Effect of Adoption of Irrigation on Rice Yield in the Municipality of Malanville, Benin

Effect of adoption of irrigation on rice yield in the municipality of Malanville, Benin

Abstract

The study employed Heckman selectivity model to determine factors influencing the adoption of irrigation and its effect on rice yield in Benin. Results from probit estimates indicate thatage of respondent, gender, extension services, access to credit, market participation, distance from home to irrigation scheme, use of tractor, and rate of fertilizer applicationare factors affecting the probability of irrigation adoption. The results from Heckman second stage estimates show that the adoption of irrigation contributes significantly to rice yield improvement. For robustness checks of the estimated effect of adoption of irrigation on rice yield, the propensity score matching method (PSM) was used. The results of the PSM indicate that the percentage increase in rice yield due to irrigation adoption varies between 63 % and 70 %. This confirms the finding of the Heckman estimates. Other variables explaining rice yield are education, extension services, access to credit, market participation, off farm income,use of tractor,labour, and fertilizer. These results imply that besides the adoption of irrigation the provision of complementary services are needed to achieve the objective of productivity improvement.

Key words: Rainfed,Irrigation, adoption, rice, yield, Heckman selectivity model, Benin

1. Introduction

Increases in rice production in Benin is often driven by an increase of the total area planted (Figure 1). Between 2000 and 2014, rice production has increased at an annual growth rate of 11 %. In the same period, the total area harvested has increased at an average annual rate of 8 %. Rice production was highly correlated (r = 0.98) withthe total cultivated area. Therefore expanding land area was the main factor for increasing rice production in Benin.

Figure 1. Trend of rice production and area harvested in Benin from 2000 to 2014.

Source: FAO (2016)

While land resources are fixed, arable land under cultivation cannot be increased indefinitely. The alternative is to improve the yield per hectare. This may be possible by adopting appropriate agricultural technologies. Irrigation has been identified as key to agricultural productivity improvement (Carruthers et al., 1997; Domenech and Ringler, 2013;FAO, 2003; Huanget al., 2006). Irrigation development increases returns to poor households in terms of their physical, human, and social capital and enables smallholders to achieve higher yields and revenues from crop production (Hussain and Hanjra, 2004). Irrigated farms also contribute to the creation of new employment opportunities through higher demand for farm labour due to additional labour needed for the construction and maintenance of irrigation infrastructure. Also during the dry season, when non-irrigated production is not possible, irrigated farms continue to operate and therefore have favorable effects on employment and wages (Hanjraet al., 2009).

There is no formal policy document available for irrigation development in Benin. However, the need for irrigation development in order to achieve high agricultural yields is clearly expressed and defined in the Plan Stratégique de Relance du Secteur Agricole (PSRSA) of Benin 2011-2015.It contains guidelines for improving the use of irrigation facilities in the country. The goal is the control and sustainable access to water for increasing agricultural productivity while safeguarding ecosystems. The main actions to be implemented focused on the support for the development of small schemes for rice development and promotion of intensive gardening, the promotion of pilot facilities for other cultures, the support to the realization of pastoral water development such as small dams, and the development of themajor valleys of Ouémé, Niger, Mono, Couffo and Pendjari. The PSRSA places particular emphasis on supporting the rehabilitation and strengthening of management capacity of the irrigation schemes. Despite this, the objective of the rice policy to be self-sufficient in rice production by 2015 was not met. National rice production in 2015 is far below the target of 600, 000 MT needed for self-sufficiency in rice production.

In this paper, the interestis focused on the following important question for irrigation policy:What informs farmers’ decision to participateinan irrigation project?Does irrigation adoption contribute to an improvement in yield among rice farmers in Benin?Analyzing farmers’ decision to adopt or not irrigation will facilitate the development of appropriate policies for sustainable development of irrigation in Benin. Theremainder of the paper is structured as follows. The nextsection reviewsfactors influencing the adoption of irrigation, followed by section 3 which provides the theoretical framework, the sampling frame and data collection procedure, and the methods of analysis. The results and discussionare presented in section 4. Finally section 5 concludes and gives policy implications of the findings.

1. Literature review on factors influencing adoption of irrigation

The decision to adopt an agricultural technology depends on a variety of factors. Grilliches (1957, 1960) studied the determinants of adoption of hybrid corn in United States of America and concluded that economic variablesuch asprofit was the major determinants for the adoption of hybrid corn. It explained the difference in adoption of hybrid corn between two regions by the difference in the average profit to be realized from the adoption. The rate of adoption tended to be faster for innovations that lead to higher profit and requiring investments that the farmer can undertake. Most of the pioneering works (Binswanger, 1974; Dinar and Yaron, 1992;Griliches, 1957) on the technology adoption were more aggregate and have concluded that the adoption of technologies is determined by the economic attributes.

