You are on page 1of 8

The Influence of Marketing Cooperatives on the Bargaining Position of Producers in Farm-Gate Banana Prices: Evidences from Smallholders in Southern

Ethiopia
G.A. Woldie and E.A. Nuppenau University of Giessen Faculty of Agricultural Sciences Institute of Agricultural policy and Market Research Senckenbergstr. 3 D-35390 Giessen Germany Keywords: asymmetric information, banana, Ethiopia, farm-gate price negotiations, marketing, sequential bargaining model Abstract This paper empirically investigates farm-gate banana (Musa spp.) pricenegotiation under asymmetric information using a bilateral bargaining model. It makes an attempt to show how memberships in cooperatives affect the bargaining power of smallholders in southern Ethiopia. A regression analysis reveals that cooperative membership positively and significantly affects the bargaining power of smallholder banana growers. The result also confirms that access to central-marketprice information and past-trade relationship enables farmers to stick to their initially asked price. INTRODUCTION It is vehemently argued in literature that agricultural development is important in playing a great role in the expansion of other sectors (Lewis, 1954; Fei and Ranis, 1964). Agriculture can be a source of (surplus) labor required for industrialization, a source of inputs for industrial processing and a potential source of demand for manufactured products. As in many developing countries, agriculture in Ethiopia remains dominant and its well-being is crucial to the economy. As the industrial sector is meager, major proportions of the population still depend on the agricultural sector for their livelihood. According to Food and Agricultural Organization (FAO) (2004), the agricultural sector on average accounts for about 45% of the GDP, 72% of total export, 80% of employment, more than 90% of the total foreign exchange earnings, 70% of the raw material supplies for agro-industries and is also a major supplier of foodstuff for consumers in Ethiopia. Hence the role of the agricultural sector, in the overall response of the Ethiopian economy to reform and adjustment policies, is important given its relative dominance. A significant positive response to such policies is needed as a means of improving the overall performance of the economy. Despite the fact that agriculture is dominant in Ethiopias economy, the sector as in many sub-Saharan economies, is subsistence. Smallholders, who produce more than 90% of the total agricultural output and cultivate close to 95% of the total cropped land, dominate the sector (CSA, 2003). Smallholder rural households in Ethiopia, as in many developing countries, often live in villages located in remote areas. They are hardly integrated to national and international markets. Among others, because of high transaction costs, poor roads, information asymmetry and other market imperfections, they remain subsistence farmers. Integration of such smallholder peasants into the exchange economy is mainly regarded as the most important task for boosting growth, accelerating economic development, food security and poverty alleviation (Kherallah et al., 2000). A shift from traditional crop production to high-value crops is considered by the Ethiopian government as an important option in increasing smallholders income, in particular and rural development in general. In this regard, the agriculture-developmentled-industrialization strategy (ADLI) of the country envisages significant scope for
Proc. IC on Banana & Plantain in Africa Eds.: T. Dubois et al. Acta Hort. 879, ISHS 2010

