Global Governance in Forestry: a Cross-National Analysis

International Journal of Sustainable Development & World Ecology

ISSN: 1350-4509 (Print) 1745-2627 (Online) Journal homepage: https://www.tandfonline.com/loi/tsdw20

Global governance in forestry: a cross-national analysis

J. M. Sommer

To cite this article: J. M. Sommer (2020): Global governance in forestry: a cross-national analysis, International Journal of Sustainable Development & World Ecology, DOI: 10.1080/13504509.2020.1714787 To link to this article: https://doi.org/10.1080/13504509.2020.1714787

Published online: 29 Jan 2020.

Submit your article to this journal

View related articles

View Crossmark data

Full Terms & Conditions of access and use can be found at https://www.tandfonline.com/action/journalInformation?journalCode=tsdw20 INTERNATIONAL JOURNAL OF SUSTAINABLE DEVELOPMENT & WORLD ECOLOGY https://doi.org/10.1080/13504509.2020.1714787

Global governance in forestry: a cross-national analysis J. M. Sommer

Department of Sociology, University of South Florida, Tampa, (USF)

ABSTRACT ARTICLE HISTORY On the one hand, researchers argue that global governance in forestry is fragmented and Received 29 November 2019 ineffective. On the other hand, some argue that global forestry governance is key to reducing Accepted 7 January 2020 forest loss related to climate change issues. Using ordinary least squares (OLS) regression for KEYWORDS a sample of 155 nations, this research tests the association between one type of global Global governance; forest governance, the number of ratifications of environmental treaties that include obligations to loss; cross-national reduce forest loss for each nation, and forest loss from 2001 to 2014. As a whole, it appears that despite a lack of unification of multilateral environmental treaties that address forest loss and the absence of a global forestry convention, multilateral forestry treaties are effective at reducing forest loss. While there are several important programs and initiatives from global forestry governance treaties impact forest loss, the effect is relatively small compared to other factors.

Introduction tools. On the other hand, global forestry governance Many researchers in social sciences and international may be key to reducing forest loss related to climate law have questioned the effectiveness of global govern- change issues. Some argue that existing global forestry ance at reducing climate change (Finkelstein 1995; governance treaties can be useful when combined with Young 1997; Canan and Reichman 2002; Virtanen and bilateral agreements and national laws to create sustain- Palmujoki 2002; Karns and Karen 2004;Kirtonand able development in forestry (Kirton and Trebilcock Trebilcock 2017). Given the persistent issue of global 2017). Though it is important that these agreements climate change, we need to continue to find ways to focus on monitoring, enforcement, and capacity build- adapt to and mitigate the damage (Biermann 2006). ing among others to be successful (Haas 2004). At the One of the largest detriments to climate change adapta- very least, environmental conferences that often lead to tion and mitigation is forest loss (Rudel 2017). Therefore, multilateral environmental agreements can help with the issue of the effectiveness of global forestry govern- the spread of norms, awareness, and practices that ance is particularly pressing. However, there are no may reduce forest loss (Haas 2002). While many multilateral treaties that fully encompass protections researchers document failures in global environmental for all forests. Instead, there are multilateral treaties governance (Park et al. 2008), there are several activities that focus on the environment or climate change as and obligations that exist in multilateral environmental a whole that have some provisions for forestry issues treaties that should reduce forest loss. However, (Kirton and Trebilcock 2017). The one exception is tro- researchers still insist that harmonizing existing global pical forestry treaties, which deals exclusively with tro- forestry treaties or using them when applicable can pical forest issues. However, because the world’sforests leave forests at risk (Ruis 2001). are not all tropical it cannot fully address forestry pro- Using ordinary least squares (OLS) regression for tection in general. Even with a piecemeal approach, a sample of 155 nations, I test the association between these forestry provisions in multilateral environmental the number of ratifications of environmental treaties treaties, when used in combination with the tropical that include obligations to reduce forest loss for each forestry treaties, are not exhaustive, which may lead to nation and forest loss from 2001 to 2014. OLS is gaps in forestry protection (Ruis 2001). a common method used in the social sciences to test On the one hand, researchers argue that global gov- how independent variables are associated with depen- ernance treaties in forestry are fragmented and ineffec- dent variables when data are only available at one time tive (Ruis 2001; Biermann 2002, 2017; Biermann et al. point, which is the case for the dependent variable 2009). According to Elliott (2017) many multilateral trea- data, forest loss. For the main independent variable ties do not have sanctions for non-compliance, lack of interest, I code original data from ECOLEX (law achievable targets, and have inadequate assessment database) for 8 multilateral environmental treaties

