,Islamic Azad University, Rasht Branch, Rasht, [email protected]@iaurasht.ac.ir

Executive Manager Dr. Hamidreza Alipour, IranIslamic Azad University, Rasht, Iran.

Editorial Board Prof. Mohammad Chizari, IranProf. Saeed Yazdani, IranProf. Ali Nourieve, AzerbaijanAssoci. Prof. R. Saravanan, IndiaProf. Hanho Kim, South KoreaAssoci. Prof. Arvind Kumar, IndiaAssoci. Prof. Nasrolah Molaee, IranProf. Ahmad S. Al-Rimawi, JordanDr. Rico Lie, NetherlandsAssoci. Prof. Murat Boyaci, TurkeyDr. Lesli D. Edgar, USADr. Mostafa Karbasioun, IranDr. Ahmadreza Ommani, IranDr. Jafar Azizi, IranDr. Nav Ghimire, USA

Assistant EditorZahra [email protected]

Abstracting/IndexingEBSCO, Agricola, CABI, AgEcon search, Ulrich's, Cabell's Directory, DOAJ, Google scholar, IndexCopernicus, Islamic World Science Citation (ISC), Scientific Information Database (SID), Open-J-Gate, Electronic Journl Library, Electronics Journal Database, Scholar, Magiran.

This journal is published in cooperation with

Iranian Association of Agricultural Economic

Analysis of the Effects of Agricultural Inputs Price Liberalization on the Production of Sunflower

in Khoy Zone.............................................................................................................................149-155

Ali Bagherzadeh and Fatemeh kazemzadeh

The Impact of Bio-Ethanol Conversion and Global Climate Change on Corn Economic

Performanve of Indonesia...............................................................................................................157-165

Yudi Ferrianta, Nuhfil Hanani, Budi Setiawan and Wahib Muhaimin

Determinants of Repayment of Loan Beneficiaries of Micro Finance Institutions in Southeast States

of Nigeria ....................................................................................................................................167-175

Stephen Umamuefula Osuji Onyeagocha, Sunday Angus Nnachebe Dixie Chidebelu and Eugene Chukwuemeka

Okorji

Analysis of Technical Efficiency of Smallholder Cocoa Farmers in Cross River State,

Nigeria......................................................................................................................................177-185

Agom Damian Ila, Susan Ben Ohen, Kingsley Okoi Itam and Nyambi N. Inyang

Comparative Cost Structure and Yield Performance Anzlysis of Upland and Mangrove Fish

Farms in Southwest, Nigeria..................................................................................................187-198

Mafimisebi Taiwo Ejiola and Okunmadewa Foluso Yinka

Effectiveness of Extension Services in Enhancing Outgrowers’ Credit System: A Case of Smallholder

Sugarcane Farmers in Kisumu County, Kenya.......................................................................199-207

Gilbert C. Abura Odilla, Hillary Kipngeno Barchok and Christopher Onyango

Applying CVM for Economic Valuation of Drinking Water in Iran...........................209-214

Morteza Tahami Pour and Mohammad Kavoosi Kalashami

Perceptions of Constraints Affecting Adoption of Women-in-Agriculture Programme Technologies

in the Niger Delta, Nigeria......................................................................................................215-222

Iniobong A. Akpabio, Nsikak-Abasi A. Etim and Sunday Okon

A Logistic Regression Analysis: Agro-Technical Factors Impressible from Fish Farming in

Rice Fields, North of Iran......................................................................................................223-227

Seyyed Ali Noorhosseini-Niyaki, Mohammad Sadegh Allahyari

PageContent

www.ijamad.com

Inte

rnat

iona

l Jo

urna

l of

Agr

icul

tura

l M

anag

emen

t &

Dev

elop

men

t, 2

(3):

149

-155

, Sep

tem

ber,

201

2.

149

Analysis of the Effects of Agricultural Inputs Price

Liberalization on the Production of Sunflower in

Khoy Zone

Ali Bagherzadeh1* and Fatemeh kazemzadeh2

Keywords: Sunflower, Liberalization,Production factor, Productionelasticity, Khoy

Received: 29 December 2011,Accepted: 18 March 2012 Sunflower is one of four main annual oil plants that cultivated

in oil and nut varieties. This plant as an important and in-dustrial food product and because of nutritional features andthe potential for earning exchange has become a valuableproduct in foreign and inner markets and has a special positionin agricultural sector. Khoy, by producing 40 percent ofsunflower productions of country annually, is the greatest sun-flower producer in Iran. The main purpose of this study is theanalysis of the effects of inputs price liberalization on productionof sunflower producers in this city. This study is according toa field research and cross-sectional data of 2009 have beenused for it. Results show input price liberalization policy byincreasing inputs prices and decreasing demand amounts ofinputs, increases the production costs and decreases the pro-duction and totally it`s harmful for sunflower producers. Forpreventing negative effects of liberalization on production,adopting necessary policies such as merging small farms andmaking big ones to profit by economies of scale and increasingproduction and productivity with the resulted incomes fromliberalization and spending them in scientific researches toproduce with low costs are suggested.

Abstract

International Journal of Agricultural Management & Development (IJAMAD)Available online on: www.ijamad.comISSN: 2159-5852 (Print)ISSN:2159-5860 (Online)

1 Assistant professor of Economics, Khoy Branch, Islamic Azad University, Iran. 2 MA in Economics, Tabriz Branch, Islamic Azad University, Iran. * Corresponding author’s email: [email protected]

Inte

rnat

iona

l Jo

urna

l of

Agr

icul

tura

l M

anag

emen

t &

Dev

elop

men

t, 2

(3):

149

-155

, Sep

tem

ber,

201

2.

150

INTRODUCTION

Oil seeds as industrial plants are one of mainand strategic products of agricultural sector thatare cultivated for providing eatable oil. The im-portance and value of oil seeds are not only fortheir oil but also for valuable material that isconsumed for nutrition after oil-pressing. Sun-flower is one of four main annual oil plants thatcultivated in oil and nut varieties. Nut varietieshave special oil acids and less oil but oil varietieshave 43-49 % oil that is full of D and E vitaminsand has an important role in human health.Sunflower oil is important in clearing vesselsand preventing brain and heart strokes. Also, itis used to make medicine, soap, colors and cos-metic materials. Because of desirable quality ofoil and desirable reaction in unsuitable envi-ronment conditions, sunflower has a special po-sition in agricultural sector and can be effectivein economies of most countries.

Khoy is the greatest sunflower producer inIran that produces 40% of sunflower productionin country. Moderate weather, susceptible land,suitable market, having high price in comparisonwith other agricultural products, being precociousand possibility for second cultivation are themost important reasons of sunflower cultivationin this area. Nut varieties of Khoy are rare incolor, taste and size and are competitive withother countries products. The products are ex-ported to Persian Gulf countries and if exportway becomes paved, thousands of tons of thisproduction can be exported to other countries.

As we know, agriculture is the economic heartof most countries and most likely source of sig-nificant economic growth (DFID, 2003). It hasbeen observed as the major and certain path toeconomic growth and sustainability. In spite ofthe dominant role of the petroleum sector as themajor foreign exchange earner, agriculture re-mains the mainstay of the economy (NEEDS,2004) as the economic of most developing coun-tries are built on agriculture. There is strong re-lationship between agricultural productivitygrowth and reduction of poverty. Sunflower inagricultural sector of Iran as an important andindustrial food product and because of nutritionalfeatures and the potential for earning exchange

has become a valuable product in foreign andinner markets. Many governments intervene,directly in producing agricultural productsthrough taxations and subsidization, so inputsprice are not real. For increasing the ability ofeconomy sector, it`s necessary to use clear andcompetitive prices of inputs to have economicefficiency. Liberalization policy of agriculturalsector as a way of development has beenproposed to developing countries by WorldBank. Liberalization simply means allowingmarket forces of demand and supply to determinewhat to provide, for whom to produce, and themethod of production to be used in an economy.Liberalization involves deregulation and the re-moval or reduction of government`s participationin the economy. Their justifying reasons forthis policy are environment protection, decreasinggovernment costs, increasing inputs productivityand stable development of agricultural sector.Since this policy has been enforced in ourcountry since some years ago and now continues,there are some concerns about loss of this sector.The analysis of input price liberalization effectson production is too important because enforcingprice liberalization policy by increasing inputsprices and decreasing their consumption affectsproduction and production costs and if these ef-fects are negative, they can make some problemsthat threat economic and political security ofcountry.

Previous studies on input consumption in agri-cultural sector and subsidies effects show thatsubsidy elimination has been one of government`simportant economic policies in recent decades.In most countries there were such experiencesthat we imply some of them.

Gulati (1990) in a study on agricultural inputprice liberalization claimed according to highand increasing costs of agricultural input subsidies,eliminating these subsidies is necessary but ithas some reactions on agricultural sector thatneeds to more study and analysis. Ready andDeshpande (1992) by analyzing the effects offertilizer price liberalization in India showedthat this policy as an agricultural developmenttool has both positive and negative effects.Elyasian and Hosseini (1996) in a research on

Effects of Agricultural Inputs Price Liberalization / Ali Bagherzadeh et al

Inte

rnat

iona

l Jo

urna

l of

Agr

icul

tura

l M

anag

emen

t &

Dev

elop

men

t, 2

(3):

149

-155

, Sep

tem

ber,

201

2.

151

the analysis of effects of agricultural inputssubsidy elimination showed in case of wheatprofitability after liberalization is twice as muchas before liberalization. In another researchAzizi (2005) studied the price liberalizationeffects of poison and fertilizer inputs on rice inGuilan. Results showed that fertilizer was usingin second area –economic area- and price liber-alization led to increase in price, decrease infertilizer consumption and so decrease in pro-duction, but poison was using in third area,price increase led to consume in second area.Then by comparing the positive and negativeeffects of liberalization, continuing the policywas suggested for this input. In another studyKarimzadegan, (2006) presented eliminatingfertilizer subsidy decreases it`s consumptionfor wheat and returning to optimal input con-sumption increases their production and profit.Ayinde et al., (2009) by studying the effects offertilizer policy on production in Nigeria in twoperiods (before liberalization and after liberal-ization) showed despite fertilizer price increase,liberalization leads to production increase. Inthis study for having more production, controllinginputs prices and educating farmers are suggested.Badmus (2010) by doing the same research oncorn production in Nigeria by using of SURmethod on time series of 1970-1998 claimedliberalization is ineffective on fertilizer priceand consumption but has positive effects onproduction. Mousavi et al., (2010) studied thewelfare effects of eliminating fertilizer subsidyon corn production in Fars. Results showed lib-eralization despite price increase did not affectfertilizer demand and so it caused high productioncosts and low profit.

The main objectives of this study are theanalysis of input price liberalization effects onsunflower production and it`s production costs.So according to our objectives, we try to deriveproduction function, cost function, productionelasticity of inputs and the comparative impor-tance of inputs to show whether the farmers be-have logical in applying inputs or not. Then bymeans of inputs demand functions and inputsdemand price elasticity, consumption and pro-duction changes are identified.

MATERIALS AND METHODS

In developing countries including Iran, betteruse of agricultural inputs like land, fertilizer,poison, water and so on, for increasing the pro-duction and development of agricultural sectorhas special importance and there are severaltools to achieve them. One of the most importanttools for choosing suitable approaches in pro-duction and optimal allocation of sources isusing production functions. In fact, productionfunctions are one of analysis ways of quantitativerelations between the amounts of inputs andproduction operation. These functions are math-ematical relations that identify input conversionrate to output. So we have:

y = f (x1,x2, ... , xn) (1)Y is the amount of production and xis are pro-

duction inputs. Agricultural production function is:

y=f(x)=Ax1α x2β , α,β>0 (2)

Total cost function is:TC = w1x1+w2x2 (3)Lagrange function is:

(4)

(5)

For minimizing agricultural sector costs, wemake derivation toward each production inputand put them equal to zero.

(6)

(7)

w1 x1+w2x2=λ (α+β)f(x) (8)

Then we give joint powers to the both sidesof first condition equations, find the amount of

𝝀 and put in cost equation as follows:

(9)

(10)

(11)


Inte

rnat

iona

l Jo

urna

l of

Agr

icul

tura

l M

anag

emen

t &

Dev

elop

men

t, 2

(3):

149

-155

, Sep

tem

ber,

201

2.

152

(12)

(13)

lnC = ln(α+β)+lnB - (1/(α+β) lnA + (1/(α+β)lny + (α/(α+β) ln w1 + (β/(α+β) ln w2 (14)

Coefficients of each parameter in above equa-tion are cost elasticities of inputs. Also we havethis result as scale elasticity:

(15)

Then we find demand functions of each in-put:

(16)

(17) This study is according to a field research and

cross-section data of 2009 have been used forit. We spent 2 months for collecting data. Atfirst we took the names of all sunflower producersof Khoy that were 5000 units then we classifiedthe statistical universe of producers to 3 classesaccording to the size of cultivated land, (0-4]hectares, (4-8] hectares and (8, more) hectares,that 60% of lands were between (0.25-4) hectares,30% of lands were between (4-8) hectares andless than 9% were more than 8 hectares. Afterclassification, by using of classified randomsampling, we could get optimal sample volumeby:

(18)

So for every class by random method wechose a volume of nh=n.Nn /N that n is optimalamount of sample for statistical universe. Sinceour method was random, for identifying optimaln, we should know the sum of all classes isequal to the optimal amount of sample of statis-tical universe. We know for collecting samples

of n>30, mean sampling distribution is normal.So:

(19)

(20)

In above equations, d is the maximum amountof authorized error, σx

2, the variance of statisticaluniverse and z, standard normal distribution byα significance level. For calculating variance,we used, σ =(Xmax-Xmin)/ε. The variance ofstatistic universe was 400, so σx=20. Then wecalculated the optimal amount of each class.

(21)

After having total optimal number and optimalnumber of sample for each class, by using ofrandom numbers table, we took the names ofsample sunflower producers and finally foundessential data for study.

RESULTS AND DISCUSSIONS

Empirical studies and sampling methods inproducing sunflower seeds show factors likeseed, labor, fertilizer and watering times are themain factors of production function in thissector. Applied production function in this studyis Cobb-Douglas. Cobb-Douglas function hasbetter goodness of fit toward other agriculturalproduction functions. So this function was usedfor estimating the relation between sunflowerproduction and inputs. Sunflower productionfunction that has been estimated by ordinaryleast squares method is presented as:

lny=-4.092+0.785 lns + 0.365 lnl + 0.751 lnw+ 0.139 lnf (22)

t- statistics: (-4.25) (2.9) (1.58) (2.67) (1.85)The numbers inside parentheses are t- ratios

that show all variables of production functionhave significance in 95% confidence level.R2=75% and Ṝ2=69% that according to cross-section data are acceptable amounts. Also F-statistic related to analysis of variance is 12.256


Inte

rnat

iona

l Jo

urna

l of

Agr

icul

tura

l M

anag

emen

t &

Dev

elop

men

t, 2

(3):

149

-155

, Sep

tem

ber,

201

2.

153

that presents the significance of total regressionin 95% confidence level. Also according to theresults of White test, there is no problem of het-eroscedasticity and not having correlation areadmitted by (D.W) statistics. Following tablepresents the results of estimation by Microfit4.1.

As we know estimated coefficients of equationare inputs production elasticities. According tocoefficients, sensitivity of production towardwatering times and seed in comparison withothers are more and production has the leastsensitivity toward fertilizer. All production elas-ticities are between zero and one that showinputs are used in second area-economic area-that is true about Cobb-Douglas functions. Pro-duction elasticity shows the sensitivity of pro-duction toward inputs demand. Here seed andwater have more sensitivity toward their demandchange and fertilizer has the least sensitivity.

In Cobb-Douglas production function, the sumof all production elasticities is the amount ofscale elasticity. We have:

E=Es+El+Ew+Ef (23)So scale elasticity is 2.04 that is more than

one and shows increasing returns to scale. In-creasing returns to scale happens in decreasingpart of long term average cost curve witheconomies of scale. On the other hand, in thisamount, firm can decrease it`s average cost by

increasing it`s production.For studying price elasticities and cost elas-

ticities, we need cost function. According toduality rule, we derive cost function as follows:lnC = 1.91 + 0.49 ln y + 0.385 ln w1 +0.179 lnw2 +0.368 ln w3 + 0.068 ln w4 (24)

wis are the price of each inputs. Now we canget demand equations.x1 (y, w1, w2, w3, w4)=1.31. 0.017-0.49 y 0.49 w1 -0.615

w20.179 w30.368 w40.068 (25)

x2 (y, w1, w2, w3, w4)=0.61. 0.017-0.49 y0.49 w10.385

w2-0.821 w30.368 w40.068 (26)

x3 (y,w1,w2,w3,w4)=1.25.0.017-0.49 y0.49 w10.385 w20.179

w3-0.632 w40.068 (27)

x4 (y,w1,w2,w3,w4)=2.230.017-0.49 y0.49 w10.385 w20.179

w30.368 w4-0.932 (28)

If we write logarithmic form of demand equa-tions, gained coefficients of inputs prices showinputs price elasticities. Following table presentsprice elasticity amounts of sunflower inputs.

Above table shows price elasticities are com-pletely according to demand rule. All inputsprice elasticities are negative. Seed has the leastand fertilizer has the most amount of elasticity.1% increase in seed price lead to 0.615%decrease in seed demand also if fertilizer priceincreases 1%, it`s demand will decrease 0.932%.Elasticity of fertilizer demand is about one andprice changes have more effect on it`s demand.Seed demand because of low severity towardprice changes takes less effect. Also this matteris true about other inputs.

Now, if price liberalization policy is enforced,the price of seed and fertilizer will increase.According to demand price elasticity of inputs,farmers demand for these inputs will decrease.


Regressor coefficient t-statistic

ConstantLnsLnlLnwLnfR-squaredR-Bar-squaredF-statisticDW-statistic

-4.0920.7850.3650.7510.139

-4.252.9

1.582.671.850.750.69

12.2561.8

Table 1: The results of estimation by Microfit 4.1

Table 2: price elasticity and intersecting elasticity of sunflower inputs

Inputs Seed Labor Water Fertilizer

Seed

Labor

Water

Fertilizer

-0.615

0.385

0.385

0.385

0.179

-0.821

0.179

0.179

0.368

0.368

-0.632

0.368

0.068

0.068

0.068

-0.932

Inte

rnat

iona

l Jo

urna

l of

Agr

icul

tura

l M

anag

emen

t &

Dev

elop

men

t, 2

(3):

149

-155

, Sep

tem

ber,

201

2.

154

On the other hand, with these amounts of pro-duction elasticities for inputs, by decreasingtheir demand, the amount of production willdecrease. Of course by using of input substitution,we can take the production level at the sameamount. So input price liberalization policy byincreasing the price of inputs, increases thecosts of production and decreases the productionand totally it`s harmful for sunflower producers.

CONCLUSION AND RECOMMENDATION

Study results show that input price liberalizationpolicy has negative effect on sunflower produc-tion. According to gained results, all inputs areused in second area of production –economicarea- so liberalization policy by increasing inputsprices, according to demand rule, decrease thedemand amounts of inputs and decrease ininputs consumption cause production decrease.

Gained price elasticity is about one for fertilizerthat shows 1% increase in fertilizer price causes1% decrease in input demand amount. Decreasein it`s consumption lead to production decreasethat is harmful for sunflower producers. Accordingto demand inelasticity of seed, sunflower pro-ducers reaction toward price increase won`t betoo severe but it will have demand decrease ofthis input and finally according to seed productionelasticity, the amount of production will decrease.In conclusion, the effect of inputs price liberal-ization policy on sunflower industry, is increasein production costs and decrease in sunflowerproduction. So for enforcing policy, we needadopting exact and planned policies. Accordingto gained results, following suggestions are pre-sented:

Agricultural organizations should have truesupervision on the amount, time and the way ofinput consumption and help farmers to increasetheir scientific informing to use inputs optimallyand therefore increase their production and ef-ficiency.

Government can compensate extra costs offarmers resulting from liberalization policy bygiving cash subsidies. These subsidies could bepaid based on the amount of production or cul-tivated land. In this way farmers distribute thecash subsidy among all inputs and decrease

severe use of one input.For preventing negative effects of liberalization

on production by merging small farms andmaking big ones, we can benefit of economiesof scale and increase our profit. It`s clear thatincrease in profit leads to increase in productionand efficiency.

Also government should pay more attentionto scientific and especially genetic researchesand increase the research budget of R&D centersand universities to produce with low costs andimprove total factor productivity.

REFERENCES

1- Atule, J., & Hense (2001). Adjustment Policies inDeveloping Countries, World Bank, Washington D.C. 2- Ayinde, O.E., Adewumi, M.O., & Omotosho, F.J.(2009). Effect of Fertilizer Policy on Crop Productionin Nigeria, The Social Sciences. (Abstract).3- Azizi, J. (2005). The Analysis of Effects of Poisonand Fertilizer Price Liberalization on Rice Productionin Guilan, Agricultural Economy & Development.(Abstract).4- Bagherzadeh, A. (2010), Comparative Analysisof Economic-Social Features on Technical Efficiencyof Sunflower Farms (The case of Khoy sunflower),Agricultural Sciences of Islamic Azad University ofKhoy.5- Badmus, M.A. (2010). Market Liberalization andMaize Production in Nigeria. Journal of Developmentand Agricultural Economics. (Abstract).6- DFID (2003). A DFID policy paper.Htpp://www.dfid.gov.uk 7- Elyasian, H., & Hosseini, M. (1996). LiberalizationEffects on Agricultural Inputs Usage, AgriculturalEconomy & Development. (Abstract).8- Gulati, A, (1990). Fertilizer Subsidy, Is the Culti-vator Net Subsidized? Indian Journal of AgriculturalEconomics. (Abstract).9- Karimi, F., & Zahedi Keyvan, M. (2010). Identi-fying Optimal Pattern of Allocating Subsidies inAgricultural Sector on Producers and Consumers,Agricultural Economy Researches. 2(4): 99-120.10- Karimzadegan, H. (2006). The Effect of FertilizerSubsidy on Unoptimal Consumption of It in WheatProduction , Agricultural Economy & Development,14, 5511- Mehrabi, H., & Pakravan, M.(2009). Calculatingthe Varieties of Efficiency and Returns to Scale ofSunflower Producers in Khoy, Economy & Agricul-tural Development. (Abstract) (In Persian).


Inte

rnat

iona

l Jo

urna

l of

Agr

icul

tura

l M

anag

emen

t &

Dev

elop

men

t, 2

(3):

149

-155

, Sep

tem

ber,

201

2.

155

12- Mhando, D.G., & Jaichi, I. (2007). Farmers`copingStrategies to a Changed Coffee Market after EconomicLiberalization: The Case of Mbinga District in Tan-zania African Study Monographs. 36: 39-58.13- Najafi, B., & Zakarya, F. (2000). Welfare Effectsof Fertilizer Subsidy Elimination on Wheat (bread)consumers. Agriculrural Economy Researches. (Ab-stract).14- National Economic Empowerment and Devel-opment Strategy (NEEDS). The NEEDS Secretariat,National Planning Commission, Federal Secretariat15- Nieuwout, W (2003). Measures of Social Costsof an Input Subsidy and Value Information, Journalof Agricultural Economics. (Abstract).16- Ready, V.R., & Deshpande, R, S. (1992). InputSubsidies: Whither the Direction of Policy Changes,Indian Journal of Agricultural Economics. (Abstract).


www.ijamad.com

Inte

rnat

iona

l Jo

urna

l of

Agr

icul

tura

l M

anag

emen

t &

Dev

elop

men

t, 2

(3):

157

-165

, Sep

tem

ber,

201

2.

157

The Impact of Bio-Ethanol Conversion and Global

Climate Change on Corn Economic Performanve of

Indonesia

Yudi Ferrianta1, Nuhfil Hanani2, Budi Setiawan2 and Wahib Muhaimin2

Keywords: The energy crisis, Climatechange, Corn

Received: 30 January 2012,Accepted: 25 May 2012 Many studies conclude that the rise in global food prices

due to higher demand from the development of bio-fuels, climate anomalies, and increased of oil prices. Not onlythe food commodity index rose more than 60 percent, non-food commodity price index also rose over 60 percent andcrude oil price index has increased even further above 60percent. The purpose of this study is to analyze the impact ofbio-ethanol conversion and global climate change on corneconomic performance of Indonesia. The results showed thatthe food crisis caused by climate anomalies lead the worldcorn prices rose 50 percent, impact on Indonesia corn importsfell by 11.86 percent. And the other hand, the energy crisisthat caused the corn used as feedstock for ethanol that causedU.S. corn exports only 20 percent of their products have animpact on Indonesia on maize imports fell 32.4 percent.

Abstract


1 Department of Agriculture Economic, University of Lambung Mangkurat, Indonesia.2 Department of Agricultural Economic, University of Brawijaya, Indonesia. * Corresponding author’s email: [email protected]

Inte

rnat

iona

l Jo

urna

l of

Agr

icul

tura

l M

anag

emen

t &

Dev

elop

men

t, 2

(3):

157

-165

, Sep

tem

ber,

201

2.

158

INTRODUCTION

Corn is the third largest crop after wheat andrice, most of the corn products are used andtraded as feed material in addition to a staplefood. In addition to food and feed, corn has awide range of industrial applications such asmaterials for the manufacture of ethanol.

Over the last decade of global corn productionhas shown increasing growth, the global cornmarket generally divided into two issues, first,the conversion of the global corn used as bio-ethanol industry, second, the share of globallytraded corn are relatively constant.

The main cereal market - corn, wheat and rice- has shown some major adjustments in recentyears. Since 2008 the global food crisis resultedin a large spike in corn prices. On the demandside of high oil prices encourage the developmentof bio-fuel which resulted in increased demandin addition to dietary changes and income andpopulation growth. High oil prices also putpressure upward on the cost of crop production(e.g. fertilizer, tillage). On the supply side withlow cereal stocks, exacerbated by a policy oftrade restrictions on cereal and speculation incommodity markets. (Flammini, 2008).

The food crisis followed by the global financialcrisis in the second half of 2008, high oil priceswhich led to concerns about the security of na-tional oil and concerns about the environmentalimpact of fossil fuel use resulting in searchingalternative energy sources, one of the interestingissues is the development of bio-fuels that affectthe global corn market.

In the United States, the enhanced productionof bio-ethanol because corn prices are relativelylow, In the year 2007-2008, as many as 82million tones of corn used for ethanol, whichrepresents a quarter of U.S. corn productionand 12% of global production (DEFRA, 2008).Besides the development of bio-ethanol, oneof the factors that cause serious problems forthe production of corn from time to time is theoccurrence of El Niño weather phenomenonassociated with an abnormal warming of seasurface temperatures in the Pacific Ocean. Cornplants are most affected by El Niño (mostly inthe form of prolonged dry conditions) are con-

centrated in the southern hemisphere, particularlyin southern Africa. During El Niño events ofthe 1980s and the 1990s, for example, corn pro-duction in the Republic of South Africa fell by40 to 60 percent. Also in Brazil, corn producerssuffered from floods and droughts driven by ElNiño situation in the past. adverse weather con-ditions caused by the events of the last major ElNiño of 1997/98 are located mostly in EastAsia and led to a sharp decline in production incountries such as Thailand and Indonesia.

In Indonesia, corn has a very strategic role,especially for the farm development and otherindustries. In past, corn mainly used as staple.However, currently, corn mainly used as an in-dustrial material. In line with the rapid growthof livestock industry, it is estimated more than55% of domestic corn needs is used for feed,while for food consumption is only about 30%,and the remainder for other industrial needsand seeds (Indonesia Department of Agriculture,2010).

Currently, the development of corn productioncan not meet high demand. Therefore, the gov-ernments meet the shortage of these needsthrough imports. For 2010 forecast figures, witharea of 3 million ha of crops, it is estimated toproduce 12.1 million tons.. Meanwhile, maizedemand in the country reached 13.8 milliontons, resulting in a shortage about 1 million tonto be imported (Ferrianta, 2012). If the importincrement increase was not controlled, it willcause a reduction in foreign exchange, and canlower the domestic maize price, where the pricewas relatively low. Based on these facts, thegovernment is trying to meet the domestic maizeneed through maize self-sufficiency program.

Maize self-sufficiency effort must be directedto external factors, not only change in domesticpolicy but also external shock e.g bio-ethanoldevelopment and global climate change. Inline with the development of world economy,maize commodity will face a different environ-ment. External and internal shock will affectcorn economic performance of Indonesia.

Based on these facts, it is deemed necessaryto conduct research on the impact of bio-ethanoldevelopment and global climate change on the

Impact of Bio-Ethanol Conversion and Global Climate Change on Corn / Yudi Ferrianta et al

Inte

rnat

iona

l Jo

urna

l of

Agr

icul

tura

l M

anag

emen

t &

Dev

elop

men

t, 2

(3):

157

-165

, Sep

tem

ber,

201

2.

159

economic performance of corn in Indonesia.


Indonesia maize economic model is a simul-taneous equations consisting of three sub-models:sub production, sub domestic market and subworld markets. The data collected is secondarytime series data.

Model estimation is done using re-specificationmodel. The goal is to obtain good models basedon economic and econometrics criteria. In theestimation of these models studied the problemof identification, aggregation and the degree ofcorrelation between explanatory variables.

Evaluation conducted to know the impact ofinstrument change simulation variable on thefuture endogenous variable. The evaluationmodel is based on economic theory and infor-mation related to the research phenomenon. Amodel is good if it meets the following criteria:

1. Economics, in association with signs andestimation parameters,

2. Statistics, relating to statistical tests, and3. Econometrics, related to the model assump-

tions (Baltagi, 2008)For unbiased and consistent estimations, si-

multaneous systems require a more complexprocedure for estimation than single equationmodels, which can generally be estimated byregression with ordinary least squares (OLS).The most frequently used method of estimatingsimultaneous systems is the two-stage leastsquares (2SLS) method (Studenmund, 1997;Greene, 1993).

Furthermore, because the model contains asimultaneous equations and lagged endogenousvariables, serial correlation test is performedusing statistical dw (Durbin-Waston Statistics)in each equation. (Gujarati, 2004)

Model validation performed to analyze howconstructed model could to represent the realworld. In this study, statistical validation criteriafor value estimate econometric model is RootMeans Squares Error (RMSE), Root MeansSquares Percent Error (RMSPE) and Theil's In-equality Coefficient.

Econometric modeling and estimation can beuseful in providing a retrospective look at the

economic effects of a policy change or externalshock (MCDaniel, 2006). To simulation the im-pact of external shock to the import corn In-donesia, this study was used ex-post econometricsanalysis to see changes in the value of endogenousvariable due to changes in exogenous variables.(T. B. Palaskas 1988 ; Baumann, 2011).

