This book examines the historical and contemporary experience of community forest management in Mexico from a variety of perspectives. As this volume makes clear, the community forest sector in Mexico is large, diverse, and has achieved unusual maturity doing what communities in the rest of the world are only beginning to explore: the commercial production of timber. In most of the world, community forest management refers to the management of recovering forestlands or non-timber forest products on government lands. The achievement of Mexican communities in the commercial production of timber from common property forests was largely accomplished over the last 30 years, but has roots deep in Mexico's twentieth-century history.
Despite these achievements, the community forest sector in Mexico is still little known outside of Mexico, and insufficiently recognized even within Mexico. It also has many challenges and deficiencies. This volume joins other recent research efforts to begin to address this lack of recognition for an important global model (Bray et al. 2003) and to document and analyze both its achievements and shortcomings. We have here collected a series of articles by established researchers in the field, some presenting new data from research commissioned especially for this book, that examines the phenomenon from historical, policy, economic, ecological, sociological, and political perspectives, frequently in ways that integrate these disciplines. The book also contains accounts by some of the important practitioners from the Mexican nongovernmental organization (NGO) sector, which has been involved in promoting community forestry for over two decades.
A few terminological notes are in order. Throughout this book we will refer to community forest management (CFM) as the general phenomenon and to community forest enterprises (CFEs) in specific reference to communities that are commercially producing timber with varying levels of integration. The Mexican Revolution in the second decade of the twentieth century left a strong mark on land tenure, creating or reinforcing community properties known as ejidos and indigenous or agrarian communities. While there are some differences in origins and governance, both forms establish collective governance of a common territory or property. While these community lands were long defined as held in usufruct from the state, reforms to the Mexican Constitution in 1992 strengthened community ownership of these lands. Unless it is important to distinguish them, the generic term communities will be used to refer to both of the common property community land tenure systems that exist in Mexico, ejidos and agrarian communities, as defined in Mexican agrarian law. Individual forest smallholder private properties exist in Mexico, and are probably more important than realized in the forest sector, but are not covered in this book.
The Forests of Mexico: Extent, Ecology, and Deforestation
According to the 2000-2001 National Forest Inventory, 32.75% of the national territory of Mexico is covered by "forests and rainforests," corresponding to 63.6 million hectares. Of this, 32.9 million hectares (52% of total forests and rainforests) are temperate zone forests and 30.7 million (48% of the total) are tropical forests, both tropical dry forests and rainforests (INEGI 1997).
The temperate pine-oak forests of Mexico cover the Sierra Madre Occidental and the Sierra Madre Oriental, the mountain ranges in western and eastern Mexico, the Volcanic Axis, which joins the two ranges in central Mexico, and the Sierra Madre del Sur along the Pacific coast of Guerrero and Oaxaca. In the south, after breaking at the Isthmus of Tehuantepec, the mountains rise again in the Sierra Madre de Chiapas and the Mesa de Chiapas in southeastern Mexico. It is on the slopes of the Sierras that Mexico's pine and oak forests are found, with the greatest number of pine species of any country in the world, some 72 in two major groups (Perry 1991). There are also some 130 species of oak, with both pine and oak having rates of endemism of over 70% (Castilleja 1996).
Castilleja (1996) has divided the two principal forest vegetation classifications for Mexico--Miranda/Hernández and Rzedowski--into the following classification scheme: Tropical Rainforests (selva alta perrenifolia, selva alta subperrenifolia, and selva mediana subperrenifolia), Tropical Seasonal Forests (including Tropical Dry Forests), Tropical Montane Forests, and Coniferous and Oak Forests. It is the Tropical Rainforests and Tropical Seasonal Forests that have been most heavily impacted by deforestation. Today, Tall Tropical Rainforest (selva alta perrenifolia), with canopy heights of over 30 meters and annual rainfall over 3 meters, are estimated to be at about 10% of their original extension, and the largest remaining masses are confined to the Lacandon region of Chiapas and the northern Chiapas-southern Oaxaca region, known as the Chimalapas. Medium-high deciduous Tropical Rainforest extends from northern Veracruz through most of the southern and central Yucatán Peninsula.
Most of the tropical community forest projects in Mexico have developed in selva mediana subperrenefolia (medium semideciduous forests) in southern Campeche and southern and central Quintana Roo. Tropical Seasonal Forests (selvas subcaducifolias and selvas caducifolias) lose up to 50% of their leaves in prolonged dry seasons, and may be less than 10 meters in height. Few if any forest management projects for timber have been developed in these forests. Tropical Montane Forests can be found in an altitudinal belt from 1,000 and 1,500 meters on the western slopes of the Sierra Madre Oriental, parts of the Sierra Madre del Sur, and north and central Chiapas. Because of the relative lack of commercial species in these forests, few community forest projects are found in these forests, either. The Conifer and Oak Forests extend throughout the Sierra Regions, with pines dominating in the higher, colder altitudes and oaks being more common at lower and drier altitudes (Castilleja 1996). The majority of all forest management communities in Mexico are found in the Conifer and Oak Forests, particularly in the states of Chihuahua, Durango, Michoacán, Guerrero, Puebla, and Oaxaca.