161

achieving greater commercialization of smallholder agriculture (MoFED, 2002). The objective of the recent economic policy of the government of Ethiopia under a plan for accelerated and sustained development to end poverty (PASDEP) is also the diversification and specialization of production and exports ( MoFED, 2006). Fruits and vegetables in this regard are among the main agricultural products that are targeted by the development strategy and plan. After the end of the planning period of PASDEP, fruit and vegetables are expected to be the third important crop in terms of foreign exchange earnings next to coffee and oil seeds. To this end production of banana (Musa spp.), for instance, has recorded a tremendous growth in the last 15 years despite its market share is mainly limited to the domestic markets and exports to Djbouti and to lesser extent to the Middle East. Accordingly banana production has increased from 80,000 t in 1993 to 215,000 t in 2005 (FAO, 2007). In this regard, it has been argued that Ethiopia has a huge potential in high-value crop production particularly in the banana sector (Gabre-Mariam, 1999). The country is considered to have the potential to achieve trade gains in this sub-sector because of agroclimatic advantages, existence of cheap and surplus labor, and proximity to the Middle East and the major European markets. Reports show that fresh fruit production in general has been showing a modest rise in the last decade (FAO, 2007). Of particular importance in the fresh fruit sector is the potential of banana production as well as its marketing. The huge demand for banana and the market potential in the Middle East and Europe make the sector more significant than others. For instance, a study by the World Bank (2004) has shown that banana constituted 42% of EU imports of fruits from 19952001. In fact banana production and marketing is still traditional and faces different constraints. Past studies showed that marketing constraints to smallholders are diverse and interrelated (G.A. Woldie, pers.commun., 2006). Among others, farmers face: (1) lack of integration of smallholders to regional and export markets; (2) weak bargaining power; (3) absence of competitive buyers; (4) entry barriers to new entrants both into the regional and central markets; (5) price information asymmetry between the central traders and the regional traders, and between the regional traders and the producers; (6) low and unstable farm-gate prices; (7) absence of strong market institutions; and (8) high transaction costs. They are the major marketing constraints that widen the marketing margin and reduce the benefit farmers would reap from the sector. All the above factors contribute to smallholder-income diversification and poverty alleviation. A policy that loosens those constraints enough to improve farm-gate prices, farmers response to policy incentives and in turn improve productivity and increase the benefits is needed. However, the fact is that designing leveraged intervention measures that help the growth of income of smallholders in the sub-sector requires among others a detailed understanding of the existing production and marketing structures, their constraints, and the role of information and transaction costs as well as farmers organizations in improving the bargaining power of smallholder producers. Analyzing existing potential opportunities and identifying key intervention points and potential partners is therefore crucial for farmers. In general, moving from subsistence towards cash-crop production also requires a a better approach from the producer side to identify, study and then decide on the best market. This however requires a special role of market information and other marketrelated facilities. As imperfections in markets like asymmetric market-price information hinder possible gains that one could otherwise obtain, information asymmetry has to be resolved. For small-scale households, getting a reasonable price for their produce depends on farmers bargaining power in price negotiations. Market prices usually appear as an outcome of the price-negotiations process and it is deemed to affect next time production decision. Better prices are also incentives that shift resources towards more commercially-oriented agriculture. This is the case in Ethiopias fruit sector, particularly in small-scale banana production, where information asymmetry is prevalent. Both buyers and sellers do not have equal market information on the banana price in the central and other regional markets. Information affects the bargaining power of smallholders in price 162

negotiation and in turn determines the net profit of the households. Cognizant of the fact that asymmetric information is prevalent in the real world, and based on theoretical and methodological foundations of game theory, a number of experimental and empirical works on price formation have been conducted. In this regard, empirical literature on bargaining and asymmetric information is vast: Srivastava et al. (2000) deliver experimental studies on price and margin negotiations in marketing channels; Zusman (1989) on peasants risk aversion and the choice of marketing intermediaries as well as contracts by using a bargaining theory of equilibrium marketing contracts; Srivastava and Chakravarti (2002) wrote on the influence of explicit communication and trustworthiness reputations on bargaining using a bargaining model under asymmetric information; Jaleta and Gardebroek (2007) documented farm-gate price negotiations for tomato growers in Ethiopia by using a sequential game theoretic model under asymmetric information; and Sieg (2000) gave a case of medical malpractice disputes by estimating a bargaining model with asymmetric information. In spite of the fact that the fruit and vegetable sector is important in enhancing smallholder income, the literature that investigates problems and constraints affecting the sector is scanty, especially in the case of Ethiopia. A single work by Jaleta and Gardebroek (2007) provided an empirical investigation of farm-gate tomato price negotiation under asymmetric information in central Ethiopia by using a sequential bargaining model. They show that farmers are less committed to their initially asked price when the buyer speaks out the transaction price first and when their quality perceptions of the tomatoes being transacted differ from those of the buyers, as well as when tomato farms are at a large distance from the main road. Sellers stick more to their initially asked price when they know that the central market price is high. The initial ask-offer price spread decreases if the buyer speaks out the initial negotiation price first, but increases in the difference in quality perception between buyer and seller, and in the quantity of tomatoes being transacted. We assume that the behavior of tomato producers is different from that of fruit producers. Therefore this research undertakes a detailed investigation of the behavior of fruit growers in Ethiopia. It investigates how price negotiations under asymmetric information in the banana sub-sector is done, and how to identify factors that affect sellers bargaining position at farm-gate transaction, and at the same time, how the flow of market information facilitates transaction to take place in a short period. In addition, Jaleta and Gardebroek (2007) failed to consider the role of marketing institutions in affecting the bargaining and negotiation process at farm-gates. Cooperatives and producer associations are usually expected to play a great role in increasing market power of smallholders (Kherallah et al., 2000). This eventually also applies to Ethiopia and hence a need arises to test the validity of such arguments. The objective of this study is to analyze how banana prices are determined at farm gates and to identify factors that affect the bargaining power of smallholder banana growers in the negotiation process. To fulfill this objective a bilateral bargaining model under asymmetric information is adopted. The current study will help us in bridging gaps in previous studies and in giving a full picture of high-value crops production and marketing in Ethiopia. Finally, looking at such relationships empirically has policy implications on the institutional setups that improve contracts and the supply chain in the banana sub-sector. The scope of the study is limited to outlining a few factors that explain bargaining power. Transaction costs and the role of credit that could affect bargaining power are not considered in the current study. MATERIALS AND METHODS For conducting the research, primary data is used. Particularly, farm-gate banana price negotiations from 118 transactions from June 25 to July 28, 2007 were recorded. The survey was aimed at fulfilling the objective of assessing factors that affect bargaining power in the banana sub-sector. Issues such as: (1) personal characteristics of buyers and sellers; (2) relationships between them; (3) price and market information; and (4) quantity traded at a time of transaction were exclusively addressed. The record of farm-gate price 163