Dynamic simultaneous equations system usedto develop econometric model. Models specifi-cation used are described as follows: 1.QJ = AJ * PRJ 2. AJ = a1 PJ + a2Pkdlt-1 + a3AJt-1+ U1

3. PRJ = b1 Pp + b2 i + b3AJ+ b4W + b5PRJt-1+b6CH+ U2

4. DIT = DIP + DIL + DK 5. DIP = c1Ppk + c2Pj + c3Pkdl + c4DIPt-1 + U3

6. DIM = d0 + d1Pop + d2PJ + d3Pni + U4

7. DK = e0 + e1PJ + e2Y + e3DKt-1 + U5

8. MIT= MIAS + MICH+ MITH + MIO9. MIAS = f1(PIAS-PIASt-1) + f2QJ+ f3DIT +f4ERI + f5(RISTI-RISTIt-1) + U6

10. MICH = g1PICH + g2QJ+ g3DIT + g4RISTI + U7

11. MITH = h1PITH + h2QJ + h3DIT + h4RISTI + U8

12. MIO = MIT- (MIAS + MICH+ MITH)13. RISTI = (PJ – PWJ)/ PWJ14. PIAS = PWJ + RISTAS 15. PICH = PWJ + RISTCH 16. PITH = PWJ + RISTTH 17. PJ = i1MIT + i2DIT + U9

18. XAS = j0 + j1QAS + j2DAS + j3XTH +j4XCH + j5MJJ + j6MJK + j7PETH + U10

19. XCH = k1QCH + k2DCH + j3XAS + j4XTH+ j5MJJ + j6MJK + U11

20. XTH = l0 + l1PWJ + l2QTH + l3DTH + U12

21. MJJ = m0 + m1PWJ + m2NPRJj + m3ERj + U13

22. MJK = n0 + n1PWJ + n2DJk + n3MJKt-1 + U14

23. XW = XAS + XTH + XCH + XRO24. MW = MJJ + MJK + MRO25. PW = o1XW + o2MW + U15

Note: • AJ = acreage of corn harvested (ha) • PRJ = productivity corn of Indonesia (tones / ha) • QJ = corn production of Indonesia (tones) • PJ = corn prices of Indonesia(US $ / tone) • i = Indonesia interest rate (%) • W = Indonesia wage labor (US $ / day) • Pp = the price of fertilizer (US $/ tone) • CH = climate change (oceanic nino index)


Inte

rnat

iona

l Jo

urna

l of

Agr

icul

tura

l M

anag

emen

t &

Dev

elop

men

t, 2

(3):

157

-165

, Sep

tem

ber,

201

2.

160

• DIT = total corn demand of Indonesia (tones) • DIP = Indonesia corn demand for feed

industry (tones) • DIM = Indonesia corn demand for food in-

dustry (tones) • DK = Indonesia corn demand for direct con-

sumption (tones) • KDP = feed prices of Indonesia (US $ / tone) • Pkdl = soybean price of Indonesia (US $ / tone)• Pop = population of Indonesia (people) • MIT = Total Imports corn of Indonesia (tones)• MIAS = Indonesia corn imports from US. (tones) • MICH = Indonesia corn Import from China (tones) • MITH = Indonesia corn imports from Thailand

(tones) • MIO = Indonesia corn imports from other

countries (the rest) • PIAS = the price of corn imports from US

(US $ / ton) • PITCH = the price of corn imports from

China (US $ / ton) • PITH = the price of corn imports from Thai-

land, (US $ / ton) • RISTI = corn trade restrictions of Indonesia • ERI = exchange rate of Indonesia (rupiah / US $) • XAS = US corn exports (thousand tones) • XTH= Thailand corn exports (thousand tones) • XCH= Chinese corn exports (thousand tones) • QAS = U.S. corn production (thousand tones) • QTH = Thailand corn production (thousand tones) • QCH= Chinese corn production (thousand tones) • MJJ = Japan corn imports (thousand tones) • PET = ethanol price (US$/bushel) • MJK = Korea corn imports (thousand tones) • DJ = corn demand of Korea (thousand tones) • NPRJ= corn trade restrictions of Japanese

(thousand tones) • ER = exchange rate of Japan (Yuan / US $) • XW = world exports (thousand tones) • XRO = corn exports of other country (thou-

sand tones) • MJW = world corn imports (thousand tones) • MRO =corn exports of other country (thou-

sand tones) This study used time series data’s, starting in

1983 until 2010. The data is obtained from In-donesia Department of Agriculture, Bureau ofIndonesia Statistics, Indonesia Ministry of Agri-

culture, Directorate General of Food Crops andHorticulture, Food and Agriculture Organization,United States Department of Agriculture, UnitedNations Commodity Trade Statistics Database,and International Monetary Fund.

RESULTS

International corn economy has undergonemajor changes over the past two decades interms of production, utilization, trade and mar-keting structure. This change was driven by anumber of factors ranging from rapid advancesin seed technology and production, changes innational policy and international trade, expansionalmost without interruption from the use offeed throughout the world and recently hugedemand for ethanol.

Production

Over the past two decades, global corn pro-duction has increased nearly 50 percent, or 1.8percent growth rate per year. Most of the increasein world corn production over the past decadecan be attributed to rapid expansion in Asia.

Asian corn production grew nearly 35 percentover the past decade, nearly 30 percent of theglobal increase. The increasing expansion ofacreage and yield contributed to high growthrates, like China that makes the most significantprogress with contributions as much as 60percent of total corn production of Asia overthe past decade.

Although progress is associated with varietiesthat have high productivity, it is likely to increasecorn production in many countries remains largealong with the good level of production efficiency,especially in developing countries are still under


Figure 1: World Corn Production (Sources:USDA, 2010)

Inte

rnat

iona

l Jo

urna

l of

Agr

icul

tura

l M

anag

emen

t &

Dev

elop

men

t, 2

(3):

157

-165

, Sep

tem

ber,

201

2.

161

major manufacturers. Average corn yields amongdeveloping countries about a third of the countriesmajor corn producer. Among some of the coun-tries largest producer of corn (Figure 2), Argentinaapproximately 5.6 tones / ha, China about 5tones / ha, while Indonesia about 3 tones / ha.This is much compared to the United Statesabout 10 tones / ha.

Corn as Biofuels Material

Bio-diesel is an alternative diesel fuel energysources are derived from vegetable oil (vegetableoil) and animal fats (animal fat) where corn isone potential source of bio-diesel product. Worldprice of bio-diesel (FOB Central Europe) in-creased to $ 4.14 per gallon by 2010, driven byhigh oil prices and prices of edible oils. Increasedcrude oil prices and the existence of tariffbarriers in Argentina, Brazil, European Union,as well as the U.S. led to an increase in worldprices.

With the huge consumption demand for localindustrial production of bio-diesel will increaseproduction by 5% in 2010 and estimated pro-duction continues to increase and reached 3.5billion gallons by 2019, on the other hand theconsumption continues to grow to 4.0 billiongallons by 2019 so that the net import growduring the outlook period and reached 559million gallons by 2019.

General Estimation of Econometrics Model

The empirical result of prediction models inthe study is good. All exogenous variables in-cluded in the structural model has a parameterthat the sign suitable with the theory and logical.Statistical criteria used in evaluating the prediction

is quite good. Coefficient of determination (R2)value in each behavioral equations ranged from0,38 to 0.99. From 15 behavioral equations,there is only one behavioral equation with R2values of 31 percent and 14 other equation isabove 64 percent. This shows that, in general,the exogenous variables included in the structuralequation model can explain variance rightly foreach endogenous variable.

The value of statistic F test generally high.There are 12 of 15 equation had value greaterthan 11.22. Meanwhile, only two equations haveF-value 8,50 and a 1,38. That is, simultaneously,explanatory variable variance in each equationbehavior are able to explain the variance of en-dogenous variable, at α= 0.0001; α = 0.0003and α=0.2744. Detailed econometric model es-timation for maize are presented in Table 1.

Simulation of External Shocks on the Eco-

nomic Performance of Indonesia corn Per-

formance

The world has been experiencing a globalcrisis caused by global warming, energy crises,and monetary crisis. Global warming has causedclimate anomaly, resulting in a sharp decline inworld agricultural production resulting foodcrises, including maize.

Global food price index increase has reached120 percent, where about 60 percent in just thepast two years, while the World Bank statedthat the price index of food crops increased 86percent between 2006 to 2008. Agriculturalcommodity prices rose in 2006 and 2007 andcontinued to increase even more sharply in2008. Meanwhile, according to the World Bank,global wheat prices increased by 81 percent(World Bank, 2008), and 83 percent increase inoverall global food prices.

The energy crisis has led to the developmentof corn as a bio-fuel feedstock, resulting in adecrease in world corn exports, especially inthe US. The figure below shows the extent ofthe use of corn for the bio-fuels industry theUnited States, 1984-2009 a huge surge in theuse of corn as an ethanol feedstock domesticproduct, this indication will be a large drop inexports US, in addition to other major exporting


Figure 2: Corn Productivity in Some Countries(Sources: FAOSTAT, 2010)

Inte

rnat

iona

l Jo

urna

l of

Agr

icul

tura

l M

anag

emen

t &

Dev

elop

men

t, 2

(3):

157

-165

, Sep

tem

ber,

201

2.

162


Model Variable coefficient t-statistic statistic

AJ

PRJ

DIP

DIM

DK

DITMIAS

MICH

MITH

MITPJ

XAS

XCH

XTH

XWMJJ

MJK

MWPW

PJPKDLLAJLF-test= 1016.30 PUPUKIAJWPRJLCHF-test = 2640.65 PPKPJDIPLPKDLF-test= 92.47 InterceptPJPNIPOPF-test=12.62InterceptPJYDKLF-test= 58.46 DIT = DIP + DIM + DKPIASHQJDITERIRISTIHF-test= 13.82 PICHQJDITRISTIF-test= 8.28 PITHQJDITRISTIF-test= 8.76MIAS + MICH+ MITH+MIOMITDITF-test= 86.26 InterceptQASDASXTHXCHMJJMJKPETH F-test= 1.52QCHDCHXASXTHMJJMJKF-test=11.23InterceptPWJQTHDTHF-test= 612.58XAS +XTH +XCH + XOInterceptPWJNPRJERJF-test=14.61InterceptPWJDJKMJKLF-test= 130684 MJK + MJJ + MIT +MJOXWMWF-test= 227.42

5.3656-0.379240.962391

R2= 0.99284 -5.87E-10-0.04547

5.259E-08-0.000181.246336-0.02339

R2= 0.9988 41.06936-34.46080.881228-0.33793

R2= 0.94628 -15280000-128.92656.187530.023537

R2= 0.64325 769281.5-22.447

-62.24210.652367

R2= 0.89306 -50301.5-0.397670.403125-16.0265-75.6419

R2= 0.77557 -540.767-0.159330.195036-624179

R2= 0.61206 -819.626-0.028010.046871

-45811R2= 0.62518

-0.000930.001993

R2= 0.88237 92622480.061165-0.02246-0.03033-0.536693.2263940.971053

-23310000R2= 0.38465

0.174571-0.23146-0.23542-0.743191.1949960.638035

R2=0.78010 -4272222148.1761.011109-0.93046

R2= 0.98870 21603517-8992.84-148722-24401.7

R2=0.67603 -122641-234.0221.0057250.003463

R2= 0.99995 -1.41E-071.72E-06

R2= 0.95187

0.41-0.757.61

DW = 1.73635 -2.57-3.010.31-0.4114.97-1.21

DW = 1.2869322.19-1.154.48-0.17

DW = 1.241174-3.86-1.135.593.18

DW = 0.8805931.18-0.94-0.457.14

DW = 0.985704-0.32-5.135.33-0.84-0.21

DW = 2.307239-0.35-1.271.9

-1.83DW = 1.874842

-1.62-0.731.41-0.37

DW = 1.429704-0.36.89

DW = 0.1363750.110.53-0.14-0.01-1.570.610.91-1.38

DW = 1.9373231.2

-1.26-1.76-0.662.271.02

DW = 2.12621-1.982.73

20.58-40.62

DW = 2.07918110.37-1.44-1.49-4.73

DW = 2.337765-7.82-1.85

536.152.03

DW = 1.885428-0.182.2

DW =1.126065

0.68820.4618<.0001

0.01860.00720.75840.6869<.00010.2416

0.03970.26420.00020.8647

0.00090.2709<.00010.0045

0.24960.35780.654

<.0001

0.7514<.0001<.00010.41180.8352

0.7280.21830.07160.0815

0.12060.47370.17370.7169

0.7638<.0001

0.91480.59990.89190.991

0.13520.54870.37380.1848

0.24360.22320.09480.517

0.03480.3217

0.06140.0124<.0001<.0001

<.00010.165

0.15040.0001

<.00010.0789<.00010.0549

0.85660.0382

Table 1: Econometrics Model Estimation

Source: Research findings.

Inte

rnat

iona

l Jo

urna

l of

Agr

icul

tura

l M

anag

emen

t &

Dev

elop

men

t, 2

(3):

157

-165

, Sep

tem

ber,

201

2.

163

countries, especially in the European Union andLatin America jointly develop bio-ethanol in-dustry. (Figure 3).

Simulation aims to analyze the impact ofvarious changes in the exogenous variables.However, before doing the simulation, modelvalidation must be done to look at the suitabilityof the predicted value in accordance with theactual value of each endogenous variable (Pindyckand Rubinfield, 1991).

Table 2 presents the results of the validationof the economic corn model. Based on Table 2can be found, only three equations in the modelhas a RMSPE value of more than 50 percent,only one equation is greater than 100 percentand the rest have RMSPE value of less than 50percent. U-Theil criteria there are 13 equations

have a U value of less than 0.20, and 5 theequation has a value of U between 0.24 to 0.50.The highest value of the Theil-U in the equationis 0.5, and RMPSE value greater than 100percent, is owned by the Indonesian corn priceequation but there is no systematic bias, becausethe value of Um more than 0.20. Overall, thismodel is suitable for use as predictive models,so the structural model has been formulatedwhich can be used for various simulations.

Simulation is used with the assumptions:

(1) climate anomalies lead the world corn pricesrose 50 and the energy crisis that caused the cornused as feedstock for ethanol, as a result theworld corn prices rose 2.9 percent. Ex ante analysisfor simulation model presented in table 3. Based


Figure 3: US Corn used for Ethanol Production1984-2009 (Sources: USDA, 2010)

Figure 4: Estimated of Indonesia Corn Import atThe Event of External Shock (Simulation analysis)

No. Variable RMSPE Reg (UR) Var (US) Covar (UC) Coef U

123456789101112131415161718

AJPRJDIPDILDK

MIASMICHMITH

PJPWJXASXTHXCHMJJMJKDITQJMIT

0.590411.68641.8582

22.899938.976752.987865.778424.6001

262.437.36817.4399

30.544521.80083.05890.3193

16.917812.002730.1795

0.020.020.010.000.050.990.250.140.070.080.220.100.000.070.080.010.030.59

0.040.020.000.000.060.620.110.000.000.080.220.100.000.010.080.010.030.18

0.760.000.370.000.000.370.140.140.070.000.000.000.000.060.000.000.010.41

0.00280.05580.00890.12930.25090.26360.46130.14450.50180.24110.03990.19090.12260.01500.00160.09260.05740.1761

Table 2: Result of Validation Dynamic Econometric Models


Inte

rnat

iona

l Jo

urna

l of

Agr

icul

tura

l M

anag

emen

t &

Dev

elop

men

t, 2

(3):

157

-165

, Sep

tem

ber,

201

2.

164

on analysis can be show that the food crisis dueto climate anomalies lead the world corn pricesrose 50 percent impact on Indonesia corn importsfell by 11.86 percent. While the energy crisisthat caused the corn used as feedstock for ethanol,causes U.S. limit maize exports only 20 percentof their products have an impact on corn importsIndonesia fell 32.4 percent. (Figure 4)

Ex-post simulation analysis results indicatethat both the external shock of climate changeand bio-ethanol conversion have an impact thedecline in the amount of corn traded in worldmarkets, and this has also impacted on thedecline in imports of corn Indonesia. Based onthe decrease in the value from the two simulationsshows that the U.S. has a significant role in In-donesian corn economy, where if the U.S.lowered its export causes a considerable impactfor more decline in Indonesia imports. This in-dicates Indonesia has a large dependence on theU.S for maize domestic supply, therefore theneed for policy in terms of increased productivityand the expansion of planting area by utilizingthe technology package and the existing landuse to reduce dependence on other countries.

CONCLUSION AND RECOMMENDATIONS

The main objective of this study was to knowsthe impact of bio-ethanol development and

global climate change on the economic per-formance of corn in Indonesia. This study usedthe annual time series data (1983-2010) anduse a dynamic simultaneous equations system.

Ex- post simulation analysis results show thatclimate change, bio-ethanol conversion as amajor determinant of import corn Indonesia.The results of simulation analysis of global cli-mate change and the conversion of corn forbio-ethanol have an impact on the fall to importcorn in Indonesia. This situation is expected toincrease competitive advantage and comparativeof Indonesian corn farming. Nevertheless thereare still many problems faced by Indonesia suchas corn farm land issues, technology, human re-sources, capital, fertilizer; rural infrastructure;and distortion distribution.

Some policies that are needed include (1) theexpansion of planting area by increasing croppingindex (IP) and extensification by making use ofidle land, (2) suppress the difference in resultsbetween regions and agro-ecosystems throughthe use of new high yielding varieties and hybridcomposites as well as site-specific applicationof the PTT model, (3) suppress the loss of theharvest and post harvest, and (4) increase thestability of the results between seasons andregions through the implementation of integratedpest management wisely. (5) human resource


Variable Base World price rise 50% US. Export only 20% from their product

AJPRJDIPDILDKMIASMICHMITHPJPWJXASXTHXCHMJJMJKDITQJMIT

21561425.4348

5338632683866784286234587915093211942024919.9

160.952286034

5346736605682

170557109501793

13020161117478621106718

21565845.4348

5335799682806684101733876110388452425.825002.2

24050281511704465

6090199163449149483296

1300488111750272985558

21574465.4348

53302586807336837408

014104211711325162.9

165.69242391544645

16681027170139659500706

1297500211754655748642

Table 3: The Ex-Post Analysis for Simultaneous Simulation


Inte

rnat

iona

l Jo

urna

l of

Agr

icul

tura

l M

anag

emen

t &

Dev

elop

men

t, 2

(3):

157

-165

, Sep

tem

ber,

201

2.

165

development of farmers through Farmer FieldSchools (human capital) and also involve farmersin innovation (joint innovation), (6) institutionaldevelopment (social capital) farmers as FarmerField School activities continued; (7) irrigationinfrastructure investment and drainage are moreflexible (physical capital), and (8) investmentin infrastructure and rural economy.

REFERENCES

1-Anderson, T.W. & Hsiao, C. (1981). Formulationand Estimation of Dynamic Models Using PanelData. Journal of Econometrics. 18: 570-606.2- Gulati, A., & Dixon, J. (2008). Maize in Asia:Changing Markets and Incentives, CIMMYT andIFPRI. New Delhi.3-Armstrong, J., Scott & Collopy, F. (1992). ErrorMeasures for Generalizing about Forecasting Methods:Empirical Comparisons. International Journal ofForecasting. 8: 69-80. 4- Baltagi, B.H. (2008). Econometrics. SpringerPublishing, Germany.5-Blundell, R. and S. Bond (1998). Initial Conditionsand Moment Restrictions in Dynamic Panel DataModels. Journal of Econometrics. 87: 115-143.6- Mcdaniel,Ch., reinert, K., & Hughes, K. (2006).Tools of the Trade:Models for Trade Policy analysis.Science, Technology, America, and the Global EconomyWoodrow Wilson International Center for Scholars.Pennsylvania Avenue Washington. DC.7-Department for Environment Food and RuralAffairs (DEFRA). (2008). The Impact of Biofuelson Commodity Prices. Nobel House. London 8-Evans, A. 2008. Rising Food Prices. “BeefingPaper”. Chatham House, London. 9- FAO. (2010). Food Outlook. FAO Publications,Rome. 10- Ferrianta, Y. (2012). Impact of Trade LiberalizationAsean-China Free Trade Area (ACFTA) on The Per-formance of Indonesia Maize Economy. (Dissertation).University of Brawijaya Malang. Indonesia.11-Flammini, A. (2008). Biofuels and the UnderlyingCauses of High Food Prices. Rome, Global BioenergyPartnership.12-Greene, W.H. (1993). Econometric Analysis (2nd

ed.), New York: Macmillan.13-Giampiero, M. Gallo & Marcellino, M, (1998).Expost and Exante Analysis of Provisional Data.Institute for Economic Research (IGIER) WorkingPaper no 141.14-Gujarati, D.N. (2004). Basic Econometrics, 4th

ed. McGraw-Hill, New York. 15-Im, K.S., Pesaran, M.H. & Shin, Y. (2003).Testing for Unit Roots in Heterogeneous Panels,Journal of Econometrics. 115: 53-74. 16- Indonesia Department of Agriculture ( 2010).Prospects and Corn Development in Indonesia.Agency for Agricultural Research and Development,Ministry of Agriculture, Jakarta.17- Labys, W.C. (1984). Commodity Models forForecasting and Policy Analysis. Nichols PublishingCompany, New York.18- Levin, A., Lin, C.F. & Chu, C.S.J. (2002). UnitRoot Tests in Panel Data: Asymptotic and Finite-SampleProperties, Journal of Econometrics. 108: 1-24.19- Pindyck, R.S & Rubinfeld, D.L. (2005). EconomicModel and Economic forecasts. McGraw Hill, Inc.New York.20- Baumann, R., & Matheson, V.A. (2011). Esti-mating Economic Impact using Expost EconometricAnalysis: Cautionary tales, College of The HolyCross, Department Of Economics Faculty ResearchSeries, Paper No. 11-03.21- Studenmund, A. H. (1997). Using Econometrics:A Practical Guide (3rd ed.). New York: Addison-Wesley.22-T. B. Palaskas (1988). An Econometric Analysisof EEC Policy Ex-post and Ex-ante. An Applicationto Dried Vine Fruits in Greece. Journal of OxfordDevelopment Studies. 17(1): 142-154.23-UNDP. (2010). Global Maize Production, Envi-ronmental Impacts and Sustainable Production Op-portunities-A Scoping Paper. UNDP Green Com-modities Facility. 24- USDA (2010). Grain: World Markets and Trade.Circular Series, FG 06-10, USA.25-USDA-ERS (2008). Cost of Production Estimates.Economic Research Service. USDA Publications.Washington. DC.26-USDA. (2008). Global Agricultural Supply andDemand: Factors Contributing to The Recent Increasein Food Commodity Prices. Economic ResearchService, USDA Publications. Washington, DC.27-World Bank (2008). Rising Food Prices: PolicyOptions and World Bank Response. Word BankPublications, Washington. D.C. 28-Von Braun, J. (2007). The World Food Situation,Food Policy Report. IFPRI. Washington. D.C.


www.ijamad.com

Inte

rnat

iona

l Jo

urna

l of

Agr

icul

tura

l M

anag

emen

t &

Dev

elop

men

t, 2

(3):

167

-175

, Sep

tem

ber,

201

2.

167

Determinants of Repayment of Loan Beneficiaries of

Micro Finance Institutions in Southeast States of

Nigeria

Stephen Umamuefula Osuji Onyeagocha1, Sunday Angus Nnachebe Dixie Chidebelu2 and Eugene ChukwuemekaOkorji2

Keywords: Determinants of Repaymentof MFIs Loan Beneficiar-ies, Nigeria

Received: 14 August 2011,Accepted: 8 April 2012 The study investigated the loan repayment, its determinants

and socio-economic characteristics of microfinance loanbeneficiaries in the Southeast states of Nigeria. It was carriedout in three states of the five southeast states. Using a multistagesampling technique, a total of 144 loan beneficiaries in thethree segments of MFIs, namely; formal (commercial and de-velopment banks); semi-formal (NGOs-MFIs) and informal(ROSCAS, “Isusu” and co-operative societies) were randomlyselected and interviewed in the three states. An ordinary leastsquare (OLS) multiple regression analysis was carried out toisolate and examine the determinants of loan repayment fromthe respondents’ perspective. Results showed that beneficiarieshad low level of education, operated enterprises at a relativelysmall scale, had large family size and were of middle age.Further, it was found out that the majority of the respondentswere involved in farming enterprise (crop and poultry) eventhough trading was the most prominent single non-farmingenterprise (trading, processing and artisanship). The result af-firmed that the informal sector respondents recorded the bestrepayment rate, followed by the respondents of semi-formaland the banks brought the rear. Outstanding among the deter-minants of loan repayments from the respondents’ perspectivewere; loan size, level of education, experience, profitabilityand portfolio diversity. These, therefore deserve special attentionin loan administration of MFIs.

Abstract


1 Department of Agricultural Economics, Federal University of Technology, Owerri, Imo State . * Corresponding author’s email: [email protected]

Inte

rnat

iona

l Jo

urna

l of

Agr

icul

tura

l M

anag

emen

t &

Dev

elop

men

t, 2

(3):

167

-175

, Sep

tem

ber,

201

2.

168

INTRODUCTION Microfinance institutions are those institutions,

which provide micro-credit, savings and otherservices to the productive poor. The focal pointof many studies on microfinance dwells in thedomain of poverty (Kanbur, 1987). Poverty isinsufficiency of means relative to human needs(Encyclopedia Am, 1981). It is estimated thatabout 70% of Nigeria’s population was poorand most of them live in rural areas and theirmajor occupation is farming (CBN, 2002).Nigeria ranks as one of the 25 poorest countriesin the world, having ranked 148 out of 173countries surveyed (UNDP, 2002).

Inadequate infra-structural facilities, poorsocial services, low technical education, unstablegrowth patterns of the economy and neglect ofagriculture, among other factors are largely re-sponsible for the despicable poverty situationin Nigeria. The fall in the quality of life ofNigerians to a reasonable extent is traceable tothe neglect of the agricultural sector and theoverdependence of the oil sector. The role ofsmall-scale farming in economic developmentof developing countries such as Nigeria is ines-timable. Apart from providing employment op-portunities to about 80% of rural population,they supply food, fiber and raw materials forthe populace, local industries and exporters.Production is characterized by small size ofland (often less than one hectare) and use ofcrude implements, poor yielding seedlings, in-efficient techniques, poor storage facilities, lowlevel of education, to mention but a few. Allthese cumulated to poor income and resilientvicious circle of poverty. Similarly, micro-en-terprises suffer from income anemia and viciouscircle of poverty of the owners.

There is concern that poverty reduction strategy(PRS) to date have tended to emphasize thepublic provision of goods and services (roads,water, etc) and paid less attention to productivesectors (Cabral Lidia, 2006). To break thesechains of poverty, ensure food security and in-dustrial growth of developing nations, there isneed for increase investment in the agriculturalsector by both the government and the farmers.It therefore becomes imperative to expand andstrengthen the financial institutions to play cat-alytic roles in this regard, especially in the areaof providing machinery and tools, improved in-puts and farmers’ education. Several studies,

including Feijo (2001) and Oyeyinka and Bo-lalarinwa (2009) have identified the positive im-pacts of credit in the operations of rural farmers.

Unfortunately, the formal financial institutions,especially the banks that are equipped to carryout these functions shy away from financingthese farmers, on grounds that they are highrisk ventures and involve huge administrativecosts. This provided the opportunity for the in-formal financial sector such as money lenders(with its obnoxious interest rates), local co-op-erative societies, credit unions and thrift schemesthat are less equipped to carry out this inter-mediation function, to key in and intensifycredit delivery functions. Confirming this, theCentral Bank of Nigeria (2005) noted that theformal financial system provides services toabout 35% of the economically active populationwhile the remaining 65% are excluded from ac-cess to financial services. According to the apexfinancial body, these 65% are often served bythe informal sector through NGO-MFIs, friends,relations and credit unions.

Surprisingly, these informal institution apartfrom their high cost of credit, are performingexceedingly well in terms of loan repayment(which is the nightmare of formal financial in-stitution). Also, their strong attribute is fast andefficient credit delivery with much less bureau-cracies like collaterals which is replaced withtrust and faith.

Loan repayment has been a critical problemof formal financial institutions in Nigeria. Studiesin Imo State by Njoku and Odii (1991) recorded27% repayment rate of the farmers, Njoku andObasi (2001) in which 33.72% was recorded asrepayment rate. This situation weakens thevirility of the MFIs. According to CBN (2005),the weak capital base of the existing financialinstitutions, particularly the community banks(now transformed to micro finance banks),cannot adequately provide a cushion for therisk of lending to farmers and micro entrepreneurswithout collateral. Further, poor repayment rateof credit reduces lenders net return thereby de-creasing the ability of the institution to generateresources internally for institutional growth. Inextreme cases, this may result in distress conditionor outright liquidation of the institution. Besley(1994) affirmed that the issue of enforcing re-payment constitutes a major problem in creditmarket. According to the author, enforcement

Determinants of Repayment of Loan Beneficiaries/Onyeagocha et al

Inte

rnat

iona

l Jo

urna

l of

Agr

icul

tura

l M

anag

emen

t &

Dev

elop

men

t, 2

(3):

167

-175

, Sep

tem

ber,

201

2.

169

problem arises in a situation in which theborrower is able but unwilling to repay theloan.

One way to tackle the loan repayment problemis to investigate the factors which affect theloan repayment of MFIs. Eze and Ibekwe (2007)in their study on determinants of loan repaymentin Orlu Local Government of Imo State, South-east, Nigeria, identified; loan size, age of bene-ficiaries, household size, and number of yearsof formal education and occupation as the keydeterminants. Similarly, Dayanandan and Welde-selassie (2008) in their study on loan determinantsof small farmers in Northern Ethiopia, agreedwith Eze and Ibekwe (2007) that amount ofcredit, educational status and occupation (non-farm income) were potent factors in loan repay-ment. Other factors they isolated as potent were;experience, repayment period and ownership oflivestock.

This study is aimed at providing answers tothe hydra-headed repayment problem. It is rea-sonable to expect that an impressive loan re-payment would be mutually beneficial to boththe farmers/micro-entrepreneurs and the loaninstitutions. On the part of the farmers andmicro entrepreneurs, good credit ratings woulddefinitely attract more loans with which toprocure improved inputs and implements. Insuch situation, efficiency would improve aswell as profitability and these are capable oflifting them out of the vicious circle of poverty.For the financial institutions, which dependmainly on interest income for their institutionalgrowth, prompt loan repayment would meanreduced cost and enhanced profitability androbust growth.