These forests contain much of Mexico's vaunted biodiversity. "Although Mexico covers only one percent of the earth's land area, it contains about one tenth of all terrestrial vertebrates and plants known to science. The meeting of the nearctic and neotropical biotic regions, the abundance of topographic islands and the wide climatic variation across its territory are significant factors in Mexico's biodiversity" (Castilleja 1996). Of some 25,000 vascular plant species and 1,352 vertebrate species found in Mexico, 81% of the plant species and 75% of the vertebrates are found in the four types of forest mentioned.
However, as in the rest of the tropics, Mexico's forests have been under assault in recent decades, although studies of deforestation suggest that deforestation rates have varied between tropical and temperate zone areas. As elsewhere in the tropics, deforestation in southeastern Mexico, where the tropical forests are concentrated, has occurred at alarming rates. In the mid-1980s deforestation rates in Mexico's tropical forests was estimated at around 2% a year, but regional studies showed local rates that range from 4.3 to 12.4% annually (World Bank 1995). De Jong et al. (2000) found that from the mid-1970s to the mid-1990s, a block of the Lacandon rainforest had lost nearly one-third of its mature forest, although there had been very little decline within protected areas. It has been estimated that 40% of the historic Lacandon was lost by 1995 (O'Brien 1998). Trejo and Dirzo (2000) have demonstrated that only 27% of the original cover of seasonally dry tropical forest remains in Mexico.
The rates of deforestation mentioned above include estimates of annual losses of forest cover that range from 365,000 annually to 1.5 million hectares annually. However, a comprehensive recent study of land use/land cover change (LUCC) in Mexico has produced the most authoritative figures to date. The National Land Use Inventory of 2001, carried out by the Instituto de Geografía of the Universidad Nacional Autónoma de México (UNAM) and commissioned by INEGI and SEMARNAT, indicates that the rate of annual forest loss in the 1976-2000 period was .25% for forests, .76% for tropical forests, and .33% for semi-arid scrub forests (matorral). This implies an average annual loss of 86,718 hectares for temperate forests, 263,570 hectares for tropical forests, and 194,502 hectares for matorral, for a total average annual loss of 545,000, with a 50,000-hectare margin of error. This is based on the most rigorous and comprehensive study to date and incorporates the most trustworthy satellite images from earlier periods (Velázquez et al. 2002).
The principal proximate drivers of this deforestation are thought to be agricultural and livestock expansion (Barbier and Burgess 1996), which are linked to colonization processes. The factors that drive these proximate causes, such as population density, population growth, increased food production, high agricultural export prices, exchange rate devaluation, increased debt-servicing ratios, and roundwood production, can be considered underlying causes, but "exact magnitudes of relationships cannot be reliably estimated" (Barbier and Burgess 1996). Attributing deforestation to particular policies can be challenging since "In practice . . . it is very difficult to determine the overall effects of policy changes on deforestation, as it is likely that a given policy change will have both positive and negative impacts on forest conversion and degradation" (Barbier and Burgess 1996).
Until recently, little effort has been made to link deforestation or stability in forest cover with community forest management (however, see Durán-Medina et al., this volume). The vast majority of community-managed forests are in the mountainous, coniferous zones that in recent decades have shown lower rates of deforestation, although no cause-and-effect relationship has been argued. In addition, tropical land use change has been well below national rates in two areas where community forestry has been prominent, southern Campeche and central Quintana Roo (Palacio-Prieto et al. 2000). It is also noteworthy that, as historical processes, the rise of community forestry in Mexico in the 1970s occurred precisely during the period when tropical deforestation was at its most intensive, and its proponents at the time and since have promoted it as an alternative to deforestation.
Economic Dimensions of Mexican Forests and the Role of CFM in the Forest Sector
Forest production plays only a very minor role in the overall Mexican economy. In the early 1990s the commercial production of timber was slightly less than 1% of the GDP of Mexico, with its share declining nearly 25% since 1987. Historically, very little investment has been made by the government in forestry, and as little as 4% of the total budget for agriculture has been allocated to the forestry agency (World Bank 1995). The forest industry is heavily concentrated in the three states of Durango, Chihuahua, and Michoacán, which have 63% of all industrial installations. Mexican timber production and the forest industry are not considered to be internationally competitive because, according to the World Bank, "production costs (including transport) are high, community-managed forests are inefficient, few forests are actively managed, and lack of infrastructure makes most of the timber inaccessible" (World Bank 1995). Although the World Bank lays part of the responsibility for underperformance of the Mexican forest sector on community forests, this volume will explore some of the factors that mitigate this presumed underperformance and how some aspects of it could become a source of competitive strength in the marketplace.