CONTACT J. M. Sommer [email protected] © 2020 Informa UK Limited, trading as Taylor & Francis Group 2 J. M. SOMMER that include protections for forests from 1992 to 2014. these conversations in 1992 (Ruis 2001). In fact, it has The earliest treaty that includes forestry provisions was been argued that these conversations and agreements created in 1992. I stop collecting ratifications after 2014 arose not for forestry protection and to reduce forest because forest data are only available until 2014. This loss, but for commodification (i.e. treating forests as analysis will assess the effectiveness of multilateral a commodity). It is suggested that any aims for forest environmental treaties at controlling forest loss in conservation are only for the purposes of continuing a cross-national comparative context. The main trade, rather than seeing forests and the biodiversity research question of this article is as follows: is the they inhabit as important for the natural environment number of ratifications of environmental treaties that and the people, animals, and other wildlife that live include obligations to reduce forest loss for each within them (Lipschutz 2001; Schama, 1995; Peluso nation associated with the change in forest loss from 1992; Scott 1998). In short, some have argued that 2001 to 2014? there is little political will to actually create an agree- The organization of this article is as follows. First, ment that protects forests for environmental reasons I give the background to multilateral forestry agree- (Ruis 2001). ments to present the history on global governance on Though the following review of existing agree- forestry. After I review how each treaty included in the ments is not exhaustive, it shows that there has been analysis should relate to reduced forest loss and limited development of multilateral forestry law. include literature that suggests favorable impacts of However, there are several recommendations in multi- treaties on forests. Next, I discuss the potential limita- lateral environmental agreements (which are binding) tions of existing treaties that deal with forest protec- that have evolved during this period of time tion. The previous two sections are necessary to (MacKenzie 2012). In the next section, I go over each identify reasons why the treaties should impact forests, relevant multilateral agreement identified through the both positively and negatively or not at all. These Ecolex database on environmental law. In particular, reasons are important to identify to be able to inter- I analyze each document for its relevance to forests pret the findings of the statistical analysis and makes and what, if any, enforcement mechanisms exist for the claims on why the findings are what they are. activities outlines in each document. Following these sections, I review the methods and data used in this analysis. Finally, I describe the results, interpret the findings and discuss the implications of Multilateral environmental treaties that this research. include forest protections Using Ecolex, I identified 8 multilateral treaties that are Background relevant to forests. Table 1 lists the treaties in the analysis. There have been several non-binding forestry agreements and instruments beginning in 1992 at the United Nations Conference on Environment and United Nations framework convention on climate fi Development (UNCED). The rst major document, change generally regarded as the first global consensus reached on forests is called the Non-Legally Binding This treaty was established to regulate greenhouse gas Authoritative Statement of Principles for a Global emissions and was ratified by 198 nations. There are Consensus on the Management, Conservation and Sustainable Development of All Types of Forests (1992), which is also referred to as The Statement of Table 1. Environmental treaties that include obligations to Forest Principles. reduce forest loss. Following the work of the UNCED in 1992, the Treaty Year Commission on Sustainable Development (CSD) cre- United Nations Framework Convention on Climate Change 1992 Establishment Agreement for the Center for International 1993 ated the Intergovernmental Forum of Forests (IFF). Its Forestry Research purpose was to continue to propose how to act on Constitution for the Center for International Forestry 1993 Research international forestry issues, which led to the establish- International Tropical Timber Agreement 1994, 2006 ment on the United Nations Forum on Forests (UNFF) United Nations Convention to Combat Desertification in 1994 those Countries Experiencing Serious Drought and/or in 2000 to develop a legal framework for forests (Lanly Desertification, particularly in Africa 1992; Palmer 1992; MacKenzie 2012). The most recent Kyoto Protocol to the United Nations Framework 1997 Convention on Climate Change UNFF contribution was the non-legally binding forest Protocol on Strategic Environmental Assessment to the 2003 instrument in 2007. Negotiations to continue this work Convention on Environmental Impact Assessment in for a binding agreement were to commence in 2015. a Transboundary Context Amendment to Annex B of the Kyoto Protocol to the 2006 The inclusion of the term ‘non-binding’ is thought United Nations Framework Convention on Climate to reflect the disagreement of the parties involved in Change INTERNATIONAL JOURNAL OF SUSTAINABLE DEVELOPMENT & WORLD ECOLOGY 3 several commitments outlined in Article 4, page 5 that constitution for the CIFOR documents to create an inter- relate to forests (emphasis added): national research center for all forests with the intention of improving ‘the productivity of agriculture, forestry, (c) Promote and cooperate in the development, appli- and fisheries in developing countries in ways that cation and diffusion, including transfer, of technologies, practices and processes that control, reduce or enhance nutrition and well-being, especially among prevent anthropogenic emissions of greenhouse low-income people’ (Establishment Agreement for the gases not controlled by the Montreal Protocol in all CIFOR 1993: 1). The agreement aims to establish relevant sectors, including the energy, transport, a research center that is ‘concerned with forestry industry, agriculture, forestry and waste management research that benefits developing countries’ (1). Such sectors; research should lead to the conservation of forests, (d) Promote sustainable management, and promote especially in developing nations. and cooperate in the conservation and enhancement, For example, according to Article 6 of the as appropriate, of sinks and reservoirs of all green- Constitution for the CIFOR, their purpose is ‘to conduct, house gases not controlled by the Montreal Protocol, including biomass, forests and oceans as well as other promote and support research that can provide the terrestrial, coastal and marine ecosystems basis for sustainable forestry and forest systems in developing countries.’ It is also obligated to ‘formulate The mention of forests above concerns their protec- a research program to underpin the science of forestry, tions to reduce anthropogenic emissions. Paragraph with a view to contributing to increasing the forestry c promotes the development of technologies to limit research capacity of developing countries’ (2). To sup- emissions in the forestry sector, and paragraph d wants plement its activities, CIFOR is also expected to ‘create to ensure sustainable management of forests for their and use partnerships with all relevant nations and agen- ability to act as carbon sinks. If followed, these commit- cies to reach its goals’ (see Article 19 and 20, page 10). ments should reduce forest loss because forests need to be preserved so they can continue to hold carbon dioxide in their roots and continue to sequester car- International tropical timber agreement (1994, bon. Cutting down forests would be contradictory to 2006) these goals. On page 8, there is a similar mention of the fi necessity in preserving forests. The rst International Tropical Timber Agreement in 1983 was negotiated by the United Nations Conference “In the implementation of the commitments in this on Trade and Development (UNCTAD). However, it was Article, the Parties shall give full consideration to not established as a multilateral treaty until 1994, where what actions are necessary under the Convention, fi including actions related to funding, insurance and it was rati ed by 45 nations. The language of the docu- the transfer of technology, to meet the specific needs ment suggests that ‘all types of forests’ need ‘conserva- and concerns of developing country Parties arising tion and sustainable management’ (10), even though from the adverse effects of climate change and/or the document only ‘protects’ tropical timber, defined the impact of the implementation of response mea- as ‘non-coniferous tropical wood for industrials uses, sures, especially on: which grows or is produced in the countries situated 1. (c) Countries with arid and semi-arid areas, between the Tropic of Cancer and the Tropic of forested areas and areas liable to forest decay; Capricorn’ (Article 2, Definitions, page 17). This article is primarily concerned with making sure While some parts of the document suggest forests developing country Parties have access to funding and need to be conserved and sustainably managed, it is other capacity requirements to reduce the impact of clear that these documents are meant to sustain for- climate change, including those nations with forested ests for trade, rather than some greater environmental areas. It is important to note that this document later or human rights purpose. In Article 1 (pages 15 and 16), lead to the establishment of the Green Climate Fund the objectives of the document are stated as such: (GCF) in 2010 which provides financial assistance to countries for climate change adaptation and mitiga- (a) To provide an effective framework for consultation (GCF 2018). Arguably these efforts to finance cli- tion, international cooperation and policy mate change adaptation and mitigation should development among all members with regard protect forests rather than harm them. to all relevant aspects of the world timber economy; (b) To provide a forum for consultation to promote Establishment agreement for the center for non-discriminatory timber trade practices; international forestry research (CIFOR) AND (c) To contribute to the process of sustainable constitution for the CIFOR development; (d) To enhance the capacity of members to imple- Australia, Sweden, Switzerland, and the United States ment a strategy for achieving exports of tropical signed both the establishment agreement and the 4 J. M. SOMMER

timber and timber products from sustainably Though it is important to promote non-discriminatory managed sources by the year 2000; trade practices, it may do so to avoid nations discrimi- (e) To promote the expansion and diversification nating against nations known to be trading illegal tim- of international trade in tropical timber from sus- ber (as such was a case in Austria) (see Virtanen and tainable sources by improving the structural con- Palmujoki 2002, where Austria violated GATT when they ditions in international markets, by taking into passed a law prohibiting import of forests that were not account, on the one hand, a long-term increase certified, i.e. not illegally obtained). in consumption and continuity of supplies, and, While it promotes ‘sustainable management’ of for- on the other, prices which reflect the costs of ests, the focus is mostly on timber for export purposes, sustainable forest management and which are and even calls for ‘expansion’ and ‘diversification’ to remunerative and equitable for members, and grow timber trade. However, there are objectives the improvement of market access; which seem more positive given our global reliance (f) To promote and support research and devel- on forestry products such as investigating how to use opment with a view improving forest manage- trees most efficiently for their resources. Additionally, ment and efficiency of wood utilization as well the document calls for transparency in forestry but also increasing the capacity to conserve and within its staff. Article 16 page 24, states that, ‘Neither enhance other forest values in timber- the Executive Director nor any member of the staff producing tropical forests; shall have any financial interest in the timber industry (g) To develop and contribute towards mechan- or trade or associated commercial activities.’ If there isms for the provision of new and additional was a conflict of interest, this could be damaging to financial resources and expertise needed to forests. enhance the capacity of producing members The 2006 agreement was ratified by 74 nations, and to attain the objectives of this Agreement; obviously has similar objectives and proceedings as (h) To improve market intelligence with a view to the 1994 document. They are included below, as ensuring greater transparency in the interna- there have been a few key additions that should pro- tional timber market, including the gathering, tect forests: compilation, and dissemination of trade-related data, including data related to species being “(c) Contributing to sustainable development and to traded; poverty alleviation; (i) To promote increased and further processing (e) Promoting improved understanding of the struc- of tropical timber from sustainable sources in tural conditions in international markets, includ- producing member countries with a view to ing long-term trends in consumption and promoting their industrialization and thereby production, factors affecting market access, con- increasing their employment opportunities sumer preferences and prices, and conditions and export earnings; leading to prices which reflect the costs of sus- (j) To encourage members to support and tainable forest management; develop industrial tropical timber reforestation (k) Improving marketing and distribution of tropical and forest management activities as well as timber and timber product exports from sustain- rehabilitation of degraded forest land, with ably managed and legally harvested sources and the regard for the interests of local commu- which are legally traded, including promoting nities dependent on forest resources; consumer awareness; (k) To improve marketing and distribution of tro- (l) Strengthening the capacity of members for the pical timber exports from sustainably managed collection, processing and dissemination of sta- sources; tistics on their trade in timber and information (l) To encourage members to develop national on the sustainable management of their tropical policies aimed at sustainable utilization and forests; conservation of timber-producing forests and (n) Strengthening the capacity of members to their genetic resources and at maintaining the improve forest law enforcement and govern- ecological balance in the regions concerned, in ance, and address illegal logging and related the context of tropical timber trade; trade in tropical timber; (m) To promote the access to, and transfer of, tech- (o) Encouraging information sharing for a better nologies and technical cooperation to implement understanding of voluntary mechanisms such the objectives of this Agreement, including on as, inter alia, certification, to promote sustainable concessional and preferential terms and condi- management of tropical forests, and assisting tions, as mutually agreed; and members with their efforts in this area; (n) To encourage information-sharing on the inter- (q) Promoting better understanding of the contribu- national timber market.” tion of non-timber forest products and INTERNATIONAL JOURNAL OF SUSTAINABLE DEVELOPMENT & WORLD ECOLOGY 5

