Democratic institutions and economic success in rural regions: the case of Fairtrade coffee co operatives in Latin America

Democratic institutions and economic success in rural regions: the case of Fairtrade coffee co operatives in Latin America By Anna Milford Department of Economics Norwegian School of Economics anna.milford@nhh.no April 2009 Introduction Many of the world s poor belong to households of agricultural smallholders such as coffee producers. Their situation is made difficult by price fluctuations and an overall trend of price decrease, as well as lack of access to markets (Markelova et al 2009). An opportunity which has lately opened up for small scale coffee producers is to join a Fairtrade labelled marketing cooperative. This provides them with access to the international export market, and the possibility to obtain the Fairtrade guaranteed minimum price and premium. These organisations are member owned and run according to democratic principles. The development of democratic institutions is important for the control of the leaders of the cooperatives, but can also be seen as an aim in itself. At the same time the contradiction between democracy and economic success is also one of the Fairtrade certified organisations key challenges (Raynolds et al 2004). In this paper the aim is to find out if it is possible for Fairtrade certified cooperatives to fulfil the democratic requirements they are obliged to keep, and at the same time have a prosperous economic development. A second objective is to see if democratic institutions are more important for economic success in the cooperatives with the smallest coffee producers. In order to shed light on these questions, I make use of a unique data set from the Fairtrade Labelling Organisation (FLO). The question how democracy and economic efficiency are linked has been raised and investigated both in relation to the organisation of firms and the ruling of nations. Theoretical predictions provide arguments for both a positive and a negative correlation between the two, and there are several hypotheses explaining which way the causal link goes. Empirical studies of democracy and economic welfare in different countries find that there is a positive correlation, but there is a lack of consensus concerning which is causing which. Lipset (1959) and Barro (1999) claim that economic development causes democracy, while Acemoglu et al (2008) find no evidence for this. They believe that the positive correlation observed is due to omitted variables, and that historical events has made countries take different paths leading either to both economic growth and democracy, or dictatorship and limited growth. They also do not exclude the possibility that democracy may have a positive effect on economic development. At the firm level, most empirical 1

studies are concerned with either labour managed firms, or worker participation in capitalist firms. Also here, authors generally report a positive correlation between worker participation and firm performance (Cable and FitzRoy 1980, Jones and Svejnar 1985, Doucouliagos 1995, Yates 2006). But there are less studies aiming to find out which way the causal link goes. Concerning agricultural cooperatives, there are few studies that investigate the correlation between efficiency and democracy, although several studies compare their performance relative to non cooperative companies (Hind 1994, Mosheim 2002). One exception is a recent study of cereal cooperatives in Ethiopia, which concludes that for these cooperatives there seem to be a trade off between performance and member inclusion and participation (Bernard and Spielman 2009). There has been little formal investigation of the relationship between democracy and economic success in Fairtrade certified cooperatives. The Fairtrade Labelling Organisation (FLO) has for many years measured the democratic and economic performance of coffee cooperatives in developing countries. The evaluations written by Fairtrade inspectors have been used for deciding on certification or recertification of the cooperatives and comprise a relatively large database with detailed information about each cooperative. To my knowledge, the present study is the first one to make use of this data for statistical analysis of any kind. In general, the democratic aspect of agricultural cooperatives is particularly important in the context of developing countries. In these countries not only economic success, but also democracy is an important outcome from these cooperatives. Economic success is clearly important because it increases the welfare of poor farmers, but democracy and active participation of members can be seen as a goal in itself, because the members are educated and become a resource for society at large (Bowles and Gintis 1993). The Fairtrade labelling system aims to achieve both. For many the main idea is to increase the level of economic welfare for underprivileged producers, but according to Raynolds et al. (2004), the empowerment and capacity building nature of Fairtrade will in the long run prove to be the most important. According to Thorp et al. (2003), groups formed to enforce rights instead of economic returns, are likely to improve the economic, social or political position of the members, which in turn is likely to lead to economic benefits. The Fairtrade cooperatives are all run by members in poor, marginalised areas, they have little education and they are excluded from political decision making processes. Being part of democratic cooperatives is likely to be beneficial directly to the members who increase their knowledge and social capital, and it may have long term, wide reaching positive effects in terms of increased influence on political processes with consequences such as increased transfers of benefits to marginalised groups and decreased public corruption because of improved citizen supervision. However, while democracy might have a positive effect on economic success, for instance because it leads to better control of managers, it may also be particularly difficult for uneducated, marginalised farmers to reach consensus and make the best collective decisions. In particular, when rules for transparency and democracy are imposed from above, as they might be through the Fairtrade requirements, rather than developed within the group, members in developing countries will often not understand them (Markelova et al 2009). It is therefore interesting to see if democracy has a positive or negative effect on the economic performance of the Fairtrade cooperatives. If democracy is in itself a good, then one could argue that it should be promoted, regardless of whether it may lead to economic success or not. However, if democracy inhibits economic success in the Fairtrade cooperatives, then there is reason to reconsider the way it is being promoted. The main 2