The exact contribution of community-managed forests to the overall forest sector is not clear. This is not surprising, since there is much confusion even about such basic facts as the total roundwood and sawnwood production in the overall forest sector. FAO and Mexican government figures show dramatically different production levels for both roundwood and sawnwood. One observer of the Mexican forest sector has noted that "the SEMARNAP numbers are probably understated and the FAO numbers overstated. I doubt we could sort this out with a month of hard study" (H. L. Arnold, email, 22 August 2002). Arnold has suggested that a figure of around 25 million cubic meters of roundwood production and 8 million cubic meters of sawnwood production may be close for the 1990s, but despite scattered guesses no one knows the magnitude of production from the community forests.
The Magnitude and Characteristics of Mexican Community Forests and Community Forest Enterprises (CFEs)
It is commonly noted that as much as 80% of Mexico's forests are in the hands of ejidos and indigenous communities. This figure reportedly was first used in a 1980 publication by the National Institute of Statistics and Geography (Instituto Nacional de Estadística, Geografía e Informática; INEGI), with no reference to the empirical foundation of the number, but it has become the single most commonly cited figure on the forest sector in Mexico (J. M. Torres-Rojo, personal communication, 2003). It is not known how forest might have been defined and thus neither the absolute numbers of communities with forests on their lands nor the total number of hectares of that forest are known with any degree of precision. Table 1.1 shows three different estimates of how many "forest communities" exist in Mexico, ranging from 7,000 to 9,047. Presumably many of these forest communities have only small and degraded patches of forest, and they cannot be regarded as communities where the forest is an appreciable economic resource. The National Ecology Institute (INE) is currently developing a more precise estimate of how many communities have forests in their community lands.
A 1991 INEGI ejido census with information on 30,000 ejidos indicates the persistent problems with estimating forests on community lands, since it shows that all ejidos in Mexico have less than 15 million hectares of forests, an impossibly low figure (Torres-Rojo, personal communication, 2003).
But if we take the 7-9,000 range of forest communities, how many of these can be said to operate CFEs or engage in the legal commercial production of timber? Table 1.2 shows the range of existing published estimates for forest production ejidos and includes one data point on the number of logging permits issued, another elusive statistic.
The table shows a range of figures from a low of 288 to a high of 740. Since most of these figures come from the late 1980s to mid-1990s, it may be useful to compare these numbers to the logging permit data point from 1992. If we assume that the 1992 figure for number of logging permits is typical, this would suggest that the number of CFEs constitute from 16 to 42% of the total logging permits. The remainder would presumably be mostly private property owners. A new national study is currently under way to determine at a greater level of precision how many communities are producing timber in Mexico, their relative degree of control over the process, and their degree of vertical integration. Early incomplete results from this study suggest that the total number may be over 2,400 communities producing timber legally, well above existing estimates (Octavio Magaña, personal communication, 2004).
The significance of community forestry for overall forest management also varies from state to state, and in some states it may not currently be an important part of overall forest management. For example, in Michoacán, only 10% of the forest area is thought to be under management, and only 3% of the forest ejidos carry out authorized extraction, two-thirds of them as stumpage communities, the lowest level of vertical integration (see below) (Merino-Pérez et al. 2000).
Typologies of Mexican CFEs
For many years, there was little concern about creating classifications or typologies for Mexican CFEs, since it was a relatively undifferentiated phenomenon. This is no longer the case. The size and complexity of the sector demands a typology of CFEs, but this turns out to be a challenging task. The underlying criteria for classifying the type of communities has been the degree of processing of the tree when sold--from standing to sawnwood--and/or the proximity to the final sale to the buyer. Until the 1970s, almost all Mexican forest communities that produced timber were considered either rentistas, a term that refers to the fact that communities "rented" their forests to outside loggers, whether contractors or concessionaires, or empresas ejidales forestales (forest ejido enterprises). As early as the 1940s and continuing into the early 1970s, various government agencies promoted community sawmills under the term "forest ejido enterprises." These sawmills were not independent businesses, since they were almost always forced to sell to only one buyer, the concessionaire, at the price it set, and the government agrarian reform agency had a strong hand in administering the mill.
Beginning in the 1970s, as more CFEs began to emerge, and the era of the concessions came to a close (see Merino-Pérez and Segura-Warnholtz, this volume), almost all forest communities were allowed to sell their timber and receive the full market price, not a government-set stumpage fee. Beginning in the 1970s a large number of CFEs emerged, with varying degrees of vertical integration that needed classification. Informal classifications came into popular usage as to whether a community sold timber "on the stump" or "standing timber" (al tocón); "at the head of the road" (a pie de brecha--they logged and took it to a forest logyard, but a buyer came to take it to the sawmill); "delivered to the patio" (al patio) (those that delivered roundwood to the sawmill, implying they had their own logging trucks); and "sawmill" communities, which had their own sawmills.