environmental services to the sustainable manage- United Nations convention to combat desertification ment of tropical forests with the aim of enhancing in those countries experiencing serious drought and/ the capacity of members to develop strategies to or desertification, particularly in Africa strengthen such contributions in the context of This document was ratified by 195 nations, with the sustainable forest management, and cooperating intention to combat desertification. In Article 8 it includes with relevant institutions and processes to this end; national action programs to ‘conserve natural resources’ (r) Encouraging members to recognize the role of and ‘ensuring integrated and sustainable management of forest-dependent indigenous and local commu- natural resources, including forests’ (37). Several articles nities in achieving sustainable forest management are able to grasp the complexity of issues relating to and develop strategies to enhance the capacity of desertification. Article 2 states (emphasis added): these communities to sustainably manage tropical timber-producing forests; and “Particular conditions of the Latin American and (s) Identifying and addressing relevant new and Caribbean region The Parties shall take into considera- emerging issues” (4 and 5). tion . . . :

ff b) the frequent use of unsustainable development prac- Some of the major di erences between the two tices in affected areas as a result of complex interactions agreements is the mention of improving forestry among physical, biological, political, social, cultural and governance, reducing corruption and illegal log- economic factors, including international economic fac- ging, and recognizing local people in sustainable tors such as external indebtedness, deteriorating terms of ff forestry management. Since corruption, poor gov- trade and trade practices which a ect markets for agricultural, fishery and forestry products; ernance and illegal logging are some of the main issues concerning forest loss, reducing some of (c) a sharp drop in the productivity of ecosystems being these issues may help reduce forest loss (Sommer the main consequence of desertification and drought, 2017). taking the form of a decline in agricultural, livestock and The 2006 document also includes important forestry yields and a loss of biological diversity; from the fi social point of view, the results are impoverishment, mentions identi ed in the 1994 document includ- migration, internal population movements, and the dete- ing the executive director not having conflicts of rioration of the quality of life; the region will therefore interest, and the organization having legal person- have to adopt an integrated approach to problems of ality. One key difference is that a non- desertification and drought by promoting sustainable ” discrimination clause is added (in addition to the development models (51). objectives) in Article 34 on page 23, which states While it is cited that unsustainable development practices ‘ that: Nothing in this Agreement authorizes the have complex impacts, it is clear that the disturbances of use of measures to restrict or ban international concern are ‘markets’ for forestry products, rather than trade in, and in particular as they concern imports the forests themselves. The next paragraph also includes ’ of, and utilization of, timber and timber products. the sentiment that desertification has led to ‘asharpdrop’ This clause would make it unlawful for a Party to in productivity, such as a decline of forestry yields. Similar not trade with a nation because they know or classification of forests as a product is also clear in Article suspected it has illegal logging, which is similar 4, which states: to protections given to Parties of the World Trade Organization (WTO). “In the light of their respective situations, the affected As a whole, it appears that while these documents country Parties of the region may take account, inter alia, focus on timber trade, they aim to make it more of the following thematic issues in developing their national strategies for action to combat desertification sustainable to protect forests. Still, the document and/or mitigate the effects of drought, pursuant to article contains no language to suggest that forests should 5 of the Convention: be protected outright and for a common good, but instead includes language that suggests forest need (c) achieving food security and sustainable development and management of agricultural, livestock- to be sustainably managed so that they can continue rearing, forestry and multipurpose activities” (52). to be exported as a commodity for future generations. However, because forests are an important It is positive that calling for the sustainable management resource for everyone, having documents that at of forests presupposes that forests are not a constantly least aim for sustainable forestry management is renewable resource, and that they will run out if not used a realistic solution that could help prevent unneces- appropriately. Additionally, there are other parts of the sary forest loss. If followed, there could be reduc- document that seem to attribute value to forests beyond tions in forest waste and illegal logging, which could their relation to development. For example, in Article 2 on prevent additional forest loss. page 55, the document highlights the Mediterranean 6 J. M. SOMMER

Region as at risk for extensive forest coverage losses due REDD+, which also aims to foster conservation and the to frequent wildfires.” Similarly, a later paragraph urges sustainable management of forests. affected Parties to include in their national action plan REDD+ aims to incentivize conservation by giving measures to protect ‘against forest fires’ (Article 6 payments-for-results to those who have reduced their page 57). net carbon emissions by reducing forest loss. While there is much debate for the provisions and implementation of these payments, it is important to note that Kyoto Protocol to the United Nations framework the Norwegian International Climate and Forests convention on climate change AND Amendment to Initiative has disbursed over 1 billion USD to Brazil’s Annex B of the Kyoto Protocol to the United Amazon Fund, and the German REDD for Early Movers Nations framework convention on climate change Programme has also sent funds (Berg 2018). Though it is unclear whether these efforts can reduce forest loss, 193 nation states have ratified the protocol, while 30 it seems that, at the very least, putting value in the ratified the Amendment to Annex B of the protocol, environmental utility of forests should result in their which includes emission limitations or reduction com- protection. mitments. In this protocol, there are several calls to protect forests as carbon sinks and for reforestation. For example, in Article 2, page 2, ‘Protection and Protocol on strategic environmental assessment enhancement of sinks and reservoirs of greenhouse to the convention on environmental impact gases not controlled by the Montreal Protocol, taking assessment in a transboundary context into account its commitments under relevant international environmental agreements; promotion of sus- This protocol was ratified by 31 nations. It is important tainable forest management practices, afforestation to note that this agreement is considered as multilateral and reforestation.’ Similar arguments exist to measure or global by the Ecolex database used to collect the changes in greenhouse gas emissions through the use treaty data even though only members of the Economic of forests. Article 3, page 3 and states. Commission for Europe and those that have consulta- tive status with the Economic Commission for Europe “ The net changes in greenhouse gas emissions by were invited to be a part of the agreement. One of the sources and removals by sinks resulting from direct main activities the document calls for is a strategic human-induced land-use change and forestry activities, limited to afforestation, reforestation and defor- environmental assessment. Article 4 page 4 states: estation since 1990, measured as verifiable changes in carbon stocks in each commitment period, shall be “A strategic environmental assessment shall be used to meet the commitments under this Article of carried out for plans and programmes which are each Party included in Annex I. prepared for agriculture, forestry, fisheries, energy, industry including mining, transport, regional devel- In the first quantified emission limitation and reduc- opment, waste management, water management, tion commitment period, from 2008 to 2012, the telecommunications, tourism, town and country assigned amount for each Party included in Annex planning or land use, and which set the framework I shall be equal to the percentage inscribed for it in for future development consent for projects listed Annex B of its aggregate anthropogenic carbon diox- in annex I and any other project listed in annex II ide equivalent emissions of the greenhouse gases that requires an environmental impact assessment listed in Annex A in 1990, or the base year or period under national legislation.” determined in accordance with paragraph 5 above, multiplied by five. Those Parties included in Annex While forestry is considered an area of industry to be I for whom land-use change and forestry constituted evaluated for environmental harm, there are also men- a net source of greenhouse gas emissions in 1990 shall tions to create projects to limit ‘deforestation of large include in their 1990 emissions base year or period the ’ ‘ ff aggregate anthropogenic carbon dioxide equivalent areas (Annex I, page 10) and Initial a orestation and emissions by sources minus removals by sinks in deforestation for the purposes of conversion to 1990 from land-use change for the purposes of calcu- another type of land use’ (Annex II, page 10). lating their assigned amount.” Generally, according to Article 3, each Party is responsible for implementing appropriate actions as outlined The use of forests to tackle carbon emissions was later in the protocol, which states: formalized through a specific program. REDD, or reduce emissions from deforestation and forest degradation is “1. Each Party shall take the necessary legislative, reg- currently an important aspect of forestry governance. ulatory and other appropriate measures to implement Though the idea became formalized at the 13th UNFCCC the provisions of this Protocol within a clear, transpar- conference of the parties (COP), it began in the Kyoto ent framework. Protocol (Holloway and Giandomenico 2009). As men- 2. Each Party shall endeavour to ensure that officials tioned previously, the UNFCCC led to the development and authorities assist and provide guidance to the of the Green Climate Fund, which is a main funder of public in matters covered by this Protocol. INTERNATIONAL JOURNAL OF SUSTAINABLE DEVELOPMENT & WORLD ECOLOGY 7