Therefore, the broad objective of this paper isto determine factors affecting repayment rate ofloan beneficiaries of MFIs in the SoutheastStates of Nigeria. The study specifically inves-tigated the social-economic characteristics ofthe respondents; determine their loan repaymentrate and its determinants.

MATERIALS AND METHODSThe study was carried out in Southeastern

states of Nigeria comprising of Abia, Anambra,Ebonyi Enugu and Imo States. The area had apopulation of 25.9 million, which is about 30%of the national population (2006). The Southeaststates are among the mostly densely settled area

of the country, with average population densityof 247 persons per square kilometer as againstthe national average of 96 persons per squareKilometer (NPC, 2006).

The choice of the area was because of intenseactivities of self help groups in various economicactivities, including agriculture in the area. Also,there is a high degree of socio-cultural homo-geneity in the study area as the inhabitants aremainly Igbos, known mainly for their hardwork, self-reliance and economic

prowess.Multi-stage sampling technique was employed

in the selection ofrespondents who were mainly loan beneficiaries

of commercial, development, community (microfinance) banks, NGO-MFIs groups and the localIsusu, co-operatives, ROSCAS members. Thesample frame was provided by the Central Bankof Nigeria for NGO – MFIs; the banks and thelocal extension agents of the local governmentcouncil.

In stage one, three out of five south-east stateswere purposively selected based on intensity ofMFIs activities.

Stage two involved the selection of MFIs,which were stratified into formal, semi-formaland informal. From each stratum, four institutionswere selected randomly. Thus, giving a total of12 MFIs per state and 36 MFIs for the threestates selected.

Finally, from each of the 12 MFIs in a state,four respondents were selected, randomly. Thus,giving a total of 48 respondents per state, and144 respondents for the three states, representingthe south-east states. The respondents were se-lected from 28 out of 57 LGAs of the threestates and this represented about 49% coverageof the total number of the LGAs. The 28 LGAscame into the sample by chance factor as no de-liberate effort was made to choose them.

From the selected respondents, which involvedfive enterprise-types namely; crop and poultryfarmers, traders, agro-processors and artisans;calibrated as farming and non-farming activities.Primary data were collected with the aid of astructured and pre-tested questionnaire. Thesecondary data were collected from journals,textbooks, annual accounts, return from banks,UNDP and CGAP (the consultative group toassist the poorest) websites.

The data collected were subjected to both de-


Inte

rnat

iona

l Jo

urna

l of

Agr

icul

tura

l M

anag

emen

t &

Dev

elop

men

t, 2

(3):

167

-175

, Sep

tem

ber,

201

2.

170

scriptive and quantitative techniques, to realizethe objectives of the study. The OLS multipleregression analysis was used to determine factorswhich affected repayment rate of loan benefici-aries. The linear functional form was adjudgedthe most appropriate for a repayment function.The model is stated as follows:

Y = f (X1, X2, X3---X13, e)Y1 = Repayment rate (%)X1 = Loan size (N)X2 = Dependency ratio (children as percentage

of total households size)X3 = Level of education (year of formal edu-

cation)X4 = Age (years)X5 = Enterprise type (dummy variable: farming

enterprise =o, and non farming enterprise = 1)X6 = Experience (years)X7 = Profitability of respondents enterprises (N)X8 = Training (total no. of days per year)X9 = Interest rate (%)X10 = Repeat loan (%)X11 = Gender factor (percentage of group mem-

bers who are female)X12 = Shocks (No. of family emergencies,

crop/income loss due to

incidence of pests and diseases, major socialevents that occurred in the previous 18 months)

X13 = Portfolio Diversity (proportion of mem-bers that have secondary

occupation).e = error term.

RESULTS AND DISCUSSIONSocio-economic characteristics

The socio-economic characteristics are im-portant sign posts in explaining the behaviourof the farmers and micro entrepreneurs in certainactions such as management and loan repaymentdecisions. They complement the results of thetechnical or quantitative analysis such as OLSmultiple regression. Some of these characteristicsare summarized in the tables.

Table 1(a) is the distribution of the meanvalue of some economic indices of the respon-


Socio-economic Characteristics Mean Value

Majority of sex: female Marital Status Family sizeAge (years)Experience (years)

638010418.8

No of Years Spent in School Frequency %

None1 – 67 – 1213 -16Total

43314624

144

29.921.531.916.7

100.0

Table 1(a): Distribution of the mean values ofsome economic indices of the respondents

Table 1(b): Distribution of Number of Years Spentin School

Table 1(d): Distribution of respondents by enter-prises size (turnover: Naira for traders, agro proces-

sors and artisans)

Class Frequency %

Less than 20,00021,000 -51,00052,000 – 82,00083,000 – 113,000114,000 – 113,000114,000 – 144,000145,000 – 175,000176,000 – 206,000Greater than 206,000Total

122

10102118121076

1.322.642.64

13.1513.1527.6323.6815.7913.15100.0

Occupation Frequency %

TradingCrop farmingAgro – processingPoultry farmingPoultry farmingTotal

4841172711

144

33.328.511.818.87.6

100.0

Table 1(c): Distribution of Primary Occupation

Table 1(e): Distribution of poultry farmers by enter-prises size (stock of birds)

Class Frequency %

less than 5051-101102 -152152-203Greater than 203Total

115821

27

3.755.629.67.43.7

100.0

Class Frequency %

01 or less0.2 -0.40.5-0.70.8-1.0Greater than 1.0Total

1218731

41

29.2743.9017.077.322.44

100.0

Table 1(f): Distribution of crop farmers by farm size(hectare)

Inte

rnat

iona

l Jo

urna

l of

Agr

icul

tura

l M

anag

emen

t &

Dev

elop

men

t, 2

(3):

167

-175

, Sep

tem

ber,

201

2.

171

dents. The majority of the respondents (63.2%)were female and male constituted only 36.8percent. Eighty percent of the respondents weremarried and by implication were likely to havefamilies, while 20% were single. On age, about55% of the respondents were of middle agebracket and above, with about 45% being youths.The respondents have relatively large familywith 10 as mean family size as against the rec-ommended national figure of six. Over 70% ofthe respondents had eight years and above inexperience in work with a mean figure of 8.8years.

Table I (b) is the distribution of the respondentsby level of education. It showed that about 70%of the respondents are literate and about 30%were not literate. This suggested that educationwas still a problem. Literacy level impacts pos-itively in productivity and efficiency of farmersthrough adoption of technology and innova-tions.

Table 1(c) is the distribution of respondentsby primary occupation. It suggested that tradingwas the primary occupation of the greatestnumber (33%) of the respondents. However,on the aggregate, farming constituted about60% of the respondents’ primary occupationwhile non-farm enterprise constituted about40%. About 40% of the non-farming respondentshave farming as secondary occupation.

Tables 1(d), (e) and (f) are the distribution ofrespondents by enterprise size (Naira) fortraders/processors/artisans; stock of birds forpoultry farmers and farm size (hectares) forcrop farmers, respectively. Table 1(d) showedthat over 71% of the respondents had a turnoverof less than N144, 000 per annum. This suggestedthat the respondents were of low income group.Table 1(e) showed that over 80% of the poultry

farmers had not more than 152 birds in theirstock. The mean stock of birds for these farmerswas 102 birds suggesting small-scale operations.Table 1(f) showed that over 90% of the cropfarmers owned or cultivated not more than 0.7hectares of land. The mean size of farm of therespondents was 0.46 hectare, suggesting thatthey were operating mainly on a small-scale.

Sources of Loan and Repayment RateTable 2, showed the distribution of respondents

according to sources of their loan. The NGO-Microfinance Institutions provided loan to 38.8%of the respondents. This was followed by RotationSavings and Credit Association (ROSCAS)(16.7%), NACRDB (14.6%), co-operative so-cieties (11.8%), commercial banks (10.4%) andCommunity Banks (or Microfinance Banks)(9.7%).

On loan repayment, this was segmented intoprompt repayment (for those repayments thatwere effected as scheduled) and overall repayment(for those repayments that were effected not asscheduled and of course, which involved recoverycosts on the part of the financial institutions) asindicated on Table 3(a). On prompt repayment,the respondents of informal institutions recordedan average of 90%, repayment rate followed by


Class Frequency %

Co-op Soc.NGO/MFIsCommercial BanksROSCAS DFI (NACRDB)Community Banks (MFBs)Total

175315242114

144

11.836.810.416.714.69.7

100.0

Table 2: Distribution of Respondents by Sourcesof Loans

Table 3: Loan Repayment of Respondents

Enterprise (or MFI Category) Frequency Repayment(%)

Prompt Overall

a) MFI CategorizationFormal Semi- Formal Informal

b) Enterprise TypeCrop Farming Poultry Farming Trading Agro-processing Artisans

121212

4127481711

43.04 73.57 90.00

55.4741.20 78.7870.7961.50

56.5884.91

100.00

48.33 (AV.)

70.35 (AV.)

Inte

rnat

iona

l Jo

urna

l of

Agr

icul

tura

l M

anag

emen

t &

Dev

elop

men

t, 2

(3):

167

-175

, Sep

tem

ber,

201

2.

172

semi-formal institutions (NGO-MFIs) 73.57%and formal institutions (banks) 43.04%. Onoverall repayment, the respondents of informalfinancial institutions recorded 100% repaymentrate. This was followed by semi-formal institu-tions 84.91% and formal institutions 56.58%.Table 3(b) indicated that the respondents intrading repaid about 79% of their loan promptly.This was followed by agro-processors (about71%). In general, non-farming enterprises onthe average repaid about 70% of their loan asagainst 48% of farming enterprises. This couldbe attributed to the complex and risky nature offarm enterprises.

Table 4 showed the reasons for default. Itshowed that family commitments ranked highest(35%) among the reasons adduced for default.This was followed by low market prices (28.5%),incidence of pests and diseases (16.3%), untimelydisbursements (13.2%) and crop failure (7%).Family commitments (like school fees, extendedfamily problems, burial and other cultural cere-

monies) were a big burden on the respondentsas well as low market prices, especially duringharvest, occasioned mainly from lack of poorstorage facilities.

Determinants of Loan Repayment of Respon-dents

Table 5 showed the factors, which affectedloan repayment and were calibrated as determi-nants of loan repayment. It indicated that out of13 explanatory variables, five were the mostpotent factors. The Coefficient of Multiple De-termination (R2) was 0.5022, suggesting thatabout 50% in the variation of loan repaymentwas accounted for by the variations of the ex-planatory variables. This suggests that theremay be other factors not included in the model.If R2 = 1, it implies that there was 100% expla-nation of the variation in loan repayment by theexplanatory variables or regessand. However,if R2 = 0, it means that the explanatory variablesdo not explain any changes in the criterion vari-


Item Frequency %

ItemPoor harvest due to crop failure Low market priceIncidence of Pest and Diseases Untimely loan disbursementFamily commitments

171041241950

144

11.87.0

28.516.313.235.0

100.0

Table 4: Distribution of Respondents by reasons for Default

VARIABLE UNIT COEFFICIENT T-RATIO

Loan SizeDependency ratioLevel of educationAgeEnterprise typeExperienceProfitability IndexTraining period Interest rateRepeat loanGender factorShocksPortfolio diversity ConstantR2F-valueNd.f.

NairaPercentYearsYearDummyYearsNumberDaysPercentDummyPercentLikert RankingDummy39.91330.502210.0884144130

12.0318-7.104315.9122-6.03598.213410.449417.03189.4227-5.03899.116311.0295-15.02146.9943

2.9272*-1.14222.6372*-1.07511.03593.3368*4.0632*1.1725-1.22601.13391.0870-1.00193.3928*

Table 5: Determinants of Loan Repayments of Respondents:

LOS = *5%

Inte

rnat

iona

l Jo

urna

l of

Agr

icul

tura

l M

anag

emen

t &

Dev

elop

men

t, 2

(3):

167

-175

, Sep

tem

ber,

201

2.

173

able or loan repayment. The F-value is used totest whether or not there is significant impactbetween the dependent variable and the inde-pendent variables. In regression equation, if F-calculated is greater than F-tabulated, then thereis significant impact between the dependentvariable and the independent variables. If oth-erwise, the reverse is the case.

The five potent variables which affected loanrepayment were; loan size, level of education,experience, profitability and portfolio diversityand they are subsequently discussed.

(a) Loan SizeLoan size was significant at the 5% LOS and

was positively related to repayment rate. Thisimplies, the greater the size of the loan, thelower the default. This was true up to a certainpoint as there was an optimum amount of loan(or funds) that would be required to break evenin projects. Moreover, it is contended that biggerloans make possible larger investments withpotentially higher returns. About 75% of theloan beneficiaries indicated that the sizes oftheir loans were inadequate, thus supporting thisviewpoint. Also, Njoku and Obasi (2001) isolatedloan size, among two other variables, that areimportant and have positive relationship withloan repayment under ACGFS in Imo State.

Similarly, Olagunju (2007) in his study on theimpact of credit use, agreed with this viewpoint. However, Zeller et al., (2001) in theirstudy of group based financial institutions forthe rural poor in Bangladesh found out that thegreater the loan size, the greater the probabilityof default.

The second perspective to this variable wasthe larger the loan, the

higher is the borrower’s cost of delayingpayment. A larger loan is more difficult torepay if allowed to accumulate especiallywhere there are compounding interest andsanctions. This second factor puts pressureon the borrower to reduce late payments andserious default. In the sample, recorded in-cremental penalty rate of interest for delaypayment was minimal.

(b) Level of EducationThe level of education was significant at the

5% level, and waspositively signed as hypothesized. This sug-

gests that as the level of education improvedthe beneficiary also improved the ability to readand write and in the process, improved dexterityin the occupation, which concomitantly improvedprofit and the capacity to repay loans. This is inagreement with Coelli and Battese (1996) inIndia.

respondents were 6.4 while the figure for nonliterate respondents was 30%, which suggestedthat there were lots of room for improvement intheir education status.

(c) Experience The coefficient of experience was positive

and significant at 5% level suggesting that thelength of experience in occupation was a potentfactor in loan repayment. This was because ex-perience provided the compass with which theentrepreneur navigated the turmoil business en-vironment and was a veritable decision tool.The result and that of Parikh and Mirkalan(1995) supported this hypothesis. The respondentshad eight or more years in terms of business ex-perience. However, Ogundare (2009) reporteda negative coefficient of age and farming expe-rience, which implies that output decreased aseach of these variables increased. It suggeststhat the more the years of experience of thefarmer and by implication, the older in age andthe less productive and the tendency of increasingrisk aversion.

(d) ProfitabilityThe coefficient of profitability index was

positive and significant at 5% level and wasin consonance with hypothesis, which statedthat profitability index (ratio of income tocosts) had direct and strong relationship withrepayment. This was because difficulties inrepayment arose whenever a business in un-profitable. It is an indication or index of man-agement ability. In the event of not makingprofit, enterprises including NGOs (whichare expected to break-even), become unsus-tainable.

(d) Portfolio DiversityThis indicates the proportion of beneficiaries

who have secondary occupation. It is thereforean indicator of asset portfolio diversity withinthe group/respondents. The study showed thatthe majority (66%) of the respondents have


Inte

rnat

iona

l Jo

urna

l of

Agr

icul

tura

l M

anag

emen

t &

Dev

elop

men

t, 2

(3):

167

-175

, Sep

tem

ber,

201

2.

174

trading as their secondary occupation. Due todiversity, income within groups tended to beless covariant, thus making it easier to bail outerrant members. As hypothesized, the coefficientof the variable was positively signed and signif-icant at 5% level, indicating strong relationship.

The linear equation can generally be representedthus:Y1 = 39.9133 + 12.0318X1 – 7.1043X2 +15.9122X3 – 6.0359X4 +

(2.93*) (2.64*) 8.2134X5 + 10.44946 + 17.0318X7 + 9.4227X8– 5.0389X9

(3.34*) (4.06*)+9.1163X10 + 11.295X11 – 15.0214X12 +6.9943X13 + 6.0038

(3.39)*R2 = 0.5022 F-Value = 10.0884 *1%LOS

CONCLUSIONThe respondents are certainly micro/small

scale operators with low income, poor educationalbackground and relatively large family size andits attendant burden and challenges. The re-spondents were of middle age and females werepredominant. Farming was the main occupationand trading constituted a third of the respondents’occupation. Nevertheless, half of the tradingrespondents have farming as their secondaryoccupation.

The major source of their loans were the in-formal sector namely; NGOs-MFIs and ROSCAS.The respondents of the informal sector performedmost creditably in terms of loan repayments.This was followed by the semi-formal (NGOs-MFIs) and the banks brought the rear. This per-haps may be due to the fact that screening,monitoring and enforcement of payment werecarried out by the group members themselves.In terms of enterprise type, trading was foundto be the most important with respect to loanrepayment. This was followed by agro-processorsand artisan (others). Crop and poultry farmingbrought the rear. In general, non-farming enter-prises performed better than farming enterprisesin terms of loan repayment. The differencecould be attributed to the complex and riskynature of farming, hence the need for extra or-dinary support for farming enterprises.

In terms of loan administration and repayment,adequate attention should be paid to loan size,level of education, experience, profitability,

portfolio diversity. These constituted the deter-minants of loan repayment from the respondents’perspective and therefore deserve more focusand attention.

Further, formation of autonomous cooperativesocieties, provision of storage facilities and re-duction of some associated expenses that affectfamily commitments (e.g. school fees) will helpreduce loan default.

REFERENCES 1- Besley, T. (1994). How Do Market FailuresJustify Interventions in Rural Credit Markets? WorldBank Research Observer. 9(1): 27-48.2- Central Bank of Nigeria (2002). Baseline Survey ofMicro Finance Institutions in Nigeria. CBN Abuja. P2.3- Central Bank of Nigeria (2005). Micro FinancePolicy, Regulatory and Supervisory Framework forNigeria. CBN Abuja P. 2 4- Coelli T and G.B. Battese (1996). Identificationof Factors which Influence the Technical Inefficiencyof Indian Farmers. Australian Journal of AgriculturalEconomics. 40: 103-128.5- Dayanandan, R., & Weldeseassie, H. (2008).

Determinants of Loan Repayment Performanceamong Small Farmers in Northern Ethiopia. Journalof African Development Studies. (Abstract).6- Eze, C.C., & Ibekwe, U.C. (2007). Determinantsof Loan Repayment under Indigenous FinancialSystem in Southeast, Nigeria, Medwell Journals.2(2): 116-120. 7- Feijo, R.L.C. (2001). The Impact of a FamilyFarming Credit Program on the Rural Economy ofBrazil. www.anpec.org.br/encontro 2001.8- Kanbur, R.S.M. (1987). Measurement and Alle-viation of Poverty, with an Explanation to the Effectof Microeconomic Adjustments. Cabral Lidia (2006).Poverty Reduction Strategies, and the Rural ProductiveSectors: What Have We Learnt, What Else do Weneed to ask? ODI Natural Resources Perspective100, UK.9- Njoku, J.E. and M.R. Odii (1991). Determinantsof Loan Repayment under the Special EmergencyLoan Scheme (SEALS) in Nigeria: A Case Study ofImo State. African Review of Money, Finance andBanking. 1: 39-52. 10- Njoku, J.E. & Obasi, P.C. (2001). Loan Repaymentand its Determinants under the ACGS in Imo State,Nigeria, Africa Review of Money Finance andBanking, (Abstract).11- Ogundare, K. (2009). Technical Efficiency ofFarmers under Different Multiple Cropping Systems


Inte

rnat

iona

l Jo

urna

l of

Agr

icul

tura

l M

anag

emen

t &

Dev

elop

men

t, 2

(3):

167

-175

, Sep

tem

ber,

201

2.

175

in Nigeria. Journal of Tropical and Sub-TropicalAgro Ecosystem 10: 117-120.12- Ohajianya, D.O., & Onyenweaku, C.E. (2003).Demand for Community Bank Credit by small HolderFarmers in Imo State, Nigeria. Journal of SustainableTropical Agricultural Research. (Abstract).13- Olagunju, F.I. (2007). Impact of Credit Use onResource Productivity of Sweet Potato Farmers inOsun State Nigeria. Journal of Social Sciences 14(2): 175-178.14- Oyeyinka, R.A., & Bolarinwa, K.K. (2009).UsingNigerian Agricultural and Cooperative Bank, SmallHolder Loan Scheme to increase agricultural pro-duction in rural Oyo State, Nigeria”. InternationalJournal of Agricultural Economics and Rural De-velopment. (Abstract).15- Parikh, A.E., & Mirkalan, S. (1995). Measurementof Economic Efficiency in Pakistan, AmericanJournal of Agriculture. 32:53-61.16- National Population Commission (2006). Popu-lation Census of Federal Republic of Nigeria, Ana-lytical Report of the National Level, Abuja, NPC.


www.ijamad.com

Inte

rnat

iona

l Jo

urna

l of

Agr

icul

tura

l M

anag

emen

t &

Dev

elop

men

t, 2

(3):

177

-185

, Sep

tem

ber,

201

2.

177

Analysis of Technical Efficiency of Smallholder

Cocoa Farmers in Cross River State, Nigeria

Agom, Damian Ila1, Susan Ben Ohen1, Kingsley Okoi Itam1 and Nyambi N. Inyang2

Keywords: Cocoa production, Techni-cal efficiency, Stochasticfrontier, Likelihood ratio,Maximum likelihood ratio

Received: 5 February 2012,Accepted: 30 April 2012 The technical efficiency involved in cocoa production in

Cross River State was estimated using the stochastic frontierproduction function analysis. The effects of some selected so-cio-economic characteristics of the farmers on the efficiencyindices were also estimated. The study relied upon primary datagenerated from interviewing cocoa farmers using a set ofstructured questionnaire. A multi-staged random sampling tech-nique was adopted in selecting two hundred (200) cocoa farmersfrom Ikom Agricultural Zone in the state. The data on the socio-economic characteristics of the farmers were analyzed usingdescriptive statistics, while the stochastic production function,using the Maximum Likelihood Estimating (MLE) techniqueswas used in estimating the farmer’s technical efficiency andtheir determinants. Result of the analysis showed that farmerswere experiencing decreasing but positive returns to scale in theuse of the farm resources. The efficiency level ranged between0.20 and 0.93 with a mean of 0.69. The result of the generalizedLikelihood Ratio (LR) tests confirmed that the cocoa farmers inthe area were technically inefficient. The major contributingfactors to efficiency were age of farmers, farm size, level of ed-ucation, sex of farmer and age of the farms. The study observedthat there is enough room to improve efficiency with thefarmers’ current resource base and available technology andconcluded that policies that would directly affect these identifiedvariables should be pursued.

Abstract


1 Deptartment of Agric Economics, University of Calabar, Calabar, Nigeria.2 Commercial Agric Development Project (CADP), Cross River State, Nigeria.* Corresponding author’s email: [email protected]

Inte

rnat

iona

l Jo

urna

l of

Agr

icul

tura

l M

anag

emen

t &

Dev

elop

men

t, 2

(3):

177

-185

, Sep

tem

ber,

201

2.

178

INTRODUCTION

Agriculture in developing countries is char-acterized by low productivity leading to lowfarm incomes. In Nigeria, cocoa production ischaracterized by several problems that lead tolow productivity. This has resulted to a fall inpercentage share of cocoa output. As Amos(2007) notes, two reasons are said to be respon-sible for the fall in percentage share of cocoaoutput. First is the negligence of the agriculturalsector by the past administration due to the dis-covery of the petroleum resources that now ac-counts for the bulk of foreign exchange earnings.Second is the endemic problem in the cocoa in-dustry. Therefore, increasing productivity willincrease the percentage share of cocoa production.Analysis reveals that increasing agriculturalproduction has probably been the simple mostimportant factor in determining the speed andextent of poverty reduction. Most of these evi-dences are derivable from the Green Revolutionin Asia (Adeniran, 2007). In China rapid pro-ductivity gains achieved largely through tech-nological advances of the Green Revolution di-rectly increased producers income and labourers’wages by lowering the price of food and by gen-erating new livelihood opportunities as successin agriculture provides the basis for economicdiversification. The importance of productivityis that it gives a measure for efficiency.

In Nigeria, there have been studies on farmlevel efficiency in tree crop production andvery few have focused on cocoa production.Among these are studies by Giroh et al., (2008)who carried out analysis of the technical ineffi-ciency of gum Arabic based cropping patternsamong farmers in the gum Arabic belt of Nigeriaand that of Amos (2007) whose study is analysisof productivity and technical efficiency of smallholder cocoa farmers in Nigeria. The authorsemployed the stochastic frontier productionfunction analysis in their studies. However wedo not have such studies in Cross River Statewhich is a major cocoa producing area in Nigeria.Recent studies carried out on cocoa productionin Cross River State examined the Socio-eco-nomic variables and cocoa production (Oluyoleand Sanusi, 2009). Fertilizer use and cocoa pro-

duction and Investment in cocoa production inNigeria: .A Cost and Return analysis of threecocoa production management system in CrossRiver State cocoa belt by Nkang et al., (2009).None of these studies examined the technicalefficiency of the cocoa farmers.

Objectives:

This study was carried out to provide estimatesof levels of technical efficiency of cocoa farmersin Cross River State using farmers in IkomAgricultural zone, where there is a high con-centration of cocoa farmers in the state. Thestudy was interested in whether the cocoa farmerswere fully technically efficient, the current levelof efficiency and factors that influenced efficiency.The study therefore was to estimate the level ofand determinants of technical efficiency amongcocoa farmers in Cross River State.

MATERIALS AND METHODES

Study area

This study was conducted in the two majorCocoa Producing Local Government Areas inCross River State; Etung and Ikom. CrossRiver State is located in the Niger Delta regionof Nigeria. It is bounded in the North by BenueState, in the South by Atlantic Ocean, in theEast by Cameroon Republic and on the West byAkwa Ibom State, Abia and Ebonyi States. Thestate lies within latitude 40o 4, South and 60o30,North and between longitude 8o and 9o 00” Eastof the equator. It has three distinct ecologicalzones, the mangrove forest to the south, thetropical rainforest in the middle and the guineasavanna to the north. The annual mean rainfallranges between 1500mm and 2000 mm.

Sampling procedure and sample size

The multistage random sampling techniquewas adopted for this study. The first stageinvolved a purposive selection of two (2) LocalGovernment Areas in the Ikom AgriculturalZone- Ikom and Etung. This is because Ikomand Etung are the major cocoa producing areasin Cross River State. The second stage involvedthe random selection of Five (5) villages fromeach of the selected Local Government Areas,

Analysis of Technical Efficiency of Smallholder Cocoa Farmers/ Agom et al

Inte

rnat

iona

l Jo

urna

l of

Agr

icul

tura

l M

anag

emen

t &

Dev

elop

men

t, 2

(3):

177

-185

, Sep

tem

ber,

201

2.

179

giving a total of Ten (10) villages. The villagesselected in Ikom were Akparabong, Ayukasa,Okondi, Alok and Ajassor, while those selectedin Etung were Effraya, Agbokim, Bendeghe Ekim,Abijang and Abia. A simple random selection oftwenty (20) farmers from each of the selected vil-lages was carried out making up a total of onehundred (100) farmers from each of the twoLocal Government Areas which gave us 200cocoa farmers for the study. Information was ob-tained on socio-economic characteristics of thefarmers, output, labour, farm size and prices ofvariables using a set of structured questionnaire.

Model specification

Descriptive statistics including mean, standarddeviation and variances were used to analysethe socio-economic characteristics of farmerswhile the stochastic production function wasused to analyze the level of technical efficiency.The production technology of the cocoa farmerswas assumed to be specified by the Cobb-Douglas frontier production function (Tadesse,and Krishnamorthy, 1997; Amos, 2007).

The specified cocoa production function wasgiven as follows;In Y = In βo+β1InX1i + β2InX2i +β3InX3i β4InX4i

+ Vi- Ui (1)Where; Y = Quantity of cocoa produced (kg)X1 = Farm size (hectares)X2 = Quantity of fertilizer (kg)X3 = Quantity of fungicide (litres)X4 = Labour (man days)Βo = Y - intercept Β1 to β4 are coefficients to be estimated and Vi

and Ui are error terms. It is expected that β1, β2,β3, and β4 will have positive signs.

Determinants of technical efficiency

The influence of some socio-economic factorson the computed technical efficiency was de-termined by incorporating the socio-economicfactors directly in the frontier model, becausethey have influence on efficiency (Kalirajan,1981). The technical efficiency model was spec-ified as:

Ui = ąo + ą1Z1ί + ą2Z2ί + ą3Z3ί + ą4Z4ί +ą5Z5ί + ą6Z6ί + ą7Z7ί (2)

Where Uί = Technical efficiencyZ1 = Sex of farmer (dummy)Z2 = Marital status (dummy)Z3 = Age of farmer (years) Z4 = Education (years spent in school)Z5 = Family size Z6 = Age of farm (years) Z7 = Farm size (ha)ąo = y - intercept ą1 to ą7 are coefficients that were estimated. It

was expected that ą3 would have a negative sign,while ą2, ą4, ą5, ą6 and ą7 would have positivesigns. The sign of ą1 was indeterminate.

In determining the level of technical efficiencyof the cocoa farmers and analyzing the determi-nants of technical efficiency among the cocoafarmers, a generalized likelihood ratio (LR) testwas used to test the hypothesis of full technicalefficiency effects defined as

LR = -2 In (logH1 – logH2) (3)Where, H1 is the log – likelihood function of

the average function. H2 is the log- likelihoodfunction of the frontier function. The value hasa mixed chi-square distribution with degrees offreedom equal to the number of parameters plusone. A computer programme frontier version4.1 by Coelli (1994) was used in the computation,while the testing of the parameters was done at1 and 5 percent levels of significance.

RESULTS AND DISCUSSION

Socio-economic characteristics of the sampled

cocoa farmers

The distribution of sampled cocoa farmers ac-cording to their sex, age, marital status, educationallevel, family size and farming experience is pre-sented in Table 1. The results indicates that ma-jority (88.5%) of the farmers were males whilefew (11.5%) were females. The cultural settingof the area allows the males to have easy accessto land especially, where majority of them arethe heads of their respective households.

The table also indicates that majority of thefarmers (44.5%) were within the 47 to 57 yearsage bracket. This was closely followed by the


Inte

rnat

iona

l Jo

urna

l of

Agr

icul

tura

l M

anag

emen

t &

Dev

elop

men

t, 2

(3):

177

-185

, Sep

tem

ber,

201

2.

180

farmers with age 36-46 years (32%). Farmersthat were in the minority constituted 1.5% andthese farmers were above the 68 years agebracket. This shows that about 86% of thefarmers were in their most economically activeage bracket (25-57) years. However, there wasa widespread of farmers among all the agegroups, implying that cocoa farming was em-braced by all the age groups. The results alsoshowed that most of the farmers (68.5%) weremarried while 31.5% were single (Table 1).