The first formal effort at classification was carried out by the World Bank (1995), which proposed a very complicated classification scheme with multiple criteria in each category. Although the numbers of CFEs in each category came to be widely quoted, the categorization appears to have been based on questionable assumptions. A much simpler scheme was proposed and developed by the Programa para la Conservación y Manejo Forestal (PROCYMAF), a World Bank/government of Mexico project to finance projects in community forest management. The PROCYMAF classification, developed in 1997-1998, is shown in Table 1.3.
As Table 1.3 indicates, Type I communities have forest resources but are not legally exploiting them for timber; Type II communities have their forests logged by outside contractors, with varying levels of direct participation in the process; Type III communities have some form of CFE where they control the logging process; and Type IV communities have sawmills and do their own marketing. Most of the focus has been on Types II-IV, since Type I communities by definition are not exploiting their forest and do not have any kind of CFE. For ease of reference, and after Antinori (2000), we shall refer to Type II communities as stumpage communities (they only sell timber on the stump and may have little involvement with the extraction process, Type III communities as roundwood communities ("roundwood" being an industry term for sawn tree trunks), and Type IV communities as sawmill communities (they have their own sawmills). Antinori (2000) has also proposed a Type V, which she calls finished products communities, communities that produce products elaborated from sawnwood, which may include dried sawnwood, furniture, and plywood.
Drawing on Antinori's modifications, we propose a modification of the PROCYMAF classification, which appears in Table 1.4.
In this proposed modification, Type I remains the same and Type II remains essentially the same, with the additional criterion that the community may be employed as laborers by the contractors and still be regarded as a Type II community. Type III is modified to include two phases. In Phase I the communities assume direct control of the extraction process by having their own logging team that works with the contractor, under the direction of the ejido president. This logging team is normally directed by a person known as jefe de monte (logging foreman), who also has other specialized functions.
In Phase II, the community may begin to capitalize itself further by acquiring extraction machinery such as skidders, tractors, winches, and trucks, and may also begin to acquire more specialized administrative functions, especially in accounting. Type IV, sawmill community, remains the same, and a fifth type, "finished products community," is added, with the criteria noted above. A final issue that should be mentioned is, "When do we consider a community to have a CFE?" It has been suggested, for example, that Type II stumpage communities are not operating CFEs, since they do nothing but take the money for selling the timber off their land. It has even been suggested that many Type III communities don't really have CFEs since they keep no books and have no capitalization. In this interpretation, a CFE emerges only when the business is formally incorporated, when a manager or managerial council is established, and when other aspects of a formal business operation are achieved.
It is here argued that all logging communities are CFEs. At the lower levels of integration, they may be enterprises in which no operating capital is maintained, in which all profits are immediately distributed, and which shut down entirely between logging seasons, but these are nonetheless income-generating operations based on common property that make production decisions, even if the decision is to contract out all activities. Thus, you may have poorly capitalized and structured or highly capitalized and structured CFEs, with most in-between, but they all are CFEs.
Mexican CFEs: The Common Property Basis
Common property has been defined as one among three major forms of property: private, government, and common properties. The discussion of historical cases of common property have tended to focus on traditional, local, and indigenous forms of governing natural resource extraction from territories held in common or, in a modern context, natural resources which by their nature do not lend themselves to either private or government tenure forms, such as groundwater or the atmosphere (Ostrom 1990). An important distinction has been made between open-access and closed-access common property.
Open-access refers to situations where a resource is genuinely without owners, and where no one feels responsible for the maintenance of the resource. It has been suggested that this is not even really a form of common property, but rather of "propertylessness." By contrast, a closed-access common property has a clear set of owners. Thus, it has been argued that closed-access common property should more properly be regarded as a form of jointly held private property, like a corporation (McKean 2000). Here the importance of common property terms such as excludability (the right and capacity of owners to exclude others from the resource) and subtractability (access must be controlled because use by one reduces the capacity of others to use it) are particularly relevant in defining the characteristics of closed-access common property.
The term common pool resources refers to the physical dimensions of a resource while common property regime (both referred to as CPR) refers to the property rights arrangements or the rules which have been developed to govern access to this physical resource (McKean and Ostrom 1995). Another important distinction in common property theory is that between the management of the stock and the flow of a common pool resource. In the case of forests, the stock is the standing forest while the flow is the outputs that come from it. The management issues around the stock and the flow may be quite different (Arnold 1998).
Most of the common property literature focuses on traditional forms of CPR management. These are most commonly characterized as being in a process of dissolution or disappearance, with governments now trying to recover modernized versions of them in the growing perception that they may confer some management advantages at the community level (Arnold 1998). However, Mexico presents a most unusual and little noted case in the common property literature. Some analysts have rather mechanically applied Ostrom's (1990) "design principles" to the Mexican case (Vargas-Prieto 1998), but Mexico actually represents what may be a unique case in the common property literature. While Mexico is rich in indigenous forms of common property management, these indigenous forms were both imitated and overlaid by the massive agrarian reforms that came out of the Mexican Revolution in the second decade of the twentieth century.