negotiation data was supplemented by a wholesale-banana-price record from the Addis Ababa central fruit and vegetables market. The methodological framework used in the study is illustrated below following the procedure of Jaleta and Gardebroek (2007). The proxy for bargaining power s that shows how the seller sticks to his initial asking price is first defined as: P * pb ,1 s = p s ,1 pb,1 where ps,1 and pb,1 define the sellers initially asked price and buyers initially offered price respectively at period 1, while P* defines the final agreed price. If s = 1, the seller is said to be fully stacked to his initial asking price which implies P* = ps,1 = ps,N, where N referes to time period. Conversely, the buyer is fully committed to his initial price if s = 0, implying P* = pb,1 = pb,N. Based on the above formula and once the important parameters are obtained, the empirical model can be formulated to estimate to what extent banana sellers and buyers at farm-gate transaction stick to their initial price quotes. Here the model can be specified in such a way that parameter s can be regressed against different economic and noneconomic variables that are deemed to affect bargaining power of the seller. Cognizant of the above facts the functional form of the empirical model can then be put as: s = 0 + 1 Z + I + R + Q + u where Z is a set of different personal characteristics of buyers and sellers, I explains information related variables that include access to price information and number of potential buyers that visited the seller, R refers to variables that explain relationship between the two in the form of frequency of earlier trade relationship, while Q defines the quantity under transaction, which is fixed during the transaction period. 0, 1, , , and refer to parameters to be estimated, and u is an error term. RESULTS AND DISCUSSION Before we go to the empirical results and discussions, a summary of different descriptive statistics of both sellers and buyers characteristics, used in the empirical model is given in Tables 1 and 2. With regard to sellers characteristics, the mean age was 48.1 years while the education level (interms of year of schooling) was 4.7 years. The corresponding figures for the buyers were 30.5 and 10.9 years, respectively, which implies buyers are younger and more educated than sellers (farmers). The average quantity of banana traded in a single transaction was about 494 kg. The average distance of the banana farm to the main road was about 0.78 km. The value of the average final price both parties agreed at the end of negotiation was 1.02 Birr/kg (1,000 Birr; 0.125 at the time of the survey). The result also showed that about 89% of sellers have centralmarket-price information, although the reliability of the obtained information is in question as the main sources of central-market-price information are traders. Sellers ranked such information as less reliable (Table 2). In terms of how recent the information they obtained was, sellers used information which was about 14 days old, and there was on average a 0.13 Birr/kg difference in price between what they heard and what was actually recorded at the central market by the authors. With regard to initially asked and offered prices, sellers asked an average price of 1.14 Birr/kg while buyers initially offered an average price of 0.94 Birr/kg. This was more or less near to the final price that sellers agreed selling their produce. The finding implied that there was a 0.20 Birr/kg difference between what sellers initially ask and what buyers offered. This implies sellers do not get the price they asked for. One explanation for this could be the low baragining power of sellers in the negotatiation process ceteris paribus. Alternatively, the average relative bargaining power of farmers could be observed from the measure of s, which equalled 0.38, implying that the final agreed price on average depended for 38% on sellers initial ask price and for 62% on the price what buyers initially offered.