Furthermore, the distribution of the cocoaframers according to their educational level (num-ber of years spent in school), shows that majority(76.5%) of the farmers had attained one level offormal education or the other. The mean level ofeducational attainment (years spent in school) ofthe farmers in the area was about 6 years, with11% of the farmers having tertiary education.This is an indication that some graduates wereinvolved in cocoa farming in the study area. Thisis a good pointer to improved productivity as thelevel of education is a tool with which an indi-vidual could be efficient at whatever endeavourbeing undertaken by the individual (Oluyole andUsman, 2006). As regards family size, a highproportion (86.5%) of the farmers had familysizes of 5 persons and above, while 13.5% hadless than 5 persons in their household. The meanfamily size of the cocoa farmers was 7 persons.Effiong (2005) reported that a relatively largehousehold size enhances the availability of familylabour which reduces constraints on labour costin agricultural production.

The table also shows that a very high proportion(99%) of the farmers had between 5 and abovethirty (30) years of experience in cocoa farming.The mean farming experience was about 15years. Farmers sometimes count more on theirexperience than educational attainment in orderto increase their productivity (Nwaru, 2004).The result implies that a good number of thefarmers are experienced farmers and thereforeare expected to have higher technical efficiencies.

Mean output and other production variables

in cocoa production in Cross River State

The statistics of the production variables ob-

tained from cocoa farmers in the study area aresummarized in table 2. The mean output of cocoafarmers in the area was 2428.10kg/annum/farmer.This is relatively high compared to figures ofless than two tonnes recorded elsewhere. Thismay be related to the age of the farms as mostfarms in Cross River State are within the pro-ductive age of 11 to 40 years (Table 3) compared


Variables Frequency Percentage

Sex: Male Female Total Age of farmer (years):25 – 3536 – 4647 – 5758 – 68>68Total Means Marital status:Single Married Total Mean Educational level (school years):068121416Total MeanFamily size:

<55 – 7 8 – 1011 – 1314 – 1617 – 19>19Total Mean Farming experience (years)<55 – 1011 – 1516 – 2021 – 2526 - 30>30Total Mean

17723200

196489253

20047.70 (8.91)

63137200

2.02(1.61)

47811436148

2006.45 (4.49)

27765424973

2007. 11 (4.66)

274543319612200

15.22 (7.70)

88.511.5100

9.532.044.512.51.5100

31.568.5100

23.540.57.0

18.07.040

100

13.53827124.53.51.5100

1.03.72.7

16.59.53.06.0100

Table 1: Distribution of socio-economic characteristicsof sampled cocoa farmers

Source: Field Survey 2010Note: Values in parentheses are standard deviations.

Inte

rnat

iona

l Jo

urna

l of

Agr

icul

tura

l M

anag

emen

t &

Dev

elop

men

t, 2

(3):

177

-185

, Sep

tem

ber,

201

2.

181

to farms in other parts of Nigeria. For labour, themean man- days used by the farmer during theproduction season were 51.50 man days. Farmsizes in Nigeria have been described as small,medium or large scale, if they fall into categoriesof less than 5ha, between 5ha and 10ha, ormore than 10ha, respectively (Upton,1972). Mostof the farms in Nigeria are of small to mediumscale categories. The average farm size amongthe cocoa farmers in the study area is 6.90hectares scattered in different locations in the lo-cality, hence the small holdings. It was observedthat majority of the cocoa farmers in Cross RiverState did not use fertilizer in cocoa production.The mean fertilizer used by the farmers was18.50, which is very low. The result is in linewith Oluyole and Sanusi, (2009) which reportedthat 98.13% of cocoa farmers in Cross RiverState did not use fertilizer in cocoa production.

An average of 1142.80 litres of fungicide wasapplied by the cocoa farmers (Table 2). Thehigh rate of application may be due to the lessresistance of the cocoa variety to infection.

Classification of cocoa farmers according to

age groups and their mean output in the

study area

The sampled farms were grouped accordingto their age groups and mean output and majority(92%) of the farms are within the productiveage of 11 to 40years age group (Table 3). Themean output initially increased within this ageas the cocoa trees get fully established, andthereafter output declines as shown in the table.

Maximum likelihood estimates of stochastic

production frontier function for cocoa farmers

in Cross River State

The coefficient of the Maximum LikelihoodEstimates (MLE) of the parameters of the sto-chastic frontier models of cocoa farmers isshown in Table 4. The variance parameters ofthe stochastic frontier production function arerepresented by sigma squared (δ2) and gamma(y). From the table, the sigma squared is 0.171and significantly different from zero at onepercent level. This indicated a good fit and cor-rectness of the distributional form assumed forthe composite error term. Gamma (y) indicatesthat the systematic influences that are unexplainedby the production function are the dominantsources of random error. The gamma estimatewhich is 0.327 and significant at five percentlevel shows the amount of variation resulting


Variables Minimum Maximum Mean Standard Deviation

Output (kg)Farm size (ha)Fertilizer (kg)Fungicide (litres)Labour (man days)Age of farmer (years)Family size Farming experience (years)

760.001.000.000.00

18.0025.001.004.00

76200.0040.00

200.002500.00

93.0071.0026.0050.00

2428.106.90

18.501142.80

51.4947.707.00

15.22

2343.536.35

37.09644.3714.658.914.667.70

Table 2: Summary statistics of output and other variables for sampled Cocoa farmers

Source: Derived from Field Survey Data 2010

Age group Frequency Percentage Mean output Standard deviation

5 - 10 11- 20 21-30 31- 40 41- 50 >50

581812283

200

2.540.540.511.04.01.5100

6305.28064.18797.0

10448.07465.16573.0

(2428.1)

6305.28064.18797.0

10448.07465.16573.0

(2428.1)

Table 3: Age group of farmers and mean output

Source: Field Survey Data 2010Note: The values in parenthesis are the mean output and standard deviation of the farms respectivelyand do not represent the total mean output and standard deviation of the groups.

Inte

rnat

iona

l Jo

urna

l of

Agr

icul

tura

l M

anag

emen

t &

Dev

elop

men

t, 2

(3):

177

-185

, Sep

tem

ber,

201

2.

182

from the technical efficiencies of cocoa farmersin the study area. This means that more than32% of the variation in farmer’s output is dueto difference in technical efficiency.

The result further shows that the signs of allthe estimated coefficients of the stochastic pro-duction frontier are positive which is consistentwith a priori expectation. This implies that thereis a positive relationship between the level ofoutput of cocoa and farm size, the quantity offertilizer, fungicide and labour used. This is ex-pected as the level of production depends largelyon the quantities of these inputs used on thefarm. This can only be up to a level that is con-sidered optimal after which farmers will be op-erating at sub optimal level. However, the co-efficients of the slope, farm size and labourwere significant at one percent indicating thatfarm size and labour are important determinantsof cocoa output.

Determinants of technical efficiency in cocoa

production

The analysis of the efficiency model shows

that the signs of the estimated coefficients inthe efficiency model have important implicationson the technical efficiency of cocoa farmers inthe study area. The coefficient of sex (z←←1)had a positive sign indicating that the malefarmers obtain higher levels of technical efficiencythan their female counterparts in the area. Cocoafarming is dominated by males in the area. Thisis so because cocoa farming is a tedious job andrequires more strength which females may notbe able to provide (Oluyole and Sanusi, 2009).

The coefficient of marital status (Z2) is inde-terminate. However, it is negative as shown inTable 5.

The coefficient of age (z3) is also positive.This does not agree with a prior expectation. Asfarmers age, there is a tendency that productivitywill continue to fall owing to their decliningstrength. However, this result could be attributedto the fact that most of the farmers in the studyarea started farming at early age. Hence, theolder they are the more experienced and efficientthey would be, since farmers’ experience in-creases with the number of years spent in farm-


Variable Parameters Coefficients Standard errors Standard errors

Constant Farm size Quantity of fertilizerQuantity of fungicide LabourDiagnostic statistics Gmma (Y)Sigma square Log likelihood function Likelihood ratio (LR)

βo

β1

β2

β3

β4

Yδ2

λ

7.4500.1550.0130.0020.009

0.3270.171-88.0871.42

0.1410.0160.0010.0040.003

0.1440.037

52.837***9.688***

1.300.50

3.00***

2.71**4.622***

Table 4: Maximum likelihood estimates of the stochastic production function for cocoa production

Source: Computed from Field Survey Data 2010 using Frontier 4.1Note: *** P ˂ 0/01, ** P ˂ 0/05

Variable Parameter Coefficient Standard error t-ratios

Constant Sex of farmer Marital status Age of farmerEducational level (sch yrs)Family size Age of farm Farm size

ąo

ą1

ą2

ą3

ą4

ą5

ą6

ą7

-1.2780.009

-0.2090.0110.054

-0476-0.0120.081

0.6990.4150.1350.1360.2040.2360.7230.010

-1.8280.022

-1.5480.8010.265

-2.016**-0.017

8.10***

Table 5: Determinants of Technical Efficiency

Source: Computed from Field Survey Data 2010 using Frontier 4.1Note: *** P ˂ 0/01, ** P ˂ 0/05

Inte

rnat

iona

l Jo

urna

l of

Agr

icul

tura

l M

anag

emen

t &

Dev

elop

men

t, 2

(3):

177

-185

, Sep

tem

ber,

201

2.

183

ing, it implies that the longer the time spent infarming the more experience they are.

Furthermore, the coefficient of educationallevel (z4) was positive but not significant.This implies that the level of technical effi-ciency of the farmer increases with the levelof education but not significantly. However,the result agrees with a prior that technicalefficiency should increase with increases inyears of schooling of the farmers since edu-cation and adoption of innovation were ex-pected to be positively correlated.

The coefficient of family size (z5) has anegative sign and is significant at the fivepercent level. This implies that increase infamily size will lead to decrease in technical ef-ficiency. This result does not agree with a priorexpectation. However, given that the cocoafarmers utilized more of hired labour than family

labour and less labour (weeding and pruning)is required once the cocoa has been estab-lished, it could be acceptable. The farmerspay wages that are more than the value oftheir marginal production and hence wouldbe inefficient as a result of allocative ineffi-ciency (Idiong, 2006).

The negative coefficient of age of farm (Z6)implies that efficiency decreases as the farmsget older. Amos, (2007) had a similar result.Lastly the coefficient of farm size (Z7) was pos-itive and significant at the one percent level.This implies that technical efficiency increaseswith the size of farm. This result agrees withthose of Giroh et al., (2008) and Amos (2007).Large farm sizes if properly managed shouldhave higher efficiency. However, there is athreshold where returns to scale decreases withincrease in farm size.

Elasticity of production and returns to scale

Typical of the power function (Cobb-Douglas),the estimated coefficients for the specified func-tion can be explained as the elasticities of theexplanatory variables.

The values of the variables indicates that a10 percent increase in the size of farm, fertilizer,fungicide and labour will lead to a 1.5, 0.9,


Variables Elasticity

Farm size Quantity of fertilizer Quantity of fungicideLabour RTS

0.1550.0860.4200.0940.179

Table 6: Elasticity of production and returns toscale

Source: Field Survey, 2010

Efficiency Likelihood function (λ) Log likelihood ratio (LR) Critical X2 0.05 Conclusion

Technical -88.08 10.712 10.371 Reject

Table 7: Test of hypotheses that cocoa farmers in Cross River State are fully technically efficient (У= 0)

Source: Derived from Table 4. Critical X2 was obtained from Kodde and Palm (1986).

Efficiency level Frequency Percentage

0.20 – 0.300.31 – 0.400.41 – 0.50 0.51 – 0.600.61 – 0.700.71 - 0.800.81 – 0.90>90Total MeanMinimum Maximum

23

20173891263

2000.690.200.93

11.5108.519

45.5131.5100

Table 8: Frequency distribution of technical efficiency estimates

Source: Derived from output of the computer programme,Frontier 4.1 by Coelli (1994)

Inte

rnat

iona

l Jo

urna

l of

Agr

icul

tura

l M

anag

emen

t &

Dev

elop

men

t, 2

(3):

177

-185

, Sep

tem

ber,

201

2.

184

4.2 and 0.9 percent increase respectively inoutput of cocoa (Table 6). The value of thereturns to scale (RTS) was 1.79 indicating thatthe farmers were operating in the region of de-creasing but positive returns to scale (stage IIof the production function). Therefore, increasingthe units of inputs will not be the best option tothe farmers as it will add less to total cocoaoutput (Table 7).

Technical efficiency estimates of the sampled

cocoa farmers in Cross River Sate

The predicted farm specific technical effi-ciencies (TE) ranged between 0.20 and 0.93(Table 8). The mean efficiency of the cocoafarmers was 0.69. The 69% mean efficiency in-dicates that in the short run, there is a scope ofincreasing cocoa production by about 31% byadopting the technologies and techniques prac-ticed by the best cocoa farmers in the studyarea (Table 8).

The efficiency distribution also indicates thatmajority of the cocoa farmers (79%) were havingefficiency of between 61% and 90% while afew of them (21%) were less than 60% efficientin their production process. The high levels ofefficiency may be due to the long years offarming experience of the farmers.

CONCLUSION

This study reveals that, cocoa farmers inCross River State are not fully technicallyefficient in their resource use. The policyvariables that were identified as having sig-nificant effects (positive and negative) onthe efficiency levels of the cocoa farmersare farmers age, family size, farm sizes, ed-ucational level, and age of the farm. It isbelieved that farmers’ technical efficiency inresource use could increase since cocoa farm-ers in the area were not fully technically ef-ficient; hence, there is room for improvementin the level of this efficiency if the importantpolicy variables are addressed. Majority offarmers were in the productive age bracketand this was directly related to technical ef-ficiency of cocoa farmers in the area. It istherefore important that a policy that would

make cocoa farming attractive to personswithin this age bracket. The 69% mean effi-ciency indicates that in the short run, there isa possibility of increasing cocoa productionby about 31% by adopting the technologiesand techniques practiced by the best cocoafarmers in the study area.

REFERENCES

1- Adeniran, K. O (2007). Perspective on the role ofAgriculture in meeting the Millennium DevelopmentGoal. World Bank Report. 2- Amos, T. T. (2007). An Analysis of Productivityand Technical Efficiency of Samallholder CocoaFarmers in Nigeria. Journal of Social Science. 15(2): 127-133. 3- Coelli, T. J. (1994). A Guide to Frontier 4.1: AComputer Programme for Stochastic Frontier Pro-duction. Department of Economics. University ofNew England, Amirdale. 4- Effiong, E. O. (2005). Efficiency of Productionin Selected Livestock Enterprises in Akwa IbomState, Nigeria. Unpublished Ph.D Dissertation, De-partment of Agricultural Economics, Michael OkparaUniversity of Agriculture, Umudike, Nigeria. 5- Giroh, D. Y., Valla, W., Mohammed, A., & Peter,O. (2008). Analysis of the Technical Inefficiency ofGum Arabic Based Cropping Patterns among Farmersin the Gum Arabic Belt of Nigeria. Journal of Agri-culture and Social Science. 4:125-128. 6- Idiong, I. C. (2006). Evaluation of Technical Al-locative and Economic Efficiencies in Rice ProductionSystems in Cross River State, Nigeria. UnpublishedPh.D Thesis, Department of Agricultural Economics.Michael Okpara University of Agriculture, Umudike,Nigeria. 7- Kalirajan, K. (1981). The Economic Efficiencyof Farmers Growing High Yielding, Irrigated Ricein India. American Journal of Agricultural Economics.63(3): 566-569. 8- Kodde, F.C., & Palm, D.C. (1986). Wald Criteriafor Jointly Testing Equality and Inequality Restrictions.Econometrical. 54: 1243- 1248. 9- Nkang, N. M., E. A. Ajah, S. O. Abang and E. O.Edet (2009). Investment in Cocoa Production inNigeria: A Cost and Return Analysis of Three CocoaProduction Management Systems in Cross RiverState Cocoa Belt. African Journal of Food AgriculturalNutrition and Development. 2(2): 35-40.10- Nwaru, J. C. (2004). Rural Credit Market andArable Crop Production in Imo State of Nigeria.


Inte

rnat

iona

l Jo

urna

l of

Agr

icul

tura

l M

anag

emen

t &

Dev

elop

men

t, 2

(3):

177

-185

, Sep

tem

ber,

201

2.

185

Unpublished Ph. D Dissertation, Michael OkparaUniversity of Agriculture, Umudike, Nigeria. 11- Oluyole, K. A. and J. M. Usman (2006). “As-sessment of Economic Activities of Cocoa LicensedBuying Agents (LBAs) in Odeda Local GovernmentArea of Ogun State Nigeria”. Akoka Journal ofTechnology and Science Education. 3(1): 130-140.12- Oluyole. K. A. and R. A. Sanusi (2009). Socio-economic Variables and Cocoa Production in CrossRiver State Nigeria. Hum Ecol. 25 (1): 5-8.13- Tadesse, B. and S. Krishnamurthy (1997). Tech-nical Efficiency in Paddy Farms of Tamil Nadu: AnAnalysis Based on Farm Size and Ecological zone.Agricultural Economics. 16: 185 – 192. 14- Upton, M. (1972). Farm Management in Nigeria.Occasional Report Department of Agricultural Eco-nomics, University of Ibadan Nigeria.


www.ijamad.com

Inte

rnat

iona

l Jo

urna

l of

Agr

icul

tura

l M

anag

emen

t &

Dev

elop

men

t, 2

(3):

187

-198

, Sep

tem

ber,

201

2.

187

Comparative Cost Structure and Yield Performance

Anzlysis of Upland and Mangrove Fish Farms in

Southwest, Nigeria

Mafimisebi Taiwo Ejiola1 and Okunmadewa Foluso Yinka2

Keywords: Fish farming, Cost struc-ture, Gross revenue,Invest-ment yield, Upland andmangrove areas, Nigeria.

Received: 31 July 2011,Accepted: 20 December 2011 The bias against mangrove areas in siting fish farms

prompted a comparison of the cost structure and yieldperformance in upland and mangrove locations. Tools utilizedincluded descriptive statistics, budgetary and cash flow analysesand profitability ratios. Empirical results revealed that substantialrevenue could be realized from both farms. While the uplandfarms yielded average gross revenue per hectare per year of$9,183.53, the mangrove farms made $8,135.93 revealing aslight difference. Results of combined cash flow and sensitivityanalysis buttressed that of budgetary analysis. NPVs were$10,888.11 and $10,375.84, B/Cs were 1.28 and 1.29 and IRRwere 48.55% and 48.51% for the upland and mangrove farms,respectively. Profitability ratios were also comparable butslightly higher in the upland farms. The conclusion is thatthere was little or no difference in yield performance. However,the high risk of investment loss in years of excessive floodshould prompt investors in mangrove farms to compulsorilyinsure their farms.

Abstract


1 Department of Agricultural Economics and Extension, The Federal University of Technology, Akure, Nigeria.2 World Bank Country office, Asokoro, Abuja, Nigeria.* Corresponding author’s email: [email protected]

Inte

rnat

iona

l Jo

urna

l of

Agr

icul

tura

l M

anag

emen

t &

Dev

elop

men

t, 2

(3):

187

-198

, Sep

tem

ber,

201

2.

188

INTRODUCTION

The widening demand-supply gap for fish inNigeria attests to the fact that there is the needto explore all avenues to increase and sustainfish supply. The factors implicated in the de-mand-supply deficit situation include water pol-lution from perpetual oil spillages which resultsin dwindling catches from capture fisheries,constant upward reviews of the prices of petro-leum products which depress profit from capturefisheries, and over-fishing which involves largequantities of by-catch sold along with targetspecies (Mafimisebi, 1995, FAO 2000 and Del-gado, et al., 2003). A right step towards arrestingthe demand-supply deficit for fish is aquaculture,which involves raising fish under controlledenvironment where their feeding, growth, re-production and health can be closely monitored.Such farm-raised fish is already accounting fora considerable and rising proportion of totalfish consumed in Nigeria and other developingcountries (Delgado, et al., 2003, Mafimisebi,2007). The rapidly growing field of aquaculturehas been recognized as a possible saviour of theover-burdened wild fisheries sector and an im-portant new source of food fish for the poor(FAO 1995, Williams, 1996).

Most parts of the maritime (coastal) regionof Nigeria (about 800km coastline, FDF, 1979)are suitable for aquaculture. The coastal area isin two parts; the upland communities (whichmake up about 25.0% of the total land area)and the mangrove swamp areas, which areperennially inundated by flood or flooded formost parts of the year. The upland communitiesare characterized by fresh water while the man-grove areas have brackish water. Aquaculturefirst started in the upland parts of the coastalareas of Southwest, Nigeria. Today, a goodnumber of private commercial fish farms arefound there. However, owing to the relativescarcity and high cost of acquiring land in theupland areas, many prospective fish farmershave not been able to commence the businessof fish farming. In comparison, only a fewprivate commercial fish farms are found in therelatively land-surplus mangrove parts. The par-ticularly difficult terrain of the mangrove areas

especially with respect to its physiographic nature,water quality, distance from input source, problemof fish pond construction and the possibility offlooding, were cited as reasons why prospectiveinvestors shy away from locating their fish farmsin the mangrove areas (Mafimisebi, 1995).

The result of this is a vast expanse of mangroveland lying unused while there is serious compe-tition for land in the upland parts. For example,Ondo State, one of the coastal states in Southwest,Nigeria, has an estimated 850 and 2450 hectaresof exploitable fresh and brackish water fishinggrounds. While more than 80.0% of the freshwater fishing grounds is being exploited withabout 10.0% of this under aquaculture, lessthan 4.0% of the brackish water grounds isbeing exploited with only about 1.0% underfish farming (Ondo State Agricultural Develop-ment Project, 1996). In Nigeria as a whole, thesame situation holds. Fish farming is the leastexploited fishery sub-sector with the vast brackishwater fishing grounds almost unexploited. Lessthan 1.0% of the fresh water grounds and about0.05% of the brackish water grounds are underaquaculture to produce a current average yieldof 20,500 tonnes of fish per annum. This repre-sents only 3.12% of the estimated fish culturepotential of 656,815 tonnes per annum. Whenthe current output is compared with potentialyield, one will immediately appreciate the needfor increased effort at bringing most, if not allland suitable for aquaculture under cultivation(Ajayi and Talabi, 1984, Tobor, 1990, Falusi,2003). In fact, apart from increasing the landarea under aquaculture, astute management systemtargeted at doubling the present national aquacultureproduction rate of 1.5tonnes/ha/yr should be em-ployed. If this can be achieved, total potentialyield will increase to 1,831,000 tonnes per year,which will exceed the projected fish demand ofbetween 1,562,670 tonnes and 1,609,920 tonnesper annum by the year 2010 and beyond (Tobor,1990, Dada, 1996, FOS, 2005).

The observation that investors are biasedagainst the mangrove areas of Southwest, Nigeriain siting of their fish farms, was the motivationto compare the yield performance and profitabilityin the two fish farm locations of upland and

Comparative Cost Structure and Yield Performance Anzlysis/ Mafimisebi Taiwo Ejiola et al

Inte

rnat

iona

l Jo

urna

l of

Agr

icul

tura

l M

anag

emen

t &

Dev

elop

men

t, 2

(3):

187

-198

, Sep

tem

ber,

201

2.

189

mangrove. This is necessary because commercialfish farmers have two major objectives; theprovision of fish for human consumption andemployment opportunities, which can only berealized when maximum income and profitabilityare achieved in farmed fish production (Fagbenro,1987). The specific objectives of the study areto (1) describe the operational characteristics ofthe fish farms (2) compute and compare indicesof yield performance in the two sets of farmsand (3) identify the constraints encountered byfarmers in the two locations.


Study Area and Sampling Technique

The study was carried out in Southwest,Nigeria. Multi-stage sampling technique wasused in collecting the data analyzed in thisstudy. Two out of the six states that make upSouthwest, Nigeria; Ondo and Ogun States,were purposively selected on the basis of havingthe highest aquaculture production figures. Four(4) Local Government Areas (LGAs), two fromeach state, were also purposively selected forhaving both upland and mangrove communitieswith exploitable fishing grounds. The two LGAsselected from Ogun State were Ijebu Watersideand Ijebu East while Ilaje and Ese-Odo LGAswere selected from Ondo State. From a list ofregistered commercial fish farms got from theAgricultural Development Programme (ADP)offices in the two states, thirty (30) and fifteen(15) commercial fish farms from the upland andmangrove areas were systematically selected.

Data and Data Collection

A well-structured questionnaire was used toobtain information on characteristics and man-agement of fish farms, fish species cultured anditems of costs and returns on the productionprocess between years 2002 and 2006. Farmrecords of the farms surveyed were also madeavailable to the data collectors. The fixed costsincurred in production were calculated as annualcost or rental values of fixed items. The depre-ciated cost (obtained through the straight-linemethod) represents annual lost in value of thefacilities and equipment arising from wear and

tear. The expected useful life (in years) of fixeditems are indicated as follows: fish pond (20),boat/canoe (10), net (5), wheel-barrow (5), bowl(5), refrigerator/pumping machine (10), weighingscale (10), outboard engine (10), farm building(25) and hatchery (10). Copies of a set of ques-tionnaires were administered to the owners orfarm managers of the farms surveyed. The ques-tionnaire was earlier pre-tested on fish farmersin the riverine areas of Irele and Ado-Odo/ OtaLGAs of Ondo and Ogun States respectively.In all, forty-five (45) fish farmers provided thedata analyzed in this study.

Analytical Techniques

The data collected were analyzed using de-scriptive statistics which included frequencycounts, percentages and tables. The budgetarymodel was used to determine the level of profitgenerated. The budgetary analysis was firstcarried out for all the five years (2002-2006)pooled together and then on a year-by-yearbasis. From the results of the yearly budgetaryanalysis, certain ratios of profitability and effi-ciency were obtained. These are:

i) Operating Ratio (OR) TVC/ GR (1)ii) Returns on Sales (ROS) NP/GR (2)iii) Returns on Assets (ROA) GM/ TCA (3)Where TVC = Total Variable Cost, GR= Gross

Revenue, NP = Net Profit, GM = Gross Marginand TCA = Total Cost of Assets.

The combined cash flow and sensitivity analysiswas done to ascertain the extent of profitabilityof the aquaculture business and the factor(s) towhich profitability is responsive. The profitabilityindicators used to measure the extent of returnsfrom aquaculture are:

i) Benefit-Cost Ratio (B/C): This is the ratioof discounted costs to discounted revenue. AB/C of greater than unity is desirable for a busi-ness to qualify as a good one. Mathematically,B/C is stated as:

(4)

where


Inte

rnat

iona

l Jo

urna

l of

Agr

icul

tura

l M

anag

emen

t &

Dev

elop

men

t, 2

(3):

187

-198

, Sep

tem

ber,

201

2.

190

Bt = benefit in each project year Ct = cost in each project year n = number of years r = interest or discount rate ii) Net Present Value (NPV): This is the value

today of a surplus that a project makes over andabove what it would make by investing at itsmarginal rate. Alternatively, it is defined as thevalue today of all streams of income which aproject is to make in future. For a good business,NPV must be positive at the chosen discountfactor. Mathematically, NPV is given as:

(5)

Where Bt, Ct, n and r are as earlier defined. iii) Internal Rate of Return (IRR): It is the

rate of return that is being expected on capitaltied down after allowing for recoupment of theinitial capital. The IRR is the rate of interestwhich equates the NPV of the projected seriesof cash flow payments to zero. It is also calledthe yield of an investment. Mathematically, it isgiven as:

(6)

Practically, the IRR is usually obtained througha series of manipulations where two discountfactors give rise to two NPVs. The NPV must bepositive at the lower discount factor and negativeat the higher discount factor indicating that theproject can earn higher than the lower discountfactor and lower than the higher discount factor.In this trial and error method, according toAdegeye and Dittoh (1985), the IRR is given as:

(7)


Operational and Farm Characteristics

The total farm size (area covered by fish

ponds) of the 30 upland farms was 332,558m2

while the total number of pond units was 168.Therefore, the average size of an upland fishpondwas 1978m2. The total size of the 15 mangrovefarms was 1,226,575m2 and the number of fish-pond units was 154 giving an average size of7,965m2. Thus, fishponds have bigger sizes inthe mangrove areas. The two possible reasonsfor this finding are that land is relatively cheaperin the mangrove areas and also that most man-grove farmers use the polyculture method whilemajority of upland farmers, use the monoculturemethod. About 96.0% and 68.2% of upland andmangrove farmers respectively, were engagedin purely table fish production while the balancein each case combined table fish withfingerlings/post-fingerlings (jumpers) production.The higher proportion of mangrove farmerscombining both table fish and fish seeds pro-duction compared with the upland farms isprobably owing to the presence of larger waterbodies from which seeds of spawning fish canbe harvested and further reared to jumpers beforebeing used to stock ponds. This is a saving oncost of inputs but has a negative effect on capturefisheries since the fingerlings are the onesexpected to grow into table fish in the naturalwater bodies (Touminen and Esmark, 2003).

Majority (60.0%) of the farmers procures theirfish seeds from the wild (Table 1). Accordingto the farmers interviewed, the natural sourceof fish seeds is cheaper and more readilyavailable. The farmers in the upland partscontract out fish seeds procurement to peopleto whom they give part payment to facilitatetimely delivery. Alternatively, fish farms some-times assign that duty to some of their workersif they have enough workforce. In comparison,some workers of mangrove farms have fishseeds procurement as their major responsibility.They harvest fish seeds of various fish duringthe spawning seasons. Such harvested seedstocks are furthered reared in a special pond.During this period, stunted, deformed and un-healthy seed stocks are removed and the re-maining used for stocking fish ponds. A higherproportion of mangrove farmers got their fishseeds from the wild. Procuring fish seeds from


Inte

rnat

iona

l Jo

urna

l of

Agr

icul

tura

l M

anag

emen

t &

Dev

elop

men

t, 2

(3):

187

-198

, Sep

tem

ber,

201

2.

191

the wild by most mangrove farmers and someupland farmers, is an economic response to theproblem of acute shortage of high quality fin-gerlings from government or private hatcherywhich is capable of crippling production. Theyclaimed that where fish seeds produced inmodern hatcheries are available, the cost is pro-hibitive more so because of the transportationcost incurred. Thus, despite the fact that farmersusing fish seeds from the wild are aware of thenegative impact of their action on yield fromcapture fisheries, they asserted that they willcontinue to use this source until there is an al-ternative arrangement that is acceptable to them.There had also been frequent bloody clashesbetween fish seeds harvesters and capturefisheries fisher folks in the study area. Thismeans that an urgent alternative has to be foundto procuring fish seeds from the wild if theproblem of threatened stocks of wild fish reportedby FAO (1995, 1998 and 2000) is not to befurther compounded in Nigeria.