This agrarian reform had as its principal land tenure expression the implementation of the two previously mentioned common property forms, ejidos and comunidades indígenas or agrarias (indigenous or agrarian communities), which came to cover about half of the national territory. These land tenure reforms were enshrined in Article 27 of the Constitution of 1917, and remained unchanged until 1992 (Ibarra Mendivel 1996). These agrarian reforms produced a situation where the state created both a common property regime and a common pool resource. The reform of Article 27 in 1992 presented sweeping changes in the ejido system, while still retaining state control in reduced ways. The reforms ended the distribution of rural lands; allowed private enterprises, through a mechanism of stocks and bonds, to become owners of rural lands; established the foundation that allowed ejidos and communities to exercise greater autonomy in their affairs; established new regulations governing use of property within the ejidos; and established new processes for resolving problems (Ibarra Mendivel 1996). This greater autonomy sanctioned a variety of forms of communal management, which were both already present and emergent (see Wilshusen, this volume). Thus, the reform to Article 27 of the Mexican Constitution may be thought of as a form of devolution or decentralization of control over natural resources as it is occurring elsewhere, but marked by the very particular agrarian history of Mexico, where significant state control is still exerted over the use of natural resources, now less out of concern for political control than for environmental protection. Thus, Mexican common property in general, and common property forests in particular, are unique in that in an era with many governments trying to institute new forms of common property, Mexico embarked on a reform, but not dissolution, of a massive state-directed effort to create common property that goes back to the third decade of the twentieth century.
This legal background, and the specific development of community forest management in Mexico, thus distinguishes it from other cases in the common property literature. For example, (McKean and Ostrom 1995) argue that "in most instances common property regimes seem to have been legislated out of existence," yet in Mexico a massive common property regime was legislated into existence. McKean and Ostrom (1995) title their article "Common Property Regimes in the Forest: Just a Relic from the Past?" but in Mexico common property regimes in the forest are the widespread legal present, not at all a relic from the past, and have in fact been reinforced by the reforms. In the same vein, contemporary common property regimes have also been characterized as those that have "endured" and those that have "emerged" (Arnold 1998), but Mexico is neither. It represents an ongoing, solid, widespread institutional reality and what has "emerged" over the last two decades may be thought of, in global development terms, as the "second stage" of what common property arrangements may lead to in terms of the erection of community enterprises on a common property base. It is in this sense that Mexico may be said to be "in the vanguard" or "the face of the future" in community forest management globally (Stone and D'Andrea 2001).
In Mexico, we are confronted with a massive, state-structured form of common property management where change in rules over management of forest resources has been driven either by government policy or by changes emanating from formally constituted organizations. Therefore, the more informal institutions-as-rules approach common in the CPR literature has relatively little applicability to the Mexican case. Some of the "rules" of resource use for Mexican ejidos are found in the agrarian laws of Mexico, while changes or additions to these rules are driven by training and technical assistance coming through government programs or second-level community organizations. Formal organizations do not play a prominent role in most of the CPR literature, but in Mexico they have been key, including both government organizations and formal civil society small farmer organizations. As Antinori (2000) has noted, "The Mexican agrarian communities are formal institutions that are adapting to a new role in resource production, whereas much of the common property literature assesses informal institutions not recognized by the state apparatus."
There are few examples in the world of formal, market-oriented community enterprises established on the basis of a common property resource, yet Mexico's forest communities have thousands of examples of this phenomenon. Common property administration by local communities is almost always seen in the context of subsistence economies. Almost entirely missing from the common property literature is a "systematic focus on stakeholders in a common property resource responding to larger market opportunities as an alternative source of benefits provided by the common property asset" (Antinori 2000). Yet this is a common occurrence in Mexican CFM.
As governments throughout the world have attempted to decentralize the administration of forest resources, various practices known as "co-management" and "joint management" have emerged (McCay and Acheson 1987). However, the term has usually referred to a mixture of local and state governance over a publicly owned resource. In Mexico, we have a case where community autonomy over lands has been strengthened in some aspects, while a strong government presence has been maintained in other aspects, particularly in the regulation of forest extraction. As will be explored further in the conclusions, the Mexican case may be thought of as a form of "joint management" or "co-management," but on the basis of privately held communal property. That is, Mexican communities manage their forests for timber with many decisions being made autonomously, but also under a strong regulatory framework provided by Mexican forestry law and the Mexican environmental agency, the Secretaría de Medio Ambiente y Recursos Naturales (SEMARNAT).
Arnold (1998) has conducted one of the most recent and comprehensive reviews of managing forests as common property. The only references to Mexico in the study are in a description of the Plan Piloto Forestal (PPF) in Quintana Roo, and some of its most salient aspects within the common property are not fully delineated. In a later section, the author concludes that the Quintana Roo case, "in which local communities are engaged in commercial logging and processing, shows that complex processes and sophisticated technologies can be handled at this level, given an appropriate institutional framework" (Arnold 1998:54). This is one of the lessons of the Mexican experience, but the fact that the common property system was created by the state early in the twentieth century with, as we shall see, massive transfers of natural forest assets taking place in later periods, and that these are community enterprises based on a common property resource, is not emphasized for the unique case that it is.
