Collective Entrepreneurship: A Comparison between two Producer Organizations in the Ethiopian Honey Sector

Paper prepared for the EMNet 2011 in Cyprus (Dec. 1 – 3)

C. Chagwiza 1, R. Muradian 1, R. Ruben 1, W. Tessema 2

1* Centre for International Development Issues Nijmegen, Radboud University, the Netherlands Email:[email protected] 2 Department of Agricultural Economics, University, P.O Box 05, Awassa,

ABSTRACT Collective entrepreneurship has emerged as a viable strategy for producer enterprises to accrue economic benefits and improved market access. Producers’ participation and collective action is the core of the value chain development. This paper focuses on the comparison of 2 honey producers` enterprises in district, Ethiopia. We argue that the collective entrepreneurial capacity of collective producer groups may vary depending on the form of organizations involved (our case: cooperatives and private limited companies (PLCs)). The producer enterprises considered are operating in the same area; they both received training on honey production and quality management provided by SNV and most importantly, are both suppliers of a lead firm, honey processing company and exporter to the European Union. By comparing the groups across a number of variables, our results show that as compared to cooperatives, PLCs have succeeded in their economic performance, concerted upgrading and integration into the value chains. Our findings are consistent with collective action theory, which claims that the likelihood of collective action is higher in small user groups. Additionally, the PLCs have managed to establish good relations with the processor which can be seen in the value chain financing. This paper further suggests that engaging the private players like processors is pivotal in facilitating innovativeness and entrepreneurship among producers, rural development and socio-economic betterment of smallholder producers.

KEYWORDS: Collective entrepreneurship, collective action, cooperatives, honey, PLCs, pro- poor, value chain development, and Ethiopia 1 THEORETICAL BACKGROUND The manner in which producer groups functions and relate with the downstream value chain players (private processor-our case) determines whether or not they are able to accrue economic rents. As Mangus and Piters (2010) put it, the private sector is interested in the output whilst the small scale producer is interested in improved market access. Organising the producers into groups bridges these two symbiotic needs, but it is not a matter of simply being “organised”, how producers are organised is essential. This reflects the importance of the type/form of producer institution. Market intermediaries, for example downstream actors like processors, can play a significant role in stimulating entrepreneurship among small scale producer groups. In this study, we seek to explain the process by which individuals collaborate as a group to seize wealth creation opportunities by comparing 2 different collective enterprises, cooperatives and private limited companies (PLCs 1). The main important differences between these 2 groups are: (i) the size of membership; primary cooperatives can have several hundreds of members while the PLCs are allowed to have a maximum of 50 members (ii) the cooperatives are strongly regulated by the cooperative offices whereas PLCs does not have much external influence and finally (iii) unlike in cooperatives, PLC members can buy shares.

Collective entrepreneurship is a key component of economic growth and development. The organization`s entrepreneurship co-aligns it with the market conditions by building and modifying organization`s resources, competences, and organizational architecture to respond (Bratnicki, 2005) and influence market conditions and development This phenomenon is described as a process by which entrepreneurs engage in collective action for the economic and social betterment of a locality by means of some transformation of social norms, values, and networks for the production of goods or services (Connell, 1999) and taking risks together (Trompenaars and Hampden-Turner, 2002). Stewart (1989) suggests that collective entrepreneurship implies an increase in the ability of each member of the group whenever they are confronted by the continued demands and opportunities from the environment. The concept of collective entrepreneurship has now evolved as a strategy for producers to overcome business challenges and accrue economic benefits. However, Cook and Plunkett (2006) proposed that collective entrepreneurship begs for more research.

Enhancement of entrepreneurial capacities has recently become part of the agenda of global rural development interventions. Having said this, we argue that the producers’ participation and collective action is the core of the value chain development 2 since they are the producers of the commodities. However, Cook and Plunkett (2006) points out that for any form of a collective organization to achieve the highest performance, members` decisions about their own on-farm activities and investments should be aligned with the cooperative. This is a risk associated with collective entrepreneurship that producers should be willing to take.

1 Despite the differences highlighted between the 2 groups, it is important to note that a PLC is also a form of a producer group formed as an alternative to a cooperative. Just like cooperatives, a PLC is formed and owned by a group of farmers 2 Value chain development implies the inclusion of upstream value chain players through guaranteed marketing outlets and the appropriate response to quality and standards requested by downstream value chain players

Vast literature on natural resource management has already demonstrated the advantages of collective action. The literature on the factors that are likely to affect collective action, particularly in natural resource management, offers numerous lessons that can be applied to collective action in marketing (Markelova and Meinzen-Dick, 2009). Social and economic heterogeneity, group size, and the mediating role played by institutions (coops and PLCs – our case) have been highlighted in literature (Poteete and Ostrom, 2004; Agrawal, 2000) as factors that influence collective action. Among the other factors that can affect collective action, group size has been mentioned by several researchers (Olson, 1965; Agrawal and Goyal, 2001; Hussi et al. 1993). For instance, Agrawal and Goyal (2001) argue that the costs of monitoring rise more than proportionately as group size increases. In the same vein, Bandiera et al (2005) argues that on one hand, the institutional features that make collective action successful, such as monitoring, are more easily implemented in small groups. On the other hand, as compared to small groups, there might be economies of scale in large groups.

Olson (1965) addressed the “free rider” problem, whereby the contributions offered by group members toward collective goals decline with increases in group size. Olson hypothesized that “unless the number of individuals in a group is quite small . . . , rational, self-interested individuals will not act to achieve their common or group interests unless certain conditions are present.” He further mentioned “privileged group and special incentive mechanisms” as some of the conditions. A privileged group is described by Olson as one that functions in a manner that aligns member preferences and reduces organizational inefficiencies. In larger groups, every member benefit even without their contributions to collective action. Olson (1971) proposes a strategy to solve this problem by creating individual selective incentives that will induce individuals to contribute to a collective good as a by-product of their contribution.

