African Journal of Food Science Research Vol. 2 (2), pp. 051-058, February, 2014. Available online at www.internationalscholarsjournals.org © International Scholars Journals Full Length Research Paper The impact of rural Agricultural development projects on agricultural products in preferred area of Benin Kashez Asiamah, Awuah Nulty and Kubby J. Department of Agricultural and Resource Economics, University for Development Studies, Tamale, Ghana. Accepted 22 January, 2014 In this study, data collected from 120 rural households located in two distinct socio-cultural locales of Benin was used to assess the impact of 20 development projects on agricultural productivity. A ‘with- without’ approach of impact evaluation is followed using ANOVA and econometric regressions. Results reveal no significant differences of projects on agricultural productivity between participants in the two study zones. Econometric regression estimates show significantly positive impacts on agricultural productivity for two selected project indicators in the two study zones. However, the goal achievement index was more remarked in the Adja area, where the projects were found to have better addressed development problems and provided higher impact. The results suggest the need to improve management of agricultural projects to enhance their impact. Likewise, objectives and activities of the projects should be oriented to deal better with development problems of rural people, in particular those of the poorest and marginalized communities. Key words: Productivity, rural projects, impact, Benin. INTRODUCTION Most less developed countries depend on rural areas for most of their survival and development resources. Agri- culture, which is the main activity of these areas, employs between 70% and 80% of the working population. Be- sides, local rural populations ensure food consumption from agricultural production, thereby guaranteeing house- hold food security (World Bank, 2008). In spite of this key role of rural areas in development, they are confronted more and more with severe problems, which retard their progress with consequent problems for livelihood survival of the local people. Natural resources such as land, forest and water do not stop degrading gradually. The decline in land fertility has resulted in decreased agricultural pro- ductivity. For decades, developed countries, international Corresponding author. E-mail: Kashez360@yahoo.com institutions such as the World Bank, FAO, UNDP, as well as non-governmental organizations NGOs) have been fighting ceaselessly to contribute to the development of rural areas via actions and interventions implemented through rural development projects. Such projects have often been designed, planned and implemented to help the rural people to develop their agriculture and to have better access to farm inputs to increase their income. In addition, most projects tend to strengthen the capacities of rural farmers through education, training and institutional support. Given the key role that project interventions play in the rural development process, it is imperative to assess both the specific and overall impacts of implemented projects. Several approaches to evaluate rural development projects have evolved over time. According to Kirkpatrick (1994), various appraisals of most projects have focused on cost-benefit or cost-effectiveness approaches by assessing project costs (monetary or non-monetary), in particular, their relation to alternative uses of the same http://www.internationalscholarsjournals.org/ Kashez et al. 051 resources and to the benefits being produced by the projects. However, some outputs of rural development projects such as capacity building and improvement of food security are sometimes difficult to measure and/or provide unsatisfactory results via the cost-benefit approach. Indeed, the decisive issue of a measure of project success is not whether the planned results have been achieved, but what impact the activities of the project have provided and whether they satisfy all the stakeholders. Consequently, project evaluations in recent times have focused on the impact evaluation approach, whereby project success emphasizes more broadly on whether the project had the desired effects on individuals, households and institutions and whether those effects are attributable to the project intervention. Accordingly, evaluating the impact of rural development projects on agricultural pro- ductivity becomes a challenge to deal with (GTZ, 2008). In evaluating projects, the central problem is how to isolate and to estimate their impacts on target groups. Since many other exogenous factors that are not related to project execution (government policy, market condi- tions, former experiences, etc.) also have an influence on target groups’ evolution, appraisal approaches of projects seem to be difficult. The literature proposes two main approaches with different concepts of measurement: the ‘before-after’ and ‘with-without’ approaches as illustrated in Figure 1 (Bauer, 2000) . According to Kerr and Kolavalli (1999) and Adekambi (2005), if the ‘with-without’ approach is designed in a consequent way to isolate the exogenous influences and to carry out the project impact only; it may provide more reliable results. This paper therefore focuses on the use of the ‘with-without’ approach to estimate the impact of rural development projects on agricultural productivity of farmers in two regions of Benin. Review of impact of projects on household livelihood Various authors have in the past focused on impact evaluation of development projects on sustainability. In Central America for example, a number of projects have promoted soil conservation or soil recuperation technologies that benefitted farmers through increase in productivity (Brunch, 2001). Likewise, Doppler and Bothe (1999) showed that the adoption of Cassia siamea in rural Benin improved soil fertility and agricultural productivity and led to an increase in the overall family income. This helped to reduce poverty of many rural farming households. As a result of increase in productivity and income due to the adoption of new technology for bean growing in South-Benin, Allogni et al. (2008) found that food expenditure increased and food security in households showed some improvement. Data from (2006) also show that, increases in agricul- agricultural productivity and income over the years due to rural development projects have undoubtedly raised food availability and kept food prices low, providing critically important benefits for extremely poor households that spend more than half their income on food (Kerr and Kollavali, 1999). Arguing in the same way, IFPRI (2001) reported that the project “Improving Food Security in Bangladesh” implemented since the 1980s resulted in a significantly increased availability of and access to food in rural areas of Bangladesh. In countries where starvation is disastrous for rural people, various