Thus, the articles in this volume are part of new, more systematic, efforts to understand the dimensions, unique characteristics, achievements, and challenges of Mexican community forest management. As the attentive reader will note, there is uneven coverage of some of the important community forestry regions of Mexico, with a bias toward studies of Oaxaca, and a relative paucity of studies from northern Mexico. There is an easy explanation for this. Oaxaca, with its rich social and ecological matrix of indigenous groups and biodiverse forests, has long been a magnet for student activists, NGOs, and academics. There are probably more NGOs in some neighborhoods of the city of Oaxaca than in all of Chihuahua.
The book is divided into five sections: Introduction, History, and Policy; Social Processes and Community Forestry; Ecology and Land Use Change in Community Forestry; The Economics of Community Forestry; and Global Comparisons and Conclusions. The Introduction, History, and Policy section lays important groundwork for the other articles. The emergence of such a large and relatively consolidated community forestry sector has deep historical roots in twentieth-century Mexico, and these articles outline some of the agrarian and forest policies which created the sector.
Following the present chapter, Christopher R. Boyer takes a detailed look at the local-level historical roots of community forestry, in this case in Michoacán. This is an important first step toward filling in the details on the little known but rich regional histories of community forestry in Mexico. Boyer's article is also a reminder that not all the regional paths of forest communities have led to successful outcomes, making the need to explain how so many have had relative "success" even more compelling. Boyer examines the community of El Rosario, Michoacán, and its neighbors, which today are best known as being a troubled locus of mass tourism that has grown up around visitations to one of the overwintering sites of the Monarch butterfly (Danaus plexippus). Boyer demonstrates that El Rosario's current state of diminishing forests and clandestine logging, which threatens the Monarch Butterfly Reserve, is a result of a series of failures to consolidate community forest management in the region. While much of this book is aimed at understanding why and how CFM has been successful in Mexico, Boyer's article joins the crucial literature that helps us understand why so many other efforts at CFM have become beset by disorganization, violence, and corruption, despite the expressed interest of many community members in the sustainable management of their forests (Klooster 2000; Vázquez León 1992).
In the next chapter in this section, Leticia Merino-Pérez and Gerardo Segura-Warnholtz analyze the evolution of Mexican forest policy and the emergence of CFEs. They show how policies of forest concessions and bans slowly began to give way in the 1970s to complex strands of supportive state policy, intertwined with more hostile state policies emanating from different points in the bureaucracy, intensified grassroots mobilizations against forest concessions, and the emergence of urban activists who took to the forests to help communities learn how to manage their own forests. Mexican forest policy has always been beset with tensions between forces that want to realize the ideals of the Mexican Revolution in terms of peasant empowerment and those that regard peasants as incapable of managing their forests for the good of the nation. In the 1930s, there was an ineffectual turn toward empowerment, followed by the long 1940-1970 period, when forest community empowerment was most clearly not on the agenda. Beginning in the 1970s however, forces within and outside the government began to come together to propel a substantial community forest management sector into existence.
In the late 1980s and the first half of the 1990s, the policy pendulum once again swung against community forestry, although by that time the sector had become relatively consolidated and could survive, albeit with difficulties, without consistent government support. However, as the authors detail, beginning with new pro-community forestry programs in the Zedillo administration (1994-2000) and continuing with major new levels of support from the Fox administration (2000-present), we are currently in a period of highly significant state support for community forestry.
In the final article in the History and Policy section, Patricia Gerez-Fernández and Enrique Alatorre-Guzmán analyze the difficult history of forest certification in Mexico. Certification represents an unusual effort to implement forest management policies through the marketplace by encouraging consumers to choose timber products from well-managed forests. Mexico's achievements in forest certification clearly show the strength of the sector in global terms. Only 3% of the 29.63 million hectares which have been certified worldwide are considered "communal," but within this category, Mexico clearly dominates, with nearly half of the communities and half of the certified community forests worldwide. Durango has taken the lead here, with a sectorwide stakeholders effort resulting in 19 certified communities (see Taylor, this volume). Despite this notable achievement, Mexico's communities have little to show for their efforts in terms of increased markets. There have been only a few sporadic sales of certified timber, with most certified communities having sold no certified timber at all. Further, as the big northern timber companies and retailers move toward embracing certification, this will once again leave small producers and communities with no particular competitive edge in the marketplace. As the authors note, new efforts by the Forest Stewardship Council (FSC) to adjust their criteria for small and community producers are under way, and will be crucial if community forestry and other small producers are to gain anything from the certification process.