Hussi et al (1993) points out that transaction and managerial costs of cooperation increase faster than the gains as group size increases beyond a certain level. In the same line, Hardin (1982) argue that the larger the number of people who must be coordinated, the higher the costs of organizing them to an effective level. Thus basically, transaction costs are expected to be proportionally higher in larger groups as compared to smaller groups. Additionally, resource characteristics and diversity among members influence transaction costs of collective action (Ostrom, 1994). There are social norms and values like trust, loyalty, social capital etc, which can play a big role in facilitating the success of collective action. Bonding social capital (involved in the linkages between different social groups, such as between suppliers and buyers) typically is formed in small groups (O`Brien et al 2005) and the potential of social capital as highlighted by Kirsten (2004) is that each partner will forego opportunistic behaviour and thereby lowering transaction costs.

In addition to factors influencing the success of collective action discussed above, Aiken and Hage (1971) identified another important factor that can hinder innovativeness of a collective group or organization. In the current study we refer to technological innovation which is defined by Devaux et al (2007) as improvements in the way commodities are produced or transformed. According to Aiken and Hage, the older the organization, the more bureaucratic and the less receptive it is to innovation. Younger organizations are believed to be in a better position to embrace new technologies and more willing to innovate than older organizations.

The current study focuses on the collective entrepreneurial capacities of different honey producing enterprises. We seek to examine how collective entrepreneurship capacities of these enterprises can transform honey production in Masha area into a competitive sector that could promote higher exports and stable income growth among beekeepers. Two types of producer groups (cooperatives and PLCs) exist in the study area and they are both producer collective enterprises. The area of study was an interesting choice due to various reasons. NGOs like The Netherlands Development Organization (SNV) and Non-Timber Forest Products (NTFP) are working in partnership in promotion of honey in the area. Since the forest in the study area play a very crucial role in the livelihoods of the local people, NTFP project aims to contribute to poverty alleviation and environmental conservation by facilitating certification of the forest products like honey and coffee. The area is characterized by producing “Tropical Forest Honey”, which allows the processor, Bezamar, to supply a niche market with organic honey. The two producer marketing organizations, cooperatives and private limited companies (PLCs), were considered for the study because they are both the main suppliers of the lead firm, Bezamar. In addition, both groups received training on quality financed through SNV and NTFP grants and also from the NTFP project. Hence, it is interesting to assess the effects of the interventions on collective entrepreneurship capacities of the 2 groups. The following conceptual framework (Figure 1) is proposed to guide and explain this phenomenon among honey producing groups using different dimensions:

Collective entrepreneurship

Dimensions of entrepreneurship

 Incentive schemes for board members  Price  Marketing  Productivity  Innovation and technological transfer (Adopting transitional beehives )

 Income

Figure 1: Conceptual Framework

The framework presented above focuses on how collective enterprises, through various dimensions of entrepreneurship, aim to be profitable, transfer competitive and responsible business practices along the corporate value chain. Thus we can argue that the cooperatives and PLCs are especially important for capacity-building and for empowering small producers to engage in market forces. Muradian et al. (2011) identified the alignment of incentives; the possibilities of coordination (organizational structure and value chain governance); attitudes towards risk; psychological traits and social abilities of managers as determinants of collective entrepreneurship. On the other hand, Yan and Sorenson (2003) points out that the collective capability of both identifying and responding to opportunities are important components of collective entrepreneurship.”

The main objective of this study is to empirically compare how the value chain development intervention has facilitated collective entrepreneurship between cooperatives and PLCs and how that has translated to productivity and commercialization of honey between these 2 groups. The general objectives of this study are:  To compare the entrepreneurial capacities between the coops and PLCs  To evaluate the influence of different forms of producer organizations (cooperatives and PLCs) on honey commercialization and innovativeness among producers  To identify constraints and opportunities for honey marketing among the cooperatives and PLCs members  To assess the producers` perceptions on the transitional beehives as compared to traditional hives

The remainder of the paper is structured as follows: the next Section 2 provides a brief background of the Ethiopian honey sector. Section 3 presents the sampling techniques, type of data collected and the methods used for data analysis. Empirical results are presented in Section 4. Finally, concluding remarks are presented in Section 5.

2 THE ETHIOPIAN HONEY SECTOR Beekeeping is a traditional important farming activity to harvest honey for many rural people in Ethiopia (Agonafir, 2005). Ethiopia`s total honey production stands at approximately 39,700 tons per year (GDS, 2009). The country is one of the five biggest wax producers with an average annual export estimated at 3000 tones (EEPD, 2006). This makes the country the leading honey producer in Africa and one of the ten largest honey-producing countries in the world. Moreover, it stands as the world’s 4th largest beeswax producer. The main importers of honey are the USA, Japan and EU countries.

Seeing the potential of the sub-sector, different stakeholders (the government, non-governmental organizations, etc.) have initiated development interventions in the country as a whole and the study area (Masha) in particular. The government has given more attention to the promotion of improved hives (transitional and modern) provided with subsidized prices through the Ministry of Agriculture and Rural Development extension system. This policy has triggered the participation of women in beekeeping activities. The NGOs have facilitated the adoption and use of low-cost and appropriate hive technologies and provided training to the beekeepers. However, this rapid promotion of improved beekeeping technologies has not been well-embraced by the smallholder producers and has been constrained by different problems which are not well documented especially for the Masha area.

Small-scale producers are the most important honey producers in Ethiopia. The main buyers for the Masha honey are private traders (local merchants), local Tej (Ethiopian traditional honey wine) brewers and processors, for example, the lead firm Bezamar, a honey processing, trading and exporting company. A lead firm can be described as a firm that have forward and backward commercial linkages with a number of small-medium enterprises within the value chain and holds a significant market share in the sector. The linkage of the private sector and honey producers and their institutions (e.g. farmers groups) is considered essential along the value chain to effectively link production, processing, value additions and the market. Thus, to satisfy the market requirements on quality and volumes producers (suppliers) and buyers (processors) need to closely align their activities in order to coordinate their transactions in an efficient manner and become profitable. As a result, their degree of interdependence is increasing. Bezamar is one of the key entrepreneurs who have assisted in facilitating the transformation of the honey sector. SNV promoted a mutually beneficial relationship between the producer groups and the lead firm through providing for the producer groups, a training on quality and technology transfer; and business development services to Bezamar.

The sector is still underdeveloped as small scale farmers often lack access to improved hives and international markets. Ethiopia’s honey production remains traditional in the sense that farmers tend to stick to the traditional way of beekeeping and thus productivity is low. This is evidenced from Kerealem et al (2009) who highlighted that about 94 to 97 percent of bees are still kept in traditional hives. Improved way of beekeeping is generally low and thus the overall potential yield of the sector is relatively low. Hence, the products obtained from this sub sector are still low as compared to the potential of the country.