In addition to the economic variables, others factors influence the technology adoption decision making process. Employing the Heckman two-stage procedures, Adeoti (2009) demonstrated that availability of labour and increases in number of extension visits per year are factors that increase the probability of adoption of the treadle pump technology in Ghana. Kamwamba-Mtethiwa et al.(2012) have conducted similar study in Malawi but using a simple logit model. Their results indicate that relatively well-off farmers have a significantly higher probability of adopting the treadle pumps than poor farmers. Although the above studies confirm the role of the economic factors in technology adoption decision making process, this raises questions about dissemination approaches and targeting. Treadle pumps are typically geared towards poor smallholders because it costs much less to operate, no need for fuel and having only limited repair and maintenance costs.

Getacheret al.(2013) also used a logit model to analyze the determinants of adoption of motor pumps for lifting irrigation water in Tigray, Northern Ethiopia. The results indicated that the most important determinants of the adoption of the small-scale irrigation technology include access to ground and surface water, yearly availability of water, sex of household head, level of education, access to credit and number of adult family members. Similar result was found by Namaraet al.(2007) in India. In their study the principal component analysis was used to classify farmers into five groups: very poor, poor, middle, rich and very rich. The multiple regression model indicates that the largest adopters of micro-irrigation belong to the middle and rich group of farmers. The most important determinants of micro-irrigation scheme adoption in India include access to groundwater, cropping pattern, availability of cash, and level of education, the social status and poverty status of the farmers.Only one source of water is considered in their study while access to surface water source like streams, dams, rivers and lakes should also be considered. The availability of these sources of water play significant role in irrigation technology adoption (Getacher et al., 2013).

A study carried out by Kohansal and Hadi (2009) reveals that variables such as farm size, educational level, farming as the first job, land slope, heterogeneity of soil and access to loan are the factors that influence the adoption of sprinkler irrigation in Iran. Besides that, Moreno and Sunding (2005) argue that sprinkler technology is an intermediate option in the sense that it is an improvement over gravity, but not as efficient as drip irrigation. Using nested logit model they analyzed the factors which influence irrigation technology (gravity, sprinkler anddrip) choice in California. Their results show that the irrigation technology adoption in California is more sensitive to financial incentives affecting input price and technology cost. An increase in the price of water appears to encourage adoption of the most efficient technology (drip) and discourage adoption of the least efficient option (gravity). The study of Moreno and Sunding has confirmed previous results found by Caswell and Zilberman (1985)whohave estimated the likelihood of using drip, sprinkler and surface irrigation by fruit growers in the central valley of California. They applied a multinomial logit using the land shares of each of the technologies as estimates of adoption probabilities. They concluded that higher water costs increase the likelihood of moving from the use of traditional irrigation technologies (furrow or flood irrigation) to the use of modern technologies such as sprinkler or drip. Although the studies ofCaswell and Zilberman (1985) and Moreno and Sunding (2005) have used different method of analysis and were done 20 years apart, they lead to the same results in term of the sensitivity of irrigators to the increase of water price.

In Greece, Koundouriet al.(2006) proposed a theoretical framework under which a farmer facing production uncertainty and incomplete information will adopt a more efficient irrigation technology. The condition of adoption was derived under the assumptions of farmers’ risk aversion, and assuming that uncertainty may come from two sources: randomness in climatic conditions and uncertainty of future profit flows associated with the use of the new technology.Koundouri et al. (2006) have concluded that a farmer will decide to adopt a more efficient irrigation technology if this provide him with an expected utility greater than in the case of non-adoption. They used the reduced form of the model to empirically test for the factors that affect adoption of a more efficient irrigation technology. The information variables proxies that were used are the educational level and the number of extension visits. The results indicate that the farmer’s human capital plays a significant role in the decision to adopt modern and more efficient irrigation equipment. The younger and the more educated the farmer is, the higher the probability that he/she will adopt new irrigation technologies.The works of Genuiset al.(2013) have analyzed the role of information transmission in adopting more efficient irrigation technology in Greece. They develop a theoretical model of technology adoption based on the maximization of the expected utility of profit.Then empiricalresults from olive-producing farmers in Crete, Greece indicate that both extension services and social learning are strong determinants of technology adoption and diffusion. Genuis et al. (2013) argue that the effectiveness of each of the two informational channels is enhanced by the presence of the other. Moreover, Abdulaiet al.(2011) have found from probit model estimates that access to extension agents, belonging to a farmer's organization and educational level, tend to influence adoption of irrigation technologies for safer vegetable productionin Kumasi (Ghana). Their result confirms the role of social learning in the adoption of a new agricultural technology in Ghana. New technologies are introduced either by farmers’ own experimentation or through formal sector intervention and the process of social learning encourages their diffusion (Conley and Udry, 2001, 2010; Genius et al., 2013; Koundouri et al., 2006; Marra et al.,2003; Rogers, 1995).

In summary, the factorsincreasing the probability of adoption of irrigation technologies includefarms and farmers characteristics, and institutional variables. Among these variables,level of education, extension visits and access to funds were found in majority of studies reviewed above as main factors explaining the adoption of irrigation.The price of water was also found as a strong determinant for the efficiency use of irrigation technology. Farmers are very sensitive to the increase of water price. In term of analytical framework, several methods were used in the literature to identify factors influencing the adoption of a new agricultural technology, especially factors affecting farmers’ decision to adopt irrigation technology. As much as several methods of analysis were available, the most used in irrigation technology adoption studies are a simple logit or probit model, Nested logit model and a multivariate choice model (Abdulai et al., 2011; Getacher et al., 2013; Kamwamba-Metethiwa, 2012; Kohansal and Hadi, 2009;Koundori et al., 2006; Moreno and Sunding, 2005; Richefort, 2008).