In the Social Processes and Community Forestry section, the authors review the recent history of the emergence of some of the best-known community forestry organizations in Mexico, with a focus on institutional processes. Francisco Chapela, who first arrived in the Sierra Juárez of Oaxaca in the early 1980s as a student activist and stayed to found with others the NGO of which he writes, Rural Studies and Advising (Estudios Rurales y Asesoría Campesina; ERA), gives a detailed account of some of the institutional factors that led to emergence of the Union of Zapotec-Chinantec Communities (Unión de Comunidades Zapoteco-Chinanteca; UZACHI). Chapela shows how an organizing and training strategy that built human and social capital, and recognized the integration of household, community, and organizational needs was able to support one of the best-known experiences in forest management in southern Mexico.
Rodolfo López-Arzola was also an important community forestry organizer, working as the leader of a government team that also landed in Oaxaca in the early 1980s. López-Arzola gives a personal view of the rise and fall of one of the first autonomous (i.e., not directly sponsored by the government) second-level organizations in Mexico, the Union of Forest Communities and Ejidos of Oaxaca (Unión de Comunidades y Ejidos Forestales de Oaxaca; UCEFO). He details UCEFO's difficult emergence as an organization, its major accomplishments in organization and improved forest management (it introduced the more silviculturally sophisticated Method of Silvicultural Development to Oaxaca), as well as the internal and external tensions that led to its effective dissolution after 10 years. Like Boyer's piece, this is a cautionary tale, warning that apparent consolidation can become dissolution if appropriate support is not provided.
Peter Taylor takes us to Durango in northern Mexico to look at how community forestry is evolving there, and shows how CFEs are evolving into new organizational forms. Most community forest advocates, whether government or nongovernment, have usually promoted the model whereby a CFE administers a common property forest where the community as a whole holds both the stock and the flow of the forest. But Taylor, like Peter Wilshusen in Quintana Roo (see below), shows how communities are now taking institutional issues into their own hands, dividing up the stocks and the flows of the forest and structuring their CFEs in the most diverse ways. The emergence of the so-called work groups throughout Mexico represents a genuine grassroots innovation in CFE organization, and shows that there is no one right way to organize a CFE. Although some community forestry advocates find the emergence of work groups to be troubling, few can be troubled by what Taylor also shows is emerging in Durango, probably the first genuine stakeholder community in Mexican forestry. The private timber industry and CFEs in Durango have understood that they have a common stake in improving production and quality of timber, and are banding together in innovative ways with, for example, timber buyers providing subsidies to communities so that they can become certified.
Peter Wilshusen gives a richly detailed ethnographic analysis of how work groups have evolved in two contrasting communities in the tropical forests of southern Quintana Roo. He shows how in the community of Caobas, despite having low volumes of timber, the work groups have become motors of local economic diversification, while in the much higher volume community of Petcacab, work groups have become a source of great community conflict with little of the proceeds being harnessed to foster genuine community development. He also uncovers the emergence of a fascinating timber futures market in Petcacab as an unexpected by-product of the work groups.
In the Ecology and Land Use Change in Community Forestry section, we enter into one of the most important new areas for research on community forestry. Frequent claims have been made that communities will manage their forests more sustainably than private enterprises, but until now there has been little hard evidence to back up that assertion. In this vein, Henricus F. M. Vester and María Angélica Navarro-Martínez take on one of the most contentious issues in tropical silviculture: What are the factors that encourage the regeneration of mahogany and what is the impact of logging on the population of mahogany in the forest? As they note, most silvicultural prescriptions have argued that mahogany needs to have relatively large clearings in order to be able to successfully to grow to canopy-dominating size (although not necessarily to germinate in the first place), and that current silvicultural practices do not permit a sustainable harvest of mahogany over the long run. However, on the basis of new research in the forests of Quintana Roo, they argue that current silvicultural practice will allow continual harvests at current levels--they are more or less "sustainable," an assertion that is certain to spark more debate and research on the subject.
Elvira Durán-Medina, Jean-François Mas, and Alejandro Velázquez present one of the first systematic efforts to examine the impact of community forestry on land use change and to compare it to land use change in another major tenure regime more associated with conservation, formal protected areas, or parks. In a conclusion that will surprise some, they find that the differences between rates of land use change in protected areas and community-managed forests, especially in the tropical area of central Quintana Roo, are not statistically significant, and that there are stronger tendencies for recovery of deforested lands in community forests than in protected areas. Although this does not necessarily say anything about what is going on beneath the forest canopy, it is an important argument for the role of community-managed forests in landscape-level management.
In The Economics of Community Forestry section, the authors began to look at another kind of sustainability, the sustainability of the CFEs as financial and economic entities. Camille Antinori undertakes a brief historical review of the institutional underpinnings of the evolution of CFEs and their roots in the ejido and agrarian community governance forms. She analyzes the issue of what she has elsewhere called the "community as entrepreneurial firm," asking how community enterprises operating with a common property resource compete in the marketplace. Antinori's article is based on the first large-scale comparative and quantitative survey-based study undertaken specifically of the CFM sector, in this case 42 CFEs in Oaxaca. Her examination of transaction costs, contractual hazards, and vertical integration demonstrates that CFEs appear to be profitable at all levels of integration and that "communities prefer to integrate forward to avoid contractual hazards with outside entities like private logging firms." She also concludes on the basis of quantitative evidence that the common pool resources contribute to community welfare and that community ownership and control over production assures access to these benefits.