Beekeeping requires relatively simple techniques that can be easily managed and it doesn't require investment to acquire big land size, which is often a problem for the poor rural dwellers (Debela, 2010). However, this depends on the type of honey-bee production system used. In Ethiopia, there are 3 types of bee husbandry systems being exercised namely; traditional, transitional and modern beekeeping. In the traditional way of beekeeping, the hives are made out of logs, bark, reeds, gourds and clay pots. The hive has to be hanged on top of a tree. The traditional beehives are costless but yield very low output. Transitional beehive refers to a hive technology that is between the traditional and the modern one and is managed at the backyard. Transitional hives have a higher honey yield over the traditional hives as well as provide a mechanism for monitoring the maturity of honey thus enabling harvest at optimal time. Finally, the modern beekeeping involves the frame hives which are created from rectangular and square boxes. These include Langsroth and Top Bar hives. The hives are more complex and difficult to build but they are easily transportable and generate greater quantities of better quality honey, which will command higher prices (Mehari, 2007).

The three types of beekeeping systems described have different costs, harvesting techniques and productivity expectations (GDS, 2009). By adopting the transitional and framed types of hives, alongside with proper training on management of the honeybees, producers can harvest high yields. The average yield potential for each hive is shown in Table 1 below!

Table 1: Average Yield Potential per each hive type Hive type Farmer’s Average Yield Research Center 3 Yield (kg/hive) (kg/hive) Traditional 5.0-7.0 NA Transitional (intermediate) 15.0-25.0 25 Framed (boxed) 30.0-45.0 40 Source: Global Development Solutions, LLC (2009)

3 Research centre refers to the centre built for scientific research As indicated in Table 1 above, transitional beehives have the potential to increase productivity of the producers from 5-7 kg/hive/harvest from the traditional beehives to 15-25 kg/hive/harvest. However, the probability of adoption of a new technology will depend on the difference in profitability between the new and old technologies and the ability of the farmer to perceive the advantages and efficiently utilize the new technology (Schultz, 1995).

3. METHODOLOGY This section describes the study area and the sampling strategy applied in selection of the respondents. Furthermore, the procedure for data collection, source of data and the kind of data collected are presented. Statistical methods used for data analysis are also presented.

3.1 Description of the Study Area This study was carried out in Masha district, South west part of Ethiopia. Masha is one of the 77 woredas in the Southern Nations, Nationalities and Peoples` Region of Ethiopia (SNNPRs) (CSA, 2005). Different nationalities are found in this zone and these include Shekicho, , Megzenger, Keficho, Amhara, Oromo and Guraghe. This woreda falls under the Sheka Administrative Zone (see map below in Figure 2).

Figure 2: Map showing the study area (Masha-Sheka Zone) Source: Abbute (2003)

More specifically, Masha is the administration center of Sheka Zone and is located 676km south west of Ethiopia from Addis Ababa along Addis-Jimma road. Geographically, the Zone lies between 7°24’–7°52’ N latitude and 35°13’–35°35’ E longitude. The altitudinal range of the areas in the Zone falls between 900–2700 m above sea level, and it receives high amount of rainfall, with an average of 1800–2200 mm annually. This woreda is notable for its relatively high forest cover as compared to other parts of Ethiopia. In general, the area is characterized by dense forests and woodlands that contain diverse plant species that provide surplus nectar and pollen to foraging bees. However, this important attribute is threatened by the rate of deforestation in the area that has aggravated in recent years due to increased rate of conversion to agriculture and monoculture plantations of coffee and tea.

3.2 Sampling Strategy and Data Collection Out of the 5 cooperatives / kebeles 4, 3 primary cooperatives (Genobay, Akach and Degele) were selected for the study. Due to the nature of the study as well as the topography of the study area, accessibility was taken into consideration in selecting these primary cooperatives. From these cooperatives, 60 producers were selected randomly from a list (obtained from the cooperative marketing office). However, 58 producers participated in the survey and the remaining two could not participate due to various reasons. As for the PLCs, all the members from the 3 PLCs (Chiefdale, Gada and Shatto) were included in the study since PLCs consists of relatively smaller numbers. Their list was also obtained from the cooperative marketing office. 43 honey producers were interviewed from the PLCs. Table 2 provides total membership (group size) and number of participants of all the groups that participated in the survey.

Table 2. Number of members in the producer groups that participated in the study

Organization Total no. of members No. of participants/group Gada PLC 14 12 Chiefdale PLC 17 15 Shatto PLC 19 16 Akach primary cooperative 445 19 Degele primary cooperative 270 19 Genobay primary cooperative 451 20

Data for this study were collected from March to April 2011 through a household survey from a total sample of 101 households. Generally, the semi-structured questionnaire was designed to generate information in the following areas: duration of membership, main source of income, dividend, annual income from honey, number and type of beehive owned, constraints of beekeeping, perceptions of the producers on the transitional beehives, amount of honey yielded/hive, year of adoption of transitional beehives and beekeeping experience.

Face-to-face interviews were also conducted with the management team of the producer organizations (cooperatives and PLCs), private traders, the processor (Bezamar), and personnel from NGOs. More relevant secondary information and data were gathered from various institutions including the zonal and woreda sector offices. Documents from NGOs (SNV and NTFP) active in the study area were also consulted and reviewed.

3.3 Data Analysis Data were analyzed using the Statistical Package for Social Scientists (SPSS) 19.0 for Windows and Generalized Linear Models (GLM) of Statistical Analysis System (SAS) (2006). The effect of organization system (whether PLC or cooperative) and some demographic data (age, sex,

4 A kebele is defined in Ethiopia as the smallest administrative units, below the municipality-district level. education, training) on adoption of transitional beehives and quantities of honey produced from both transitional and traditional beehives were analyzed. The first analysis consisted of descriptive statistics in order to have a comprehensible profile of the sampled respondents. Descriptive statistics (percentages, frequencies, mean and standard deviations) were generated for gender, age, years of experience in beekeeping, education level and duration of membership. T-test and χ2 were employed to test the significance of continuous and discrete variables, respectively. Based on the theoretical framework we provided in the first section, our main hypotheses to be tested includes: (1) PLCs as institutions are more entrepreneurial and innovative than cooperatives and (2) More honey is commercialized through PLCs as compared to cooperatives.

