1. INTRODUCTION

Coffee produced in Mexico mostly comes from small producers-with varying socioeconomic and biophysical characteristics-who tend to own less than two hectares of land and operate under marginalizing conditions. Furthermore, they face high production costs, volatile coffee prices, substantial difficulties accessing the market, little bargaining power, a lack of training and a lack of financing (Bernard and Spielman, 2009; ^{Robles Berlanga, 2011}). Under these conditions, the attainable benefits of coffee bean production are limited; such conditions also limit the adoption of production practices that improve productivity and quality and prevent diseases and pests (^{Barham et al., 2011}; ^{Beuchelt and Zeller, 2011}; ^{Barham and Weber, 2012}; ^{Bravo-Monroy et al., 2016}). These circumstances have been a determining factor in the reduction of coffee production and cultivation in Mexico, as compared to the 1980s (^{CEDRSSA, 2018}).

The present article seeks to determine whether more consolidated coffee organizations actually increase producers' benefits, regardless of their socioeconomic conditions and the biophysical characteristics of their agricultural land. To do this, the case of the Santa María Yucuhiti municipality, one of the most important coffee production regions of the state of Oaxaca, Mexico, was studied.

In this case, possible benefits include increased access to credit and improved management practices, as well as entering an organic market context that allows producers to obtain higher prices, thereby increasing their production and reducing the presence of diseases. This study focuses specifically on the presence of rust¹ as a way to capture the vulnerability of coffee plantations to disease, given that in 2017-the year in which data was collected-the amount of rust recorded in the area of study was quite severe.

The hypothesis presented here is based on neo-institutional theory, which establishes that individuals socialize and organize in order to undertake shared actions and function as collective actors in the market. This is done with the aim of obtaining goods and services at a lower cost by exploiting economies of scale, managing with incomplete markets or internalizing other market failures (^{Téllez-Iregui and Cubillos-González, 2009}; ^{Narváez-Rodríguez, 2014}; ^{Caldentey, 2003}). All of the above is translated to the case of the coffee sector, under the assumption that consolidated coffee organizations can effectively offer more benefits to their affiliates because they have the scale and organizational structure to reduce transaction costs, improve their bargaining position vis-à-vis buyers, and exploit economies of scale (^{Bray et al., 2002}; Bernard and Spielman, 2009; ^{Wossen et al., 2017}; ^{Ma and Abdulai, 2016}; ^{Milford, 2012}; ^{Abebaw and Haile, 2013}). In the event that such an organization does not grant more benefits to certain types of producers, they can abandon it and join another which does.

Various studies indicate that small producers in general, and coffee growers in particular, have found in rural producer organizations an alternative to high production costs, volatility of coffee prices, substantial difficulties in accessing the market, poor bargaining power, and a lack of training and financing (^{Bray et al., 2002}; Bernard and Spielman, 2009; ^{Ito et al., 2012}; ^{Bravo-Monroy et al., 2016}; ^{Ma and Abdulai, 2016}; ^{Verhofstadt and Maertens, 2016}). Such organizations offer various financial, commercial and productive strategies, such as obtaining a certification in organic or fair-trade production (^{Perfecto et al., 1996}; ^{Bray et al., 2002}; ^{Barham and Weber, 2012}; ^{Haggar et al., 2017}; ^{Wossen et al., 2017}).

There is also evidence that these strategies improve prices and net income in particular (^{Valkila and Nygren, 2009}; ^{Barham et al., 2011}; ^{Beuchelt and Zeller, 2011}; ^{Barham and Weber, 2012}; ^{Ito et al., 2012}; ^{May Abdulai, 2016}; ^{Haggar et al., 2017}), and that through them producers have more access to credit (Bernard and Spielman, 2009; ^{Donovan and Poole, 2014}; ^{Wossen et al., 2017}), in addition to improving certain social conditions (^{Barham et al., 2011}; ^{Donovan and Poole, 2014}; ^{Bravo-Monroy et al., 2016}; ^{Ibañez and Blackman, 2016}; ^{Wossen et al., 2017}).

However, other studies imply that there are no benefits to belonging to an organization (^{Attwood and Baviskar, 1988}) and that certified organizations do not improve the conditions of most producers or certain groups of producers (^{Beuchelt and Zeller, 2011}; ^{Lyngbæk et al., 2001}).

To test the hypothesis, coffee producers were first grouped according to their biophysical characteristics and production management strategies, and in terms of access to credit and the income they receive from the sale of coffee beans.