question is therefore to find out if there are signs that the promotion of democracy inhibits economic success. The paper starts with a brief introduction to the concept of Fairtrade, followed by a literature review of the different hypotheses linking democracy and economic success. The second part is a qualitative field study of various Fairtrade coffee cooperatives in Chiapas, Mexico. Then follows the statistical analysis of the FLO evaluations from the Spanish speaking Latin American Fairtrade coffee cooperatives between 2001 and 2005. 1. Fairtrade and the Fairtrade co-operatives Fairtrade is a labelling scheme based on values of solidarity. It makes it possible for consumers to choose products made by disadvantaged producers, and be sure that they receive a better deal in terms of price and working conditions. There are different certification schemes that make similar claims, but in this case focus is on the original Fairtrade labelling system which started in 1989 and which is run by the Fairtrade Labelling Organisation (FLO), with headquarters in Bonn, Germany. The Fairtrade label exists for a range of different products, but coffee, the first product to become certified, is the most important in terms of sales. For coffee, only farmers organised in co operatives can become certified and receive the Fairtrade price. These co operatives are expected to fulfil certain requirements. They should be generally small scale producers, provide members with economic and social benefits, and be democratic, transparent organizations run by the producers themselves. There are also environmental criteria on the use of chemical inputs, and social criteria on the use of hired labour. At the end of 2006 there were 241 Fairtrade certified coffee producer organisations in Africa, Asia and Latin America. Producer organisations or co operatives become certified once an inspector from FLO has made a first visit and found that the requirements are fulfilled. After this, a control visit is made every year or every second year 1. The co operatives cover the costs for certification. Once they are certified, they may sell their product under Fairtrade conditions, which includes a minimum price of 121 USD/lb and a premium of 10 cents/lb. But this requires that they are able to find an importer wanting to buy coffee under these conditions. The importer will in turn market the coffee with the Fairtrade label, for which it pays a licence fee to FLO. These cooperatives have different levels of economic success. Some prosper, they offer members a good price and invest in income generating activities, while others have financial problems and struggle to keep their member base. There are also differences concerning their democratic levels: some have high participation levels and well functioning democratic structures, while others are dysfunctional to the point that they are excluded from the Fairtrade system. In the next section we will look at some theories explaining how these two aspects may be linked to each other. Thorp p 11: trade off between discipline needed for economic viability and objective of generating genuine empowerment for all members. Montgomery, p. 20 1 Up until 2005 the control visits were rather irregular, afterwards the policy has been that there should be one visit every year 3

2. Democracy and economic success: some issues Much of the literature on co operatives and their level of efficiency compared to investor owned firms is concerned with producer co operatives, or labour managed firms. In the case of the Fairtrade coffee co operatives, they are agricultural co operatives, or marketing co operatives, run by the members who produce coffee on individual farms, then market it jointly through the co operative and share the profits thus generated according to quantities delivered. These co operatives therefore do not have collective action problems such as free riding when it comes to the production of coffee. The collective task that the members of the cooperatives have to perform is to administer the different activities of the cooperatives, related mainly to marketing. It is not clear beforehand what the relationship between economic success and democracy should be in these co operatives. Firstly, to the question whether the two are positively or negatively related previous research find several explanations arguing for both outcomes. Secondly, if one can establish such a relationship, it is not clear what are the causal links, whether it is democracy that leads to, or inhibits economic success, or the opposite. In this section we will look at the different hypotheses that have been raised in the literature. In the following section the hypotheses will be compared with the findings from the field research performed among Fairtrade coffee co operatives in Chiapas, Mexico. Positive correlation between democracy and economic success The level of democracy measures to what extent the members are in control of their co operative and how much they influence the decisions taken by its leaders. If there is little democracy and members do not control their leaders, it means that the leaders have more possibilities to shirk, to spend less time doing their job and therefore doing it worse. They can also spend more money unnecessarily on for instance travels and meals for themselves, or even become downright corrupt. This hypothesis is related to principal agent theory and it means that in the less democratic cooperatives, more resources are lost by managers, or agents, who are not being supervised sufficiently by their principal, the members (Bowles and Gintis 1993, Hansmann 1996). Another argument for a positive correlation, is that the members of a cooperative may have important knowledge which is relevant for making the best decisions. According to Hansmann (1996), members of agricultural co operatives are well positioned to be informed about co operative activities since they, as producers of a single product, have focused incentives and accumulate information as a by product of their main productive activity. This means that the participation of members in decision making processes can be important for economic success. If decisions are instead based mainly on externally hired staff members ideas and opinions, they could be misplaced and therefore decrease economic success in the co operative. Another reason why democracy may increase economic success is that decisions taken democratically are more legitimised than decisions taken by leaders without consulting members. Decisions taken jointly increase the trust in the fairness of these decisions, and they are therefore easier accepted and implemented (Cable and Fitzroy 1980). In many cases the motivation of members is important, for instance when it comes to technical change and new innovations (Vanek 1970). One could also say that democracy creates a more positive perception of the organisation for the members, which gives them a greater feeling of responsibility for its well functioning and has a positive effect on morale and quality of work (Jones and Svejnar 1985). A positive perception of the 4