164

Bearing in mind the facts which were obtained from the descriptive analysis above, an empirical investigation was made to identify factors that determined the bargaining power of smallholder banana sellers at farm-gate price negotiation. The regression result is given in Table 3. The estimation result, which explains the bargaining power of sellers using the proxy, shows the relative deviation of initial ask price from the final price. It reveals that the bargaining power of sellers is positively and significantly influenced by access to central-market-price information. The bargaining power of smallholder banana grower increases with access to central-market-price information. It is higher (by 10.5%) compared to those who had no central-market-price information during transaction. Previous trade relationships with a given buyer (measured in terms of frequency of previous transactions with the same buyer) has a positive impact on the bargaining power of seller. The result reveals that a one unit increase in the frequency of trade relationship leads to a 0.25 unit increase in the bargaining power of sellers. This may be due to the fact that, over time, the seller got to know the behavior of the buyer and learnt from previous expriences with him, which enables the seller to stick more to the initial asking price. Whether a buyer or seller declares the transaction price first has also significant impacts on explaining the bargaining power parameter. Sellers usually let buyers declare the transaction price first in order to get an idea on the value of their product, which is because buyers are deemed to have better central-market-price information. Sellers (farmers) then can estimate their own price based on what the buyers (traders) quoted. The result depicts that the bargaining power of sellers was negatively affected if the buyer declares the transaction price first. There is a 33% bargaining power loss in those cases where the buyer declares the transaction price first during the negotiation process. This implies that the seller may look for a higher price, assuming that the price declared by the buyer may be the lowest possible price while alternatively buyers do not deviate much from what they initially quoted. This is a typical case in the banana sub-sector and in most small-scale markets where trading partners are always suspicious about the behaviour of their opponents. The other important variable that significantly explains bargaining power is cooperative membership. Recently, farmers started to organize themselves and form marketing cooperatives to mitigate the negative impact that they would have faced from individual merchants. In spite of the fact that these cooperatives have: (1) organizational problems in terms of materials, finance and human resource; (2) face sabotage from the regional traders; and (3) are unable to penetrate the central market, they are becoming an alternative for farmers and a threat for the organized merchants as an increased trend in farmers organization means a decline in the market share of the merchants. The existence of cooperatives therefore expected to enable farmers to receive a slightly improved price by improving bargaining power. Cooperative membership was included in the empirical model to investigate whether a cooperative member seller adheres to the initial ask price or not. The bargaining power of farmers who are members of cooperatives was higher than those who are not (on average by 10%). Other parameters such as age of seller and buyer, education level of buyer and seller, quantity traded at the time of transaction, and distance to the main road could not explain sellers commitment to the asking price. Distance to the main road was expected to affect bargaining power negatively as distant farmers are usally visted by few traders and face high transportation costs. However, farmers included in the study were on average only 0.78 km from the main road, which could be the reason why distance did not explain bargaining power. CONCLUSIONS For countries like Ethiopia, where smallholder agriculture is dominant, integrating farmers into markets by promoting production of cash crops is usually regarded as an important task for achieving the objective of food security and rural development. 165

Enhancing the income of smallholders mainly depends on how much farmers receive from their produce, which is influenced by how markets and institutions are organized. Taking into consideration that there is asymmetric information in the banana subsector, which complicates transactions, an attempt was made to identify factors, which determine the relative bargaining power of smallholder banana producers in southern Ethiopia at farm-gate. The results suggested that, among others, access to information, cooperative membership and frequency of earlier trade relationships explains bargaining power. The implications of the findings are that improving access of farmers to market price information and facilitating the information transmission would minimize information asymmetry and in turn improve bargaining power of smallholders. Supporting local marketing institutions by giving organizational and capacity building (such as training and finance) will improve the bargaining power of smallholders and enhance market integration. The existing marketing cooperatives should be strengthened to expand actively their outreach to those areas where smallholders are not a member so that smallholders actively participate and improve their farm-gate price. ACKNOWLEDGMENTS Special thanks to the German Academic Exchange Service (DAAD) for sponsoring this study. We are also very grateful to Arba Minch University, Ethiopia for technical support during data collection. The authors would also like to thank two anonymous reviewers for helpful comments. Literature Cited Central Statistical Agency of Ethiopia (CSA). 2003. Agricultural Sample Enumeration 2001-2002. CSA, Addis Ababa. Food and Agricultural Organization (FAO). 2004. Statistical year book. FAO, Rome. Food and Agricultural Organization (FAO). 2007. faostat.fao.org Fei, J.C. and Ranis, G. 1964. Development of the Labor Surplus Economy-Theory and Policy. Richard D.Irvin, Homewood. Gabre-Mariam, S. 1999. Banana production and utilization in Ethiopia. Research Report 35, EARO, Addis Ababa. Jaleta, M. and Gardebroek, C. 2007. Farm-gate tomato price negotiations under asymmetric information. Agri. Econ. 36:245251. Kherallah, M., Delgado, C., Gabre-Madhin, E., Minot, N. and Johnson, M. 2000. The road half traveled: Agricultural market reform in Sub-Saharan Africa. IFPRI Issue Brief No. 2. IFPRI, Washington DC. Lewis, A. 1954. Economic development with unlimited supplies of labour. Manchester School of Economics and Social Studies 22:139191. Ministry of Finance and Economic Development (MoFED). 2002. Ethiopia: Sustainable Development and Poverty reduction Program. Federal Democratic Republic of Ethiopia, Addis Ababa. Ministry of Finance & Economic Development (MoFED). 2006. Ethiopia: Building on Progress A Plan for Accelerated and Sustained Development to End Poverty (PASDEP). Federal Democratic Republic of Ethiopia, Addis Ababa. Sieg, H. 2000. Estimating a bargaining model with asymmetric information: evidences from medical malpractice dispuites. J. Political Economy 108:10061021. Srivastava, J., Chakravarti, D. and Rapoport, A. 2000. Price and margin negotiations in marketing channels: An experimental study of sequential bargaining under one-sided uncertainty and opportunity cost of delay. Mktg. sc. 19:163184. Srivastava, J. and Chakravarti, D. 2002. Bargaining under assymetric information in marketing channels: Influence of explicit communication and trustworthiness reputations. Working Paper. Robert H. Smith School of Business, University of Maryland Maryland. World Bank, 2004. Background Report: Opportunities and Challenges for Developing High-Value Agricultural Exports in Ethiopia. World Bank, Washington D.C. 166