From Table 1, it is obvious that about 88.9%of farmers depend on the wild for their fishseeds. Only about 6.7% and 4.4% of farmersprocure their fish seeds from own modern hatch-ery and government-owned hatchery. There is aneed to re-orientate farmers away from usingfish seeds from the wild to stock their fishponds.Most of the farmers which depend on the wildindicated that they would have preferred seedstocks from specialized private or public hatch-eries because of their high quality if the pricescan be reduced and if some hatcheries can besited close to them.

Fish Species Cultured

The species commonly reared in the uplandfarms were Tilapia, Alestes, Heterotis and Catfish

and these fish species were raised by over 70.0%of the farmers. The other fish species which arereared but not as frequently as the above speciesare Mudfish, Heterobranchus, Ophiocephalus,Aeroplane fish and Mormyrus. These specieswere reared by less than 25.0% of the samplefarms. The reasons given for preference for themost popularly cultured species include (1) easeof procurement and high rates of survival of theseeds (2) easy culturing (3) fast growth and re-productive rates when supplementary feedingis practised (4) high yield and (5) high demandand price in the study area. In the mangrovefarms, however, the commonest fish speciesreared in order of frequency are Alestes (78.13%),Tilapia (60.75%), Gymnarchus (56.25%) andHeterotis (46.88%). Other fish species wereCatfish, Aeroplane fish and Ophiocephalus.Only about 30.0% of the sample mangrovefarms had these fish species reared in them.

Cost Component Analysis for Fish Farms

In both farm situations, the single most ex-pensive item of variable cost was fish feeds. Itaccounted for 51.43% of variable cost in uplandfarms and 57.79% in the mangrove farms. Thisdifference is probably owing to the astutenessof the upland farmers who formed small groupsof 5-10 and pooled their money together to buyfish feeds directly from feed milling companieswhereas most mangrove farmers bought theirfeeds from the dealers in the headquarters oftheir LGAs. Travelling long distances to theheadquarters of their LGAs to buy small quantitiesof feeds, each time they run out of stock, leadsto increased transportation cost. Added together,seed stocks and fish feeds accounted for 68.73%and 69.67% of variable cost in the upland andmangrove farms respectively. This high proportion


Source of Fingerlings Frequency Percentage

From wild From own modern hatcheryFrom other farmers rearingwild fish seeds From government hatchery Total

2703130245

60.006.67

28.894.44

100.00

Table 1: Distribution of Farmers by Source of Fish Seeds

Source: Survey data, 2007

Inte

rnat

iona

l Jo

urna

l of

Agr

icul

tura

l M

anag

emen

t &

Dev

elop

men

t, 2

(3):

187

-198

, Sep

tem

ber,

201

2.

192

accounted for by seed stocks and fish feeds asitems of variable expenses is in accordancewith the findings by Zadek (1984), Inoni, (1992)and Mafimisebi, (2003) that cost of feeds andseed stocks accounted for more than 50.0% oftotal production cost in aquaculture (Table 2).

The depreciated average fixed cost, per hectareof upland fish farm was $8,175.23 in the fiveyears covered by the study. The correspondingvalue for the mangrove farm was $10,890.00.The depreciated cost of pond construction andvehicle/boats carried 10.67% and 11.98% re-spectively in the upland farms while the sameitems accounted for 8.08% and 16.18% respec-tively in the mangrove farms. The cost of pondconstruction was higher in the mangrove farmsbecause the mangrove species have had to becleared first before pond construction properbegins. Not only that, the problem of pond edgestabilization gulps a lot of money comparedwith upland ponds. This is more so becausefish pond construction in the mangrove areasinvolves putting special structures in place toprevent escape of fish into the wild during slightor excessive flood. Table 2 also shows that landis cheaper in the mangrove areas. However, the

cheap cost of land as an item of fixed cost iseroded by the heavy expense on pond constructionin the mangrove parts.

For the two farms, labour was the single mostexpensive item of fixed cost. While labour ac-counted for 41.83% of fixed cost in the uplandfarms, the value for the mangrove farms was48.24%. Also revealed in Table 2 is the fact thatseed stocks constituted about 17.00% of variablecost in the upland farms while the correspondingvalue in the mangrove farms was 12.00%. Thisis attributable to the fact that more uplandfarmers than mangrove farmers procured seedstocks from sophisticated private and publichatcheries. The seed stocks bought from suchhatcheries were more expensive than the onesbought from local hatcheries.

Gross Revenue

Gross Revenue (GR) is the amount realizedfrom sale of table fish, fingerlings and jumpers.However, because revenue from sales of fingerlingsand jumpers is negligible, only revenue fromtable fish production is considered in this study.The information on GR from the various fishspecies cultured is provided in tables 3 and 4.


Fixed Items Upland Farms

($)% Composition

Mangrove Farms

($)% Composition

LandPond constructionFarm buildingsVehicles + boats Nets Boreholes + Water pumpsWheel Barrow + Basins Generators/Deep freezers/ Weighing ScaleLocal hatchery + Fencing materialsLabour (permanent)Sub-totalVariable ItemsFingerlings & Jumpers Fish feedsFertilizer + Other ChemicalsTransportation + FuellingRepairs + Maintenance Casual labourSub-total

335.54872.41374.58979.64183.54

1,814.9896.6435.9835.10

3,419.838,175.23

5,721.4717,012.708,740.024,352.903,034.722,081.96

33,078.49

4.1010.674.5811.982.25

22.531.180.040.43

41.83100.00

17.3051.432.64

13.169.186.29

100.00

120.36875.57202.85

1,761.66624.54

1,103.20116.65344.37483.12

5,252.9510,890.00

3,549.0617,264.761,614.423,797.432,479.211,171.51

29,876.39

1.108.081.86

16.185.74

10.131.073.164.44

48.24100.00

11.8857.795.40

12.718.303.92

100.00

Table 2: Fixed and Variable Cost for One-Hectare Upland and Mangrove Fish Farms

Source: Survey data, 2007.Note: For the period covered by the data used for this study, the average exchange rate was N127= $1 whilethe value fluctuated between N125 and N129.

Inte

rnat

iona

l Jo

urna

l of

Agr

icul

tura

l M

anag

emen

t &

Dev

elop

men

t, 2

(3):

187

-198

, Sep

tem

ber,

201

2.

193

While the upland farms made net revenue of$87,171.39 per hectare for the period studied,the mangrove farms got $81,446.06 per hectarefor the same period. Therefore the average netprofit per hectare per year was $9,183.53 and$8,135.93 in the upland and mangrove farmsrespectively. Table 3 shows that in the uplandparts, Heterotis contributed the highest proportionof GR followed by Gymnarchus and Alesteswhile Ophiocephalus, Heterobranchus andMormyrus together amounted to just about one-fifth of GR. It can thus be concluded that Het-erotis, Gymnarchus and Alestes were the majorcommercial species cultured in the upland areas.Table 4 showed that Alestes accounted for thehighest proportion of GR followed by Heterotis,Gymnarchus and Tilapia, in that order in themangrove farms. Catfish, Ophiocephalus andAeroplane fish contributed less than one-fifthof GR. The major commercial species in themangrove parts were Alestes, Heterotis, Gym-narchus and Tilapia

For both farms, there was positive net revenueindicating that aquaculture is operating at aprofit in the two locations. This finding ofpositive net revenue in the mangrove farmscontradicts earlier findings by Inoni (1993),Falusi (2005) and Zadek (1984) that mangrovefarms in Delta and Ogun States, Nigeria andPort Said, Egypt respectively, sustained lossesduring an operational period of between 1987and 2003.

Profitability and Efficiency Ratios

The year-by-year results of the budgetaryanalysis are shown in Table 5. From the valuesgiven in the table, profitability ratios that enabledus to arrive at a conclusion as to the efficiencyof operation of the two fish farms, were calculated.

The data presented in Table 6 showed theprofitability ratios by year of the two sets offarms. A decrease in OR over time is an indicationof a good and efficient business. A decline inOR in the study indicates either increasing TR


Fixed Items Gross Revenue per year ($)

Heterotis Gymnarchus Alestes Tilapia Catfish Heterobranchus MudfishOphicephalus Mormyrus Total

2002

4,336.183,869.692,543.661,146.18905.79735.33248.05659.290.44

14,444.61

2003

4,711.203,710.661,687.221,100.481,020.44821.90185.26584.210.32

13,821.70

2004

4,701.754,152.843,150.561,425.831,267.511,170.59

321.26818.12

0.5717,014.75

2005

5,475.054,475.543,799.761,297.551,529.091,261.75

401.57924.16

0.6919,165.17

2006

6,544.365,565.614,302.561,333.611,705.111,604.48

354.331,314.05

1.0622,725.17

Total2002-200625,768.5421,774.3415,483.76

6,309.386,427.945,594.051,510.474,299.83

3.0787,171.39

Table 3: Gross Revenue on a One-Hectare Upland Farm

Source: Survey data, 2007.

Fixed Items Gross Revenue per year ($)

Heterotis Gymnarchus Alestes Tilapia Catfish Ophiocephalus Aeroplane fish Total

2002

3,032.482818.323,830.151,926.611,529.54

733.536.90

13,877.53

2003

2,869.132,583.133,520.861,637.281,338.63618.66

6.0212,573.72

2004

3,618.943,276.634,125.832,191.201,710.50867.52

9.6315,800.24

2005

3,834.443,784.874,821.942,792.601,993.951,009.74

14.0118,251.54

2006

4,466.873,961.485,624.023,065.332,487.391,321.36

16.5620,943.02

Total2002-200617,821.8716,424.4321,922.8011,613.029,060.164,550.81

53.1181,446.06

Table 4: Gross Revenue on a One-Hectare Mangrove Farm

Source: Survey data, 2007.

Inte

rnat

iona

l Jo

urna

l of

Agr

icul

tura

l M

anag

emen

t &

Dev

elop

men

t, 2

(3):

187

-198

, Sep

tem

ber,

201

2.

194

or decreasing TVC. For the upland farms, ORwas 0.447 in 2002 which decreased to 0.397and 0.308 in 2003 and 2004. The value tookan upward turn to 0.379 which was maintainedin 2006. The same pattern was observed inthe mangrove farms. OR fell from 0.418 in2002 to 0.405 and 0.292 in 2003 and 2004 re-spectively but picked up to 0.367 in 2005which remained same in 2006. For the periodstudied, average OR was 0.382 in the uplandfarms and 0.370 in the mangrove farms. Judgingby these ratios, the mangrove farms seemedto promise a better efficiency in future yearsas OR was falling faster than in the uplandfarms. The increase in OR in years 2005 and2006 on both farms is clearly not a desirablesituation. The farmers must do all that ispossible to achieve a consistently decreasingOR. This can be achieved by a more efficientuse of farm resources. For example, feedingfish beyond a stipulated market weight shouldbe avoided as the rate of growth slows downcompared with the quantity of feeds consumed.Also, farmers should explore avenues for widermarket outlets so that mature fish can bepromptly disposed off. This scenario will leadeither to a decreasing TVC or an increase inTR which will depress OR.

An increasing return on sales over time in-dicates a stable, profitable and efficient business.Return on sales was constant in the periodstudied on both farms. It was 0.527 for theupland farms and 0.499 for the mangrovefarms. Fish farmers in both farm locationsneed to take steps to ensure an increasingreturn on sales.

The indication that assets are being more in-creasingly utilized is increasing returns on assets.On the upland farms, there was an increasingtrend of returns to assets from 2002 to 2005 butthere was a fall in the value in 2006. On themangrove farms, the trend in returns on assetswas towards an increase except in years 2003and 2006 in which the figures fell below theyear preceding them.

Combined Cash Flow and Sensitivity Analysis

Some assumptions were necessary in carryingout this analysis. These assumptions are as fol-lows:

(1) The average bank lending rate to agriculturein the thirteen (13) years covered by the analysisis 25.0%.

(2) A risk-discounted factor of 5.0% is addedto the bank lending rate meaning that a discountfactor (DF) of 30% is used.


Total Variable Cost ($) Gross Revenue ($) Gross Margin ($) Net Profit ($)

Year20022003200420052006

Upland6,456.505,481.225,244.867,272.518,623.40

Mangrove5,795.915,090.614,612.346,695.127,682.41

Upland14,365.8313,821.7017,014.7519,165.1722,725.16

Mangrove13,877.5312,573.7215,800.2418,251.5420,943.02

Upland7,988.078,340.4811,769.8911,892.6614,101.76

Mangrove8,081.177,483.11

11,187.9011,556.4413,260.62

Upland7,608.727,280.608,962.55

10,095.2811,970.52

Mangrove6,931.376,280.167,891.719,116.05

10,460.36

Table 5: Year by Year Budgetary Analysis of a One-Hectare Farm

Source: Survey data.

Operating ratios Return on sales Return on sales

Year20022003200420052006Average

Upland0.4470.3970.3080.3790.3790.382

Mangrove0.4180.4050.2920.3670.3670.370

Upland0.5270.5270.5270.5270.5270.527

Mangrove0.4990.4990.4990.4990.4990.499

Upland1.6801.7542.4752.5012.9652.275

Mangrove1.4341.3271.9852.0502.3521.830

Table 6: Profitability and Efficiency Ratios for a One-Hectare Farm

Source: Survey data.

Inte

rnat

iona

l Jo

urna

l of

Agr

icul

tura

l M

anag

emen

t &

Dev

elop

men

t, 2

(3):

187

-198

, Sep

tem

ber,

201

2.

195

(3) There is a 20.0% and 10.0% projected an-nual increase in variable cost and unit price offish between 2006 and 2014. This is in accordancewith the farm management maxim which saysit is better to be optimistic about cost rise andpessimistic about revenue increase in the esti-mation of future profitability of a business(Adesimi, 1985).

The result of the combined cash flow andsensitivity analysis for the upland and man-grove farms are shown in Tables 7 and 8 re-spectively. The results indicate that aquacultureis profitable at both locations at the assumedbank lending rate in spite of prices of keyproduction inputs rising faster than outputprice. For the upland farms, the NPV stoodat $10,887.24, the B/C was 1.28 and IRRwas 48.55%. The corresponding values forthe mangrove farms were $10,375.84, 1.29and 48.51%. Thus, the results are comparableand do not show any considerable differencein yield performance between the two typesof farms. While at the assumed bank lendingrate, the upland farms would return $0.19 forevery $0.79 invested, the farms located inthe mangrove areas will also return approxi-mately $0.19.

Constraints to Upland and Mangrove Fish

Farming

Fish farmers in the two farm locations wereasked to rank the constraints identified in theirbusiness. The problems encountered by themangrove farmers in rank order were (1) financialconstraints; (2) high and rising cost of feeds;(3) flooding which leads to total loss of investmentwhenever it happens as fish escape into thewild. Numerous studies have named potentiallynegative effects of escaped farmed fish on wildpopulations (Naylor et al., 2000); (4) silting upof ponds which result in massive death ofcultured fish; (5) pests which include snakes,water-dogs and piscivorous birds; (6) attack bycapture fishermen during sourcing of fish seedsfrom wild; (7) water pollution and (8) inadequateaccess to extension services. Only about 30%of mangrove farmers had had a contact with ex-tension agents since commencement of business.

The problems commonly encountered by farm-ers in the upland areas were (1) financial con-straints occasioned by high running costs; (2)drying up of ponds owing to seepage of waterthrough dykes; (3) massive loss of fish owingto polluted or high-temperature water; (4) scarcityof high quality seed stocks and (5) problems of


Table 7: Cash Flow and Sensitivity Analysis for a One-Hectare Upland Farm (2002-2014)

Year Cost ($) Revenue

($)

Incremental

Benefit ($)

DF30%

NPV

30% ($)

DF50%

NPV50% ($)

DiscountedCost ($)

DiscountedRevenue ($)

20012002200320042005200620072008200920102011201220132014

21,774.186,456.505,481.225,244.867,272.518,623.40

10,348.0812,417.7014,901.2417,881.492,142.822,576.11

30,899.2137,079.47

-14,444.5713,821.7017,014.7419,165.1722,725.1724,997.4527,497.1930,246.9133,271.6036,598.7640,258.6444,284.5048,712.95

-21,774.187,988.078,340.4811,769.8911,892.6614,101.7614,649.3715,079.4915,345.6715,390.1215,140.9814,509.3013,385.2911,633.90

0.7690.5920.4450.3500.2690.2070.1590.1230.0940.0730.0560.0430.0330.025

-16,744.354,728.943,711.514,119.463,199.132,919.062,329.251,854.781,442.491,123.48847.89623.90441.71290.85

10,888.11

0.6670.4440.2960.1980.1320.0880.0590.0390.0260.0170.0120.0080.0050.003

-14,523.383,546.712,468.782,330.441,569.831,240.96864.31588.10398.99261.63181.69116.0766.9334.90

-854.04

16,744.353,822.252,439.141,835.701,956.301,785.041,645.351,527.381,400.721,305.511,201.641,107.221,019.67926.98

38,717.24

-8,551.196,150.665,955.165,155.434,704.113,974.593,382.152,843.212,428.832,049.531,731.121,461.391,217.84

29,031.78

Source: Field data and projected figures Notes: (1) 2001 is the investment year (year zero), so there is no revenue

(2) Costs and Revenues for 2002-2006 are actual flows recorded by the fish farms(3) Cost and revenues for 2007 – 2014 are projected figures NPV at 30% = 10,888.11 IRR = 48.55% B/C = 1.28

Inte

rnat

iona

l Jo

urna

l of

Agr

icul

tura

l M

anag

emen

t &

Dev

elop

men

t, 2

(3):

187

-198

, Sep

tem

ber,

201

2.

196

theft which can lead to over-night harvesting offish ponds with marketable fish if security isnot beefed up around ponds. These are the prob-lems that must have solutions proffered to themfor the operation of these farmers to be en-hanced.


The study explored the operational character-istics of upland and mangrove farms, comparedcosts and profitability of investment in the twolocations and determined the production variablesto which profitability is more sensitive. Thestudy also examined the constraints to fishfarming in the two locations.

Empirical results show that mangrove farmsare about four times bigger in size than theupland farms. Monoculture method was prevalentamong mangrove farmers while the uplandfarmers mostly practised polyculture. The com-mercial species reared in the study area wereTilapia, Alestes, Heterotis and Gymnarchuswhile Ophiocephalus, Catfish, Heterobranchusand Mormyrus were the minor commercialspecies. The depreciated fixed cost in the uplandfarms was lower than that of the mangrovefarms. However, the level of variable cost inthe upland farms was greater than that of themangrove farms. In both farms, the cost oflabour was the single most expensive item of

variable cost. G.R per hectare was comparablein the two farm situations but slightly higher inthe upland farms. Profitability ratios which in-dicate efficiency did not show any considerabledifference in the yield of investment from thetwo farm locations. The result from the combinedcash flow and sensitivity analysis shows thatinvestment in aquaculture is profitable at bothfarm locations. All performance indicators showthat profitability is not different between farmsin the two locations to justify the avid preferencefor the upland locations in the siting of fishfarms in the coastal areas of Southwest, Nigeria.

The magnitude of cost involved in establishingand managing a fish farm is clearly beyond thataffordable by a peasant farmer. Investible fundsin form of loans should be made available toprospective investors wishing to site their farmsin the mangrove areas at affordable interestrate. The study has shown that fish farmers canrepay loans advanced to them conveniently ifgiven a moratorium of two years.

There is also the need to encourage investorsin hatcheries to produce fish seeds for use byfish farmers especially in the mangrove areaswhere majority of the farmers depend on thewild for their fish seeds. Once the governmenthas succeeded in attracting investors in hatcheriesto the mangrove areas, a campaign against theuse of fish seeds from the wild in the study area


Table 8: Cash Flow and Sensitivity Analysis for a One-Hectare Mangrove Farm (2002-2014)

Year Cost ($) Revenue

($)

Incremental

Benefit ($)

DF30%

NPV

30% ($)

DF50%

NPV50% ($)

DiscountedCost ($)

DiscountedRevenue ($)

20012002200320042005200620072008200920102011201220132014

20,898.935,795.915,090.614,612.346,695.127,682.419,218.7311,062.6713,275.2015,930.2419,116.2922,939.5527,527.4633,032.95

-13,877.5312,573.7215,800.2418,251.5420,943.0223,037.3325,341.0627,875.1630,662.6830,662.6833,728.9537,101.8440,812.13

-20,898.938,081.627,483.11

11,187.9011,556.4413,260.6213,818.4414,278.3914,599.9614,732.4414,612.6614,162.3013,284.5711,860.28

0.7690.5920.4450.3500.2690.2070.1590.1230.0940.0730.0560.0430.0330.025

-16,071.284,784.323,329.983,915.773,108.682,744.952,197.131,756.241,372.401,075.47818.31608.98438.39296.51

10,375.84

0.6670.4440.2960.1980.1320.0880.0590.0390.0260.0170.0120.0080.0050.003

-9,215.183,588.242,215.002,215.201,525.451,166.93815.29556.82379.60250.45175.35113.3066.4235.58

-835.92

16,071.283,431.182,265.321,614.321,800.981,661.121,465.801,360.711,247.871,162.911,070.51986.40908.41825.82

35,801.77

-8,215.505,595.305,530.094,909.674,335.213,662.933,116.952620.272,238.391,888.821,595.381,346.801,122.33

46,177.60

Source: Field data and projected figures Note: NPV at 30% =10,375.84 IRR = 48.51% B/C = 1.29

Inte

rnat

iona

l Jo

urna

l of

Agr

icul

tura

l M

anag

emen

t &

Dev

elop

men

t, 2

(3):

187

-198

, Sep

tem

ber,

201

2.

197

should be launched. As soon as the hatcheriescan produce enough seed stocks to satisfy allidentified fish farmers, the practice of procuringfish seeds from the wild should be banned. Thisis a matter of priority if yield from aquacultureis to be increased and natural fisheries resourcesconserved. Finally, since the level of capital in-vestment for establishing fish farms is veryhigh, the government can subsidize cost of fishfeeds for new investors in the mangrove areasonly in the first year of operation. This mayserve to attract investors into the area so thatthe vast mangrove land can be put to productiveuse. There is also the need to step-up extensionvisits to fish farmers.

It is also recommended that farmers in themangrove areas take policy with the NigerianAgricultural Insurance Company so that theycan be indemnified if there is loss of investmentfrom fish escapes during periods of excessiveflood.

All performance indicators show that prof-itability is not different between farms in thetwo locations to justify the bias against themangrove areas in the siting of fish farms in thecoastal areas of Southwest, Nigeria. Solvingsome of the identified problems of fish farmersin both locations is a step towards cheap and af-fordable animal protein production especiallyin the vast mangrove areas with its hydrographiccharacteristics.

ACKNOWLEDGEMENT

Thanks are due to Dr. I. Ajibefun of TheFederal University of Technology, Akure, Nigeria,Professor Yoshi Matsuday and Ms. Ann Shriver,and two other anonymous reviewers who aremembers of International Institute of FisheriesEconomics and Trade (IIFET) that revieweddeveloping countries’ aquaculture economicspapers for various awards and travel grants forthe 2006 conference in Portsmouth, U.K. IIFETis gratefully acknowledged for a partial travelgrant of $900 to enable the first author attendand give a work-in-progress version of thispaper in IIFET 2006, Portsmouth. Various othermembers of the Aquaculture Economics Sessionin the said conference are thanked for raising

useful comments that have been addressed inthis revised version of the paper.

REFERENCES

1- Adegeye, A.J. & Dittoh, J.S. (1985). Essentialsof Agricultural Economics. Impact Publishers Nig.Ltd., Ibadan. p 251.2- Adesimi, A.A. (1985). Farm Management Analysis.Tintner Publishing Company, pp 105-206.3- Ajayi, T.O. & Talabi, S.O. (1984). The Potentialsand Strategies for Optimum Utilization of theFisheries Resources of Nigeria. NIOMR TechnicalPaper, No. 18, P 24.4- Dada, B.F. (1976). Present Status and Prospectsfor Aquaculture in Nigeria. NIOMR Technical Paper52: 5-12.5- Delgado, C.L., Wada, N., Rosegrant, M.W.,Meijer, S. and M. Ahmed. (2003). Fish to 2020:Supply and Demand in Changing Global Markets.International Food Policy Research Institute, Wash-ington D.C, World Fish Center, Penang, Malaysia P223.6- Fagbenro, O.A. (1987). A Review of the Biologicaland Economic Principles Underlying CommercialFish Culture. Journal of West African Fisheries,11(2): 171-177.7- Falusi, O.A. (2003). Poverty Analysis of FishFarmers in Abeokuta Metropolis. An UnpublishedProject Report, Department of Agricultural Economicsand Farm Management, University of Agriculture,Abeokuta, Nigeria. p 57.8- FAO (1995). The State of World Fisheries andAquaculture, Rome.9- FAO (1998). The State of World Fisheries andAquaculture, Rome.10- FAO (2005). The State of World Fisheries andAquaculture, Rome.11- Federal Department of Fisheries (1979). FisheriesStatistics of Nigeria. pp 1-32.12- FOS (2000) Projected Fish Demand and Supplyin Nigeria (1999-2010). Federal Office of Statistics,Abuja, Nigeria. 13- Inoni, O.E.,(1992). Financial Analysis of FishFarming in Delta State, Nigeria. Unpublished M.ScDissertation, University of Ibadan. p 88.14- Mafimisebi, T.E. (1995). Profitability and YieldPerformance of Selected Fish Ponds in Ilaje Ese-Odo Local Government Area of Ondo State, Nigeria.Unpublished M.Sc Dissertation, University of Ibadan.p. 83.15- Mafimisebi, T.E. (2003). Yield Performance ofCommercialized Upland Fish Farms in Ondo State


Inte

rnat

iona

l Jo

urna

l of

Agr

icul

tura

l M

anag

emen

t &

Dev

elop

men

t, 2

(3):

187

-198

, Sep

tem

ber,

201

2.

198

of Nigeria. Nigerian Journal of Animal Production.30 (2): 217-228.16- Naylor, R.L., Goldburg, R.J., Primavera, J.H.,Kautsky, N., Beveridge, M.C.M., Clay, J., Folke,C., Lubchenco, Mooney J. H., & Troell, M. (2000).Effects of Aquaculture on World Fish Supplies.Nature 405: 1017-1024.17- Ondo State Agricultural Development Project(1996). Aquaculture Potentials of Ondo State ofNigeria. pp 1-10.18- Tobor, J.G.(1990). The Fishing Industry inNigeria: Status and Potentials for Self-Sufficiencyin Fish Production. NIOMR Technical Report. No.54: 1-23.19- Touminen, T.R., & Esmark, M. (2003). Foodfor Thought: The Use of Marine Resources in FishFeeds. WWF Report 02/03. Norway: World WildlifeFund.20- Williams, M.J. (1996). Transition in the Contri-bution of Living Aquatic Resources to SustainableFood Security”. In Perspectives in Asian Fisheriesed Sena S. De Silva. Makati City, The Philippines:Asian Fisheries Society.21- Zadek, S.,(1984) Development de l’Aquaculturein Egypte. Reference a la Farme de Reswa (PortSaid et Froposition d’une Politique National Equacole.Ph.D Thesis, Toulouse Institut National Polytechniquede Toulouse P. 151.


Inte

rnat

iona

l Jo

urna

l of

Agr

icul

tura

l M

anag

emen

t &

Dev

elop

men

t, 2

(3):

199

-207

, Sep

tem

ber,

201

2.

199

Effectiveness of Extension Services in Enhancing

Outgrowers’ Credit System: A Case of Smallholder

Sugarcane Farmers in Kisumu County, Kenya

Abura Odilla Gilbert 1, Barchok Kipngeno Hillary1 and Onyango Christopher Asher2

Keywords: Public extension service,Private extension service,Outgrowers’ Credit sys-tem, Effectiveness, WesternKenya

Received: 20 January 2012,Accepted: 20 May 2012 The purpose of this study was to investigate the role of ex-

tension services in enhancing effectiveness of outgrowers’credit system in Kisumu County, Kenya. The study specificallysought to determine whether public and private extensionservices play a significant role in enhancing effectiveness ofout-growers’ credit system among smallholder sugarcanefarmers. A total of 110 small scale farmers were randomlyselected for the study. A closed ended questionnaire was usedto collect data from farmers. Both descriptive and inferentialstatistics were used for data analysis. The findings indicatedthat both public and private extension services were insignificantin enhancing effectiveness of outgrowers’ credit system.Further, the findings indicated that there was no significantdifference between public and private sector in provision ofextension services. The findings suggest that for outgrowers’credit system to be effective in terms of creation of awarenessabout credit, accessibility, timely supply of credit, supervisionof credit and provision of extension advice on credit utilization,both public and private extension services should be intensifiedand coordinated to avoid duplication. The results also suggestthat sugarcane factory extension division should be strengthenedjust like in the coffee and tea sub-sectors.

Abstract


1 Chuka University College, P. O. Box 109, Chuka, Kenya. 2 Egerton University, P. O. Box 536, Egerton, Kenya.* Corresponding author’s email: [email protected]

Inte

rnat

iona

l Jo

urna

l of

Agr

icul

tura

l M

anag

emen

t &

Dev

elop

men

t, 2

(3):

199

-207

, Sep

tem

ber,

201

2.

200

INTRODUCTION

The importance of agriculture to the Africaneconomies is stressed due to the fact that agri-culture remains the principal occupation of themajority of people, constitutes the largest pro-duction sector, and produces an average of 32%of GDP, major sources of raw materials for in-dustries and a significant purchaser of the coun-tries manufacturers and services (Agbamu,2005). In Kenya for example, the economy isheavily dependent upon agricultural sector andas the World Bank (2007) report indicates, thecountry’s future will considerably depend onproductivity of smallholder farms. Agricultureis by far the single largest economic sector inKenya and accounts for about 30% of GDP,over 60% of the exports, 75% of the total labourforce and provides 80% of industrial raw materials(Economic Survey, 2007; Kenya Sugar ResearchFoundation, 2007, Government of Kenya, 2005).Since independence, smallholder agriculturegained ground from mere provision of subsistenceand minimal marketed surplus to account forover three quarters of agricultural productionand 85% of agricultural employment (GoK,2005, World Bank, 2007). Sugarcane farmingis one such subsector that contributes to the na-tional economy. According to Guda et al., (2001)smallholder farmers accounts for 89% of thetotal area under sugarcane farming in Kenya.This provides an investment opportunity. How-ever, this is only possible if the problemsaffecting it are addressed. Some of the mainproblems include shortage of sugarcane due tolack of systemic and synchronised sugarcanedevelopment, poor crop husbandry practices,poor cane varieties and qualities in some factoryzones, poor harvesting methods, poor manage-ment in some factories affecting factory efficiencyand output and inadequate agricultural productioncredit among others (Guda et al., 2001). Howeverthe most notable problem is complaints due todelayed payment dues to the farmers after canedelivery. These problems have elicited diversereactions from the farmers. The most severe re-actions are cases of burning the cane crop bythe farmers in their own farms, so as to turn toother lucrative enterprises like maize or bean

seed production (Agribusiness DevelopmentSupport Project Annual Report, 2001).