1. Materials and methods
2. Theoretical framework

This section presents a model that describes a farmer decision facing the adoption of irrigation under the expected utility maximization theory.This is used in the study for two reasons. First, it is more widely used in technology adoption literature (Caswell and Zilberman, 1985;Genius etal., 2013;Just and Peterson, 2010;Koundouri et al., 2006; Richefort, 2008). And second in the adoption of irrigation, capital and other financial investments are made based on the farmer’s expectation of a net profit or benefit in monetary or physical terms.

The farmer decision is modeled as a discrete choice. Generally the farmer can decide to participate or not to in irrigation project . The distribution of is as follow: for the case of non-participation and for the case of participation. Let assumed that the farmer iproduce one single output at price. In our case the output is rice. The production function is assumed continuous and twice differentiable. is the vector of inputs and r the vector of input prices. All prices are assumed non-random and the farmers are price-takers both in input and output markets. Water () is assumed to be an essential input in the production process. is the vector of the other inputs such as seed, labour, fertilizer, and agrochemicals.

Based on the above assumptions, the production function can be written as follow:

(1)

Then the production function of farmer i before adoption is and after adoption is . Equation (1) states that the production depends on the availability of water and others inputs. Indeed, water inputs is assumed to be the main factors for agricultural production as it makes difference between irrigated farms and non-irrigated farms. The change in the production function due to the adoption of irrigation is observed through the following equation.

- (2)

Where denotes the change observed in the production after the introduction of irrigation. The necessary condition for adoption is . In other words with the same inputs level the output will increase because of the adoption of irrigation. Or the same output level can be produced with a lower level of inputs.

The farmer’s decision to participate to an irrigation project implies an additional investment cost which in the case of this study include basically the irrigation water fees, membership feesand other services provided in the scheme. for the participants and for the non-participants .The profit maximization function of the farmer corresponding to the case of irrigation adoption is given by:

(3)

The first-order condition for irrigation water input derived from equation (3) is:

(4)

Equation (4) shows that the expected marginal productivity of the water equals the water input price over output price.

Formally the producer decision to participate to an irrigation project will verify the following inequalities:

0 (5)

Thus farmers adopt irrigation only when this could provide them an expected utility of profit greater than is the case without it.

3.2. Study area and sampling technique

3.2.1.Study area

The study was carried out in Benin in the municipality of Malanville. Malanville is borded to the North by the Republic of Niger, to the South by the municipalities of Kandi and Ségbana, to the West by the municipality of Karimama, and to the East by the Federal Republic of Nigeria. It covers an area of 3,016 km² of which 8,000 hectares is arable land (Figure 2).

Figure 2. Map of Benin and municipality of Malanville showing the study sites

The climate is Sudano-Sahelian and the area has only one rainy season which lasts for 5-6 months from May to October with a rainfall range between 700 mm and 1000 mm. This low rainfall negatively affects agricultural production. Malanville is characterized by high level of food insecurity and poverty (Table 1). The majority of its inhabitants are involved in subsistence agriculture and other economic activities such as fishing, livestock rearing, small business, trade and crafts. The major crops grown are maize, rice, millet, sorghum, cotton, and vegetables.

Malanville
Population (2013)
Religion (%)
Child schooling rate (%)
Literacy rate (%)
Main economic activities
Major crops
Food insecurity (%)
Poverty incidence (%) / 168, 641
Muslims: 80, Others: 20
28.4
14.1
Agriculture, fishing, livestock, small business, trade and crafts
Maize, rice, millet, sorghum, cotton, and vegetables
35
42.5

Table 1. Socio-economic characteristics of the municipality of Malanville

Source: Institut National de la Statistique et de l’Analyse Economique (INSAE) (2011; 2013).

The municipality of Malanville was chosen for this study because (1) the municipality is known as the largest rice producing area in Benin, (2) it is crossed by the Niger River and its tributaries which offer an important opportunity for rice production, and (3) also among the rice irrigation schemes developed by the State, the rice irrigation scheme of Malanville is the most important in terms of size, cropping season, and yield. The total irrigable land under the scheme is 516 ha and 400 ha were used in 2015 with rice produced twice in a year. The average rice yield under the scheme is about 5.7 MT/ha. The irrigation scheme of Malanville was constructed in 1970 and the water used is pumped from the Niger River and distributed into the farms through surface canals.There are approximately 1054 rice farmers in 2015operating on the scheme.

3.2.2.Sampling technique

Four districts out of five in the municipality were selected for the survey based on the criteria of the proximity to the irrigation scheme located in the district of Malanville and on the size of rice production. In each of these districts, two villages, one high rice producing village and one low rice producing village were purposively selected with the help of the extension officers in the municipality of Malanville. In total, eight villages within the municipality were covered during the survey.