3. Each Party shall provide for appropriate recognition illegal forest activities. In addition, unfortunately, of and support to associations, organizations or ‘much of the existing regional law enforcement and groups promoting environmental, including health, regulatory infrastructure is inadequate to prevent envir- protection in the context of this Protocol. onmental crimes’ (Rose 2014, p. 4). Many environmental 6. Each Party shall ensure that persons exercising their damages through corruption are transnational crimes rights in conformity with the provisions of this that span multiple nations, even beyond regions. This is Protocol shall not be penalized, persecuted or har- arguably why the multilateral agreement ITTA is so assed in any way for their involvement. This provision important, due to its focus on the elimination of unre- shall not affect the powers of national courts to award reasonable costs in judicial proceedings. gulated trade and recommending all parties agree (even those in transit) before trade commences. Still, the 7. Within the scope of the relevant provisions of this inability to discriminate against trade partners can Protocol, the public shall be able to exercise its rights allow illegal forestry trade to continue. While there are without discrimination as to citizenship, nationality or domicile and, in the case of a legal person, without some multilateral treaties that deal with this, such discrimination as to where it has its registered seat or actions are difficult to catch and prosecute due to an effective centre of its activities” (4). many factors, including the lack of state to state cooperation and an environmental international court. Additionally, the use of incentive mechanisms to Summary reduce carbon emissions and increase carbon sinks Given the above overview, it appears that multilateral with forests beginning in the Kyoto Protocol should environmental law has the potential to protect the be critical to forestry conservation. It is important that natural environment, especially vulnerable popula- multilateral agreements exist that do not just treat tions and forests (Daniel Bonilla-Maldonado 2015). forests as a natural resource or industry, but also For example, in Colombia, multilateral agreements a valuable resource to environmental health and well- such as the Framework Convention on Climate being. Still, there is not much in the documents to Change, were used to declare the General Forestry protect forests for the health and well-being of people. law (GFL) unconstitutional. According to Bonilla- Trees are therefore never really viewed as a ‘right’ to Maldonado (2015), ‘the Framework Convention on humans or similar. Below I review some of the major Climate Change and the Convention on Biological criticisms of these documents in an attempt to outline Diversity also recognize the differentiated obligations why these multilateral agreements may have no of the parties in the protection of the environment, impact on forests. I do so to explain why multilateral protect the principle of participatory democracy and agreements may have no impact or a detrimental promote the participation of vulnerable and histori- impact on forests. cally marginalized groups in the processes of making decisions that seek to prevent climate change and Potential limitations of existing treaties defend biodiversity’ (83). It is important to note that the Convention on Biological Diversity was not According to MacKenzie (2012), ‘the conjunction of included this analysis because it does not contain spe- legal, political and economic forces works in favor of cific provisions on forests. In short, multilateral law was continuing forest loss and degradation’ (115). Given necessary to protect the environment of minorities and this current climate, the targets of multilateral agree- indigenous peoples in Colombia. ments seem unachievable. For instance, we treat envir- Though the ITTA tends to treat forest as onmental products such as wood as plentiful and a commodity, it does so for sustainable management renewable resources, which can help explain the low- so that the resource can be used in the future. However, monetary value of forestry products on the global it does include important provisions on reducing envir- market, which reinforces forest destruction. When this onmental crimes against forests. Environmental crimes is coupled with increasing agricultural expansion into are seen as responsible for immense environmental forested area and the subsequent devaluation of agri- issues including forest loss. According to Rose (2014) cultural products, increased forest loss is imminent there are both environmental crimes to the living (Rudel et al. 2009; Austin 2010). Unfortunately, it is world (illegal exploitation such as poaching, harvesting common that forestry laws at the national level are during or beyond limits, and introducing biological accepted and enforced only after severe deforestation pests) and inanimate world (waste dumping, pollution, has taken place, such as in Indonesia (see Palmer 2001; habitat alterations and use of damaging/hazardous sub- Brockhaus et al. 2012; Abood et al. 2015). Therefore, stances), which can be considered illegal, though illegal MacKenzie (2012) may be correct when saying that, actions are growing faster than national, regional, and ‘countries in which deforestation rates have fallen especially international regulation. However, only the have achieved this outcome simply because they International Tropical Timber Agreement (2006) and have no forest left to destroy’ (115). In short, nations CIFOR covers illegal logging with policies to control with forests left are seemingly incapable of reducing 8 J. M. SOMMER forest loss on their own given poor national laws and ecosystems and endangered species, or that people international pressures to export forestry products or all around the world have lived in them and build their clear forests for agricultural expansion. Therefore, we livelihoods off of them for millions of years. need multilateral agreements and international pres- According to Ruis (2001), ‘Increased coordination’, sure to reduce forest loss. ‘avoidance of duplication’, ‘fostering synergies’, and However, many argue that multilateral action in ‘holistic approach’: these terms represent the new forestry conservation has been inadequate and incon- way of thinking in international environmental law. It sistent (MacKenzie 2012). Though multilateral law is is dangerous, however, and potentially detrimental to necessary for environmental health, the agenda has forests, to focus attention only on synergies among been ‘dominated’ by wealthy nations, which leaves existing treaties (11). Frankensteining existing docu- poorer nations at risk for continued environmental ments together also poses the risk of global forestry damage, poverty, and economic insecurity (Atapattu governance being too fragmented. This can be detri- and Gonzalez 2015, p. 1). For instance, various multi- mental to forests because less appreciated parts of national corporations from rich nations have violated their value may be missed or overlooked for a more human and environmental rights of those in poorer convenient benefit. Additionally, countermeasures nations, such as the Ogoniland case in Nigeria and may not be enough to ensure nations follow their Chevron in Ecuador (Gomez 2013; Atapattu and agreements. This can leave forests unprotected and Gonzalez 2015). Multilateral corporations can often at risk for forest loss (Shandra et al. 2008). Though evade scrutiny and sanctions for their environmentally there are incentive mechanisms built into the Kyoto damaging behavior because they ‘operate in the gray Protocol and REDD+, these are not yet fully developed zone between international law and national law’ and very few monetary disbursements have taken (Atapattu and Gonzalez 2015, p. 3). This is one of the place (Berg 2018). major fallbacks of multilateral treaties in protecting the Unfortunately, there is little political will to create natural environment. a new document that covers all aspects of the global Additionally, international trade and investment has good of forests, nor even one that can cover the gaps. drastically harmed the environment and people in the In fact, no legal framework has ever been adopted even Global South through structural adjustment lending, though according to research by MacKenzie (2012) indi- forestry and agricultural investment and so on through cate there are countless legal scholars that have tried to the World Bank, the International Monetary Fund, and tackle designing a multilateral agreement on forests (see the World Trade Organization (Bryant and Bailey 1997; MacKenzie 2012 for reference to the following legal ana- Peet 2003; Atapattu and Gonzalez 2015; Richardson lyses; Schally 1993;Skala-Kuhmann1996;Boer1997; 2015;Alametal.2015;Alam2015; Gonzalez 2015). Tarasofsky 1999;Lipschutz2001;Steiner2001, 2002; Unfortunately, these actions are also in the gray area in Davenport and Wood 2006; Glück et al. 2008;Boyd multilateral agreements (Atapattu and Gonzalez 2015). 2010; Srivastava 2011). Despite these exceptional outlines Even if corporations were held accountable, there are for a binding multilateral forestry treaty, a fragmented severe gaps and contradictions that exist within current global governance to protect forests remains. multilateral agreements concerning forests. For example, in ITTA agreements, forests are treated only as Hypothesis: Although there are many issues with multi- a commodity, whereas in the Kyoto Protocol they are lateral environmental treaties that focus on forests, seen as necessary for the world to mitigate and adapt to there are several important initiatives within the agree- climate change. These two roles are contradictory to ments that should be related to reduced forest loss. each other, as the former calls for ‘sustainable forestry Therefore, I hypothesize that multilateral treaties that management’ so there are still forests left to trade, and focus on forests is associated with decreased forest loss. the latter understands, at least in part, the critical role forests play in ecosystem conservation and carbon sequestering. Methods and data If we ignore the inherent contradictions of the existing documents, and try to ‘harmonize,’ or ‘foster Following previous research on forest loss, I use synergy’ between the documents, we would still end Ordinary Least Squares (OLS) regression to analyze for- up with huge gaps (Ruis 2001). The document would est loss from 2001–2014. Forest loss is measured by only address the decidedly ‘useful’ aspects of forests, dividing the change in forest loss from 2001–2014 by which (after careful analysis of each document) only total forest area in 2000. These data are from Global fuel capitalism or try to dig the world out of the envir- Forest Watch from the World Resources Institute. onmental degradation caused by capitalism. No docu- Collection and measurement of data follow Rudel ment reviewed, to my knowledge appreciates the (2017). Due to measurement, the forest loss data are inherent value of forests, that the majority of our med- not suitable for longitudinal comparisons. However, this icines come from them, that they are home to is consistent with previous studies in this area and these INTERNATIONAL JOURNAL OF SUSTAINABLE DEVELOPMENT & WORLD ECOLOGY 9