organisation could in turn influence member commitment, which is important for the functioning of an agricultural co operative, particularly when price offers from private competitors are at similar or higher levels (Fulton 2004). Another argument raised by Hirschman (1970) is that in co operatives where members have a voice and a saying in the administrative matters, they will inform the leadership about their dissatisfactions, and the leadership can search for possible cures. If the members do not have this voice option, they will simply exit the co operative. Wade (1987) has also found that the better developed the arrangements for discussion of common problems, the greater the chances of success for collective action projects. So far the argument has been that democracy leads to economic success. But it is also possible that the causal link goes from economic success to democracy. In the literature on democracy and economic performance on a national level, many have argued that only countries with a relatively high standard of living can maintain democracy in the long term (Lipset 1959). The argument is that education levels in the population, an enlarged middle class and private organisations and institutions are important elements that lower the power of the landlord class (Barro 1999). Transferring this argument to the cooperative sector, would be to say that only members with enough education and resources will be able to contribute with their active participation in the democratic running of the organisation. They will also have enough insight and power to counteract despotism if this should develop among members. Another argument would be that there are no causal links going from one to the other, the explanation for a positive relationship between democracy and economic success is that the two are linked to the abilities of the members of the co operative and their leaders. Coffee farmers with more social capital, who are motivated and talented, and leaders and staff members that are dedicated, hard working and highly competent, will produce positive results both on democratic and economic measures of the co operative s performance. This is related to the argument raised by Acemoglu et al (2008), that there is a third factor influencing both economic and democratic performance. Negative correlation between democracy and economic success In relation to the issue of collective decision making in co operatives, several authors have emphasised the problem of conflicting objectives among members. There are many potential sources for tensions to arise, such as between keeping the organisation small or allowing to grow. The profit maximising objective could also conflict with other objectives such as improving services or increase employment in the area (Le Vay 1983). Other conflicting economic objectives includes the location of facilities (Staatz 1983) and full distribution of profits versus profit retention to finance collective services and reserves (Ruben 1999). Particularly when the outcome of a decision affects members differently, there will be different opinions and therefore decisions are difficult to make (Hansmann 1996). The directors of a co operative can sometimes experience the dilemma between running a flourishing business and expressing the wishes of membership (Le Vay 1983). If objectives are conflicting to the extent that it prevents a decision from being taken, striving for democracy may inhibit progress and thereby reduce the economic performance of the co operative. Also, democratic processes, even without conflicting objectives among members, are time consuming. In many cases decisions need to be taken quickly, and if the leaders should consult members first, it may be too late and the opportunity will be lost. Hence democracy may slow down 5

efficient decision making (Jones and Svejnar 1985). Democracy is also, as mentioned before, costly. If the co operative has skilful leaders, they could take decisions that are as good as the ones that would be taken collectively, but neither important opportunities nor time and financial resources would be lost. It may even be the case that the capable leaders will make better decisions if they do not have to consult the members. Members may not be the most able to choose the most advantegous path for the organisation, because they do not see the big picture, they lack education, training and experience (Harrison and Freeman 2004, Bernard and Spielman 2009). Or, even if the members do have the capacities, the voting system is such that it is often the opinion of the median voter, not the average member that will win, or it could be the case that the political process falls into the hands of better situated patrons (Hansmann 1996). One classical problem in democratic organisations, is the lack of willingness to make investments. The argument is that the members planning horizon is too short to conform to the requirements for Pareto optimality (Furubotn 1976). Because co operative shares cannot appreciate in value, and cannot be sold on the open market, and because the claims on benefits from any investments made by the co operative usually end with the termination of membership, the investment level of a cooperative is assumed to be less efficient than that of an IOF. The horizon problem implies that members approaching retirement undervalue investments (Porter and Scully 1987). In the case of cooperatives in developing countries, the problem of under investments is larger than in industrialised countries, because less affluent people will have a comparatively strong consumption preference and therefore a general disinclination to invest (Abell and Mahony 1982). Bowles and Gintis (1993) also claim that in general members of cooperatives have a tendency to engage in insufficient levels of risk taking and innovation. Again, leaders that seek to maximise the long term profits of the co operative would do better by not consulting the members before implementing plans. Hence, theoretically there are reasons both for why democracy and economic success should be negatively and positively correlated, and the causal links may go in both directions. In the next section we will see how the findings from the literature correspond to the qualitative field research performed in Chiapas, Mexico. 3. Case study Chiapas The field research was done in Chiapas during a ten month period in 2006/2007, where interviews were made with co operative managers and other stakeholders in the coffee producing sector. General description of democratic structures in Fairtrade certified cooperatives The different Fairtrade certified co operatives in Chiapas have similar democratic structures, which again are similar to the ones described in the FLO evaluation reports from other Latin American countries. The members, who usually live dispersed in different villages, or comunidades, communities, generally have meetings once a month. In some cases meetings are compulsory and members are penalised for not showing up. Each community elects a delegate who goes once a month or every second month to meetings at the co operative headquarters, where he or she will represent the community and report their opinions and ideas. The information that the delegate receives here is in turn dispersed to the members at the community meetings. In some cases the co 6