Torres-Rojo, Guevara-Sanguines, and Bray examine the economic underpinnings of one of Mexico's most successful CFEs, that of the community of El Balcón in the state of Guerrero. El Balcón, after a difficult beginning, has stabilized and grown into a CFE with one of the best sawmills in Mexico, a stable and productive relationship with a U.S. timber company, growing export markets, and clear community leadership of the enterprise, while also leaving many decisions to professional management. However, the authors argue that there are weaknesses to the business model that must be addressed if El Balcón is to continue delivering benefits to community members and maintaining their forest as a productive ecosystem. It is suggested that greater investments must be made in the forest, that the forest is currently subsidizing the industry to an unsustainable degree, and that there must be a "reengineering" of the processes of production, investment, and distribution of profits.
Finally, in the Global Comparisons and Conclusions section, Dan Klooster and Shrinidhi Ambinakudige situate Mexican community forestry within a sweeping global review. As the authors make clear, most local communities worldwide are still struggling to organize themselves to co-manage state forest lands as woodlots for firewood, forage, and non-timber forest products, even if they are lands that they have used for hundreds of years. And although important advances are being made in these areas, they conclude that Mexico "confirms the expectations of community forestry proponents: Greater community participation in forest benefits and greater community power over forest management results in better forest use and protection and improved livelihoods for local people." David Bray then summarizes some of the lessons learned from some 30 years of community forestry in Mexico, and where it might be going from here.
Camille Antinori, in her article, comes to the following conclusion: "Therefore, Oaxaca's forests are community-managed in a strong sense of the phrase. The community governance and territorial land claims have national recognition, and a governance structure is in place where (mainly male) community members can determine the distribution of forest benefits while the elected authority on common property matters, the CBC, normally has the power to administer decisions on common property forestry matters." As Antinori is also well aware, many CFEs stumble from crisis to crisis, have serious problems in stable governance, fiscal administration, and forest management, and do indeed, as she notes, continue traditional patterns of suppression of women's rights. However, this does not take away from the fact that an impressive number of CFEs in Mexico, most likely in the hundreds, have achieved the constitution of enterprises that deliver benefits to the community, maintain the ecosystem functions of the forests, and that are, indeed, community managed in the strong sense of the phrase.
Finally, given the magnitude of Mexican community forests, both in number and in territory, they also have very significant implications for Mexican landscapes. To the extent that community-managed forests contribute to the maintenance of forest cover and other ecosystem processes, they may be said to contribute to "sustainable landscapes." The idea of landscape management is one that is frequently found in Europe but less so in Latin America. However, Mexico has many areas where there is still a significant forest matrix and which have been comparatively stable in terms of forest cover, but these regions have been little recognized. These "sustainable landscapes" may be operatively defined as those where deforestation rates are low, in equilibrium, or with net expansion of forest cover, and where multiple institutional processes are present that encourage the preservation of ecosystem structure, composition, and process (Haines-Young 2000; Shepherd and Harshaw 2001). Working landscapes can still retain many crucial ecosystem processes and biodiversity. As Chazdon (1998:1295) has argued, "A tropical landscape containing a matrix of old-growth forest fragments, second-growth forest, logged forest, and agricultural fields could conceivably protect most of the species present in the regional biota." Conservation organizations speak of "designing" sustainable landscapes by rebuilding fragmented forests in biodiversity conservation corridors (CABS et al. 2000; Sanderson et al. 2003). However, we argue in this book that many Mexican forest communities are already managing for sustainable landscapes that have been designed by both grassroots action and government policy over many decades.
As mentioned earlier, this volume attempts to pull together the existing work and reflection by its most important researchers and practitioners on Mexican community-based forestry. The breadth of the topics reflects the focus of existing work, but also reveals areas where there is a lack of systematic inquiry, despite its importance. For example, conspicuously absent is research incorporating a gender perspective in the analysis of Mexican CFM. In general within CFM communities, gender rigidity encapsulates a severe limitation of women's participation in the power structures and decision making of forestry activity. The strong traditional governance systems in the rural communities and the male dominance of the forestry sector, as a whole, often combine to suppress women's rights with little understood implications and consequences.
Deeper inquiry into the understanding the benefits of CFM at the household level is also an understudied topic, as is the impact of migration and remittances and their impact on family and community survival strategies, changes in their governance systems, and influence in their economic well-being. Lastly, the role of the generational dynamics of these communities is of crucial importance. Social sustainability must go hand in hand with the analysis of environmental sustainability, which leads us to the need for understanding how CFM is affecting the youth in forestry communities: their access to resources and expertise, and their participation in leadership.