To make our comparison easier, our research objectives were transformed into research questions. The research questions were operationalized as presented below:

Operationalization of the research questions

Research questions Indicators 1 Are there any differences in entrepreneurial capacities -Incentive schemes for board members between coops and PLCs -Dividend -Price -Productivity -Adopting transitional hives (innovation and technological transfer) 2 Which institution commercializes higher quantities of -Marketing (proportion of honey honey? marketed by groups)

4. EMPIRICAL RESULTS This section presents the findings of the research. The section specifically seeks to highlight the conditions that currently exist in the area in terms of production and marketing between 2 different producer groups in Masha district. Comparisons of the 2 samples are made with reference to the question of statistical significance, because the key objective of our study is to determine if the two groups differ from each other in any substantively significant ways. The presentation and discussion of results between and within the 2 different producer groups (cooperatives and PLCs) is guided by the hypotheses proposed in the previous section of the current study.

4.1 Socio-economic Characteristics of Respondents Most of the households interviewed were male-headed, while a small percentage was female- headed. More detailed information about the gender composition is provided in Table 3. Across all producer groups (coop and PLC), most beekeepers (83%) were 34 years or older. Table 2 shows the distribution of age across all groups. The findings are similar to the results obtained by Kugonza and Nabakabya (2008), which showed that over 82% of the beekeepers interviewed in Uganda were 30 years or older. Across all the respondents, about 78.3% of them went to school, with the majority having completed secondary school education and the others having completed high school. The remaining 22 respondents indicated that they have never been to school. As can be seen from Table 3, there were no much differences in education levels between members of cooperatives and PLCs. Table 3: Distribution of respondents according to age and education

PLCs (43) Coops (58) Number % Number % Gender Female 9 20.9 2 3.4 Male 34 79.1 56 96.6 Total 43 100 58 100

Age groups 24 & below 1 2.3 1 1.7 25-34 9 20.9 7 12.1 35-44 14 32.6 21 36.2 45-54 14 32.6 13 22.4 55-64 5 11.6 15 25.9 65 & above 0 0 1 1.7 Total 43 100 58 100

Education level Never been to school 10 23.3 12 20.7 Primary school 8 18.6 15 25.9 Secondary school 19 44.2 24 41.4 High school 6 14.0 7 12.1 Total 43 100 58 100 Source: Survey results (2011) About 72 percent of all the respondents had beekeeping experience of 20 years or above. This implies that beekeeping is a long-standing activity in the area. The minimum and maximum numbers of years of experience indicated amongst respondents were 2 years and 48 years respectively. About 49 percent of the respondents have been members of their organizations for 8 years or above. The highest duration of membership was 19 years, mainly by cooperatives members since the cooperatives existed long way back. PLCs are relatively new organizations since they were formed as an alternative to the cooperatives.

Across the whole sampled respondents in both producer organizations, honey was ranked as the highest source of income, followed by kocho and then livestock. This probably owes to the rich and thick forests which allow the residents to produce organic forest honey. Table 4 below shows the means of each of the sources of income as they were ranked by the individual households.

Table 4: Sources of income as ranked by respondents

Source (Mean rank) a 1 Honey 1.707 2 Kocho 1.802 3 Livestock 2.275

Source: Author research results a The lower the rank of a source, the greater is its importance

Between the groups, only 33% of cooperative members ranked honey as the highest income earner whereas in PLCs, 58% of the respondents indicated honey as their highest income earner. This is probably because cooperatives engage in the trading of different agricultural products which give them alternative higher incomes whereas for PLCs, their core business is on honey production. During our interviews the cooperative members highlighted that their production is diversified into different commodities, thus in addition to honey, they also trade commodities like peas and beans. The management team even highlighted this multi-purpose nature as one of the advantages of cooperatives that they serve their members in different aspects unlike in PLCs. Honey production is of high priority for PLCs which is reflected in the quantities they market.

4.2 Level of adoption of transitional beehives In the study area, the use of transitional beehives was generally low across all the respondents. However, significantly different results were recorded in the means of the number of transitional beehives adopted in the years 2007 and 2010 between cooperative and PLC members. A higher mean for the PLCs shown in Table 5 reveals that the PLCs had higher number of transitional beehives as compared to cooperatives. Similar results were obtained for the variable measuring a decrease or increase in rate of adoption of the transitional beehives. Here, the number of beehives adopted in 2007 were subtracted from those adopted in 2010 to show the increase (or otherwise) in the number of transitional hives. The results in Table 5 show a higher mean for the PLCs as compared to the cooperatives. This result is in agreement with Aiken and Hage (1971) who found out that the older the organization, the more bureaucratic and the less receptive it is to innovation. In our case, the PLCs are relatively younger organizations since they were formed later as an alternative to cooperatives. Table 5: Summary of independent-samples t test results Cooperatives (58) PLCs (43) (Difference; Coops - Variables PLCs)

Mean SD Mean SD T-stat P-value Socio-economic variable Beekeeping experience (years) 26.29 11.06 26.23 10.15 -0.028 0.978 Duration of membership (years) 13.03 4.35 6.12 1.12 -11.610 0.000***

Production and income variable Number of transitional beehives in 2007 0.31 1.08 2.09 3.06 3.654 0.001** Number of transitional beehives in 2010 1.48 1.64 4.56 3.71 5.080 0.000*** Change in number of transitional beehives 1.17 1.77 2.47 4.17 1.911 0.061* Production transitional 2007 (kilograms) 3.28 13.81 21.26 35.97 3.112 0.003** Production transitional 2010 (kilograms) 12.47 21.05 53.51 56.38 4.545 0.000*** Production change transitional (kilograms) 9.10 17.50 32.26 62.01 2.37 0.022** Productivity of honey (kg/hive) 2.47 1.83 3.48 2.14 2.560 0.012** Income honey sales 2008 (Birr) 1075.84 863.63 2542.59 2014.24 4.480 0.000*** Income honey sales 2010 (Birr) 1615.50 1229.49 4060.21 2740.78 5.456 0.000*** Dividend paid (second payment) (Birr) 14.93 46.18 276.95 308.72 5.520 0.000*** Source: Survey Data *significant at 10%; **significant at 5%; ***significant at 1% The beekeepers were asked whether they are willing to give up on traditional beehives and focus only on transitional beehives. 89% of all the respondents answered “Yes”, indicating that they are very much aware of the benefits that come along with the use of these improved beekeeping technologies. The respondents raised important reasons for their response. Among the most important were; transitional beehives are easier to manage, there is a huge increase in production, they can be managed from the backyard; and honey from transitional beehives has better quality than that from traditional beehives.