operative produces written information that the delegates can distribute in their communities, but in many cases this is meaningless since the members are illiterate. Sometimes this is the case also for the delegate, who has to try to memorize the information he or she has received. In these cases distribution of information will often be deficient. In some cooperatives one of the most important collective decisions to make is the decision of the price to pay members. In other cooperatives this decision is taken without consulting the general assembly. Other collective decisions taken in the co operative are typically concerned with implementation of new projects. Many decisions are taken at the meetings of the delegates, while supposedly the most important are saved for the general assemblies, which are usually held once a year. Every second or every third year the general assembly elects a new committee with president, secretary and treasurer. Often they also elect a control committee and sometimes a conflict resolution committee. The co operatives usually also hires staff members. Typically these are the people responsible for complex affairs such as the exporting of coffee, which requires years of experience. Most co operatives also have an advisor, often an external person who has been with the cooperative from its beginning. Very often this is a person with either a religious or political background. The advisor and the hired staff members often have years of experience in the coffee business, while the elected committee are only experienced as producers. This inequality of knowledge often creates a latent power struggle between the political leadership and the hired staff (Martinez Torres p. 94 2006). The administration of the co operative is mainly concerned with the marketing process: finding importers, organising deliveries, processing, transport, finance, payment and so on. In some cases the co operatives roast and grind the coffee and sell it locally, in addition to exporting it as green coffee beans. Furthermore the certification process must be organised in order to receive both the Fairtrade and in many cases also the organic certificate. In addition many co operatives lobby governments and NGOs for support for different projects, and they organise courses for member on coffee production and other issues. Moreover, some co operatives also diversify into other sectors such as honey, crafts or tourism. These tasks are performed by a committee with, as a minimum, a president, a treasurer and a secretary, elected for a certain period, usually 2 or 3 years. In addition there is often one or more hired staff members in the co operative. The level of democracy in this case is determined by the extent to which the members of the co operative have an influence on the decisions that their elected and hired leaders make. This will depend both on how well informed they are, how often there are meetings with delegates and general assemblies with all members, and how many of the members that participate. The level of economic efficiency, at least the static efficiency, can be described by the income generated by the co operative. But the long term performance of the co operative can also depend on how much investment they make. Democracy and economic success in Fairtrade cooperatives Chiapas That lack of democracy and efficient control by members is a problem in many co operatives in Chiapas, is clear from the many stories of corruption among co operative leaders. Most of these stories were not about Fairtrade co operatives, but about co operatives from earlier times. One story of corruption in a co operative was told in the following manner: 7