Zusman, P. 1989. Peasants risk aversion and the choice of marketing intermediaries and contracts A bargaining theory of equilibrium marketing contracts. p.297317. In: P. Bardhan (ed.), The Economic Theory of Agrarian Institutions. Oxford University Press, Oxford.

Tables

Table 1. Descriptive statistics of variables used in the empirical model on farm-gate banana (Musa spp.) price-negotiation under asymmetric information using a bilateral bargaining model. Variables Age of sellers (years) Education level of sellers (years) Age of buyers (years) Education level of buyers (years) Quantity of banana traded during a single transaction (100 kg) Distance of banana farm from the main road (km) Presence of earlier trade relationship (1= yes, 0 = no) Frequency of earlier trade relationship (0, 1, 2, 3, 4 = if >3 times) Seller (farmer) is a member of a cooperative (1=yes, 0=No) Who declared price first during transaction (1 = seller, 2 = buyer) Final price both parties agreed (Birr/kg) Seller had central-market-price information (1 = yes, 0 = no) Sellers initial ask price (Birr/kg) Buyers initial offer price (Birr/kg) Ask-offer spread () (Birr/kg) Sellers commitment (as measured by s) Mean 48.07 4.64 30.53 10.93 4.94 0.78 1.00 4.00 0.17 1.23 1.02 0.89 1.14 0.94 0.20 0.38 Std. dev. 9.18 2.38 1.38 1.62 3.15 0.57 0.00 0.00 0.38 0.42 0.04 0.31 0.08 0.06 0.12 0.14 Min. 27 0 27 5 1 0.05 1 4 0 1 1 0 1 0.9 0 0 Max. 77 12 35 12 14 3 1 4 1 2 1.2 1 1.25 1.2 0.3 0.75

167

Table 2. Descriptive statistics on banana wholesale price information of the Addis Ababa central fruit market. Variables Sellers central-market-price information (Birr/kg) Actually observed and recorded wholesale price (Birr/kg) Differences between actual and informed price (Birr/kg) Duration since price information (days) Reliability of information source as perceived by seller (1 = very reliable, 2 = reliable, 3 = less reliable, 4 = not reliable) Mean Std. dev. 1.30 0.08 1.42 0.03 -0.13 0.10 13.61 6.35 2.74 0.50 Min Max 1.2 1.5 1.4 1.5 -0.25 0.1 3 36 1 4

Table 3. Regression result of the bargaining power parameter. Explanatory variables Age of seller (years) Age of buyer (years) Education level of seller (years) Education level of buyer (years) Distance of farm to the main road (km) Earlier trade relationships (relationships) Who declared the transaction price first Knowledge of central-market-price information by seller Quantity traded in transaction (100 kg) Marketing cooperative membership Constant Coefficients 0.001 -0.006 0.007 -0.009 0.015 0.250*** -0.332*** 0.105** -0.007 0.099*** 0.483** Std. error 0.001 0.007 0.011 0.007 0.017 0.043 0.024 0.047 0.030 0.033 0.236 T-value 0.819 -0.897 1.281 -1.006 0.647 5.701 -14.532 1.991 -0.216 2.816 2.207

Adjusted R2 = 0.783. *, **, and *** indicate significance level of 0.1, 0.05 and 0.01, respectively.

168