On the other hand, only 11% answered “No” and they highlighted a number of reasons why they are not willing to give up completely on their traditional beehives. According to the beekeepers, the major constraints in adopting the transitional beehives were; absconding (the total movement of honeybee colony by leaving the hive – Workneh, 2007) of bees from the transitional beehives due of ants and pests, uncertainty and risk that comes with just abandoning a large number of traditional beehives; and shortage of labour to monitor the transitional beehives.

4.2.1 Dividend (second payment) When asked if they received dividend from their organizations in the last production season, Table 9 (showing the Chi Square tests) shows that a large proportion of the PLC respondents answered Yes, whereas in cooperatives only a few also answered Yes. Furthermore, Table 9 reveals there was a statistically significant difference of the proportion of people who received dividend between the 2 groups (p<0.041). More interestingly, Table 5 shows that there was a significant difference in the amount received by the producers between the 2 groups. PLCs had a higher mean as compared to cooperatives. Producers across the groups highlighted a number of reasons why they did not receive dividend from their organizations. Table 6 presents the reasons as they were ranked by the respondents. The information was generated only from the members (coop=40, PLC=21) who indicated that they did not receive the dividend.

Table 6: Reasons for not receiving dividend Number of Reason respondents % (coop = 40) 1 Re-invest for expansion of business 15 37 2 Not aware of the reason 12 30 3 The coop not yet audited 6 15 4 Did not sell through my organization 6 15 5 The processor delay to pay my cooperative 1 3 Total 40 100 Number of respondents % (PLC = 21) 1 Re-invest for expansion of business 10 48 2 The PLC not yet audited 4 19 3 My PLC did not make enough profits to distribute 3 14 4 Not aware of the reason 3 14 5 My PLC have just been audited, we are going to get soon 1 5 Total 21 100 Source: Survey results (2011) 4.2.2 Incentive system for management team By incentive we refer the payment or compensation given to the board members of an organization for keeping the firm running. The outcome of our interviews with the managers of the producer groups shows that the incentive schemes for board members among PLCs are more market-oriented where the board members do their job and expect to receive a proportion of the total profits made by the PLC as their payment. The board members of the PLCs get 10% of the net profit made by the organization. During the interviews, unlike in PLCs, all cooperative board managers revealed that they do not receive any payment for the work they are doing for their organization. They justified their unpaid managerial tasks rather as a contribution to the common good and a social duty. This implies that the incentive system in PLCs could be more organized and business focused as compared to that of cooperatives.

4.2.3 Honey productivity among groups Increased productivity can translate into increased producer income. Productivity is defined by Sartorius and Kirsten (2004) as the ratio of output to input or alternatively input to output. We used the quantile method of classifying data to classify the households. The data was arranged in descending order that is from the highest to the lowest. 101 households were grouped into 3 classes; high, medium and low, according to their productivity performance. Table 7 presents the productivity measures between and within the producer groups.

Productivity in honey production was calculated as follows:

Quantity of honey (kg) Honey productivity = ------Number of beehives

Table 7: Honey productivity PLCs (43) Cooperatives (58) Rank Gada (12) Chiefdale (15) Shatto (16) Akach (19) Degele (19) Genoby (20) % % % % % % 1 High 33 27 50 21 10 20 2 Medium 58 53 50 58 37 45 3 Low 9 20 0 21 53 35 Total % 100 100 100 100 100 100 Source: Own computation

The results of the statistical analysis were very interesting, showing a great variation between cooperatives and PLCs. Honey productivity levels significantly (p<0.05) varied between cooperatives and PLCs with the means (Table 5) showing that PLCs had higher productivity as compared to cooperatives. As can be seen in Table 7, all the 3 PLCs (Gada, Chiefdale and Shatto) had higher percentages of their member respondents in the high rank as compared to the 3 cooperatives. In the low productivity rank, the cooperatives had higher percentages of their members in the rank. This is in line with priori expectations that PLCs are more entrepreneurial, efficient and productive than cooperatives. However, amongst the PLCs (though lower than all the other 3 cooperatives) Chiefdale had a relatively higher percentage (20) of their respondents in the low productivity rank but the reason for this was not established. The findings of our study that productivity was higher amongst PLCs` members as compared to cooperatives` members are similar to the study findings by Bandiera et al (2005) where their study results revealed that a given worker cooperates more when she works in smaller groups than in larger groups.

4.2.4 Trend of Honey Prices between Producer Groups In this study we consider price given to the farmers as one measure of efficiency, entrepreneurship and market orientation of a producer group. To allow for easier comparisons in price trends across a period of 4 years (2007-2010) between cooperatives, PLCs and private traders, a line graph is presented. Because of lack of time series data on producer prices for analyzing price variation over time, average honey prices received by producers across four years from alternative marketing channels have been used in the analysis. Figure 3 shows the trend of how the prices have evolved amongst the three buyers across time.

18

16

14

Coop 12 PLC

Price/kg Private

10

8

6 2007 2008 2009 2010 Year

Figure 3: Price variation among buyers across years Source: Survey data (2011)

Figure 3 shows that across a period of 4 years, producers delivering their honey to PLCs are receiving better prices than those delivering to cooperatives and private traders. More interestingly, the trend shows that, except for 2007, the private traders are offering higher prices than cooperatives but less than PLCs. This probably explains why a good number of cooperative members market a proportion of their honey through private traders. In addition to paying relatively higher prices than the cooperatives, some of the cooperative producers interviewed also mentioned exceptional services they get from the private traders like advance payment. This means that the PLCs group succeeded in negotiating better market conditions for its members as compared to the cooperatives. Differences in honey prices reported by farmers may also be due to other factors (e.g. quality of honey, cleanliness, etc.) that cannot be easily measured.