There was fraud, because there were advisors, accountants, who were all from other places, and for the poor producers, who have no education, they don t know, well, they just deliver their coffee. I ll pay you this price, that s how it was, but in reality that wasn t the price, most of was left with the advisors, the accountants. 2 Co operative staff member and son of producer, Los Altos This was a story of how the members of a co operative were deceived by the externally hired advisor and accountant, who took their money. The members in this cooperative were just delivering their coffee, they had no insight to the administrative matters of the cooperative, and were therefore not in real control of the running of the cooperative. This case of corruption was discovered by the public authorities in the region 3. The lack of trust and fear of corruption is deeply rooted in many producers, often stemming from previous bad experiences, either personal or of someone they know. They are therefore often reluctant to let others take responsibility for their money. It is observed that there is a difference between the co operatives in how much effort and money the members want to put into the organisation. In particular there is variation in how willing the members are to use their income for investments in the co operative. It seems logic that democratic structures and transparency, if they increase the members feeling of control over the co operative and its administration, also will decrease their fear of corruption and thereby make them more willing to make income generating investments in capital or other projects. The difference in motivation to invest can also be seen between members in the same cooperative. A staff member of one co operative explained that the members who lived close to the cooperative offices and warehouse participated more in the democratic processes and were better informed, and they were also more willing to invest in the cooperative than the members living far away. This could be an indication that members who participate more actively, feel that they have more control of the cooperative, their sense of responsibility is stronger and they are more willing to 2 había fraude, porque habían asesores, contadores, que todos ellos son de fuera, pues, y a los pobres productores, como no tienen estudios, no saben pues, solo entregan su café, que te pago a este precio, así salió, pero en realidad no es el precio pues, ya la mayor parte se les queda a los asesores, los contadores. 3 These stories of corruption have led to a general level of mistrust in externally hired staff members. This problem has been solved in many co operatives by following a policy of hiring staff members that are either co operative members who get a paid position after being in the committee, or grown up sons and daughters of members. Possibly this is not just seen as a way of reducing chances for corruption, but also as a way to help out and create wage opportunities for co operative members and their families, as an extra benefit from membership. The dilemma is that the most qualified people for the job in terms of relevant education and experience may not be found among the members and their families, and therefore this policy may imply a loss of efficiency. But it may also be the case that the members and their families have better knowledge about co operative affairs, and therefore are better positioned to run the co operative in an efficient way. They may also benefit from more trust and legitimacy than someone entirely external to the co operative, which also will have a positive effect on implementation of policies and thereby efficiency. This aspect of increased control of the cooperative management through the choice of hired staff members is an aspect which is not covered in the evaluations made by the FLO inspectors. 8

make investments. But it also mean that members who live further away are more unsure about their future membership. The question whether economic success leads to democracy has relevance also in the case of the cooperatives in Chiapas. It was indicated that cooperative members often did not manage to control the staff because of lack of education and knowledge. It is possible that cooperatives that have been well functioning economically for some time, will also have increased the welfare level of their members and thereby their ability to monitor. But this would be in a long term perspective. A more immediate link can be seen by arguing that a strong economic base is important in order to implement the relatively costly democratic structures that the co operative needs in order to fulfil the FLO criteria. Organise meetings and elections, ensure transparency at all levels and produce written information: all of these are costly processes. In the case of a co operative which is performing badly economically, it will perhaps barely have enough resources to run the processes related to its profit generating activities, and will prioritise this to the democratic processes. But an economically well functioning co operative can more easily afford the costs of democratic structures. The interviewed co operative staff members in Chiapas affirmed that democratic procedures are costly and may put them at a disadvantage in comparison with conventional firms. The participation in a cooperative, on the one hand is more control of the cooperative, and more cohesion and more solidity. But on the other hand it also implies more costs. When we have meetings, we have to pay for the passage, we have to pay for petrol, we have to pay for people s meals. Although people don t get paid, but they still lose their day. In fact, this is one of the big differences that often the importers and Fair Trade don t understand, that being a cooperative and being more democratic, we have a lot of costs. A conventional firm doesn t have the costs that we have in terms of meetings 4. Co operative advisor, Los Altos One could therefore assume that co operatives with a bad economy will find it more difficult to maintain those structures, than co operatives that are economically well functioning. The interviewed cooperatives in Chiapas supported the view that the members and their knowledge and opinions were important to take into account in the decision making process. One cooperative staff member emphasised the quality of the suggestions made by members, claiming that this was important since they were the once who better saw what their needs were. There was also a 4 La participación en una cooperativa, por un lado es mayor control de la cooperativa, y mayor cohesión y mayor solidez, pero por el otro lado tambien implica mucho gastos. Cuando tiene reuniones, se tiene que pagar pasaje, se tiene que pagar gasolina, se tiene que pagar el alimento de las personas. Las personas aunque no cobran, pero de todos modos pierden su día. De hecho estos son unas de las grandes diferencias que muchas veces los importadores y el comercio justo no entiende. Que tenemos muchisimos gastos por ser cooperativa y por ser mas democratico. Una empresa convencional no tiene los gastos que tenemos nosotros en terminos de reuniones. 9