Table 2. Descriptive statistics. Variables Description (Source) Mean (Std. Dev) Range Dependent Variable Forest Loss (2001–2014) Forest losses from 2001 to 2014 (75% canopy density level) divided by total .106 (0.145) 0 – .693 forest area in 2000 (World Resources Institute) Independent Variables Democracy Composition The average of political competition index (measures the percentage of votes 16.590 (12.339) 0 – 45.6 gained by smaller parties in parliamentary and presidential elections) and the political participation variable (measures the percentage of the population that voted in parliamentary and presidential elections) (Vanhanen 2014) Multilateral Treaty Ratifications Number of environmental multilateral treaty ratifications that include forests 4.013 (1.238) 0 – 8 (Ecolex.org) Environmental NGOs (ln) Number of ties to environmental non-governmental organizations (sample 3.002 (1.002) 0 – 4.780 developed by Schofer and Hironaka 2005) Agricultural Land Area Agricultural land (% of land area) (World Bank) 40.439 (21.641) .564–85.465 Population Density (ln) Population density (people per sq. km of land area) (World Bank) 4.161 (1.329) .933–8.702 GDP per capita (ln) Gross domestic product per capita (current US dollars) (World Bank) 7.609 (1.583) 4.519–10.746 Forestry Exports as a % of GDP Forestry Exports (USD 1000) (% of GDP) (Food and Agricultural Organization) 159.406 (622.6582) 0 – 6985.782 Total Population Growth Total Population Percentage Change (1990–2000) (World Bank) .170 (.140) −.162 – .513 Rural Population Growth Rural Population Percentage Change (1990–2000) (World Bank) .095 (.165) −.228-.725 Urban Population Growth Urban Population Percentage Change (1990–2000) (World Bank) .277 (.268) −.168–2.045 are the best data currently available to measure this Sample phenomenon. The OLS model is as follows: The sample includes 155 nations,1 broken down into yi ¼ a þ b1X1 þ b2X2 ...þ bkXk þ ei low- and middle-income nations and high-income where, nations using the World Bank Atlas method to see if ff y = dependent variable for each country, results di er by level of income. Through this method, i fi a = the constant, low- and middle-income nations are de ned as having a gross-national income (GNI) of 12,475 USD or less in b1 to bk = unstandardized coefficients for each independent variables, 2016 and high-income nations have more (Shandra, Rademacher, and Coburn 2016; World Bank 2015). xk = independent variables for each country, and ei = error term for each county. I check several regression assumptions including Main independent variable multicollinearity, linearity, outliers, heteroskedasticity, and selection bias. Multicollinearity does not seem to The main independent variable, multilateral environ- be biasing the results as variance inflation factors mental treaty ratifications that include forests, was scores are below 2.5 (Allison 1999). I take the natural collected from Ecolex.org. Using this database, logarithm of variables when there are issues with line- I selected ‘treaties’ then clicked the boxes for ‘multi- arity and report it in Table 2. There appear to be no lateral’ and ‘global’ to include only globally relevant outliers that bias the findings given robust regression agreements to avoid potential issues with selection models (using an mm estimator) (see Frey Dietz and bias. I used the keywords ‘forest’ and ‘deforestation’ Kalof 1987 and Sommer 2018). I deal with potential to find relevant treaties in conjunction with previous issues of heteroskedasticity by reporting robust research (Ruis 2001; MacKenzie 2012). After removing standard errors. To test for selection bias, I used instru- duplicate treaties and those without any ratifications mental variable regression models. However, because (one treaty had no ratifications), I arrived at a sample of it was difficult to find a proper instrument (i.e. 8 treaties from 1992–2006. Ratifications by nations a variable that is uncorrelated with forest loss but were only counted if they were before 2014 to avoid correlated with treaty ratifications) that passed the potential issues with simultaneity bias (Shandra et al. Anderson canonical correlation, Sargan chi-square, 2016). The resulting variable is a count of the number and Cragg-Donald F-statistics (Woolridge 2015; Baum of treaty ratifications in each given country (between 0 2006) the results were inconclusive. The closest to and 8). I ran robustness checks with ratifications only proper instrumental variables were HIV prevalence until 2005 as well, because it is common that the and foreign direct investment (stock) as a percentage effects of treaties would not be seen for a few years of GDP. However, there is little theoretical reason for after their ratification. The results were substantively them to be instrumented. Therefore, because there are similar, so the data represent ratifications from potential problems with selection bias, care must be 1992–2014. In short, these data are intended to cap- taken when generalizing from this analysis. ture national ratifications of multilateral treaties that 10 J. M. SOMMER include forests in an effort to measure global govern- governmental environmental organizations (available ance in forestry. from Schofer and Hironaka 2005), forestry exports as It is important to note that including every treaty a percentage of gross domestic product (available from simultaneously in the models was chosen over analyz- the Food and Agricultural Organizations (FAO)), and ing the effect of one treaty at a time because theore- gross domestic product per capita current USD, agricul- tically, the impact of all relevant multilateral tural land area, population density, total population agreements as a whole should impact forests more growth, rural population growth, and urban population than individual agreements. This is because global growth (all available from the World Bank Open forestry governance is the topic of interest rather Database). Table 2 includes descriptive statistics and than an individual treaty. The goal is to make claims definitions for all included variables. Appendix 1 about the current global situation regarding forests to includes the sample and appendix 2 includes give recommendations on how we as a global com- a bivariate correlation matrix. munity can move forward with forestry agendas. However, if agreements are ‘contradictory’ to one another then they may cancel each other out, and Results therefore it would be important to separate them out In Table 3, equations (3.1) and (3.2) contain the ordin- in such an analysis. Despite this potential problem, ary least squares regression estimates of multilateral I include all treaties for theoretical reasons. environmental treaty ratifications that include forests on of forest loss for low- and middle-income nations and equations (3.3) and (3.4) include all nations. The Control variables first number presented is the unstandardized coeffi- Based on previous research on forest loss (Shandra et al. cient, the second is the standardized coefficient, and 2016;Sommer2018; Jorgenson and Burns 2007)Icontrol the third number in parentheses is the robust stan- for the following factors: democracy composition (avail- dard error. I report one-tailed tests because of the able from Vanhanen 2014), the number of ties to non- directional nature of the hypotheses (Shandra et al.