4.2.5 Producer Perception about the Effect of Improved Hives and Training on Income An independent-samples t test was used to compare means of the income obtained from honey for the years 2008 and 2010 between PLCs and cooperatives. Table 5 clearly shows that PLCs obtained significantly higher incomes from honey sales as compared to cooperatives. The higher incomes may be attributed to higher efficiency and productivity; and the higher prices the PLC members receive from their organizations. PLCs show higher collective entrepreneurial capacity as compared to the cooperatives.

All the respondents were asked to rank their responses to the statement “I am very satisfied with my income over the past three years as a result of the training on beekeeping”. This was ranked on a 5-Likert scale from strongly disagree to strongly agree. About 76 percent of all the respondents indicated that they in agreement (whether strongly agree or agree) with the statement. However, Table 8 indicates that significant differences (p<.05) were observed between the means of PLCs and cooperatives. PLCs had a higher mean (4.28) as compared to cooperatives (3.79), showing that the PLCs agreed more to the statement that indeed they are satisfied with their income as a result of the training as compared to cooperatives. This tend to agree with the result shown in Table 5, which shows also a significant difference (p<.05) on the means of income obtained by the PLCs and cooperatives in 2010 as well the average income.

The respondents were also asked to give their responses on the statement “As a result of the training and access to modern technologies on beekeeping, I would confidently say that my household income has…” This statement was measured on 3-Likert scale with the following items; increased, slightly increased and remained the same. Across the two groups, about 65 percent indicated that their income has increased. As shown in Table 8, significant differences (p<.05) were observed in the means of PLC producers and cooperative producers. Cooperatives had a higher mean (1.71) on the statement that is they highly considered their income to have remained the same as compared to PLC members (1.26).

However, it should be noted that probably there are a number of factors that may have indeed resulted in the increase in income among the honey producers. Hence, we cannot make a general conclusion that the higher incomes were as a result of factors limited to training and access to modern beekeeping technologies, the transitional beehives.

Table 8: Summary of producers` opinion about their incomes Strongly Disagree Neutral Agree Strongly Mean & Std. T-value P-Value disagree agree I am very satisfied PLC (43) 0 3 3 16 21 4.28 with my income over (0.882) the past three years as a result of the training

Coop (58) 0 7 12 25 14 3.79 2.619 0.010* (0.951)

Increased Slightly Remained Mean & Std. T-value P-Value increased the same As a result of the PLC (43) 34 7 2 1.26 training and access to (0.539) modern technologies on beekeeping, I would confidently say Coop (58) that my household 32 11 15 1.71 -3.234 0.002** income has… (0.859)

Source: Survey Data Means are in bold and standard deviations are in parenthesis Significant at 10 % = * Significant at 5% = ** 4.3 Honey production and marketing In the study area, the mean annual honey yield from transitional beehives in 2010 was 29.94kg/hive/annum. The yield was above the average yield that is expected from transitional beehives at farm level, 15.0-25.0kg/hive/annum. Significant differences were revealed (see Table 5) in the quantities of honey produced from transitional beehives in 2007 and 2010 between cooperatives and the PLCs. Cooperative producers indicated lower means as compared to PLC member producers. Similar results were obtained for the change in production where PLCs indicated a higher mean of 32.26 against 9.10 for the cooperatives.

In this study, 3 marketing channels were identified among respondents; own organization (cooperative or PLC), private traders and neighboring cooperatives. Selling to own organization is the most popular direct market as it offers benefits like dividend and better prices as compared to the other buyers, for example, private traders. Table 9 gives summary of Pearson Chi Square test results showing the association of numerous variables with the type of organization, i.e. PLC or cooperative.

Table 9: Summary of Pearson χ2 test results Organization variables Coop (58) PLC (43) χ2 Value Asymp. Sig (2- sided) Age 5.582 0.349 Gender 7.776 0.005** Education 0.760 0.859 Yes Yes Dividend 18 22 4.183 0.041** Marketing channels: (1)Own organization 43 42 10.26 0.001** (2) Private traders 37 8 20.41 0.000*** (3)Neighboring coops 0 2 2.75 0.097

Willingness to give up 52 38 0.042 0.838 traditional beehives Advance payment from 0 5 7.095 0.008** organization Receive equipment 10 16 5.150 0.023** Note *significant at 10%; **significant at 5%; ***significant at 1% Source: Author, survey results

As indicated in Table 9 above, 43 (74%) of cooperative producers are selling their honey through their cooperative, whereas 42 (98%) of the PLC members are selling their output through the PLCs. Choice and utilization of marketing channel varies significantly (p<0.05) across producer organization (coop or PLC). This is probably due to the fact that the PLCs member producers gets higher dividend for their patronage as compared to cooperative members. The PLCs ought to have better and efficient services for their members as compared to cooperatives. Although both producer groups have enhanced the level of market integration of the smallholder farmers, we argue that the effect varies greatly between the cooperatives and PLCs. Our data reveal that members of PLCs over the years consistently market a considerable higher amount (ranging between 80%-96%) of their honey through their own group. This is comparable to the cooperatives members who market only between 35% and 49% of their produce through their own organization.

Another important marketing channel identified in the study area is the private trader. As shown in Table 9, the cooperatives` 37 (64%) members are marketing a proportion of their honey through private traders, whereas only 8 (19%) of PLCs are marketing part of their output as well through this channel. The results further shows a significant difference in the extent to which producers from the 2 groups uses this channel. At the time of the interview, a number of cooperative member beekeepers raised an important issue on why they market their honey through private traders. They highlighted that they get an advance payment from private traders which allows them to cover urgent expenses like school fees, food, etc; a service they are not getting from their cooperatives. This is different from PLCs where some of the members (12%) indicated that they get an advance payment from their own organization.

The respondents further indicated that private selling is the simplest form of marketing their honey since the merchants just buy whatever quality of honey you provide for them. However, they expressed concerns over the private buyers` low prices and cheating in weighing the honey. Lastly, the less used channel is the neighbouring cooperatives. Only 2 (5%) of the PLC members were selling part of their honey through the neighbouring cooperatives. As for the cooperative members, none were selling their honey through this 3 rd market outlet. 4.3.1 Honey marketing by producer groups Bezamar (honey processing and exporting company) is the main buyer of honey from both the cooperatives and PLCs. The producer groups buy honey from their suppliers (members), bulky it and sell to the processor. As has been highlighted earlier on, PLC members are marketing most of their honey through their organization than the cooperatives members who sell most of their output to private traders. Figure 4 below shows the honey sales reported by both PLCs and cooperatives from 2007 to 2010.