Producers were grouped using a latent model that statistically identifies the optimal number of groups which can be obtained from a sample (^{Lo et al., 2001}). After applying this model, it became evident that there are three groups of producers, each with different features. One group is distinguished by having the highest production, procuring the highest prices, registering the most access to credit and having the lowest proportion of rust cases. Meanwhile, the second group is defined by having the lowest production and lowest prices, the least access to credit and the highest proportion of rust cases. Finally, the third group registers average levels for these variables.

Next, the relationships between different variables (both socioeconomic and biophysical) that characterize the producers and the groups obtained from the latent model were examined using a multinomial logistic regression. One of these variables is membership in a consolidated coffee organization. This allowed for an exploration of how certain variables-in particular, belonging to a consolidated organization-influence the possibility of becoming part of the group with higher production, higher prices, better access to credit and less presence of rust.

This study uses representative data from 2017 for producers belonging to various coffee organizations as well as unaffiliated producers in the municipality of Santa María Yucuhiti, Oaxaca. In particular, consolidated organizations were considered those belonging to the State Coffee Producer Coordinator of the State of Oaxaca civil association (SCPCO) because, as explained in the following section, this is the state's largest coffee producer organization and one of its most experienced.

The area of study is a coffee production region with multiple community organizations that have organic certifications, and which is characterized by producing high quality coffee (e.g., in recent years various producers in the region have won the Cup of Excellence)² . Yet it is also distinguished by high levels of social marginalization and limited options for the development of other agricultural activities.

The results of this study indicate that the strategies followed by SCPCO increase the possibility that all of the producer groups in consideration increase their production, average prices, access to credit and probability of not presenting rust. However, the benefits producers receive are inequitable, due partly to pre-existing conditions.

The details and points discussed in this article as a whole have not been explored in the existing literature. In particular, the presence of rust in coffee plantations, the heterogeneous biophysical conditions of the farmland, the adoption of production practices and results of their implementation in the municipality of Santa María Yucuhiti have not been considered.

The article is divided into five sections. The following section provides a brief description of environmental and socioeconomic characteristics of the area of study, and the third section presents the methods used for data collection and analysis. In the fourth section, the results of latent models and the multinomial logistic regression are provided. The last section presents a discussion of the results and conclusions.

2. AREA OF STUDY

The area of study is located in the municipality of Santa María Yucuhiti, part of the Sierra Madre del Sur (see Map 1); this region has temperate wet (79.11%) and temperate sub-humid (20.89%) climates. The municipality is comprised of 68.53% forest and 3.43% induced grassland; 98% of the soil is not suitable for agriculture, only 0.06% is used for year-round manual agriculture and 2.06% is used for seasonal manual agriculture (^{INEGI, 2010}).

Source: prepared by the authors.

Map 1 Chosen locations in the Santa María Yucuhiti municipality, Oaxaca 

In Yucuhiti, most coffee producers are organized through cooperative organizations and marketing networks that make up farmer enterprises. These organizations differ from each other in terms of their administrative complexity. Additionally, all of them have organic certifications issued by the Ministry of Agriculture and Rural Development (MARD); several have other international certifications, such as that of the United States Department of Agriculture or the European Union’s Commissioner for Agriculture.

SCPCO is the largest coffee commercialization organization in the state of Oaxaca and specializes in providing technical assistance, marketing services and low-cost credit to its members; it covers all areas of the state dedicated to coffee bean production. It has been recognized for its experience in integrating the production chain and for influencing public policies in the coffee sector. Its marketing strategies promote fair trade and it is one of the pioneers in access to specialized markets and certification seals (^{Calo and Wise, 2005}).³ The organization exerts an important influence in the region of study through the Tee Ne Nuu and Café Yucuhiti organizations (see Table 3).

3. METHODS

To determine whether producers belonging to SCPCO affiliated organizations have better production management, producers were divided into homogeneous groups using a latent model (^{Muthén, 2002}) according to their observed behavior regarding established production practices, prices obtained, production development and the presence of rust. The latent model was calculated using ^MPLUS version 7 software.

Using a multinomial logistic regression and STATA version 15 software, the question of whether belonging to a more consolidated organization is associated with more productive groups and better management was examined.

4. RESULTS

5. DISCUSSION AND CONCLUSIONS

The results obtained indicate that the heterogeneity which characterizes producers can be synthesized into three groups that share similar characteristics. The first group is distinguished by having higher production levels, more access to credit and a low level of tree diversity, heterogeneity and number compared to the other groups. The producers in the other groups have characteristics which oppose those favoring conservation, as indicated by ^{Perfecto et al. (1996)}. Compliance with management practices is consistent with these characteristics. That is, the first group is more compliant in terms of production management whereas the other groups are more compliant with respect to environmental management.