story of how the staff members of another co operative once had tried to save money by not borrowing from the bank. This meant that they could not pay the members at the time when they delivered their coffee, so they had to wait until the payment had arrived from the importer. This was contradictory to the members needs, who, because of the delayed payment from the co operative, instead sold their coffee to the private intermediary at a lower price. The co operative was then unable to comply with the importers orders, which meant both loss of money and reputation. Hence, this is an example of a case where consulting the members before implementing the new policy is better than not doing so. For the co operative staff members interviewed it was seen as important that the members took active part in the decision making because this would increase the legitimisation of the decisions. Decisions taken by members were always listened to, as one staff member said. Another co operative staff member explained the following about the importance of collective decisions: They are better when the members take them and because the members take them. Sometimes, when a decision is taken, or a proposal is made which is not arranged, then this proposal will fall. Sooner or later the proposal falls, it seizes to have validity. Because for the organisations, one of the most important points are the decisions agreed between everyone. If there is no consensus, there can be division. 5 Co operative advisor, Los Altos In other words, taking decisions without the members approval could be counterproductive, since these decisions would not be seen as valid in the same way as a democratic decision, and therefore they would be more difficult to implement. This, however, would only be the case for implementations that requires the members active collaboration, which might not be necessary for all kinds of decisions. Concerning not the democratic structures, but the level of participation among members, the qualitative study did not bring about any clear indications to whether economically well functioning co operatives foster more or less participation. On the one hand it was argued that a low price offer made the members feel downhearted and therefore they did not show up at meetings, and that the opposite was seen when there were good earnings from the co operative. On the other hand it was said that when the co operative was doing well, the members were happy and did not say anything, but when there were problems they would raise their voice against the leaders and hence participate more actively. Democracy and economic success negatively linked 5 Son mejores cuando los hacen los socios y porque las hacen los socios. En ocasiones, cuando se toma una decisión, o se hace una propuesta no concertada, cae esta propuesta. Tarde o temprano se cae la propuesta, deja de tener validez. Porque para la organización cooperativa, una de los puntos más importantes son las decisiones acordadas entre todos. Si no hay consensus, puede haber división. 10

The first argument in the literature review on how democracy could inhibit economic success, was that conflicting objectives would make democratic decisions more difficult to take. In general, members of coffee co operatives have homogenous interests since they are all producing the same product, coffee, and their main interest is to generate profits from selling it. They also produce on similar coffee plot sizes and have similar living standards. Nevertheless, members may differ in their opinions about what investments to make, whether to use resources for courses, collective goods or not, and a range of other issues. These conflicting objectives may make democratic processes difficult and time consuming to undertake. One example of a conflict between members was a cooperative where members had their plots at different altitudes. The ones producing at the lower altitudes had more problems in getting the quality that the co operative required than the producers at higher altitudes. The lower altitude producers therefore wanted less strict requirements for the coffee deliveries, but this was not in accordance with the wishes of the high altitude members. This was an ever returning theme and source of conflict at the meetings in this co operative. Taking into account the wishes of the low altitude members would have implied losing customers buying high quality coffee, so in this case the right was given to the high altitude members. Another cooperative told how groups of members from certain villages had decided collectively to leave the cooperative, because they disagreed with its policy. In a village in Jitotol, coffee producers who were not organised said they did not want to join the cooperative because they don t reach agreement. 6 However, in general conflicts among members were not seen as a major problem which delayed processes. It has been mentioned already that collective decisions are costly and time consuming. Possibly this may have affected some of the co operatives economies negatively. Too much democracy can turn into bureaucracy. This is a danger, a threat that there can be in cooperatives. When all the decisions need to pass through superior levels it takes away the activity and the flexibility of the work. So, on the one side, at one extreme, is the neoliberal company that does its job without taking into account the opinions of the others. At the other extreme is the one where everything needs to be decided by everyone, which makes the operations very slow. Democracy is neither one nor the other 7. Cooperative advisor, Los Altos 6 No se ponen de acuerdo. 7 Un exceso de democracia puede convertirse en buraucracia. Eso es un peligro un amenaza que puede haber en las cooperativas. ( ) Cuando todas las decisiones se tiene que pasar por niveles superiores se le quita actividad y flexibilidad al trabajo. Entonces, por un lado, en un extremo, esta la empresa neoliberal entre comidas que hace su trabajo sin tomar en cuenta las opiniones de los demás. En el otro extremo está el que todo tiene que pasar por las decisiones de todos por en consenso que ( ) hace muy lento en trabajo. La democracia no es ni lo uno no lo otro. 11