Reported Honey Sales by PLCs Reported Honey Sales by Primary Cooperative

180 140 160 120 140

120 100 Ethiopian 2007 Ethiopian Birr 100 2007 2008 Birr 80 2008 2009 (1000 ) 80 2009 (1000 ) 60 2010 2010 60 40 40 20 20

0 0 Chaffe Shatto Gada Genobay Akach Degele

Figure 4: Honey sales by PLCs and coops Source: Survey data

As can be seen from Figure 4, more sales are reported by the PLCs than the cooperatives. This is probably due to the fact that PLC members are more productive, engage more using the transitional hives which increases production and they are loyal to their organization than cooperative members. Our results contradicts the findings of Barham and Chitemi (2009) where they found no evidence in their study to support the hypothesis that smaller farmer groups will be better positioned to improve their market situation over larger groups. Another important factor is that of value chain finance. During our interviews with the managers, the PLCs indicated that they get an advance payment without interest from the lead firm, Bezamar, so that they have enough capital to buy higher quantities of honey from their members. The opposite is true for the cooperatives where the lead firm gets honey on credit basis and pays the cooperatives with interest when payment is received by the lead firm from the export market. Bridging social capital (O`Brien et al 2005) is higher between PLCs and the processor. This kind of value chain financing has tended to benefit the PLCs more than the cooperatives. Bezamar delays to pay the cooperatives which in turn compromise their relationship or trust with their current money lenders. Moreover, the cooperatives highlighted that their current creditors will seize to operate and will not be available in the future, leaving them with no source of credit. The other reason that generally affects both producer groups is the poor road networks that hinder the organizations from buying honey during the rainy season. 4.3.2 Marketing constraints Despite all the benefits that honey can bring to the beekeepers in the area, the producers are confronted with a number of challenges and constraints that can potentially hamper the future of honey production and the economic contribution it brings to their livelihoods. Identifying these marketing constraints can help increase the quantities of honey that is produced and commercialized by the producers.

Table 10: Marketing constraints as reported by the beekeepers

Number of Challenge respondents % (coop = 38) 1 Low price of honey 15 39 2 Lack of access to credit 9 24 3 Lack of support from the union 8 21 4 Private trader cheats on price and weight 6 16 5 Lack of capital for organization to buy all our honey 5 13 6 Transport problem 5 13 7 Fewer buyers 3 8 8 Unable to get timely information 2 5 9 The organization does not buy honey on time 2 5 10 The coop cheat when weighing honey 2 5 Number of respondents % (PLC = 30) 1 Low price of honey 16 53 2 Lack of access to credit 7 23 3 Fewer buyers 6 20 4 Lack of capital for organization to buy all our honey 4 13 5 Unable to get timely information 3 10 6 Private trader cheats on price and weight 2 7 7 The organization does not buy honey on time 2 7 8 Transport problem 1 3

Source: Author`s survey results (2011)

Table 10 above presents, separately, all the marketing constraints identified in the order in which they were ranked by the producers. Respondents who indicated that there are no problems were not included. Based on the results of this study, beekeepers across all producer groups (PLCs and cooperatives) ranked low price of honey as the most important marketing constraint. Lack of access to credit was ranked second in both groups. As their third marketing constraint, cooperative members indicated that they lack support from the union. This was also confirmed by the management team during our interviews with them. They highlighted that they lack financial and technical support from the union. This could probably be explained by the fact that the union is multipurpose thus it deals with numerous agricultural products like coffee, spices, etc. Hence, it is multi-focused resulting in the cooperatives not having the proper support they expect. 5 CONCLUDING REMARKS Collective entrepreneurship has evolved as a strategy for producer enterprises to exercise efficiency and accrue rents whenever they are faced with opportunities. Institutional and structural differences between groups may cause also differences in the way in which producer groups may react to opportunities and innovations. The main purpose of this study was to compare the collective entrepreneurial capacities of 2 honey producing groups; cooperatives and PLCs. Striking differences were identified across a number of aspects between the 2 producer enterprises. PLCs had a higher means for productivity, income, dividend, number of transitional beehives, and price per kilo of honey. Additionally, the incentive scheme exercised by the PLCs was market oriented in that the board members were getting payment for their services. In contrast, the cooperative board members were not getting any payment for their work. Furthermore, as compared to cooperatives, the PLC members demonstrated higher loyalty shown by the proportions of honey they market through their organization.

However, this does not mean that cooperatives are not performing good and also that PLCs are without their own disadvantages. Cooperatives have higher bargaining power due to economies of scale. For this reason, upgrading opportunities are higher in cooperatives if they are to increase their sales to the processor. Hence, they can easily move to processing (possibilities of vertical integration). The PLCs have a disadvantage as far as economies of scale are concerned. This is attributed to their size of membership (fewer members) which limits their bargaining power. Moreover, unlike cooperatives which are diversified in terms of the commodities they trade, PLCs are not diversified as they only focus on the marketing of honey. Thus, we can consider the price risk for cooperatives to be spread over a number of commodities they trade like peas and spices, as compared to that of PLCs.

Collective entrepreneurship has the potential to benefit members of producer enterprises financially and through inclusion into the dynamic value chains. This study shows that honey production in Ethiopia and particularly in Masha can serve as a reliable source of income for the rural producers. Private players are pivotal in the success of interventions set to assist rural producers in improving market access, commercialization and innovation and technological transfer. Hence, from the results we can see that through collective entrepreneurship, the PLCs have succeeded in upgrading and integration into the value chains. Policymakers aiming at promoting rural development through improved market access, livelihoods and income must ensure increased involvement of the private /downstream players like processors in such a way that facilitate collective entrepreneurship amongst groups. Also, owing to the general slower rate of adoption of the transitional beehives among beekeepers, the government should ensure that the improved hives are sold at a subsidized price or on credit to promote increased adoption and use by the honey producers.