All three groups had rust present, although the least productive groups with the most precarious management practices were most affected. This coincides with the argument of ^{Avelino et al. (2015)} regarding factors that led to the expansion of rust in Central America: lack of management, adverse economic conditions (e.g., low prices and high prices for inputs) and conducive environmental factors. In the present case, these factors are mostly identified with the least productive of the selected groups.

Meanwhile, the prices obtained by coffee growers in the least productive groups tend to be lower than those in the most productive group. Furthermore, the results indicate that producers who have organic certifications are in the most productive group (although there are some in the second and third groups). This is consistent with ^{Calo and Wise (2005)}, who note that producers receive different prices if they have an organic certification.

Although the price difference is minimal, it is significant and indicates a positive relationship between management, production, quality and price-as indicated by ^{Taugourdeau et al. (2014)}. The minimal price difference may be due, following ^{Valkila and Nygren (2009)}, ^{Weber (2011)} and ^{De Janvry et al. (2011)}, to the fact that a significant proportion of producers sell their beans to various buyers at different prices; this may be a consequence of the limited demand for certain types of certified coffee. Another probable cause is the presence of rust, which reduced the coffee supply in the region of study and Mexico overall.

The results obtained also indicate that SCPCO’s strategies increase the possibilities of all producers increasing their production, average prices, access to credit and the probability of not having rust. However, this does not mean that they necessarily improve their production management. It was also found that there are different conditions that influence the possibility of belonging to the most productive (or least productive) producer group. Examples of this include altitude and slope of the coffee plantation, in addition to the head of household’s age and the number of hectares dedicated to coffee production. The former three factors are pre-existing (the producer has no control over them) and the latter is determined by the total area that the producers have available (which also means that it is partially pre-existing).⁶ Less productive producers are more likely to have pre-existing conditions that prevent them from improving their production management, production, access to credit, and prices. Consequently, even if organizations provide goods and services to improve the least productive producers’ potential, the possibilities of achieving this are slim. The opposite is true for the most productive producers.

This finding coincides with other studies that point out the fundamental role that pre-existing conditions play in favoring or harming producers (^{Barham and Weber, 2012}; ^{Abebaw and Haile, 2013}; ^{Donovan and Poole, 2014}; ^{Ma and Abdulai, 2016}; ^{Verhofstadt and Maertens, 2016}; ^{Haggar et al., 2017}; ^{Wossen et al., 2017}). For example, age, housing capacity, gender, distance between the plot and markets, productive land area and education are important factors that influence the benefits organizations provide. Institutional arrangements and land tenure also influence the strategies that producers adopt (^{Méndez et al., 2009}).

On the other hand, public mechanisms of strengthening coffee production are not associated with better production management. This result coincides with ^{Robles Berlanga (2011)}, who explains that state intervention-especially in coffee-growing municipalities-is minimal and does not generate positive effects. Furthermore, support meant to improve agricultural production does not impact coffee cultivation.

The prevailing conditions of the coffee sector (e.g., the aftermath of the rust epidemic) and decreasing coffee prices have further reduced the benefits of coffee bean production, and therefore, have increased the opportunity cost of investing in small producers’ production management. These conditions also affect the ability of organizations to provide goods and services to their members. In addition to the factors mentioned above, it is expected that climate change will contribute to conditions which favor diseases and pests, requiring producers to improve their production management.

The consequences of these conditions may be quite negative for producers’ well-being and the conservation of biodiversity, since they could replace coffee production with activities that involve changes in land use, such as agricultural activities (indicated most of the coffee leaders interviewed in the region). This sort of situation was previously experienced in the 1999-2004 period when coffee prices saw historical lows (^{Castro et al., 2004}).

In this context, it is important to find public and private mechanisms that encourage the adoption of practices which improve productivity and quality-as well as prevent diseases and pests-by small producers who grow coffee under shade. This is particularly critical for the most disadvantaged producers, in biophysical and socioeconomic terms.

This will require changing the way governmental organizations and support programs work to promote economies of scale; also, it means avoiding the implementation of generalized production standards which assume biophysical and socioeconomic conditions that are not appropriate to local characteristics, as some certifications are known to do (^{Gonzalez and Nigh, 2005}; ^{Barham and Weber, 2012}).