Agricultural extension is considered to be animportant service in increasing agricultural pro-ductivity and attaining sustainable development(Kibet, et al., 2005). Its role is to help peopleidentify and address their needs and problems.There is a general consensus that extensionservices if successfully applied, should result inoutcomes which include observable changes inattitudes and adoption of new technologies, andimproved quality of life based on indicatorssuch as health, education and housing. It hasbeen recognized that agricultural extension ac-celerates development in the presence of otherfactors such as markets, agricultural technology,availability of supplies, production incentivesand transport (Kibet, et al., 2005). Koyenican(2008) equates help in extension to empoweringall members of the farm households to ensureholistic development. This is because agriculturalextension brings about changes, through educationand communication in farmers attitude, knowl-edge and skills.

The performance of the public agriculturalextension service in Kenya has been a verycontroversial subject (Gautam and Anderson,1999). The system has been perceived as top-down, uniform (one-size-fits-all) and inflexibleand considered a major contributor of the poorperforming agricultural sector (Government ofKenya, 2005). Thus there has been a desire toreform extension in to a system that is cost ef-fective, responsive to farmer’s needs, broadbased in service delivery, participatory, account-able and sustainable. As a result of ineptness inthe public extension system, private agriculturalextension system has emerged comprising ofprivate companies, non-governmental organi-zations (NGO’s), community based organizations(CBO’s) and faith based organizations (Nambiroet al., 2005 and Rees et al., 2000).

Agricultural extension as a public sector in-stitution has an obligation to serve the needs ofall agricultural producers, either directly or in-directly (Anderson, 2007). This is because publicsector extension is a public good. The Kenyagovernment has tried a number of extension

Effectiveness of extension services/ Abura et al

Inte

rnat

iona

l Jo

urna

l of

Agr

icul

tura

l M

anag

emen

t &

Dev

elop

men

t, 2

(3):

199

-207

, Sep

tem

ber,

201

2.

201

models and styles, including the progressive(farmer approach) model, integrated agriculturalrural development approach, farm management,training and visit, farming systems approachesand farmer field schools. All these approacheshave emerged with varying level of success fordifferent groups. However, the effectiveness ofextension services in enhancing effectiveness ofoutgrowers’ credit system among sugarcanefarmers in Kisumu County of Kenya has notbeen examined. Thus the present study was setwith the premise that both outgrowers’ creditand extension service are instruments for pro-moting agricultural development and that an ef-ficient and effective extension service is importantin enhancing effectiveness of outgrowers creditsystem. Credit to farmers is an important instru-ment in improving productivity. Indeed as Wangia(2001) noted, it is a prerequisite to the adoptionof improved agricultural technologies for thesmallholder farmers. Nevertheless, for creditsystem to help the smallholder farmers it shouldbe tied to improved technologies, remunerativeprices for the farmers’ output and good extensionnetwork (Ogunsumi, 2004).This paper presentsresults on the role of extension services in en-hancing effectiveness of outgrowers’ credit systemamong smallholder cane farmers in Kenya.

Conceptual Framework

This study formulated a conceptual modelthat encompassed major variables and their pos-sible patterns of influence on each other andeventually on effectiveness of outgrowers’ creditsystem. The effect of the extension servicesnamely public and private services are mediatedby level of education, farm size and farmersperiod of residence. What this structural modelindicates therefore is that the moderator variablesinfluence adoption of sugarcane technologiesdisseminated by extension agents whether frompublic or private sector. In view of this model,the theory underpinning this study is that,adoption is complex and multifaceted process.While the main activity of extension centers onincreasing production, this study concentratedon implementation of such activities. These are;creation of awareness about credit, accessibilityto credit, timely supply of credit, supervision ofcredit and provision of extension advice oncredit utilization (Table 1).

MATERIALS AND METHODES

Ex-post facto survey design was adopted forthis study. Kisumu county in Kenya was purposelyselected because of its uniqueness in that thecounty boasts of three major sugar factories.


Figure 1: The Conceptual Framework on the Role of Extension in EnhancingEffectiveness of Outgrowers’ Credit System

Inte

rnat

iona

l Jo

urna

l of

Agr

icul

tura

l M

anag

emen

t &

Dev

elop

men

t, 2

(3):

199

-207

, Sep

tem

ber,

201

2.

202

These are; Miwani, Muhoroni and Chemelilwhich were established in the years 1923, 1966and 1968 respectively. The County has favourablemoderate climatic conditions, with temperaturesaveraging 27o C and receives bimodal rainfallranging from (560 -1630) mm per annum.Kisumu County comprises of the main topo-graphical land formations namely, the Nandihills, the Nyando plateau and Kano plains whichare sandwiched between two hills. The Kanoplains comprise predominantly black cottonclay soils derived from igneous rocks. TheCounty’s altitude range from 1000-1860 Mabove sea level. The target population was thesugarcane farmers in Kisumu county. A total of110 smallholder cane farmers were randomlyselected for the study but only 108 farmersquestionnaire were useful for analysis. A closedended questionnaire was used to collect data by

personal interviews. The information gatheredwas analysed using both descriptive and infer-ential statistics.

RESULTS AND DISCUSSIONS

Public extension Services

Tables 1, 2 and 3 below shows data on differentextension activities done by public sector ex-tension service for cane farmers. Table 1 showresult on the effect of public extension servicesin relation to creation of awareness. Creation ofawareness was one of the activities used tomeasure public extension services. To elicit in-formation on creation of awareness, the farmerswere asked to respond to statement designed toelicit negative responses on performance resultingto creation of awareness. A 5-point likert scalewas constructed to record these responses. Table1 show results on public extension services in


Activity SD D U A SA Total

Creation of Awareness f (n=23)%

2.08.7

17.073.9

--

2.08.7

2.08.7

23.0100.0

Table 1: Creation of Awareness on credit facility

Key (SD= Strongly Disagree, D= Disagree, U= Undecided, A= Agree, SA= StronglyAgree)


Land Preparation f (n=23)%

2.08.7

14.060.2

--

6.026.0

1.04.3

23.0100.0

Table 2: Land Preparation

Key (SD=Strongly Disagree, D=Disagree, U=Undecided, A=Agree, SA= StronglyAgree)


Appropriate Input Use f (n=23)%

2.08.7

7.030.4

626.0

6.026.0

2.08.7

23.0100.0

Table 3: Appropriate Input Use

Key (SD=Strongly Disagree, D=Disagree, U=Undecided, A=Agree, SA=StronglyAgree)


Creation of Awareness f %

16.020.0

51.063.8

--

9.011.3

4.05.0

80.0100.0

Table 4: Creation of awareness on credit facility


Inte

rnat

iona

l Jo

urna

l of

Agr

icul

tura

l M

anag

emen

t &

Dev

elop

men

t, 2

(3):

199

-207

, Sep

tem

ber,

201

2.

203

relation to creation of awareness.The results in Table 1 indicated that majority

of farmers (82.6%) receive information on creditfacility from public extension providers.

Advice on land preparation was one of the ac-tivities used to measure public extension services.To elicit information on land preparation, thefarmers were asked to respond to statement de-signed to elicit positive knowledge of performanceresulting to land preparation. A 5-point likertscale was constructed to record these responses.Table 2 show results on public extension servicesin relation to land preparation.

The result in table 2 showed that land prepa-ration as an activity has not been satisfactorilyaddressed by public service extension officersas reflected by the high percentage (68.9%) offarmers who disagreed with the positive statement.This suggests that perhaps the declining caneproduction in Kisumu County is due to poorand inadequate land preparation, culminatingfrom inadequate machinery to prepare land forcane growing.

Advice on appropriate use of inputs was oneof the activities used to measure public extensionservices. To elicit information on the use of ap-propriate input, the farmers were asked torespond to statement designed to elicit positiveknowledge of performance resulting to use ofappropriate input. A 5-point likert scale wasconstructed to record these responses. Table 3show results on public extension services in re-lation to appropriate use of input.

The results in table 3 showed that 39.1% of thefarmers have not received services on appropriateuse input from public sector extension services.This suggests that farmers do not know whetherthe inputs they use are appropriate or not.

Private extension service

Private sector extension may play a predominantextension role for particular inputs, particularenterprises / commodities and for particularfarmer’s in particular geographical areas. Thisenables farmers to benefit from increased incomesand economic security. Tables 4, 5, and 6 showsdata on different extension activities done byprivate sector extension service for cane farmers.

Table 4 show result on the effect of privateextension services in relation to creation ofawareness. Creation of awareness was one ofthe activities used to measure private extensionservices. To elicit information on creation ofawareness, the farmers were asked to respondto statement designed to elicit negative knowledgeof performance resulting to creation of awareness.A 5-point likert scale was constructed to recordthese responses. Table 4 show results on privateextension services in relation to creation ofawareness.

The results in table 4 showed that majority offarmers (83.75%) received information on creditfacility.

Advice on land preparation was one of the ac-tivities used to measure private extension services.To elicit information on land preparation, the



Land Preparation f (n=80)%

9.011.25

12.015.0

1.01.25

35.043.8

23.028.8

80.0100.0

Table 5: Land Preparation



Timely Use of Appropriate Inputs f (n=80)%

3.03.75

17.021.3

--

44.055.0

16.020.3

80.0100.0

Table 6: Use of Appropriate Inputs


Inte

rnat

iona

l Jo

urna

l of

Agr

icul

tura

l M

anag

emen

t &

Dev

elop

men

t, 2

(3):

199

-207

, Sep

tem

ber,

201

2.

204

farmers were asked to respond to statement de-signed to elicit positive knowledge of performanceresulting to land preparation. A 5-point likertscale was constructed to record these responses.Table 5 show results on public extension servicesin relation to land preparation.

The results in 5 showed that majority of thefarmers (72.5%) received extension services onland preparation from private sector. Perhapsthis is due to the diverse nature of private sectorextension where land preparation machines arereadily provided to try and promote return oninvestment as well as enabling the farmers toincrease their income through increased caneproduction.

Advice on appropriate use of inputs was oneof the activities used to measure private extensionservices. To elicit information on the use of ap-propriate input, the farmers were asked torespond to statement designed to elicit positiveknowledge of performance resulting to use ofappropriate input. A 5-point likert scale wasconstructed to record these responses. Table 6show results on public extension services in re-

lation to appropriate use of input.The results in table 6 showed that majority of

the farmers (75%) receive advice on use of ap-propriate input from private sector.

A comparison of public and private extension

services

Table 7 shows a comparison of the role publicand private extension play with respect to variouscane farming activities. The relevant informationwas elicited by asking the farmers to state fromwhom they receive extension services from,followed by their responses to statements designedto elicit positive knowledge of performance tovarious cane farming activities. A 5-point likertscale was used to record these responses.

The results in table 7 indicate that, public ex-tension has a lesser role in enhancing effectivenessof outgrowers’ credit system as compared toprivate extension services. This was because themajority of the public extension recipients (61.5%)either disagreed or strongly disagreed comparedto 47% private extension recipients in terms ofadvising farmers on various farm activities.


Table 7: Comparison between role of public and private extension services in KisumuCounty

Key (SD=Strongly Disagree, D=Disagree, U=Undecided, A= Agree, SA=Strongly Agree)LP- Land PreparationUAI- Use of Appropriate InputsCA- Creation of Awareness

Public Extension Service (N=23)

LP UAI CA Average

Private Extension Service (N=80)

LP UAI CA Average

SA

A

U

D

SD

Total

f%f

%f

%f

%f

%f

1.04.36.0

26.0--

14.060.22.08.7

23.0

2.08.76.0

26.06.0

26.07.0

30.42.08.7

23.0

2.08.72.08.7--

17.073.92.08.7

23.0

2.08.75.0

21.52.08.7

12.553.81.8

7.7423.0

23.028.835.043.81.0

1.2512.015.09.0

11.2580.0

16.020.344.055.0

--

17.021.33.0

3.7580.0

4.05.09.011.3

--

51.063.816.020.080.0

13.216.528.235.300.801.027.434.310.413.080.0

Value df Asymp. Significance 2 sided

Pearson Chi-squareNo. valid cases

14.952103

18 0.667

Table 8: Chi-square test for Effectiveness of Outgrowers Credit System byType of Extension

Significance set at (α = 0.05)

Inte

rnat

iona

l Jo

urna

l of

Agr

icul

tura

l M

anag

emen

t &

Dev

elop

men

t, 2

(3):

199

-207

, Sep

tem

ber,

201

2.

205

Hypotheses testing

The null hypothesis tested stated that there isno significant difference between the publicand private extension services in terms of en-hancing effectiveness of out-grower’s creditsystem in Kisumu County. Both the public andprivate sector extension services were measuredwith respect to advice given to farmers on landpreparation, use of appropriate inputs and creationof awareness. The information on public andprivate extension services with their effects oneffectiveness of out-grower’s credit was elicitedby use of farmers’ questionnaire. Testing of thishypothesis was carried out by use of chi-squaretest and the results are presented in Table 8.

Results in Table 8 indicate that, there was nosignificant difference between public and privatesector extension services. This was because thePearson chi-square value (14.952) was not sig-nificant at α = 0.05 (p>0.05). The null hypothesiswas thus accepted. This result suggests that publicand private sector extension services are inadequatein terms of quantity despite the fact that canefarmers require it to realize a positive change.

DISCUSSION

Akroyd and Smith (2007) noted that lack ofagricultural services has negatively impactedon food production. Consequently, in manyparts of less developed countries, agriculturalextension services often bypass or do not reachthe rural farmers (FAO, 1997). In most countriesextension services provided by the governmentare supplemented by private sectors. Milu andJayne (2006) acknowledged that, in developingcountries, the private sector extension is extremelydiverse. Depending on the particular economicand political situation, the private sector mayconsist of individual farmers/ farm enterprisesof all sizes, agricultural input industries, agro-services enterprises, processing industries, mar-keting farms and multinational firms. It mayalso include a wide range of agricultural pro-duction and marketing co-operatives, farmersassociations and private and voluntary organi-zations. Despite their differences, all these or-ganizations share a common market orientation.They all try to make profit by selling goods and

services. As a result all these private sector or-ganizations have a strong incentive to delivergoods and services (including agricultural ex-tension) efficiently and effectively so as to en-hance their ability to survive. Firms that supplyagricultural inputs such as seed, chemical fer-tilizer, pesticides may provide farmers with awide range of technical and managerial infor-mation (through various outreach mechanisms)both to assure that their products are usedcorrectly and also increase agricultural productionand income to the farmers. These also motivatecustomers to buy more products in future (Miluand Jayne 2006). Examples of these private ex-tension agencies are the Muhoroni SugarcaneOutgrowers Company and Chemelil OutgrowersCompany, which are currently operating andsupporting farmers.

The findings of this study indicated thatmajority of farmers (82.6%) receive informationon credit facility from either from public orprivate extension providers. The results agreewith the findings by Khasiani (1992) who indi-cated that agricultural technologies might notbe adopted if the farmers are not aware of itsexistence. He continued that lack of awarenessacts as a hindrance to the effective participationin agricultural activities. Similarly, Madhur(2000) argued that, impact would be limited ifextension is unable to appreciably increase thelevel of farmers awareness. Further, the resultsalso supports the findings by Mbata (1991) whoacknowledged that through extension servicesthe small-scale farmers should be made to un-derstand that credit supervision is for his / herown interest and that, through supervision, creditwould be better managed and used for the in-tended purposes which in turn will increase hisproductivity and raise their capital base.

The findings of the study also showed thatmajority of the farmers received extension serv-ices on land preparation from private sector.Perhaps this is due to the diverse nature ofprivate sector extension where land preparationmachines are readily provided to try and promotereturn on investment as well as enabling thefarmers to increase their income through increasedcane production. However, among the farmers


Inte

rnat

iona

l Jo

urna

l of

Agr

icul

tura

l M

anag

emen

t &

Dev

elop

men

t, 2

(3):

199

-207

, Sep

tem

ber,

201

2.

206

who received advice from public extension of-ficers it was noted that the services were notsatisfactory. This suggests that perhaps the de-clining cane production in Kisumu County isdue to poor and inadequate land preparation,culminating from inadequate machinery toprepare land for cane growing.

The results further showed that majority ofthe farmers receive advice on use of appropriateinputs from private sector. This suggests thatprobably a few farmers may be benefiting asthe private sector normally targets potentialfarmers to maximize the profit from their prod-ucts. Absence of dependable information tofarmers on inputs, on credit and marketingwould erode the credibility of extension, hencethe rate of adoption by, farmers would be low(FAO, 1994, Khasiani, 1992). However thereshould be a positive correlation between farmers’link with information sources and adoption(World Bank, 1992 & Chitere, 1995).

This result suggests that public and privatesector extension services are inadequate in termsof quantity, that is, in terms of extension agentcontact with the farmer despite the fact thatcane farmers require it to realize a positivechange. Perhaps inadequate extension from gov-ernment is due to the retrenchment of manystaff in the Ministry of Agriculture in the studyarea (Owuor, 2002).

CONCLUSIONS

Based on the findings of the study a numberof conclusions were drawn:

- Public extension service has a role in enhancingeffectiveness of the outgrowers’ credit systemamong smallholder cane farmers in KisumuCounty. However, the sector needs to enhancethe following: provision of advice with respectto, land preparation and use of appropriateinputs.

- That except for inadequate quantity of ex-tension, private extension service plays a greaterrole in enhancing effectiveness of outgrowers’credit system.

- That in terms of enhancing effectiveness ofoutgrowers’ credits system there was no signif-icant difference between public and private

sector extension services. RecommendationsFrom the findings of the study, the following

recommendations were suggested.- Intensification of both public and private

extension services. - Strengthening factory extension division. - Increasing the number of extension personnel.- Establishing the contribution of extension

among other factors in cane production.

REFERENCES

1- Agbamu, I. U. (2005). ‘Problems and Prospectsof Agricultural Extension Services in DevelopingCountries.’ In: Adedoyin, S. F. (ed) op cit Pp 159-169.2- Agribusiness Development Support Project, (2001).Annual Report. Kisumu: Lagrotech Limited.3- Akroyd, S., & Smith, L. (2007). The Decline inPublic Spending to Agriculture – Does it Matter?Briefing Note, No. 2, Oxford Policy ManagementInstitute, Oxford.4- Anderson, J. R. (2007). Agricultural Advisory Serv-ices. Background Paper for the World DevelOpmentReport 2008. http://siteresources.worldbank.org/IN-TWDR2008/Resources/2795087-1191427986785/Anderson AdvisoryServices.pdf.5- Chitere, P. A. (1995). Extension Education andFarmers Performance in Improved Crop Farming inKakamega District (Kenya). Agricultural Adminis-tration. 18: 39-57.6- Economic Survey (2007). Central Bureau of Sta-tistics. Ministry of Planning and National Develop-ment. Government of Kenya. 2007.7- Food and Agriculture Organization. (1997). Ef-fectiveness of Agricultural Extension Services inreaching Rural Women in Africa, Volume 2. Italy,Rome: FAO.8- Gautam, M. & Anderson, J. R. (1999). Reconsid-ering the Evidence on Returns to T&V Extension inKenya. Policy Research Working Paper 1098, theWorld Bank, Washington D. C.9- Guda, E., Otieno, L.O., Ko’bonyo, P., Okumu,B., Ohito, D., Odera, J., Ogallo, O.S., Rasugu, O.,& Odudo, J. (2001). Business and Investment Insight:(Abstract). Maroko Investments Advisory ServicesPublications.10- KESREF (2007). Kenya Sugar Research Foun-dation. Strategic Plan 2009-2014.11- Khasiani, C. (1992). Towards Legitimisation ofAfrican Women Indigenous Knowledge in Natural


Inte

rnat

iona

l Jo

urna

l of

Agr

icul

tura

l M

anag

emen

t &

Dev

elop

men

t, 2

(3):

199

-207

, Sep

tem

ber,

201

2.

207

Resource Management. Award News Kenya: IssueNumber 4. December, 1993. 12- Kibet, J. K., Omunyinyi, M. E., & Muchiri, J.(2005). Elements of Agricultural Extension Policy inKenya. Challenges and Opportunities. African CropScience Conference Proceedings. 7: 1491 - 1494.13- Koyenikan, M. J. (2008). Issues for AgriculturalExtension Policy in Nigeria. International Journalof Agricultural Extension. 12:51-61.14- Madhur, G. (2000). Agricultural Extension: TheKenya Experience, an Impact Evaluation. WashingtonD. C: The World Bank.15- Mbata, J. N. (1991). Agricultural Credit Schemein Nigeria, A Comparative Study of the Supervisedand Non- Supervised Agricultural Credit Scheme asa Tool for Agricultural Development in Rivers StateNigeria. Discovery and Innovation. (Abstract).16- Milu, M., & Jayne, T.S. (2006). AgriculturalExtension in Kenya: Practice and PolicyLessons.Tegemeo Institute of Agricultural Policy and De-velopment, Egerton University.17- Nambiro, E., Omiti, J., & Mugunieri, L., (2005).Decentralization and Access to Agricultural ExtensionServices in Kenya. SAGA Working Paper. 18- Ogunsumi, L.O. (2004). Analysis of SustainedUse of Agricultural Technologies on Farmers’ Pro-ductivity in Southwest, Nigeria. Ph.D. Dissertation,Department of Agricultural Economics and Extension,Federal University of Technology, Akure, Nigeria. 19- Owuor, G. (2002). The Effect of Financial Self-help Groups Credit on Agricultural Production, ACase of Ukwala Division in Siaya District. (Unpub-lished MSc Thesis), Njoro: Egerton University.20- Rees, D. M., Wekundah, F., Ndungu, J., Odondi,A.O., Oyure, D., Andima, M., Kamau, J., Ndubi, F.,Musembi, Mwaura, L., & Joldersma, R. (2000).Agricultural Knowledge and Information in Kenya-Implications for technology dissemination and de-velopment. ODI Agricultural Research and ExtensionNetwork Paper (Abstract).21- Government of Kenya. (2005). Review of theNational Agricultural Extension Policy (NEAP) andits Implementation. Volume II – Main Report andAnnexes. Ministry of Agriculture and Ministry ofLivestock and Fisheries Development. Nairobi. 22- Wangia, C. (2001). Micro- Finance Experiencein Kenya. In: Anandajayasekeram, Dixon, Kashuliza,Ng’anjo, Tawonezvi, Torkelsson, Wanzira (Eds).Micro – Finance Experience of FARMESA MemberCountries in East and Southern Africa. Farmesa,Harare, Zimbabwe.23- World Bank. (1992). Trends in Agricultural Di-

versification: Regional Perspectives. Technical PaperNo. 180, Washington D.C: World Bank.24- World Bank (2007). World Development Report2008: Agriculture for Development, World BankWashington D.C.


www.ijamad.com

Inte

rnat

iona

l Jo

urna

l of

Agr

icul

tura

l M

anag

emen

t &

Dev

elop

men

t, 2

(3):

209

-214

, Sep

tem

ber,

201

2.

209

Applying CVM for Economic Valuation of Drinking

Water in Iran

Morteza Tahami Pour and Mohammad Kavoosi Kalashami *

Keywords: Economic Value of Water,Willingness to Pay, Mu-nicipal Water Consump-tion, Contingent ValuationMethod, Kohkiloye & Boy-erahmad Province

Received: 20 April 2012,Accepted: 12 July 2012 Economic valuation of water is useful in the administration

and management of water. Population growth and urban-ization caused municipal water demand increase in Iran.Limited water resource supply and urban water networkcapacity raised complexity in water resources management.Present condition suggests using economic value of water as acriterion for allocating policies and feasibility study of urbanwater supply projects. This study use contingent valuationmethod for determining economic value of drinking water inKohkiloye & Boyerahmad province. Required data set wereobtained from 177 questionnaires by applying stratified randomsampling in 2011 year. From 136 investigated urban households111 ones are willing to pay more for qualified drinking water.Also, from 41 investigated rural households only 3 ones arewilling to pay more for qualified drinking water. Resultsindicated that economic value of drinking water is 6877 Rialper cubic meter.

Abstract


1 Ph. D Student of Agriculture Economics, Department of Agriculture Economic, University of Tehran, Iran. * Corresponding author’s email: [email protected]

Inte

rnat

iona

l Jo

urna

l of

Agr

icul

tura

l M

anag

emen

t &

Dev

elop

men

t, 2

(3):

209

-214

, Sep

tem

ber,

201

2.

210

INTRODUCTION

Iran is one of the arid and semi-arid countriesof the world with average precipitation of 251mm per year. The total renewable water resourcesof Iran is 130 billion cubic meters, out of which92 percentages is used for agriculture, 6 per-centages for domestic use and services and 2percentages for industrial uses (Assadollahi,2009). Rapid population growth and low irrigationefficiency in agricultural sector of Iran have in-creased the demand for water resources. There-fore, rational management for water supply anddemand and optimum use of the available waterresources is necessary. Managing water as aneconomic good is an important way of achievingefficient and equitable use, and of encouragingconservation and protection of water resources.Water resources provide variety of products andservices for human being include food, drinkingwater, hygiene, entertainment and hydropower.Water demand amount and variety increasebeside water supply limitations in municipalregions cause competition among users and im-pose pressure on drinking water networks. Wide-spread network of water in cities, high cost ofdrinking water supply and important role ofwater in sustainable development raised com-plexity in water resource management in mostcities of Iran. Water allocation and using watereconomic value as a criterion for allocating isone of the mentioned complexity in waterresource management (Turner et al., 2004). Thevalue of water in alternative uses is importantfor the rational allocation of water as a scarceresource, whether by regulatory or economicmeans. Water economic value could be used asa proper framework for water pricing and waterallocating policies. Also, benefit-cost analysisof water supply projects mainly uses water eco-nomic value for economic benefit determination.Most of development documents in Iran insiston estimating economic value of water indifferent uses and applying this amount as acriterion for allocating. Fourth developmentprogram of Iran noted economic value of watershould be calculated in water basins. Also, inFifth development program of Iran using watereconomic value for allocating and supplying

water resource has been mentioned. Many studiesestimate economic value of water in agricultureusage applying parametric and non-parametricmethods like production function and mathe-matical programming (Hussain and Young, 1985,Thompson, 1988, Chakravorty and Roumasset,1991, He and Tyner, 2004). Against irrigationwater, drinking water considered as a final com-modity in economical modeling and its economicvalue determination approach completely differentfrom the situation in which water is a productioninput like in agriculture and industry (Lehtonenet al., 2003). When water considered as a finalcommodity, contingent valuation method couldproperly used for its economic value determi-nation. Many studies use CVM for estimatingand calculation economic value of drinkingwater (Gnedenko and Gorbunova, 1998, Gnedenket al., 1999, Farlofi et al., 2007, Guha, 2007).

According to the importance of economicwater valuation in Kohkiloye and Boyerahmadprovince, present study estimated economicvalue of drinking water in municipal and ruraldistricts of mentioned province based on will-ingness to pay of households for proper qualityof drinking water. Estimated WTP for drinkingwater could be used in benefit-cost analysis ofurban water supply projects in Kohkiloye andBoyerahmad province.


Several approaches applied for economic val-uation of water which categorize into two maingroups include economic valuation of water asa production input and economic valuation ofwater as an economic commodity. In agricultureand industry, water used as an input and ap-proaches like production function or mathematicalprogramming applied for economic valuationof water. In municipal usage water treated as afinal commodity and water consumers gainedutility by using water. Hence, in this case ap-proaches like contingent valuation method,choice modeling and water market transactionshave been used for estimating economic valueof water (Young, 2005). CVM studies are verypopular among mentioned approaches used foreconomic valuation of drinking water, present

Applying CVM for Economic Valuation/ Tahami Pour and Kavoosi Kalashami

Inte

rnat

iona

l Jo

urna

l of

Agr

icul

tura

l M

anag

emen

t &

Dev

elop

men

t, 2

(3):

209

-214

, Sep

tem

ber,

201

2.

211

study apply CVM for achieving to the foresaidgoals.

CVM usually used as a standard and flexibletool for measuring non-use values and non-market use values of natural resources and en-vironment (Hanemann et al., 1991 and Hane-mann, 1994). CVM determined individuals WTPin hypothetical scenarios (Lee, 1997). In thisapproach a hypothetical market for drinkingwater with proper quality established and someproposed price suggested to individuals for ac-quiring mentioned commodity. Individual's an-swer to the proposed prices or bids has beenformed based on maximizing utility. As shownbelow, accepting proposed price means thatutility gained by accepting is more than theutility of denying proposed price.

(1)

In above equation U is indirect utility, Y is in-dividual income, A is proposed price or bid andS is social-economical characteristics. Utilitydifference is as below:

(2)

According to the model characteristics logitor probit functional form has been used for es-timating valuation function. Considering Pi asthe probability of accepting proposed price (A)by individual, logit functional form could beformed as below:

(3)

In which Fη(ΔU) is cumulative distributionfunction. θ, γ and β are regression coefficients

and it is expected that θ>0, γ>0 and β≤0. Afterestimating above logit function it is possible tocalculate expected WTP using integral.

Logit regression model coefficients have beendetermined using maximum likelihood estimator(Lehtonen and et al., 2003). Integral of Fη(ΔU)between 0 to infinity has been calculated basedon below equation:

(4)

In above equation E(WTP) is expected will-ingness to pay and α∗ is adjusted constant which

is constructed as below:

(5)

One of the main advantages of logit estimationis that it is possible to investigate change invariables amount on the probability of acceptingproposed price by individual i. The probability ofbid acceptance by individual defined as below:

(6)

In which, Xi* is the vector of variables and λ

is the vector of coefficients. In order to evaluatethe effect of each variable quantity change onthe probability of bid acceptance, the derivationof above equation has been calculated (Maddala,1991):

(7)

This equation generates marginal effect ofeach variable which is very useful in policymaking analysis of model results. Requesteddata set were obtained through a survey usingquestionnaires. Mainly, information like educa-tion, age, individual satisfaction of drinkingwater quality and quantity, monthly income andexpenditure, water consumption quantity andwillingness to pay for drinking water has beenasked from each individual by interviewing.Present study used stratified random samplingmethod for determining the sample size. From177 sample size 136 and 41 questionnaires gath-ered from urban and rural districts of Kohkiloye& Boyerahmad province, respectively. Explana-tory variables which were used in logit modelfor estimating economic value of drinking watercould be summarized in below table 1.