Table 3. Ordinary least squares regression estimates of treaty ratifications on forest loss, 2001–2014. Equation (3.1) Equation (3.2) Equation (3.3) Equation (3.4) Independent Variables Multilateral Treaty Ratifications, 1993–2014 −.034* −.034* −.024* −.023* −.196 −.196 −.209 −.193 (.016) −0.017 (.011) (.011) Democracy Composition, 2000 −.001 .001 .001 .001 −.001 .008 .034 .061 (.004) (.017) (.001) (.001) EINGOs, 2000 .039** .034* .043*** .039*** .203 .175 .297 .268 (.014) (.015) (.011) (.011) Agricultural Land Area, 2000 .002** .002** .001** .001** .251 .252 .194 .211 (.001) (.001) (.001) (.001) Population Density, 2000 −.040*** −.036** −.028*** −.026** −.300 −.267 −.260 −.237 (.012) (.013) (.009) (.009) GDP per capita, 2000 −.006 −.019 −.006 −.010 −.037 −.124 −.067 −.009 (.014) (.014) (.010) (.010) Forestry Exports (% of GDP) .001 .001 −.001 −.001 .039 .036 −.014 −.009 (.001) (.001) (.001) (.001) Total Population Growth Rate, 1990–2000 .369** .298** .296 .289 (.125) (.097) Urban Population Growth Rate, 1990–2000 .057 −.023 .061 −.011 .097 −.038 .113 −.020 (.065) (.049) (.060) (.047) Rural Population Growth Rate, 1990–2000 .272** .170* .249 .194 (.104) (.073) R-Squared .285 .293 .228 .241 Number of Countries 110 110 155 155 Highest Variance Inflation Factor Score 1.69 1.85 2.30 2.33 Mean Variance Inflation Factor Score 1.35 1.39 1.61 1.75 Breusch-Pagan Test 37.64*** 41.70*** 66.87*** 71.80*** Notes: a) * indicates p < .05, ** indicates p < .01, and *** indicates p < .001 for a one-tailed test. b) The first number is the unstandardized coefficient, the second number is the standardized coefficient, and the third number in parentheses is the robust standard error. INTERNATIONAL JOURNAL OF SUSTAINABLE DEVELOPMENT & WORLD ECOLOGY 11

2016). In every equation, I include democracy com- it appears that multilateral environmental treaty position, the number of ties to non-governmental ratifications that include forests have the smallest environmental organizations, forestry exports as magnitudes (.193-.210), followed by environmental a percentage of gross domestic product, gross non-governmental organizations (.175-.297), rural domestic product per capita current USD, agricultural population growth (.193-.249), and agricultural land area, population density, and urban population land area (.194-.252). Total population growth and growth. Odd numbered equations include rural population density have the largest effect on forest population growth, and even numbered equations loss (.288-.296) and (.237-.300), respectively. Why do include total population growth. multilateral environmental treaty ratifications that Let me begin by discussing the main effects of include forests have the smallest effects? I discuss multilateral environmental treaty ratifications that why this may be the case in the following discus- include forests on forest loss. In all equations in sion section. Table 3,Ifind that the coefficients that represent multilateral environmental treaty ratifications that Discussion include forests reach levels of statistical significance. This suggests that higher levels of multilateral envir- Based on the above findings, it appears that global onmental treaty ratifications are associated with forestry governance, measured via treaty ratification, lower levels of forest loss. Put differently, the multi- does impact forest loss. However, the impact is rela- lateral agreements seem to be reducing forest loss. tively small. I argue that this may be because global Additionally, I find a number of other factors are forestry governance is too fragmented, lacks assess- associated with forest loss. First, I find that the coeffi- ment, and has difficulty to achieve targets. Going for- cients that represent agricultural land area are positive ward, it is important to support ongoing initiatives and significant in every equation. This suggests that from multilateral forestry treaties. However, they may higher levels of agricultural land area correspond with not have a sizable impact on forests unless the treaty more forest loss, which is most likely because of agri- provisions are adhered to by nations, there are suffi- cultural expansion (Rudel et al. 2009). Second, I find cient sanctions, incentives or other enforcement that the coefficients that represent population density mechanisms, and there is a global convention or for- are negative and significant in every equation, sug- estry treaty that either fills the gaps of preexisting gesting that factors such as migration to urban areas treaties or protects all forests. Because the effect is away from agricultural farms and forests may reduce small, we need to improve global forestry governance, forest loss (Jorgenson and Burns 2007; Rudel et al. so they are more effective at doing what they arguably 2016). Third, I find that the coefficients that represent were designed to do; protect the environment. population growth and rural population growth are Additionally, a multilateral treaty and convention for associated with increased forest loss. This suggests forest protection on its own may be important to that resource scarcity resulting from population improving global forestry governance to increase the growth may contribute to forest loss. However, rural impact of global forestry governance on forest loss population growth, rather than urban population cross-nationally. Although possible and necessary, har- growth seems to contribute to forest loss. This may monization or synergy of existing documents may not be due to the pressures rural populations put on for- be the more effective way to reduce forest loss (Ruis ests for additional land area (Jorgenson and Burns 2001). Therefore, rather than stretching treaties to 2007). Lastly, I find that the coefficients that represent expand coverage, it is important to call for a binding the number of environmental non-governmental orga- document for the protection, management and sus- nization ties are associated with higher levels of forest tainable development of all forests. loss. Though this contradicts some previous research While the majority of the treaties call for ‘sustainable (Shandra 2007), it is possible that the number of NGOs forestry management’ (SFM) it is unclear whether this are increasing to deal with increasing forest loss, rather will actually be sustainable solution going forward. It than the NGOs are increasing forest loss (Schofer and seems like a good concept, but it also seems like Hironaka 2005). There are also non-significant findings. deforestation with extra steps (Tacconi et al. 2003). It First, I find that the coefficients that represent GDP per also opens up a lot of room for counterfeit documents capita and forestry exports fail to reach levels of statis- and other types of corruption that plague the forestry tical significance. Second, I find that democracy is not sector (see Sundstorm 2016 for a review). Also, national associated with forest loss, despite previous research discrimination against traders that do not use SFM or (Shandra 2007; Payne 1995; Li and Reuveny 2006). The get certified does not seem possible given the WTO, coefficients that represent these variables fail to reach GATT, and ITTA agreements (Virtanen and Palmujoki levels of statistical significance. 2002). Though it is possible for this discrimination to Focusing just on the standardized coefficients for go unnoticed at the state level, which was the case for each statistically significant variable in the models, California in the United States (Virtanen and Palmujoki 12 J. M. SOMMER