This signifies that one the one hand, the costs of democracy means that only cooperatives that are economically efficient can afford it, and therefore there is a positive link. On the other hand it could mean that cooperatives that feel forced to pay the costs of democracy loose out both in terms of time, flexibility and financial resources, and their economic success level is therefore negatively affected, which implies that the correlation is negative. It is clear from the case study in Chiapas that co operative members are reluctant to make investments. The reasons for this are not only the short horizon or the fear of corruption. The most obvious reason is that the producers are small scale, sometimes with just half a hectare of land. Living in absolute poverty as they are, it is difficult to leave some of their income in the co operative, even though it implies increased earnings in the future. The main reason why many co operatives in Chiapas still make investments is that they use the Fairtrade premium for this purpose, or they receive support from governments or donors. But if the co operative receives finance, whether from Fairtrade or other sources, the members will often prefer that it is dispersed among them, no matter how small the amount is. The only way to avoid this is for the leaders and administration to present projects in a clear and intelligible way, and thus persuade the members to use the money for collective projects that in the future will benefit them all. A lot depends on who is leading the cooperative. People very often just want to allocate the money, but very often this is not the best decision 8. Co operative staff member Hence, the democratic decisions that the members make need to be guided by the leaders of the co operative if investments are to be made, indicating that the quality of leadership is more important to economic success than democracy, which may in fact hinder beneficial projects from being implemented. Another reason why the members may be unwilling to make investments, is that they do not understand that they will, in the long term, benefit from this. Some of the co operative leaders considered it a problem that the members often lacked an understanding of the co operative processes. For the members, the processes of the cooperative looks easy, but they don t understand the problems. For example they say: why doesn t my credit come? This depends on the standards of the bank, they don t understand that 9. 8 Mucho depende de los que están encabezando la organización. La gente muchas veces solo quieren repartir el dinero, pero muchas veces eso no es la mejor decisión. 9 Para los socios, los procesos en la cooperativa se ve muy fácil, pero ellos no entienden los problemas. Por ejemplo dicen por qué no llega mi crédito? Eso depende de las normas de la banca, ellos no lo entienden. 12

Co operative committee member, Centro If the members are to make decisions that will prove most beneficial for the organisation, they need to have a certain level of knowledge. If they do not, there is a danger that they will not be able to make the right decisions, which implies a loss of efficiency. This, again, indicates that democracy can sometimes be a problem. Synthesis To sum up, the case study from Chiapas indicate that cooperatives that are more democratic will also be more economically efficient because firstly, they have a stronger control over the managers of the cooperative, secondly, members have valuable information about which decisions that are the best, thirdly, collective decisions are legitimised and therefore more easily implemented, fourthly, democracy and participation increases the members trust in the cooperative, and lastly, because democracy is costly and a strong economic base is needed for its operation. On the other hand, there are also indications that there could be a negative link, because, firstly, there are a number of conflicting objectives within a cooperative which makes democracy less efficient, secondly, if costly democratic structures are implemented it may reduce the financial base and economic benefits from the cooperative, thirdly, members are generally unwilling to invest and therefore it is better that leaders decide on their behalf, and lastly, because they do not see the big picture and therefore will not make the best decisions. It seems to be the case that whether democracy is good or bad for the economic performance of the co operative, depends on whether or not the members are the most apt to make the decisions, and whether their consent is necessary for the implementation of the decisions. It also depends on whether the co operative leaders are more competent in decision making than the members, and how prone they are to mismanage the co operative if they are not properly controlled by the members. Since the members level of knowledge and the leaders capacities and honesty are external factors which are difficult to influence, the general impression from the qualitative study is that democratic structures are a safer bet than more power in the hands of the leaders. Overall the interviewed stakeholders seemed to share the view that democracy is important for the success of the cooperatives. One staff member put it like this: We have always said: everyone has the right to participate and make decicions. And from that the best solutions will come out. That helps a lot to create strength, right? Inside the cooperative 10. Co operative staff member, Los Altos 10 Nosotros siempre hemos dicho, todos tenemos derecho a participar a tomar decisiones. Y de ahí se va a sacar las mejores decisiones. Eso ayudaría mucho a construir mas fuerza, no? Dentro de la cooperativa. 13

Hence, democracy is seen as a building brick of the cooperatives, and it is believed that the active participation in decision making is important both in order to find the best solutions, and to fortify the cooperative. 4. Statistical analysis of the FLO evaluation reports 4.1. Data background Description of the data The data base comprises 285 evaluations of Fairtrade co operatives in Mexico, Guatemala, Nicaragua, Honduras, El Salvador, Costa Rica, The Dominican Republic, Columbia, Venezuela, Ecuador, Bolivia and Peru, from 157 different cooperatives. The cooperatives have been evaluated on an irregular basis, between one and four times, between 2001 and 2005 (one from 2006). These evaluations, which are from all the coffee producing, Spanish speaking Latin American countries, comprise my data base 11. The FLO evaluations have so far not been studied by independent researchers for academic purposes. This is the only data source used in the analysis. Descriptive statistics: Facts about the co operatives The oldest co operative is from 1961, the most recent was founded in 2003. About 20% of the cooperatives were founded before 1980, 60% were founded after 1990. The average number of members in the co operatives is 522, the smallest has 17 members, the largest 7503. Half of the co operatives in the data base have less than 200 members, 30% have less than 100. In other words, the majority are small cooperatives. Only 15% have more than 1000 members. Below is a table of descriptive statistics, based on the averages for the cooperatives that provide the information. Table 1: Descriptive statistics FLO cooperatives Observations Foundation year Members Plot size FLO register year Bolivia 27 1986 183 3 1998 Colombia 41 1992 397 2.1 1997 Costa Rica 3 1993 1868 2 1990 Dominican Republic 4 1989 4234 2.5 1994 Ecuador 5 2001 760 2.4 2003 El Salvador 5 1995 99 2.42 2002 Guatemala 29 1986 355 1.22 1999 Honduras 25 1989 99 2.4 1995 Mexico 76 1993 557 2.6 1999 Nicaragua 17 1996 634 2.9 1999 Peru 58 1982 1000 3.2 2000 11 I received the evaluations in March 2006, in word files. The evaluations were then entered manually into excel. The ones from Guatemala, Nicaragua, Honduras, El Salvador and Costa Rica (77 evaluations) were entered by a hired assistant (Laura Weiss). I myself have entered the other evaluations. 14