projects are implemented to avoid malnutrition diseases and death, mainly among children. For example, (IDRC, 2003) found that 30 projects implemented in Ethiopia that focused on agriculture and water management saved more than 25% of rural communities from starvation, malnutrition diseases and death. The foregoing discussion shows an optimistic view of the adoption of technology leading to poverty alleviation through positive effects on consumer food prices, producer incomes and labourer waged incomes. In this scenario, higher productivity, better natural resource management and poverty alleviation are mutually reinforced and may lead to achievement of a sustainable food system (Winkleman, 1998). In contrast to this optimistic point of view, the pessimist sees the overall process of project implementation and technology adoption in agriculture biased towards wealthy people so that the poor are made worse off. The rich get richer while the poor get poorer resulting in social unrest and a decidedly unsustainable food system. The key relationship according to this framework is that tech- nologies, policies and institutions are biased in favor of wealthy farmers who have unequal access to assets to begin with. Their income rises when they adopt the improved technologies while the income of non-adopting farmers fall, many agricultural workers are displaced and some of those who remain, suffer from overexposure to poisonous chemicals (Winkleman, 1998; Kerr and Kolavalli, 1999). Finally, in impact, evaluation of a rural development project, another decisive discourse regarding success is whether the impacts are maintained after the project is completed, or in short, whether the project is sustainable (GTZ, 2008). Sustainability is seen as a result of the impact on sustainability of the production system where the project is implemented and of a long-term duration of the impact, even after the termination of the projects. Typically, sustainable projects are those designed and financed to build local capacities and to develop the ability of local people to manage and utilize the deve- lopment activities themselves, that is institutional and empowerment supports (Clayton et al., 1998; Uphoff, 1989; McAllister, 1999) . The capacity building is parti- cularly viewed as very important for sustainability and many institutions such as GTZ, World Bank, UNDP, etc. C indicator B Impact(income) A Beginning of the Project Figure 1. Illustration of project impact (Bauer, 2000). have directed their supports towards more technical assistance to achieve better capacity building of local people (Rudovist and Woodford-Berger, 1996; GTZ, 2008). METHODS AND ANALYSIS Theoretical and empirical modelling Methods for impact evaluation provided in the literature include systematic comparison, indicator trend function, econometric models and more complex system modelling (Bauer, 2000; Yabi, 2004). To estimate direct changes in agricultural productivity when the participation in project changes, the study focussed mainly on econometric models to evaluate the impacts. Supposing that IP is an index of projects and Y represents agricultural productivity, then if a unit change in index IP induces  unit change in Y, then:  Y /  IP   (1) By taking an integral of both sides of the equation (1), we obtain:  Y dIP   dIP (2)  IP It is evident from equation (2) that Y can be a function of IP and the general mathematical form of the regression model is expressed as: Y= f(X1, …, Xn, IP, Z1, …Zm, ) (3) Where,  is the error terms supposed to be a N(0, 2 ), the X1, …, Xn are production inputs and Z1, …, Zm other explanatory variables such as economic, social or human capital variables, etc. Considering its computational ease and extensive use in many studies, the Cobb-Douglas functional form is applied to Equation (3) to estimate the impacts of participation in projects on agricultural productivity. The empirical model estimated in this paper can be 052 Afr. J. Food Sci.Res. Without the Project With the Project Status Quo Time Evaluation specified as: ln( yi )    1 ln(LANDi )  2 ln(LABORi )  3 ln(CAPI i )  IPi  1PROi 2 AGEi  3SEX i  4 EDU i  5 ALPH i  6TEN i  i (4) Where: 1n(.) = the natural logarithm; i = the i th farmer; yi = the value of agricultural productivity expressed in FCFA/ha(FCFA is the local currency for Benin. 1 =655 FCFA ). The productivity is computed for major cultivated crops such as maize, cotton, cassava, nuts, beans and yams. LAND is the overall cultivated area in hectare (ha) while LABOUR, the total family labour used is expressed in man-days per ha and CAPITAL, the total capital used in FCFA per ha. The total capital is calculated as the total amount of input expenditures (seed, fertilizer, pesticide, hired labour, etc.). PRO is a dummy variable representing the project type; PRO = 1 if the project is integrated and 0 if it is single activity project; AGE is the age of the farmer (year); SEX is a dummy variable expressing the sex of the farmer; SEX = 1 for a man and 0 for a woman; EDU is a dummy education variable; EDU =1 if the farmer is formally educated and 0 if not. ALPH is a dummy informal education variable; ALPH = 1 if the farmer had received informal education and 0 if not; TEN is a dummy variable of land tenure; TEN = 1 if the cultivated land is secured and 0 if not. Socio- economic and demographic variables that were essentially measured in this study as dummy variables can accordingly not be logged. The IP are indicators of agricultural projects at a beneficiary level. By the nature of the approaches employed during the project implementation in the study zone, most local people had the opportunity to participate in several projects at the same time without any restrictions. In order to appreciate the presence of the projects at a beneficiary level, two indicators, relative to the projects in which each stakeholder was involved, were computed, namely: contact index (IC) and goal achievement index (IS). The IC is expressed as the sum of contact frequency at stakeholder level and Kashez et al. 053 mathematically defined as: if the farmer i was involved in no project ICi = 0 n if farmer was involved in n f ki projects; n = 1, 2, …, p k 1 Where; ICi represents the contact index of stakeholder i, fki the frequency of contact this stakeholder i made per week with the team of the k th project, n the number of projects in which he participated. As defined, the contact index considers only the frequency of contacts with the beneficiaries. It fails to take into account success in achievement of activities that were implemented through projects. Conversely, the goal achievement includes the overall success in achievement of objectives and activities of the projects, and thus computing its index could help in appreciating these aspects at the beneficiary level. As suggested by Sarbeck (1994), the utility value of the projects can be defined as: 1 (6) UA  gi * GAi g i Where; UA is the utility value with 0