2002). In summary, it seems pertinent to recommend if donor nations do not make it a priority (Hein et al. changing consumer behavior to shift the market away 2018). from unsustainable forestry goods. For instance, IKEA Fortunately, the recent United Nations strategic plan uses a private company who DNA tags forestry pro- for forests, 2017–2030 takes a completely holistic ducts to ensure they are not protected species or approach to forestry. This plan treats forests as neces- logged illegally. This is not mandated by any govern- sary for survival for humans and for the environment. ment, but instead a decision from a conscientious For instance, on page 1 the document states that the UN company to only sell sustainable products (Ikea envisions: ‘A world where all types of forests and trees 2018). This should not violate free trade agreements, outside forests are sustainably managed, contribute to and also may shift in interpretations or hopefully sustainable development and provide economic, social, alterations of existing documents. environmental and cultural benefits for present and future generations.’ Still, this document is not a treaty or agreement, and has as much enforcement potential Conclusion as the UN’s Sustainable Development Goals (SDG). This study finds that multilateral environmental However, the fact that the SDGs are not legally binding treaties that include protections for forests are asso- is a little relevance because state progress on achieving ciated with decreased forest loss from 2001 to 2014. the goals is monitored and states are expected to report This research also finds that agricultural land area, on their progress regularly, which therefore may have rural population growth, and population density are better compliance than several of the treaties I discuss. associated with forest loss. Based on the main find- Therefore, this approach to forestry may be taken ser- ings from this research, it appears that environmen- iously by nations, but more research is needed to deter- tallawsthatincludeforestsareeffective at reducing mine if this is the case. forest loss, though the effect is relatively small. In sum, the findings of this study raise fundamental There are several potential factors that may have questions about global governance as it relates to forests. reduced the effectiveness of these treaties on forest This research concerns the importance of investigating loss, including a lack of unification of multilateral how existing multilateral agreements that include forests environmental laws that address forest, the absence may impact forests, and what can be done to improve of a global forestry convention, contradictory objec- their effectiveness. Going forward, we as researchers and tives, unachievable goals, and a general lack of practitioners should consider how other types of global political will. It appears that while there are several governance relate to the natural environment. important programs and initiatives from global forestry governance treaties that reduce forest loss, more needs to be done. Note There are other recommendations that emerge 1. The nations in the sample include: Albania; Algeria; from the results of the study. Including mandates for Andorra; Angola; Antigua and Barbuda; Argentina; sustainable developed forests, such as the Forest Armenia; Aruba; Australia; Azerbaijan; Bahamas; Stewardship Council (FSC), may help reduce agricul- Bangladesh; Barbados; Belarus; Belgium; Belize; Benin; Bhutan; Bolivia; Bosnia and Herzegovina; Brazil; Brunei tural and population pressures found to be associated Darussalam; Bulgaria; Burundi; Cabo Verde; Cambodia; with forest loss. Another solution may be to focus more Cameroon; Canada; Central African Republic; Chad; on forestry in Convention on International Trade in China; Colombia; Comoros; Congo, Dem. Rep.; Congo, Endangered Species of Wild Fauna and Flora (CITES). Rep.; Costa Rica; Croatia; Cuba; Cyprus; Czech Republic; CITES includes provisions for enforceable sanctions Denmark; Dominica; Dominican Republic; Ecuador; and has proven successful in making sure some species Egypt, Arab Rep.; El Salvador; Equatorial Guinea; ffi Estonia; Ethiopia; Fiji; Finland; France; Gabon; Gambia; did not go extinct. However, it is di cult to enforce as Georgia; Germany; Ghana; Greece; Grenada; Guatemala; is, and adding additional species to the list would make Guinea; Guinea-Bissau; Guyana; Haiti; Honduras; it even more difficult to enforce beyond a few endan- Hungary; India; Indonesia; Iran, Islamic Rep.; Ireland; gered tree species. Israel; Italy; Jamaica; Japan; Jordan; Kazakhstan; Kenya; Going forward, the Paris Agreement (2015) may also Korea, Rep.; Kyrgyz Republic; Lao PDR; Latvia; Lebanon; Liberia; Libya; Lithuania; Luxembourg; Macedonia, FYR; be effective at reducing forest loss in the coming dec- Madagascar; Malawi; Malaysia; Maldives; Mali; Malta; ade. The Paris Agreement includes REDD+ as critical to Mauritius; Mexico; Moldova; Mongolia; Morocco; adapting and mitigating climate change issues. This Mozambique; Namibia; Nepal; Netherlands; New treaty improves upon the Kyoto Protocol by including Zealand; Nicaragua; Nigeria; Norway; Pakistan; conservation efforts for developing nations, though it Panama; Papua New Guinea; Paraguay; Peru; still focuses a lot on preserving forests just to reduce Philippines; Poland; Portugal; Romania; Russian fi Federation; Rwanda; Sao Tome and Principe; Senegal; emissions rather than other bene ts of forests. Seychelles; Sierra Leone; Singapore; Slovak Republic; Additionally, finance for REDD+, which is largely Slovenia; Solomon Islands; South Africa; Spain; Sri through the Green Climate Fund may not be available Lanka; St. Kitts and Nevis; St. Lucia; Sudan; Suriname; INTERNATIONAL JOURNAL OF SUSTAINABLE DEVELOPMENT & WORLD ECOLOGY 13

Swaziland; Sweden; Switzerland; Syrian Arab Republic; Biermann F. 2017. A world environment organization: solu- Tajikistan; Tanzania; Thailand; Togo; Trinidad and tion or threat for effective international environmental Tobago; Tunisia; Turkey; Turkmenistan; Uganda; governance?. Routledge. Ukraine; United Kingdom; United States; Uruguay; Biermann F, Pattberg P, Van Asselt H, Zelli F. 2009. The Uzbekistan; Vanuatu; Venezuela; Vietnam; Zambia; fragmentation of global governance architectures: Zimbabwe. a framework for analysis. Global Environ Politics. 9 (4):14–40. Boer B. 1997. Commentary: developments in international Acknowledgments environmental law relating to forests. Environ Plann Law J. 14:378. I want to thank the Center for United Nations Constitutional Bonilla-Maldonado D. 2015. International law, cultural Research (CUNCR) for providing me with the resident fellow- diversity, and the environment: the case of the gen- ship necessary to conduct this research. I also want to per- eral forestry law in Colombia. Int Law. 131(27):65–107. sonally thank John Shandra, Otto Spijkers, Shahr-yar Sharei, Boyd W. 2010. Ways of seeing in environmental law: how and Andrew Hargrove for their help with comments and deforestation became an object of climate governance. suggestions. Ecol Law Q. 37(3):843. Canadian Council on Inter- national Law (ed.), Global Forests and International Environmental Law (Kluwer Law International, 1996). Disclosure statement Brockhaus M, Obidzinski K, Dermawan A, Laumonier Y, Luttrell C. 2012. An overview of forest and land allocation No potential conflict of interest was reported by the author. policies in Indonesia: is the current framework su fficient to meet the needs of REDD+?. Forest Policy Econ. 18:30–37. Bryant RL, Bailey S. 1997. Third world political ecology Funding Routledge. New York: Routledge. This work was supported by the Center for United Nations Canan P, Reichman N. 2002. Ozone connections: expert net- ffi Constitutional Research (CUNCR) [resident fellowship]. works in global environment governance. She eld (UK): Greenleaf. Davenport DS, Wood P. 2006. Finding the way forward for References the international arrangement on forests: UNFF-5, -6, -7. Rev Eur Commun Int Environ Law. 15(3):316. Abood SA, Lee JSH, Burivalova Z, Garcia-Ulloa J, Koh LP. 2015. Dietz TR, Frey S, Kalof L. 1987. Estimation with cross-national Relative contributions of the logging, fiber, oil palm, and data: robust and nonparametric methods. American mining industries to forest loss in Indonesia. Conserv Lett. Sociological Review. 52(3):380–390. 8(1):58–67. Elliott L. 2017. Introduction. In: Kirton, Trebilcok, editors. Alam S. 2015. Trade and the environment: perspectives from Hard choices, soft law: voluntary standards in global the Global South. Chapter 14. In: Alam S, Atapattu S, trade, environment and social governance. New York: Gonzalez CG, Razzaque J editors. International environ- Routledge. mental law and the global south. Cambridge, England: Finkelstein LS. 1995. What is global governance. Global Cambridge University Press; p. 297–318. Governance. 1:367. Alam S, Atapattu S, Gonzalez CG, Razzaque J, Eds. 2015. Glück P, Tarasoffsky RG, Byron RN, Tikkanen I. 2008. Options International environmental law and the global south. for strengthening the international legal regime for forests Cambridge University Press. (Euro- pean Forest Institute, 1997); K. Kunzmann, ‘The non- Allison PD. 1999. Multiple regression: a primer. Newbury Park, legally binding instrument on sustainable management of CA: Pine Forge Press. all types of forests – towards a legal regime for sustainable Atapattu S, Gonzalez CG. 2015. Introduction: the North-South forest management. Ger Law Rev. 9(8):981. Divide in international enviornmental law: framing the Gomez MA. 2013. The global chase: seeking recognition and issues. In: Alam S, Atapattu S, Gonzalez CG, Razzaque J enforcement of Lago Agrio judgement outside of Ecuador. editors. International environmental law and the Global Stanf J Complex Litig. 27:65–108., South. Cambridge, England: Cambridge University Press; Gonzalez CG. 2015. Food justice: an environmental justice p. 1–22. critique of the global food system. Chapter 19. In: Alam S, Austin KF. 2010. Soybean exports and deforestation from a Atapattu S, Gonzalez CG, Razzaque J editors. International world-systems perspective: a cross-national investigation environmental law and the Global South. Cambridge of comparative disadvantage. Social Q. 51(3):511–536. University Press. Baum CF. 2006. An introduction to modern econometrics Haas PM. 2002. UN conferences and constructivist govern- using stata. College Station, TX: Stata press. ance of the environment. Global Governance. 8:73. Berg J. 2018. Norway fulfills 1 billion dollar commitment to Haas PM. 2004. Addressing the global governance deficit. Brazil in recognition. Ministry of Climate and Environment; Global Environmental Politics. 4(4):1–15. [accessed 2018 Jan 9]. https://www.regjeringen.no/en/ Holloway V, Giandomenico E. 2009. The History of REDD aktuelt/norway-fulfils-commitment-to-brazil/id2439969/ Policy. https://redd.unfccc.int/uploads/2_164_redd_ Biermann F. 2002. Strengthening green global governance in 20091216_carbon_planet_the_history_of_redd_carbon_ a disparate world society. Int Environ Agreements. planet.pdf 2:297–315. Ikea. 2018. Forests and Wood. https://www.ikea.com/ms/en_ Biermann F. 2006. Global governance and the environment. AU/about_ikea/our_responsibility/forestry_and_wood/ In: Betsill MM, Hochstetler K, Stevis D, editors. Palgrave index.html advances in international environmental politics(pp. Jorgenson AK, Burns TJ. 2007.Effects of rural and urban London: Palgrave Macmillan; p. 237–261. population dynamics and national development on 14 J. M. SOMMER