Venezuela 4 1990 270 3.7 1993 Description of evaluations The inspectors are either locally based or people travelling from Europe. Increasingly FLO is using local people for the evaluations. The inspectors go through a course with FLO and training with an experienced inspector. 31 different inspectors have evaluated the cooperatives in the dataset. Each inspector has written between 1 and 39 evaluations. During the inspection visit at the cooperative, which usually lasts from 2 to 5 days, the inspector interview staff members, investigate administrative documents and select villages at random where they go to speak to members. The evaluations have the form of a questionnaire with approximately 200 questions each to be answered either by a quantity or by 'yes', 'no' or 'partly'. The first part of the questionnaire is about coffee production and sales. Then follows a section where the inspector gives his or her general impressions of the cooperative concerning different aspects: economy, democracy, development, environment and the use of the Fairtrade premium. The following sections cover each of the same aspects with a set of questions. The inspector also writes resumés about the cooperative and the area where it is situated. The inspector has sometimes left questions unanswered. This could be either because the cooperatives were unable to provide the necessary information, because the question for some reason was irrelevant for certain cooperatives, or for other, unknown reasons. Description of variables used in the analysis Democracy The aim with the FLO inspection is to a large extent to evaluate how democratic the organisation is, and many of the questions in the evaluation scheme are related to this aspect. I have selected the ones that reflect the structure of the democratic organisation, meaning what institutions are in place, how much information is distributed and the participation among members. I have not included variables that may be reflecting the nature of the member group more than the structures implemented. There are no variable that measures quantitatively the influence of members on decision making, such as the percentage of decisions taken collectively. Table 2: Description of democratic variables Variable Description Sdem Demostr Demelec Inflow Infwrit Infmeet Inffin Infcom General judgement on whether minimum requirements for democracy, participation and transparency are fulfilled The organisational structure as laid out in the statutes is democratic The executive (board) is democratically elected Regular updates given to members x times per year Regular updates given to members: written Regular updates given to members: in meetings Regular updates given to members about the financial situation Regular updates given to members about the state of commercial activities (contracts, prices, buyers) 15

Infcomi Regular updates given to members about committee activities Infpro Regular updates given to members about the state of projects frgenasm Frequency per year of General Assemblies of members Frmrep Frequency per year of Meetings of Representatives/Delegates Frmcom Frequency per year of Meetings on community level Atgenm Attendence in percentage General Assemblies of members Anrep Annual report and accounts presented to members Atmrep Attendence in percentage Meetings of Representatives/Delegates Atmcom Attendence in percentage Meetings on community level Anrepwr Annual report and accounts presented to members: written Anrepm Annual report and accounts presented to members: in meetings fininfexm Financial information explained to members Anrepam Annual report and accounts are approved by the members Filsys A filing system exists Booksys A bookkeeping system exists minmeet Minutes of meetings exist planapga Plans are approved by the General Assembly A correlation matrix for the democratic variables is in appendix 1. Table 3: Summary statistics democratic variables Variable Obs Mean Std. Dev. Min Max Sdem 276 1.27 0.45 1 3 Demostr 280 1.06 0.24 1 2 Demelec 279 1.08 0.28 1 3 Inflow 268 5.71 4.77 1 24 Infwrit 269 1.78 0.98 1 3 Infmeet 275 1.08 0.39 1 3 Inffin 274 1.14 0.51 1 3 Infcom 270 1.16 0.54 1 3 Infcomi 262 1.26 0.67 1 3 Infpro 227 1.84 0.99 1 3 frgenasm 264 2.28 2.89 0 22 Frmrep 233 11.03 9.62 0 52 Frmcom 197 10.65 6.74 0 50 Atgenm 239 78.88 17.44 21 100 Atmrep 216 89.58 10.83 40 100 Atmcom 174 84.03 14.12 45 100 Anrep 272 1.19 0.49 1 3 Anrepwr 268 1.58 0.91 1 3 Anrepm 268 1.28 0.69 1 3 fininfexm 275 1.27 0.55 1 3 Anrepam 272 1.32 0.65 1 3 Filsys 277 1.22 0.52 1 3 16