Findings of this study that PLCs are more entrepreneurial than cooperatives should not be generalized across all sectors and regions. Rather, our findings could shed more light on the importance of organizational features like group size for successful implementation of rural development interventions. The finding would justify, for instance, the organization of larger communities into smaller subgroups that interact with each other through representatives. Additionally, cooperatives should change their incentive mechanisms, for example, they should adopt from PLCs a system where board members benefit from the organization’s profits as a way of motivating them. References Abbute, W. S. 2003. Resettlement as a Response to Food Insecurity: The case of Southern Nations, Nationalities, and Peoples’ Region (SNNPR). United Nations. Online: http://www.ocha-eth.org/Archive/DownloadableReports/resettlementSNNPR0603.pdf Agonafir, J. 2005. Strategic Intervention Plan on Honey & Beeswax Value Chains. SNV support to Business Organizations and their Access to Markets (BOAM). Addis Ababa, Ethiopia. Agrawal, A. 2000. Small is beautiful, but is larger better? Forest-management institutions in the Kumaon Himalaya, India. In: Gibson, C.C., McKean, M., Ostrom, E. (Eds.), People and Forests: Communities, Institutions, and Governance. MIT Press, Cambridge, MA, pp. 57–85. Agrawal, A. and Goyal, S. 2001. Group Size and Collective Action: Third-party Monitoring in Common-Pool Resources. Comparative Political Studies 34 (1): 63-93 Aiken, M and Hage, J. 1971. The Organic Organization and Innovation. Sociology 5(1): 63-82. Altenburg, T. 2007. Donor approaches to supporting pro-poor value chains (2007) Donor Committee for Enterprise Development Working Group on Linkages and Value, German Development Institute. Bandiera, O., Barankay, I. & Rasul, I. 2005. Cooperation in Collective Action. Economics of Transition 13 (3): 473–498. Barham, J. and Chitemi, C. 2009. Collective Action Initiatives to Improve Marketing Performance: Lessons from Farmer Groups in Tanzania. Food Policy 34: 53–59. Bratnicki, M. 2005. Organizational Entrepreneurship: Theoretical Background, some Empirical Test and Directions for the Future Research. Human Factors and Ergonomics in Manufacturing 15(1): 15-33. Chan, Y. H. 2003. Biostatistics 102: Quantitative Data – Parametric & Non-parametric Tests. Singapore Med Journal 44(8): 391-396. Connell, D. J. 1999. Collective entrepreneurship: In search of meaning. Available at http://www.djconnell.ca/papers.html Cook, M. and B. Plunkett. 2006. Collective entrepreneurship: Am emerging phenomenon in producer-owned organizations. Journal of Agricultural and Applied Economics 38 (2): 421- 428. Debela, S. 2010. Evaluation of the Dutch Ministry of Foreign Affairs Support to Capacity Development: An Evidence-Based Case Study SNV Support to Capacity Development for the Ethiopia Honey Value Chain. Addis Ababa, Ethiopia. Devaux, A., Velasco, C., López,G., Bernet, T & Ordinola, M. 2007. Collective Action for Innovation and Small Farmer Market Access: The Papa Andina Experience. Collective Action and Property Rights (CAPRi) Working Paper 68. Ethiopian Export Promotion Department (EEPD). 2006. Exports of honey and beeswax. Draft Report. EEPD, Ministry of Trade and Industry, Addis Ababa, Ethiopia. Global Development Solutions (GDS). 2009. Integrated Value Chain Analyses for Honey and Beeswax Production in Ethiopia and Prospects for Exports. Hardin, R. 1982. Collective Action. Baltimore: John Hopkins University Press. Hussi P., Murphy, J., Lindberg, O & Breneeman, L. 1993. The Development of Cooperatives and other Rural Organizations. Technical Paper No 199. Washington DC: The World Bank. Kirsten, J. 2004. Expanding the Frontiers of Agricultural Economics to Meet the Future Challenges of Agricultural Development. Paper Presented for African Association of Agricultural Economists, Proceedings of the Inaugural Symposium, Nairobi, Kenya. Mangus, E. and Piters, B. S. 2010. Dealing with small scale producers: Linking buyers and producers. KIT (Royal Tropical Institute) publishers. Markelova, H and Meinzen-Dick, R. 2009. Collective Action for Smallholder Market Access: The Case for Collective Action in Smallholder Marketing. Policy Brief 6. Mehari, G. 2007. Impact of Beekeeping on Household Income and Food Security: The Case of Atsbi Wemberta and Kilte Awlailo Woredas of Eastern Tigray, EthiopiaM.Sc. Thesis, Mekelle University. Muradian, R., Chagwiza, C., Tessema, W & Ruben, R. 2011. Pro-poor Value Chain Interventions, Upgrading and Inclusion: Drawing insights from the SNV’s value chain development approach in Ethiopia. Paper presented at the BOAM Conference: Learning from six years of Value Chain Development in Ethiopia. 8-9 June 2011, Sheraton Addis, Addis Ababa. O`Brien, D. J., Phillips, J.L & Patsiorkovsky, V. V. 2005. Linking Indigenous Bonding and bridging Social Capital. Regional Studies 39(8): 1041-1051. Olson, M. 1965. The Logic of Collective Action: Public Goods and the Theory of Groups, Cambridge, MA: Harvard University Press. Olson, M. 1971. The Logic of Collective Action: Public Goods and the Theory of Groups, Cambridge, MA: Harvard University Press. Ostrom, E. 1994. Constituting Social Capital and Collective Action. Journal of Theoretical Politics 6(4): 527-562. Sartorius, K. and Kirsten, J. 2004. The Cost Efficiency of Small Farm Inclusion in Agribusiness Supply Chains. South Journal of Accounting Research 18(1): 87-113. Stewart, A. 1989. “Team entrepreneurship”, Newbury Park, CA: Sage. Shultz, T. W. 1995. The Value of the ability to deal with Disequilibrium. J. Econ.Literature 13:827-846. Trompenaars, F., & Hampden-Turner, C. 2002. 21 Leaders for the 21st century: How Innovation Leaders Manage in the Digital Age . New York, NY. Workneh, A. 2007. Determinants of adoption of improved Box Hive in Atsbi Wenberts District of Eastern Zone, Tigray Region. MSc. Thesis, Haramaya University, Ethiopia. Yan, J., and Sorenson, R. L. (2003). Collective Entrepreneurship in Family Firms: The Influence of Leaders` Attitudes and Behaviors. New England Journal of Entrepreneurship, 6 (2): 37-51.