Investigating sample individuals' satisfactionof drinking water quality showed that 7 per-centages of rural individuals and 40 percentagesof urban individuals satisfied with present con-dition of drinking water quality. On the otherhand, 93 percentages of investigated rural indi-viduals and 60 percentages of investigated urban


Inte

rnat

iona

l Jo

urna

l of

Agr

icul

tura

l M

anag

emen

t &

Dev

elop

men

t, 2

(3):

209

-214

, Sep

tem

ber,

201

2.

212

individuals dissatisfied with drinking water qual-ity. Reasons of dissatisfaction could be summa-rized as kidney diseases occurrence, parasite

diseases occurrence and existence of pollutionin rural drinking water.

Investigating sample individuals' satisfaction ofdrinking water quantity showed 17 percentagesin rural districts and 69 percentages in urbandistricts satisfied by present condition of drinkingwater quantity or supply. Also, 83 percentages ofinvestigated people in rural districts and 31 per-centages in urban districts dissatisfied from drinkingwater quantity or supply. Reasons of dissatisfactioninclude water quantity, dissection, and networkinsufficient pressure. Hence, households forcedto use tankers or reserve drinking water.

Investigating sample individuals' educationlevel showed that frequency of diploma degree


Figure 1: Comparing satisfaction level of drinkingwater quality in urban and rural districts (percentages).

Figure 2: Comparing satisfaction level of drinkingwater supply in urban and rural districts (percentages).

Figure 3: Comparing education level of sample indi-viduals in urban and rural districts (percentages).

Table 2: Comparing willingness to pay in urban and rural districts.

District Description Willing to pay more Will not pay more

Urban districts

Rural districts

FrequencypercentagesFrequencypercentages

111813893

251937

Source: research findings.

Explanatory variables Description

RegionWater quality

Water quantity

Monthly incomeMonthly water consumptionFamily sizeProposed price or bidFamily supervisor genderFamily supervisor ageFamily supervisor education

Dummy variable (1= urban district, 0= rural district)Dummy variable (1=satisfaction from drinking water quality,0=dissatisfaction from drinking water quality) Dummy variable (1=satisfaction from drinking water quantity,0=dissatisfaction from drinking water quantity)In 1000 RialsIn cubic meter-In 1000 RialsDummy variable (1=male, 0=female)--

Table 1

Inte

rnat

iona

l Jo

urna

l of

Agr

icul

tura

l M

anag

emen

t &

Dev

elop

men

t, 2

(3):

209

-214

, Sep

tem

ber,

201

2.

213

is more in urban districts. Education level fre-quency of sample is shown in figure 3.

From 136 investigated urban households 111ones are willing to pay more for qualifieddrinking water. Also, from 41 investigated ruralhouseholds only 3 ones are willing to pay morefor qualified drinking water.

After estimating different models with ex-planatory variables, below model select as thebest estimations.

Results revealed that bid, monthly income,family supervisor education level and dummy ofregion variables had direct and statistically sig-nificant effects on probability of bid acceptance.Marginal effect value of monthly income variableshowed that one unit increase (1000 Rials increase)in monthly income 0.0192 unit increase the prob-ability of bid acceptance. Also, marginal effectof bid showed that 1000 Rials increase in bidamount 0.024 unit decrease the probability ofbid acceptance by sample individuals. One levelincrease in family supervisor education 0.0192unit increase the mentioned probability amount.

Percentages of right prediction in logit modelwere 77 percentages. So, 77 percentages of in-dividuals' responses could be simulated bymodel. Using model coefficients, the amount ofα∗ calculated (α∗ = 2.171). Expected quantity ofWTP for drinking water consumption equals213187 Rials per month. Considering averagedrinking water consumption for each householdas 31 cubic meters, value of a cubic meterdrinking water equals 6877 Rials.

CONCLUSION

Results revealed that drinking water consumerswere willing to pay 6877 Rials per cubic meterin Kohkiloye & Boyerahmad province. This

price could be used as a good framework for al-locating planning of drinking water. Also, forcalculating economic benefits of new drinkingwater supply projects this value could be used.Applying CVM approach for economic valuationof drinking water provides good view for man-agers and planners about consumers' demandof drinking water.

REFERENCES

1- Assadollahi, S.A. (2009). Groundwater ResourcesManagement in Iran. Secretary General, IRNCIDand Deputy of Protection and Exploitation of IranWater Resources Management Company. 2- Chakravorty, V. and J. Roumasset. (1991). EfficientSpatial Allocation Irrigation Water. American journalof Agricultural Economics. 73: 165-173.3- Farlofi, S., Mabugu, R., & Ntshingila, S. (2007).Domestic Water Use and Aalues in Swaziland:Contingent Valuation Analysis, Agrekon Magazine.(Abstract)4- Gnedenk, E., Gorbunova, Z., & Safonov, G.(1999). Contingent Valuation of Drinking WaterQuality in Samara City, Moscow State University,zone “I”, room 75, Moscow, 117234, VorobievyGory.5- Gnedenko E.D., & Gorbunova, Z.V. (1998). AContingent Valuation Study of Projects ImprovingDrinking Water Quality”, Modern ToxicologicalProblems. (Abstract).6- Guha, S. (2007). Valuation of Clean Water Supplyby Willingness to Pay Method in Developing Nation,a Case Study in Calculate, India. Jabalpur University.7- Hanemann, W. M. (1994). Valuing the Environmentthrough Contingent Valuation. Journal of EconomicPerspectives. 8(4): 19-43.8- Hanemann, W. M., Loonis, J., & Kanninen, B.(1991). Statistical Efficiency of Double-bounded Di-chotomous Choice Contingent Valuation. AmericanJournal of Agricultural Economics, 73(4): 1255-1263.9- He, L., & Tyner, W. (2004). Improving Irrigation


Variables Coefficients S.D. t-statistics Marginal Effect

BidMonthly incomeFamily supervisor educationRegionConstant

-0.1070.00064

0.085-1.1131.834

0.0420.00036

0.0410.6790.639

-2.5031.76

2.061-1.6372.868

-0.0240.000140.0192-0.25

-

Table 3: Logit model estimation result.

Log of likelihood function = -86.5Percentage of Right Predictions = 0.77Source: Research findings.

Inte

rnat

iona

l Jo

urna

l of

Agr

icul

tura

l M

anag

emen

t &

Dev

elop

men

t, 2

(3):

209

-214

, Sep

tem

ber,

201

2.

214

Water Allocation Efficiency Using Alternative Policyoption in Egypt, http://econpapers.hhs.se10- Hussain, R.Z., & Young, R.A. (1985). Estimatesof the Economic Value Productivity of IrrigationWater in Pakistan from Farm Surveys, Water Re-sources Bulletin. 26(6): 1021-1027.11- Lee, C. (1997). Valuation of Nature-based TourismResources Using Dichotomous Choice Contingent Val-uation method, Tourism Management. 18(8): 587-591.12- Lehtonen, E., Kuluvainen, J., Pouta, E., Rekola,M., & Li, C. (2003). Non-market Benefits of ForestConservation in Southern Finland. EnvironmentalScience and Policy. 6: 195-204.13- Maddala, G.S. 1991. Introduction of Econometrics.2nd Edition, Macmillan, New York.14- Thompson, C.D. (1988). Choice of FlexibleFunctional Forms: Review and Appraisal, WesternJournal of Agricultural Economics. 13: 169-183.


Inte

rnat

iona

l Jo

urna

l of

Agr

icul

tura

l M

anag

emen

t &

Dev

elop

men

t, 2

(3):

215

-222

, Sep

tem

ber,

201

2.

215

Perceptions of Constraints Affecting Adoption of

Women-in-Agriculture Programme Technologies in

the Niger Delta, Nigeria

Iniobong A. Akpabio, Nsikak-Abasi A. Etim1* and Sunday Okon

Keywords: Constraints to Adoption,Awareness, AgriculturalTechnologies, Womenfarmers.

Received: 5 April 2012,Accepted: 24 July 2012 The study focused on constraints affecting the adoption of

innovative agricultural technologies disseminated by theWomen-in-Agriculture (WIA) unit of the Akwa Ibom AgriculturalDevelopment Programme (AKADEP) to its women clientele.The study also ascertained the awareness and adoption levelsof such introduced technologies. Findings revealed that re-spondents were aware of 61.9% of introduced technologies,while only 33.3% were fully adopted. The study also identifiedseven factors responsible for the non-adoption of womenfarmers’ related technologies. The three highest ranking con-straining factors were revealed as; high cost of inputs, lowincome level of women farmers and lack of regular contactwith WIA extension agents. Reasons have been proffered forthe relatively low technologies’ adoption levels. Recommen-dations have also been made to enhance the technologyadoption level. These include the necessity to introduce onlysocio- economically and culturally compatible technologies toWIA clientele, a wholesale focus on follow-up activities afterinitial group based technology introduction activities, and theattachment of a credit scheme to the WIA program.

Abstract


1 Department of Agricultural Economics and Extension, University of Uyo, PMB 1017, Uyo, Akwa-Ibom State, Nigeria. * Corresponding author’s email: [email protected]

Inte

rnat

iona

l Jo

urna

l of

Agr

icul

tura

l M

anag

emen

t &

Dev

elop

men

t, 2

(3):

215

-222

, Sep

tem

ber,

201

2.

216

INTRODUCTION

Agriculture constitutes a large share of nationaloutput and employs a majority of the labourforce in most developing countries; hence thesector has been integrated into any thinkingabout development (World Bank, 2003). How-ever, whereas agriculture-led growth played animportant role in slashing poverty and trans-forming the economies of many Asian and LatinAmerican countries, the same has not occurredin Africa, including Nigeria (Diao et al., 2007).According to Baker (2005), technical change isthe engine of long-term growth and it becomestechnically important through diffusion. This ismore so for agricultural production, where theprospect of enhanced production offered by im-proved agricultural technologies is recognized,according to the World Food Program, as essentialto improving the household food security ofsmall scale farmers, raising rural incomes andcreating national surplus that can improve thebasis for economic growth (WFP, 1998).

Baker (2005) took a retrospective view atAfrica’s lack of robust economic growth anddearth of modern technology and concludedthat technology (especially agricultural tech-nology) diffusion appears to have failed therein.Eicher (1992) revealed that nearly 100% of theincrease in food production in the West Africansub-region, since 1960, has come from expandedharvest area, rather than improvements in tech-nology, a trend which Sanders (1996) hasdeemed, inefficient and with negative long termprospects.

Jafry (2000), Brown et al., (2001) and the Di-rectorate for International Development (DFID,2004), among many other authors and researchscientists, revealed that women are the keyfarmers, food producers and natural resourcemanagers, in most countries of sub-SaharaAfrica. This is because they provide 65 – 89%of food, provide nearly half of farm labour,shoulder over 90% of domestic responsibilitiesand work twice as many hours as men. Akpabio(2005) also reported an African study which re-vealed that women carry over 80 tonnes of fuel,water and farm produce for a distance of morethan one kilometer over the course of a year.

Despite all these contribution, the TechnicalCentre for Agricultural and Rural Cooperation,asserted that women are still restricted in theirroles as farmers by unequal rights and unequalaccess to and control over resources, especiallyland (CTA, 2000). Women also carry out theirwork without much help from agriculturalsupport mechanisms such as extension agencies,input suppliers and credit institutions (FAO,2000).

The Women-in-Agriculture (WIA) sub-com-ponent of the Agricultural Development Pro-gramme (ADPs) was instituted in 1988 to addressgender specific agricultural problems. The focusis on food nutrition, processing, storage andutilization of crop and livestock produce, inorder to raise women’s income and living stan-dards through business oriented farming andprocessing strategies. Ever since the introductionof the WIA programme in Nigeria, and with thecurrent emphasis on participatory extension,various efforts have been made to elicit varioustypes and levels of information on the activitiesand effectiveness of the programme in specificlimited areas (states) of Nigeria and the NigerDelta. Akpabio (2005b) reported that the WIAprogramme in Akwa Ibom State remains lessthan effective, in terms of its contribution to theupliftment of the economic and socio-psycho-logical status of rural women while Adetoun(2000) in South Western Nigeria, and Eshiett(2007) with reference to Akwa Ibom State, re-vealed that only a few of the technologies dis-seminated to WIA clientele have been fullyadopted.

The importance of agricultural technologiesin the development process cannot be overem-phasized. It is against this background that thisstudy sought to ascertain clientele perceptionson reasons for the reported low trend of adoptionof agricultural technologies. This study howevercovers the larger South – South (Niger Delta)region of Nigeria, hence it was decided first ofall to ascertain on a wider scale of the NigerDelta, the validity of earlier reports of Akpabio(2005b) and Eshiett (2007). In essence, thestudy sought to answer pertinent questionsrelating to: (i) the level of women farmers’

Affecting Adoption of Women-in-Agriculture Programme/ Iniobong A. Akpabio et al

Inte

rnat

iona

l Jo

urna

l of

Agr

icul

tura

l M

anag

emen

t &

Dev

elop

men

t, 2

(3):

215

-222

, Sep

tem

ber,

201

2.

217

awareness of specified innovations introducedthrough the WIA programme and (ii) respondents’perceptions of constraints affecting adoption oftechnologies disseminated through the WIAprogramme in the Niger Delta.


Study Area

The Niger Delta is located in the Southernpart of Nigeria. It spreads over a total landmass of about 75,000 square kilometers. It isinhabited by an estimated 30 million population.The people are distributed into forty ethnicgroups in about 13,329 communities/settlementsin nine states. It is characterized by wetlandsand water bodies, with creeks and rivers criss-crossing the entire Southern parts and is oftenregarded as the largest wetland in Africa andthe third largest in the world (UNDP, 2006).The region is however endowed with naturalresources. It has the third largest mangroveforest, with the most extensive fresh waterswamp forest and tropical rainforest characterizedby great biological diversity. Alongside its im-mense potentials for agricultural revolution, thestudy area also hosts vast reserves of non-re-newable natural resources, particularly hydro-carbon deposits in oil and gas.

The population for the study comprised allthe leaders/representatives of different WIAgroups who attended the various one-day inter-active fora organized by the ADPs in all statesin the region. Relevant data could be collatedfor five states. These were Akwa Ibom, CrossRiver, Delta, Edo, and Rivers. All the 267 par-ticipants were purposefully utilized for the study,although responses from 250 respondents wereeventually utilized for data analysis (viz, table1). A pre-tested and validated structured InterviewSchedule and Focus Group Discussions were

utilized to elicit relevant information from theselected sample. These activities were performedwith the aid of trained enumerators.

To ascertain the level of women farmers’awareness and adoption of specified innovationsintroduced through the WIA program, a list oftechnologies disseminated through the WIAprogramme was obtained, after which awarenessand adoption scores were computed for eachtechnology. Scores of 0 and 1 were recordedfor awareness and non-awareness of disseminatedtechnologies while scores of 2, 1 and 0 wererecorded for adopted, discontinued and non-adopted technologies, respectively. A mean cut-off score of 0.5 was adopted to demarcatebetween technologies for which respondentswere either ‘aware’ or ‘not aware’, while a cut-off mean score of 1.0 was utilized to differentiatebetween technologies which have either been‘adopted’ or ‘not adopted’. In essence, respon-dents were deemed to be aware of a technologywith a mean score of 0.5 and above, while theywere not aware of technologies with meanscores of less than 0.5. Similarly, technologieswhich recorded mean scores of 1.0 and abovewere perceived as adopted by respondents,unlike technologies with mean scores of lessthan 1.0, which were regarded as not adopted.

To determine respondents’ perceptions of con-straints affecting adoption of WIA programmetechnologies, a list of possible constraints thatmay hinder the adoption of disseminated tech-nologies was drawn up with the aid of interviewsand literature search. A 3-point Likert continuumof agreed (3) undecided (2) and disagreed (1)was employed to compute responses on reasonsfor non-adoption of WIA technologies. A cut-off mean score of 2.5(3+2+1/3 +0.5) was utilizedto differentiate between ‘major’ and ‘minor’factors for non –adoption, where a score of 2.5and above, was depicted as a ‘major’ factor fornon-adoption, while items with scores below2.5 were adjudged minor factors.


Awareness and adoption levels of WIA tech-

nologies

Tables 2 and 3 show that women farmers


Table 1: Selected Sample

S/N State Population Sample

12345

Akwa IbomCross River

Delta Edo

Rivers TOTAL

5160535647

267

5153485444

250

Inte

rnat

iona

l Jo

urna

l of

Agr

icul

tura

l M

anag

emen

t &

Dev

elop

men

t, 2

(3):

215

-222

, Sep

tem

ber,

201

2.

218

were aware of 61.9% (13 of 21) introducedtechnologies, while only 33.3% (7 of 21) of thetechnologies were eventually adopted. It wasalso observed on table 3, that respondentsadopted 53.9% (7 of 13) of the technologies forwhich they were aware. A related finding in thecourse of study revealed that only 59.2% re-spondents received information on improvedagricultural technologies from extension officialsof the WIA program, while 20.8% and 20% re-spondents received information fromrelatives/friends and husbands, respectively.There is a cause for concern here. This isbecause an extension program deliberately tar-geted at women farmers reaches only 59.2% ofintended clientele. Many reasons have beenproffered for this undesirable situation. Theseinclude; long distance from meeting venues andconcomitant non-attendance at group meetings(Adetoun, 2000) lack of interpersonal contact,arising from lack of follow-up after group meet-ings (Udoh, 2001) and lack of relevance of dis-seminated messages to the amelioration offemale farmers livelihood constraints (Reij and

Waters-Bayer, 2001) among many others, mighthave led to clientele’ loss of interest in extensionofferings.

Table 3 also shows that none of the technologiesdisseminated to respondents under “input use”and “tree crops planting” classifications wasadopted. Odourless fufu/garri (fresh/dried cassavapaste) x = 1.96; cassava/maize/melon cropscombination (x = 1.76) and intercropping (x =1.45) were the most adopted technologies. Thisresult corroborates Baker’s (2005) and Swinkelsand Franzel’s (1997) assertion that compatibletechnologies and technologies that differed verylittle from the old technologies would diffusefaster since there would be less of an informationproblem associated with them. Pannell (1999)described four conditions necessary for farmersto adopt innovative technologies, two of whichare “awareness of the technology” and “perceptionthat technology promotes farmers objectives”.It may be inferred that farmers will adopt moreof the technologies for which they are aware. Inessence, awareness of technology is a motivatingfactor for the adoption of technological packages.


Table 2: Distribution of Respondents based on the extent of Awareness of WIA technologies

AKADEP Technologies Food Crops

Aware

(1)

Not aware

(0)

Means Remarks

12345

678910

11121314

15161718

192021

Cassava/maize/melon plantingYam mini setDry season vegetableWet season vegetableRice cultivationProcessing & UtilizationSoya bean milk/ flourOdorless fufu/garriFruit drinksPlantain chips processingPineapple chips processingInput useFertilizer useImproved crop varieties e.g. maizeAgro chemicals e.g. PesticidesImproved animal breedsAgroforestry technologySnail rearingPlantain /Cocoyam intercroppingAfang cultivationBee raisingTree crops plantingImproved oil palm seedlingsRubber seedlingsImproved cocoa seedlings

230 (92) *124 (49.6)170 (68)210 (84)86 (34.4)

140 (56)250 (100)116 (46.4)210 (84)

200 (80)180(72)160 (64)84 (33.6)130 (52)

124 (49.6)196 (78.4)210 (84)78 (31.2)

160 (64)90(36)70 (28)

20 (8)126 (50.4)

80 (32)40 (16)

164 (65.6)

110 (44)0 (0)

134 (53.6)40 (16)

50 (20)70(28)90 (36)

166 (66.4)120 (48)

126 (50.4)54 (21.6)40 (16)

172 (68.8)

90 (36)160 (64)180(72)

0.920.490.680.840.34

0.561.000.460.84

0.800.720.640.330.52

0.490.780.840.31

0.640.360.28

AwareNot Aware

AwareAware

Not Aware

AwareAware

Not Aware Aware

AwareAwareAware

Not Aware

AwareAwareAwareAware

AwareNot AwareNot Aware

*-Percentages in parentheses

Inte

rnat

iona

l Jo

urna

l of

Agr

icul

tura

l M

anag

emen

t &

Dev

elop

men

t, 2

(3):

215

-222

, Sep

tem

ber,

201

2.

219

Hardarker, Huirne and Anderson (1997) howevercautioned that high level awareness of tech-nologies does not necessarily translate intohigher adoption levels. This is because farmerswill only adopt those innovations which are ad-judged useful and beneficial to their particularsituation. A disaggregated analysis of table 2reveals relatively high frequencies of “non-awareness” scores that were recorded for sometechnological offerings which were generallyperceived (mean scores) as being “aware” of byrespondents (viz; items 3, 6, 12, 14 and 19).Bunch (1982) and Baker (2005) harped on theimportance of critical mass in the adoption ofinnovations. The researchers contended that thatthere is a higher level of adoption and less dis-continuance for a new technology in which thewhole community or a critical mass (propor-tionately larger than average) of farmers areaware of and in which they are interested. Theyexplain reasons for this in terms of traditionalcommunities being accustomed to living in an

environment of consensus, and that schemeswhich entail community risk sharing are moreeasily imbibed than otherwise.

Constraints affecting adoption of WIA pro-

gramme technologies

Results as shown on table 3 reveals that re-spondents perceived 7 of the 21 identified itemsas possible reasons for the relatively low adoptionlevels of agricultural technologies introducedthrough the WIA programme. These are: highcost of inputs (x = 3.0), low income level ofwomen farmers’ (x = 2.97) lack of regularcontact with extension agents (x =2.82) old ageof women farmers in the study area (x = 2.73)poor attitude towards risk and change (x = 2.55)and complexity of introduced technologies (x =2.55). The above revealed findings find relevancein related literature. Rogers (1995) identifiedfive key characteristics of innovations that de-termine their adoption potential, including:relative advantage, trialability, compatibility,


Table 3: Distribution of respondents based on the extent of adoption of WIA technologies.

AKADEP Technologies (FoodCrops)

Not Adopted

(0)

Discontinued

(1)

Adopted

(2)

Means** Remarks

12345

678910

11121314

15161718

192021

Cassava/maize/melon plantingYam mini setDry season vegetableWet season vegetableRice cultivationProcessing & UtilizationSoya bean milk/ flourOdourless fufu/garriFruit drinksPlantain chips processingPineapple chips processingInput UseFertilizer useImproved crop varieties e.g maizeAgro chemicals eg. pesticidesImproved animal breedsAgroforestry technology Snail rearingPlantain /Cocoyam intercroppingAfang cultivationBee raisingTree crops plantingImproved oil palm seedlingsrubber seedlingsImproved cocoa seedlings

0 (0) *0 (0)

60 (24)0 (0)

172 (34.4)

10 (4)0 (0)

110 (44)4 (1.6)

14 (5.6)

40 (16)24 (9.6)

34 (13.6)40 (16)

16.(6.4)12 (4.8)30 (12)

78 (31.2)

140 (56)82 (32.8)66 (26.4)

20 (8)88 (35.2)60 (24)40 (16)

0 (0)

70 (28)10 (4)6 (2.4)

16 (6.4)10 (4)

52 (20.8)60 (24)

46 (18.4)20 (8)

24 (9.6)4 (1.6)

16 (6.4)0 (0)

20 (8)8 (3.2)4 (1.6)

210 (84)36 (14.4)48 (19.2)170(68)

0 (0)

60 (24)240 (96)

0 (0) 190 (76)

176 (70.4)

88 (35.2)76 (30.4)

4 (1.6)60 (24)

84 (33.6)180 (72)

164 (65.6)0 (0)

0 (0)0 (0)0 (0)

1.760.640.621.520.00

0.761.960.021.581.45

0.910.840.210.56

0.761.451.360.00

0.080.030.01

AdoptedNot AdoptedNot Adopted

AdoptedNot Adopted

Not AdoptedAdopted

Not AdoptedAdoptedAdopted

Not AdoptedNot AdoptedNot AdoptedNot Adopted

Not AdoptedAdoptedAdopted

Not Adopted

Not AdoptedNot AdoptedNot Adopted

*-Percentages in parentheses** Mean Scores calculated, based on total no. of respondents- regardless of the no. of actual recordedresponses per innovation.

Inte

rnat

iona

l Jo

urna

l of

Agr

icul

tura

l M

anag

emen

t &

Dev

elop

men

t, 2

(3):

215

-222

, Sep

tem

ber,

201

2.

220

observability and complexity. Reed (2001) iden-tified the most significant of these characteristics,as: high relative advantage, high compatibilityand low complexity. Swinkels and Franzel(1997) agreed with the submission above, butalso opined that for the female gender, additionalincentives for adoption may include factors like,suitability to accepted gender roles, cultural ac-ceptance and compatibility with other enter-prises.

High cost of inputs for introduced technologiesand low income of respondents’ were revealedas the greatest constraints to adoption of intro-duced technologies. Obinne (1994) and Arokoyo(1996) mentioned low income level of farmersand high cost of inputs as constraints to tech-

nology adoption, especially among low incomefarmers. In that wise, Baker (2005) and Hebinck,Franzel and Richards (2007) asserted that themost successful programmes of agriculturalchange are those that tie adoption to credit pro-grammes. Udoh (2001) and Eshiett (2007) main-tained that contact with extension agents, espe-cially with respect to interpersonal contacts,relate favorably to the adoption of new farmpractices and concomitant improved agriculturalproduction. Obinne (1994) and Baker (2005)opined that poor attitude to risk, in terms of ex-cessive risk aversion may severely limit adoptionof technological innovations especially amongfemale rural farmers, while Baker (2005) opinedthat technologies that differed very little from


Table4: Distribution of respondents based on perception of factors affecting non-adoption of WIAtechnologies.

Reasons for Non Adoption/ Discontinuace

Disagreed

(1)

Undecided

(2)

Agreed

(3)

Mean

(X)

Remarks

1234

567

8

9

10

1112

13

14

15

1617

1819

2021

High cost of inputsLack of supporting inputsProblem of diseases / pestsNon-appropriateness of the technological packageto the Local environmentNon-availability of the improved package Non-Profitability of the new technology Superiority of the old technology to the newly intro-duced one.Incompatibility of the new technology with thenorms and customs of the local environmentLack of clear understanding of the newly introducedpackage Low level of educational attainment by womenfarmers in the study area.Low level of income of women farmers in the areaInsufficient Programs designed to convince and en-courage farmers to changeWomen farmers perception of the old technologyas better than the new one.Inconsistence of the innovation with the existingfarming system, values and needs of women farm-ers in the areaInadequate information about the newly introducedtechnological package.Complexity of the introduced innovation.Failure of some demonstration plots set –up by theextension agents.Lack of regular contact with extension agentsPoor attitude of women farmers towards changeand riskAge of women farmers in the study areaLack of access and control over production re-sources such as land and credit facilities.

0(0) *88(35.2)66 (26.4)60 (24.1)

84 (32.80)188 (75.2)174 (69.6)

144 (57.6)

12(4.8)

48 (19.2)

2 (0.8)12 (4.8)

138 (55.2)

156 (62.4)

50(20)

44(17.6)138 (55.2)

20(8)8 (3.2)

20 (8)8(3.2)

0(0)62(24.8)

34(13.60)68 (27.2)

16 (6.4)40 (16)

48 (19.2)

60(24)

104(14.6)

106 (42.4)

4 (1.6)144 (59.6)

68 (29.2)

36(14.4)

58 (23.2)

24 (9.6)58(23.2)

4(1.6)60 (24)

28 (11.2)96(38.4)

250 (100)100(40)150 (60)

122 (48.8)

152 (60.8)22 (8.8)28 (11.2)

46 (18.4)

134 (53.6)

96(38.4)

244 (97.6)94 (37.6)

44 (17.6)

58 (23.2)

142(59.2)

182 (92.8)54 (21.6)

226(90.4)182 (72.8)

202 (80.8)146(58.4)

3.002.052.332.25

2.281.161.42

1.60

2.48

2.19

2.972.32

1.62

1.61

2.44

2.551.66

2.822.70

2.732.55

Major FactorMinor FactorMinor FactorMinor Factor

Minor FactorMinor FactorMinor Factor

Minor Factor

Minor Factor

Minor Factor

Major Factor Minor Factor

Minor Factor

Minor Factor

Minor Factor

Major FactorMinorFactor

Major FactorMajor Factor

Major FactorMajor Factor

*-Percentages in parentheses

Inte

rnat

iona

l Jo

urna

l of

Agr

icul

tura

l M

anag

emen

t &

Dev

elop

men

t, 2

(3):

215

-222

, Sep

tem

ber,

201

2.

221

the old technologies would diffuse faster thanunrelated technologies, while the older generationmay credibly block adoption even if the youngergeneration co-ordinates. Dove (1991) contendedthat individuals with insecure tenure will generallybe less likely to invest in new technologies thatrequire complementary immobile inputs, whileDue, Mudenda and Miller (1993) asserted thatalthough women want to increase the productivityof the resources they control, they face greaterobstacles to change. One of such obstacles, ac-cording to Reij and Waters-Bayer (2001) is lackof relevance of disseminated messages to theamelioration of female farmers’ livelihood con-straints. (PLEASE, THIS IS NOT LITT RE-VIEW. THIS IS SIMPLY A COMPARISONOF FINDINGS WITH RESULTS FROM PRE-VIOUS STUDIES )

It is obvious that although poor female farmersin the study area are conscious of innovating inorder to overcome their present precarious so-cio-economic situation, they are however pre-cluded from benefiting from opportunities opento them due to various constraining factors, ashave been identified above.


It has been revealed that the WIA program, asbeing implemented in Akwa Ibom State doesnot reach out to a large number of its intendedclientele base. This has resulted in an averagelevel of awareness and concomitant relativelylower level of adoption of innovative technologiesdisseminated by WIA extension officials. Thestudy also identified seven factors which combineto hinder the adoption of disseminated WIAtechnologies. The major constraints were: highcost of inputs, low income level of womenfarmers and lack of regular contact with WIAextension agents. Many reasons, backed by lit-erature, have been proffered for this trend, in-cluding the fact that only 59.2 percent respon-dents regarded WIA extension officials as theirsource of information on innovative agriculturaltechnologies. In order to enhance the success ofthe WIA programme in Akwa Ibom State, at-tempts should be made to ameliorate constraintswhich hinder extension officials’ access to their

potential clientele. To aid in this direction, ade-quate logistic support should be

provided to WIA extension agents so as tohelp enhance the process of contacting their ex-pectant clientele Technologies slated for dis-semination should be compatible to clientelesocio-economic and cultural base and emphasisshould be focused on follow-up activities, afterinitial group meetings. This would help to prac-ticalize disseminated technologies on the farmsand in the homes of potential adopters of tech-nological innovations. It may also be necessaryto attach credit schemes to the WIA program, interms of linking the various women groups tovarious credit agencies.

REFERENCES

1- Adetoun, B. A. (2000) Organization and Man-agement of Agricultural Extension Services forWomen Farmers in South-Western Nigeria. Centrefor Sustainable Development and Gender Issues.Abner Publishing, Ibadan, (Abstract).2- Akpabio, I.A (1997) Extension CommunicationMethods as a Strategy for Agricultural Developmentin Akadep In: Issues in Sustainable Agricultural De-velopment. Fabiyi, Y.L. and M.G. Nyienakuna (eds).Vol. 1. August. Dorand Publishers. Uyo. P. 122.3- Akpabio, I.A. (2005). Human Agriculture: SocialThemes in Agricultural Development. Abaam Pub.Co. Uyo. 219 pp.4- Akpabio, I.A. (2005b). ‘Beneficiary Perceptionsof the Gender Specific Extension Delivery Servicein Akwa-Ibom, Nigeria.’ Gender and Behaviour396-405.5- Anderson, J. R and P. B. Hazell (1989). Variabilityin Grain Yields: Implications for Agricultural Researchand Policy in Developing Countries. Baltimore:John Hopkins University Press, p. 23 6- Arokoyo, T. (1996). Agricultural Technology De-velopment and Dissemination. A Case Study ofGhana and Nigeria. CTA. The Netherlands, p. 5.7- Baker, E. (2005) Institutional Barriers to TechnologyDiffusion in Rural Africa. University of Massachusetts,Amherst. 30pp.8- Bunch, R. (1982). Two Ears of a Corn. USA:World Neighbours, p. 11.9- Brown, L., Feldstein, H., Haddad, L.C. & Pena,A. Quisumbing (2001). Generating Food Securityin the year 2020: Women as Producers, Gatekeepersand Shock Absorbers” (In) The Unfinished Agenda.Pinstrup-Andersen and Pandya-Lorch (eds) IFPRI.


Inte

rnat

iona

l Jo

urna

l of

Agr

icul

tura

l M

anag

emen

t &

Dev

elop

men

t, 2

(3):

215

-222

, Sep

tem

ber,

201

2.

222

Washington D.C. pp. 205-209.10- Diao, X., P. Hazell, D. Resnick and J. Thurlow(2007). The Role of Agriculture in Development:Implications for SSA. Research Paper 153. IFPRI.Washington DC, p.511- Department for International Development (DFID2004) “Gender” Fact Sheet. Sept. London.12- Dove, M. R. (1991) Forester’s Beliefs aboutFarmers: An Agenda for Social Science Research.In: Social Forestry- EAPI Working Paper (Abstract).13- Due, J., Mudenda, T., & Miller, C. (1983). HowDo Rural Women Perceive Development? A CaseStudy in Zambia. Report no 83-E-265. University ofIllinois. Dept of Agricultural Economics, Urbana, IL.14- Eicher, C. (1992) “African Agricultural Devel-opment Strategies. In: Alternative DevelopmentStrategies in sub-Saharan Africa. Stewart, F., L.Sanjaya and S. Wengwen (eds) Macmillan. London.15- Eshiet, E. (2007) “Level of Participation ofWomen Farmers in the Women in Agriculture Pro-gramme in Akwa Ibom State”. Unpublished M. Scthesis. Dept of Agricultural Economics and Extension,Universilty of Uyo, Nigeria. 88 pp.16- FAO, (2000). Rural Women and DevelopmentAgencies. Rome. P. 2317- Hardarker J. B., Huirne R.B.M & Anderson,J.R. (1997). Coping with Risks in Agriculture. CAB.Int. Oxon. UK.18- Hebinck, P., Franzel, S., & Richards, P. (2007).Adopters, Testers or Pseudo Adopters? Dynamicsof the Use of Improved Tree Fallow by Farmers inWestern Kenya. AgroSystems 94(2): 509-515.19- Jafry, T. (2000). Women, Human Capital andLivelihood – an Ergonomics Perspective. 0D1 NaturalResources Perspectives (Abstract).20- Pannell, D.J. (1999). Social and Economic Chal-lenges in the Development of Complex FarmingSystems. Agroforestry Systems 45: 393-409.21- Obinne, C. 1994). Fundamentals of AgriculturalExtension. ABIC Publishers, Enugu, Nigeria.22- Reid, M.S. (2001). Paticipatory Technology De-velopment for Agro-forestry: An Innovation – DecisionApproach. Centre for Enviornmental ManagementWorking Paper R01/121. University of Leeds.23- Reij, C., & Waters-Bayer, A. (2001). An InitialAnalysis of Farmers. Innovations in Africa. in: ASource of Inspiration for Agricultural Development.Reij and Water Bayer eds. Earthscan Publicationspp. 3-22.24- Rogers, E. M. (1995). Diffusion of Innovations,4th Edition. The Free Press, NewYork.25- Swinkels, R., & Franzel, S. (1997). Adoption

Potential of Hedgerow Intercropping in Maize BasedCropping Systems in the Highlands of WesternKenya. Economic and farmers’ evaluation. Experi-mental Agriculture 33:211-223.26- Technical Centre for Agricultural and Rural Co-operation (CTA-2000) The Economic Role of Womenin Agricultural and Rural Development: PromotingIncome-Generating Activities. C.T.A. Wageningen.The Netherlands, 56pp.27- Udoh, A.J. (2001). Agricultural Extension De-velopment Administration. Etofia Media ServicesLtd. Uyo. p. 39.28- Udokah, E. (2004). Information Source Preferenceof Small Scale Farmers in Akwa Ibom State. Un-Published M. Sc thesis. Dept of Agricultural Eco-nomics and Extension, University of Uyo, Nigeria,71pp.29- World Bank (2003). Sudan Country EconomicMemorandum: Stabilisation and Reconstruction.Washington DC. World Bank, p. 3.30- World Food Programme (1998). Food Aid toSupport Technology Adoption Among Small ScaleAgriculturists. WFP, Rome, Italy.


Inte

rnat

iona

l Jo

urna

l of

Agr

icul

tura

l M

anag

emen

t &

Dev

elop

men

t, 2

(3):

223

-227

, Sep

tem

ber,

201

2.

223

A Logistic Regression Analysis: Agro-Technical Factors

Impressible from Fish Farming in Rice Fields, North

of Iran

Seyyed Ali Noorhosseini-Niyaki 1*, Mohammad Sadegh Allahyari 2

Keywords: Rice-Fish Farming, Tech-nical-Agronomic Factors,Pest, Weed, Plow, Fertilizer

Received:5 April 2012,Accepted: 25 July 2012 This study was carried out to identify Technical-Agronomic

Factors Impressible from Fish Farming in Rice Fields.This investigation carried out by descriptive survey duringJuly-August 2009. Studied cities including Talesh, Rezvanshahrand Masal set in Tavalesh region near to Caspian Sea, Northof Iran. The questionnaire validity and reliability were determinedto enhance the dependability of the results. Data were collectedfrom 184 respondents (61 adopters and 123 non-adopters)randomly sampled from selected villages and analyzed usinglogistic regression analysis. Results showed that there was asignificant positive relationship (p<0.05) between biologicalcontrol of pests in rice fields and the fish farming in ricefields. Also, there was a significant negative relationship(p<0.10) between the fish farming in rice fields and variablesof quantity using pesticide of Diazinon in rice fields andnumber of plows in rice fields.

Abstract


1Department of Agronomy, Lahidjan Branch, Islamic Azad University, Lahidjan, Iran2Department of Agricultural Management, Rasht Branch, Islamic Azad University, Rasht, Iran* Corresponding author’s email: [email protected]

Inte

rnat

iona

l Jo

urna

l of

Agr

icul

tura

l M

anag

emen

t &

Dev

elop

men

t, 2

(3):

223

-227

, Sep

tem

ber,

201

2.

224

INTRODUCTION

The earliest records of fish culture in rice-fields originate from China, circa 2000 yearsago, followed by India, 1500 years ago. Othercountries with a recorded history of rice-fishculture are Indonesia, Malaysia, Thailand, Japan,Madagascar, Italy, Russia, Vietnam (Rothuis,1998), Egypt, Philippines, Bangladesh, Cambodia,Korea and other countries (Saikia and Das,2008; Frei and Becker, 2005; Halwart, 1998).Also in northern Iran rice-fish farming is a newfarming system (Karami et al., 2006; Noorhos-seini and Allahyari, 2010). Integrated rice-fishfarming offers a solution to economic problemof farmers by contributing to food, income andnutrition. Not only the adequate supply of car-bohydrate, but also the supply of animal proteinis significant through rice-fish farming. Fish,particularly small fish, are rich in micronutrientsand vitamins, and thus human nutrition can begreatly improved through fish consumption(Larsen et al., 2000; Ahmed and Garnett, 2011;Noorhosseini and Mohammadi, 2010). Manyreports suggest that integrated rice-fish farmingis ecologically sound because fish improve soilfertility by increasing the availability of nitrogenand phosphorus (Giap et al., 2005; Dugan etal., 2006; Noorhosseini, 2012).The feeding be-havior of fish in rice fields causes aeration ofthe water. Integrated rice-fish farming is alsobeing regarded as an important element of inte-grated pest management (IPM) in rice crops(Berg, 2001; Hilbrands and Yzerman, 2004). Atthe farm level rice-fish integration reduces useof fertilizer, pesticides and herbicides in thefield. Such reduction of costs lowers farmer’seconomic load and increases their additionalincome from fish sale (Noorhosseini, 2010;Noorhosseini and Radjabi, 2010). Also, integratedrice-fish farming gave higher rice yields and

fetched higher gross margin than sole rice crop-ping system (Das et al., 2002; Hossain et al.,2005). Ahmed and Garnett (2011) Reported thathigher yields can be achieved by increasinginputs in the integrated farming system. Integratedrice-fish farming also provides various socioe-conomic and environmental benefits. Never-theless, only a small number of farmers are in-volved in integrated rice-fish farming due to alack of technical knowledge, and an aversion tothe risks associated with flood and drought. Inaddition, Ahmed et al., (2011) Reported thatrice-fish farming is as production efficient asrice monoculture and that integrated performsbetter in terms of cost and technical efficiencycompared with alternate rice-fish farming. How-ever, a lack of technical knowledge of farmers,high production costs and risks associated withflood and drought are inhibiting more widespreadadoption of the practice. Our objective was toidentify technical-agronomic factors impressiblefrom fish farming in rice fields in north of Iran.


Studied Location and Survey: This study wascarried out by survey during July and August2009. Studied area including Talesh, Rezvanshahrand Masal set in Tavalesh region of Guilanprovince near to Caspian Sea, north of Iran(Figure 1). Respondents were selected fromrural area and categorized into adopters andnon-adopters of integrated rice-fish farming.Totally 184 farmers were selected by stratifiedrandom sampling technique using the table fordetermining the sample from given populationdeveloped by Bartlett et al., (2001) that including61 (33.15%) adopters and 123 (66.85%) non-adopters (Table 1). This survey was conductedin using a questionnaire with open-ended ques-tions. The questionnaire was pre-tested by in-

Fish Farming in Rice Fields/ Noorhosseini and Allahyari

Talesh Masal Rezvanshahr Total

IRFF Adopters PopulationIRFF Adopters Sample SizeIRFF Non-adopters Sample Size

311938

312856

171429

171429

Table 1: Total sample size used in the study area

Source: Survey Results, 2009

Inte

rnat

iona

l Jo

urna

l of

Agr

icul

tura

l M

anag

emen

t &

Dev

elop

men

t, 2

(3):

223

-227

, Sep

tem

ber,

201

2.

225

terviewing three farmers (not included in thestudy). After some modifications, it was testedagain with 10 other respondents.

Statistical Analysis: The technical-agronomicvariables for the two groups were examinedusing logistic regression model. The dependentvariable was the adoption of rice-fish farming.The dependent variable was dichotomized witha value 1 if a farmer was an adopter of integratedrice-fish farming and 0 if non-adopter. The def-initions and measurement of variables are presentin Table 2. AF, AH, MW, AP, BP and WI wereentered in the model as dummy variables. Theother variables namely QF, QD and NP wereentered as continuous variables. Data analysiswas conducted with Statistical Package forSocial Sciences (SPSS 18).

The model was specified as follows;

Y= f (AF, QF, AH, MW, AP, BP, QD, NP, WI)


The results of the Logit likelihood regressionmodel indicated that the overall predictive powerof the model (70.1%) is quite high, while thesignificant Chi square (p<0.01) is indicative of

strength of the joint effect of the covariates onprobability of adoption among farmers in thezone. The results also showed that the decisionon adoption of rice-fish farming is determinedby biological control of pests in rice fields (BP),quantity using Diazinon in rice fields (QD) andnumber of plows in rice fields (NP) which havesignificant influence. Also, the Wald indicatingthe relative contribution of individual variableto probability of adoption of rice-fish farmingshowed that BP (4.538) was the most importantfactor determining choice of adoption of rice-fish farming among the rice farmers. Generally,the results of logistic regression show that therewas a significant positive relationship (p<0.05)between biological control of pests in rice fieldsand the fish farming in rice fields (Table 3).These results were similar to Frei and Becker(2005) and Kathiresan (2007). In other words,fish farming in rice fields reduced the use ofchemical control methods that is light reason ofrice-fish farming system sustainability. Also,there was a significant negative relationship(p<0.10) between the fish farming in rice fieldsand quantity using Diazinon in rice fields (Table3). These results are consistent with Saikia andDas (2008), Salehi and Momen Nia (2006).The use of chemical control methods reducedwith adoption of rice-fish farming which is also

compatible with sustainable agriculture. Sincemany fish species feed partly on the aquaticfauna, it has been assumed that they can act asbiological control agents in rice fields. Concurrentrice and fish culture decrease pesticides appli-


Figure 1: Site of study

Dependent variableY = Adoption

Adopters = 1, Non adopters = 0

Independent variable AF = Application of Chemical FertilizersQF = Quantity Using Chemical FertilizersAH = Application of HerbicidesMW = Mechanical Control of WeedAP = Application of PesticidesBP = Biological Control of PestsQD = Quantity Using Diazinon NP = Number of PlowsWI = Accessibility to Water Supply for Irrigation

Yes = 1, No = 0Kg/haYes = 1, No = 0Yes = 1, No = 0Yes = 1, No = 0Yes = 1, No = 0Kg/haNumber in yearVery mach = 5, Much = 4, Intermediate = 3, Little = 2, Verylittle = 1

Table 2: Definition of variables included in the regression model

Inte

rnat

iona

l Jo

urna

l of

Agr

icul

tura

l M

anag

emen

t &

Dev

elop

men

t, 2

(3):

223

-227

, Sep

tem

ber,

201

2.

226

cation compare monoculture (Berg, 2002;Noorhosseini, 2011). In general, common carp,being an omnivorous feeder, seems to be themost promising species in controlling insectsand snails [15]. In addition, there was a significantnegative relationship (p<0.10) between fishfarming in rice fields and number of plows inrice fields (Table 3). According to results, byadoption of rice-fish farming, number of plowswas reduced. It seems that the decrease of tillagefrequency among adopters, caused by farmsoccupied by fishes is the time dimension.

CONCLUSION

In general, our results suggest that biologicalcontrol of pests in rice fields, quantity usingDiazinon in rice fields, and numbers of plowsin rice fields were the most important techni-cal-agronomic factors impressible from fishfarming in rice fields. In other words, adoptersof rice-fish farming used less chemical materialsin order to control pests and reduce the numberof plows. Furthermore, society health and envi-ronment sustainability will be saved and theyreach more profit that is economical. Also, sinceaquaculture requires resources such as pond,land, water and other inputs, poor farmers cannotafford the requirements. As a target to understandand meet their needs and to access the commonwater resources available in their rice-fields,rice-fish farming is the most appropriate tech-nology in recent times.

REFERENCES

1- Ahmed, N. & Garnett, T.S. (2011). IntegratedRice-fish Farming in Bangladesh: Meeting the chal-lenges of food security. Food Security. 3(1): 81-92.2- Ahmed, N., Zander, K.K., & Garnett, S.T. (2011).Socioeconomic Aspects of Rice-fish Farming inBangladesh: Opportunities, Challenges and ProductionEfficiency. The Australian Journal of Agriculturaland Resource Economics, 55: 199–219.3- Bartlett, J.E., J.W. Kotrlik and C.C. Higgins,(2001) Organizational Research: Determining Ap-propriate Sample Size in Survey Research. InformationTechnology, Learning and Performance Journal,19(1): 43-50.4- Berg, H. (2001) Pesticide Use in Rice and Rice-Fish Farms in the Mekong Delta, Vietnam. CropProtection, 20: 897–905.5- Berg, H. (2002) Rice Monoculture and IntegratedRice–Fish Farming in the Mekong Delta, Vietnam-economic and ecological considerations. EcologicalEconomics, 41: 95– 107.6- Das, D.R., Quddus, M.A., Khan, A.H. & Nur-e-Elahi, M. (2002). Farmers Participatory ProductivityEvaluation of Integrated Rice and Fish Systems inTransplanted Aman rice. Pakistan Journal Agronomi.1(2-3): 105-106.7- Dugan, P., Dey, M.M., & Sugunan, V.V. (2006).Fisheries and Water Productivity in Tropical RiverBasins: Enhancing Food Security and Livelihoodsby Managing Water for Fish. Agricultural WaterManagement, 80: 262–275.8- Frei. M., & K., Becker. (2005). Integrated Rice-Fish Culture: Coupled Production Saves Resources.Natural Resources Forum. 29: 135–143.9- Giap, D.H., Yi, Y., & Lin, C.K. (2005). Effects of


B S.E. Wald Sig.

AFQFAHMWAPBPQDNPWIConstant

-19.1310.000

-21.47620.4930.6110.996-0.029-0.7170.112

20.822

25320.4770.001

18858.28740192.879

0.5580.4680.0160.3960.167

51110.003

0.0000.0050.0000.0001.2004.5383.3113.2730.4500.000

0.9990.9430.9991.0000.273

0.033**0.069*0.070*0.5021.000

Table 3: Logistic regression coefficients of the technical-agronomic factors affectingadoption of rice-fish farming

*** p<0.01, ** p<0.05 and * p<0.10-2 log likelihood = 204.830Chi square statistic = 28.943***Overall Correct predictions = 70.1%

Inte

rnat

iona

l Jo

urna

l of

Agr

icul

tura

l M

anag

emen

t &

Dev

elop

men

t, 2

(3):

223

-227

, Sep

tem

ber,

201

2.

227

Different Fertilization and Feeding Regimes on theProduction of Integrated Farming of Rice and PrawnMacrobrachium Rosenbergii (De Man). AquacultureResearch, 36: 292–299.10- Halwart, M. (1998). Trends in Rice-Fish Farming.FAO Aquaculture Newsletter, 18: 3-11.11- Hilbrands, A., & Yzerman, C. (2004). On-FarmFish Culture. Printed by: Digigrafi, Wageningen,the Netherlands. Second edition. 67p.12- Hossain, S.T., Sugimoto, H., Ahmed, G.J.U., &Islam, M.R. (2005). Effect of Integrated Rice-DuckFarming on Rice Yield, Farm Productivity, andRice-Provisioning Ability of Farmers. Asian Journalof Agriculture and Development. 2 (1): 79-86.13- Karami E. A., Rezaei Moghadam K., AhmadvandM., & Lari, M.B. (2006). Adoption of Rice- FishFarming (RFF) in Fars province. Iranian JournalAgriculture. Extension & Education. 2(2): 31-43.14- Larsen, T., Thilsted, S.H., Kongsbak, K., & Hansen,M. (2000). Whole Small Fish as a Rich CalciumSource. British Journal of Nutrition. 83: 191–196. 15- Noorhosseini-Niyaki, S.A. (2010). Fish Farmingin Rice Fields Toward Sustainable Agriculture: TheCase Study in Guilan Province. The First NationalConference on Sustainable and Cleaner Product,Esfahan of Iran. Nov 10-11, 2010.16- Noorhosseini-Niyaki, S.A. (2011). Ecologicaland Biological Effects of Fish Farming In RiceFields. Regional Congress of Sustainable ManagementScience-based in Agriculture and Natural Resources.Gorgan University of Agricultural Sciences and Nat-ural Resources, Iran. 21-22 May 2011. pp: 244-250.17- Noorhosseini-Niyaki, S.A. (2012) Productionof Fish Varieties in Rice Fields Simultaneously. Na-tional Conference on Food Industries, Young Re-searchers Club, Islamic Azad University, QuchanBranch, Iran.18- Noorhosseini-Niyaki, S.A., & Allahyari, M.S.(2010). Problems of Fish Farming in Rice Fields:The Case Study in Guilan Province. The FirstNational Conference of Aquatic Sciences, IslamicAzad University, Booshehr Branch, Iran. 23-24 Feb,2010. 15p.19- Noorhosseini-Niyaki, S.A., & Mohammadi, N.(2010). Environmental and Social Benefits of FishFarming in Rice Fields. The 4th Conference and Ex-hibition on Environmental Engineering, Tehran Uni-versity, Iran.20- Noorhosseini-Niyaki, S.A., & Radjabi, R. (2010).Decline Application of Insecticide and Herbicidesin Integrated Rice-Fish Farming: the Case Study inNorth of Iran. American-Eurasian Journal Agriculture

& Environ Science. 8(3): 334-338.21- Rothuis, A. (1998). Rice-Fish Culture in theMekong Delta, Vietnam: Constraint Analysis andAdaptive Research. Thesis Submitted for the Awardof the Degree of Doctor of Science. Department ofBiology, Laboratory of Ecology and Aquaculture,Katholieke Universiteit Leuven, Vietnam. 113p.22- Saikia. S.K., & Das. D.N. (2008). Rice-FishCulture and its Potential in Rural Development: ALesson From Apatani Farmers, Arunachal Pradesh,India Journal Agriculture Rural Development. 6(1&2):125-131.23- Salehi, H, Momen-Nia, M. (2006) The Benefitsof Fish and Rice Integrated Culture in Iran. IranianJournal Fish Science. 15(3):97-108.24- Kathiresan, R.M. (2007). Integration of Elementsof a Farming System for Sustainable Weed and PestManagement in the Tropics. Crop Prot 26: 424-429.


Instructions for Authors

The International Journal of Agricultural Management and l Development (IJAMAD) is an open access journal

that provides rapid publication of articles in all areas of the subject.

The Journal welcomes the submission of manuscripts that meet the general criteria of significance and scientific

excellence. Papers will be published approximately one month after acceptance.

Electronic submission of manuscripts is strongly encouraged, provided that the text, tables, and figures are included

in a single Microsoft Word file (preferably in Times New Roman font).

Submit manuscripts as e-mail attachment to the Editorial Office at: [email protected], [email protected].

A manuscript number will be mailed to the corresponding author same day or within 72 hours.

The cover letter should include the corresponding author's full address and telephone/fax numbers and should be in

an e-mail message sent to the Editor, with the file, whose name should begin with the first author's surname, as an attachment.

The International Journal of Agricultural Management and Development will only accept manuscripts

submitted as e-mail attachments.

Article Types

Three types of manuscripts may be submitted:

Regular articles: These should describe new and carefully confirmed findings, and experimental procedures should

be given in sufficient detail for others to verify the work. The length of a full paper should be the minimum required

to describe and interpret the work clearly.

Short Communications: A Short Communication is suitable for recording the results of complete small investigations

or giving details of new models or hypotheses, innovative methods, techniques or apparatus. The style of main

sections need not conform to that of full-length papers. Short communications are 2 to 4 printed pages (about 6 to 12

manuscript pages) in length.

Reviews: Submissions of reviews and perspectives covering topics of current interest are welcome and encouraged.

Reviews should be concise and no longer than 4-6 printed pages (about 12 to 18 manuscript pages). Reviews are also

peer-reviewed.

Review Process

All manuscripts are reviewed by an editor and members of the Editorial Board or qualified outside reviewers.

Decisions will be made as rapidly as possible, and the journal strives to return reviewers’ comments to authors within

3 weeks. The editorial board will re-review manuscripts that are accepted pending revision. It is the goal of the

IJAMAD to publish manuscripts within 8 weeks after submission.

Regular articles

All portions of the manuscript must be typed single-spaced and all pages numbered starting from the title page.

The Title should be a brief phrase describing the contents of the paper. The Title Page should include the authors' full

names and affiliations, the name of the corresponding author along with phone, fax and E-mail information. Present

addresses of authors should appear as a footnote.

The Abstract should be informative and completely self-explanatory, briefly present the topic, state the scope of the

experiments, indicate significant data, and point out major findings and conclusions. The Abstract should be 100 to

200 words in length.. Complete sentences, active verbs, and the third person should be used, and the abstract should

be written in the past tense. Standard nomenclature should be used and abbreviations should be avoided. No literature

should be cited.

Following the abstract, about 3 to 10 key words that will provide indexing references should be listed.

A list of non-standard Abbreviations should be added. In general, non-standard abbreviations should be used only

when the full term is very long and used often. Each abbreviation should be spelled out and introduced in parentheses

the first time it is used in the text. Only recommended SI units should be used. Authors should use the solidus

presentation (mg/ml). Standard abbreviations (such as ATP and DNA) need not be defined.

The Introduction should provide a clear statement of the problem, the relevant literature on the subject, and the

proposed approach or solution. It should be understandable to colleagues from a broad range of scientific disciplines.

Materials and methods should be complete enough to allow experiments to be reproduced. However, only truly

new procedures should be described in detail; previously published procedures should be cited, and important

modifications of published procedures should be mentioned briefly. Capitalize trade names and include the

manufacturer's name and address. Subheadings should be used. Methods in general use need not be described in detail.

Results should be presented with clarity and precision. The results should be written in the past tense when

describing findings in the authors' experiments. Previously published findings should be written in the present tense.

Results should be explained, but largely without referring to the literature. Discussion, speculation and detailed

interpretation of data should not be included in the Results but should be put into the Discussion section.

The Discussion should interpret the findings in view of the results obtained in this and in past studies on this topic.

State the conclusions in a few sentences at the end of the paper. The Results and Discussion sections can include sub-

headings, and when appropriate, both sections can be combined.

The Acknowledgments of people, grants, funds, etc should be brief.

Tables should be kept to a minimum and be designed to be as simple as possible. Tables are to be typed double-

spaced throughout, including headings and footnotes. Each table should be on a separate page, numbered consecutively

in Arabic numerals and supplied with a heading and a legend. Tables should be self-explanatory without reference to

the text. The details of the methods used in the experiments should preferably be described in the legend instead of in

the text. The same data should not be presented in both table and graph form or repeated in the text.

Figure legends should be typed in numerical order on a separate sheet. Graphics should be prepared using

applications capable of generating high resolution GIF, TIFF, JPEG or Powerpoint before pasting in the Microsoft

Word manuscript file. Tables should be prepared in Microsoft Word. Use Arabic numerals to designate figures and

upper case letters for their parts (Figure 1). Begin each legend with a title and include sufficient description so that

the figure is understandable without reading the text of the manuscript. Information given in legends should not be

repeated in the text.

References: Follow the APA Publication Manual (6th ed.) for all citations in text and reference list. It is the

author’s responsibility to insure strict adherence to APA.

For more see: http://www.ijamad.com/APA.pdf

In the text, a reference identified by means of an author‘s name should be followed by the date of the reference in

parentheses. When there are more than two authors, only the first author‘s name should be mentioned, followed by

’et al.,‘. In the event that an author cited has had two or more works published during the same year, the reference,

both in the text and in the reference list, should be identified by a lower case letter like ’a‘ and ’b‘ after the date to

distinguish the works.

Examples:

Abayomi (2000), Agindotan et al., (2003), (Kelebeni, 1983), (Usman and Smith, 1992), (Chege, 1998; Chukwura,

1987a,b; Tijani, 1993,1995), (Kumasi et al., 2001)

References should be listed at the end of the paper in alphabetical order. Articles in preparation or articles submitted

for publication, unpublished observations, personal communications, etc. should not be included in the reference list

but should only be mentioned in the article text (e.g., A. Kingori, University of Nairobi, Kenya, personal communication).

Journal names are abbreviated according to Chemical Abstracts. Authors are fully responsible for the accuracy of the

references.

Short Communications

Short Communications are limited to a maximum of two figures and one table. They should present a complete study

that is more limited in scope than is found in full-length papers. The items of manuscript preparation listed above

apply to Short Communications with the following differences: (1) Abstracts are limited to 100 words; (2) instead of

a separate Materials and Methods section, experimental procedures may be incorporated into Figure Legends and

Table footnotes; (3) Results and Discussion should be combined into a single section.

Proofs and Reprints: Electronic proofs will be sent (e-mail attachment) to the corresponding author as a PDF file.

Page proofs are considered to be the final version of the manuscript. With the exception of typographical or minor

clerical errors, no changes will be made in the manuscript at the proof stage. Because IJAMAD will be published

freely online to attract a wide audience), authors will have free electronic access to the full text (in both HTML and

PDF) of the article. Authors can freely download the PDF file from which they can print unlimited copies of their ar-

ticles.

Copyright: Submission of a manuscript implies: that the work described has not been published before (except in the

form of an abstract or as part of a published lecture, or thesis) that it is not under consideration for publication

elsewhere; that if and when the manuscript is accepted for publication, the authors agree to automatic transfer of the

copyright to the publisher.

Fees and Charges: Authors are required to pay a $100 handling fee.

www.ijamad.com

Aims and Scopes

International Journal of Agricultural Management and Development (IJAMAD) is an international

journal publishing Research Papers, Short Communications and Review on Agricultural Management

and related area. The journal covers all related topics on agricultural management and development.

Papers are welcome reporting studies in all aspects of agricultural management and development

including:

Agricultural ManagementFarm ManagementRural DevelopmentSustainable DevelopmentFarming SystemsAgricultural EconomicsAgricultural PolicyAgribusinessSocio-economic AspectsMarketingRural & Agricultural SociologyAgricultural ExtensionAgricultural Education

This journal is published in cooperation with Irainain Association of Agricultural Economic

Published by:Islamic Azad University, Rasht Branch, Iran

nuhfil.lecture.ub.ac.idnuhfil.lecture.ub.ac.id/files/2012/12/FINALseptember2012...PUBLISHER Islamic...

Documents

Transcript of nuhfil.lecture.ub.ac.idnuhfil.lecture.ub.ac.id/files/2012/12/FINALseptember2012...PUBLISHER Islamic...