deforestation in less-developed countries, 1990–2000. Schofer E and Hironaka A. 2005. The effects of world society Sociological Inquiry. 77(3):460–482. on environmental protection outcomes. Social Forces. 84 Karns M, Karen M. 2004. The politics and processes of (1):25–47. global governance. Colorado: Lynne Rienner Publishers. Scott J. 1998. Seeing like a state: how certain schemes to Kirton JJ, Trebilcock MJ. 2017. Hard choices, soft law: volun- improve the human condition have failed. New Haven: tary standards in global trade, environment and social Yale University Press. governance. Routledge. Shandra JM. 2007. Economic dependency, repression, and Lanly JP. 1992. Forestry issues at the united nations confer- deforestation: a quantitative, cross-national analysis. ence on environment and development. Unasylva. 43 Sociological Inquiry. 77(4):543–571. (171):61. Shandra JM, Rademacher H, Coburn C. 2016. The world bank Li Q, Reuveny R. 2006. Democracy and environmental and organized hypocrisy? A cross-national analysis of degradation. International Studies Quarterly. 50 structural adjustment and forest loss. Environmental (4):935–56. Sociology. 2(2):192–207. Lipschutz RD. 2001. Why is there no international for- Shandra JM, Shandra CL, London B. 2008. Women, non-gov- estry law? An examination of international forestry ernmental organizations, and deforestation: a cross- regulation, both public and private. UCLA J Environ national Study. Population and Environment. 30(1-2):48– Mental Law Policy. 19(1):153. 72. Mackenzie CP. 2012. Future prospects for international Skala-Kuhmann A. 1996. Legal instruments to enhance the con- forest law. International Forestry Review. 14(2):249– servation and sustainable management of forest resources at 257. the international level. German Federal Ministry for Economic Palmer C. 2001. The extent and causes of illegal logging: an Cooperation and Development and GTZ, 1996. Berlin. analysis of a major cause of tropical deforestation in Sommer JM. 2017. Grand and petty corruption: a cross- Indonesia. national analysis of forest loss in low-and middle-income Palmer G. 1992. The earth summit: what went wrong at nations. Environmental Sociology. 3(4):414–426. rio? Washington Univ Law Q. 70(4):1005. Sommer JM. 2018. State spending and governance: a cross- ParkJ,ConcaK,FingerM,Eds.2008.Thecrisisofglobal national analysis of forest loss in developing nations. environmental governance: towards a new political Sociological Inquiry. 88(4):696–723. economy of sustainability. Routledge. Srivastava N. 2011. Changing dynamics of forest regulation: Payne RA. 1995. Freedom and the environment. Journal coming full circle? Rev Eur Commun Int Environ Law. 20 of Democracy. 6(3):41–55. (2):113. Peet R. 2003. The unholy trinity: the international mone- Steiner M. 2001. After a decade of global forest negotiations, tary fund, world bank, and world trade organization. where are we now? Rev Eur Commun Int Environ Law. 10 London: Zed Books. (1):98. Peluso NL. 1992. Rich forests, poor people: resource Steiner M. 2002. The journey from rio to johannesburg: control and resistance in java.UnivofCaliforniaPress. ten years of forest Ne“gotiations,tenyearsofsuccess Richardson BJ. 2015. International environmental law and failures. Golden Gate Univ Law Rev. 32(4) and sovereign wealth funds. Chapter 17. In: Alam S, :629. Atapattu S, Gonzalez CG, Razzaque J editors. Tacconi, Luca, Boscolo M, Brack D. 2003. Illegal Logging. International environmental law and the Global http://www.cifor.org/publications/pdf_files/events/Illegal- South. Cambridge: Cambridge University Press; p. logging.pdf 356–379. Tarasofsky RG. 1999. The international forests regime: legal Rose G, (Ed.). 2014. Following the proceeds of environmental and policy issues (IUCN/WWF, 1995). In: Tarasofsky RG, crime: fish, forests and Filthy Lucre. Routledge. editor. Assessing the International Forest Regime (IUCN, Rudel TK. 2017. The dynamics of deforestation in the wet and 1999). Gland, Switzerland: IUCN. dry tropics: a comparison with policy implications. Forests. Vanhanen T. 2014. Global inequality as a consequence of 8:108. human diversity. London, UK: Ulster Institute for Social Rudel TK, Schneider L, Uriarte M, Turner BL, DeFries R, Research. Lawrence D, . . . Birkenholtz T. 2009. Agricultural intensifi- Virtanen P, Palmujoki E. 2002. Sustainable Forest Management cation and changes in cultivated areas, 1970–2005. Proc Through Multilateral Environmental Agreements and Natl Acad Sci. 106(49):20675–20680. Market-based Mechanisms: Final Report of a Study on Rudel TK, Sloan S, Chazdon R, et al. 2016. The drivers of tree International Efforts to Promote the Sustainable cover expansion: global, temperate, and tropical zone Management of Tropical Forests in Developing Countries. analyses. Land Use Policy. 58:502–513. Ministry for Foreign Affairs, Department for International Ruis BM. 2001. No forest convention but ten tree treaties. Development Cooperation. Unasylva. 52(206):3–13. World Bank. 2015. World development indicators. Schally H. 1993. Forests: toward an international legal regime. Washington, DC: World Bank. Yearbook Int Environ Tal Law. 4:30. Young OR, Ed. 1997. Global governance: drawing Schama S. 1995. Landscape and memory. London: Harper insights from the environmental experience. Collins Publishers. Cambridge, MA: MIT press. INTERNATIONAL JOURNAL OF SUSTAINABLE DEVELOPMENT & WORLD ECOLOGY 15

Appendix 1

1). The nations in the sample include: Albania; Algeria; Andorra; Angola; Antigua and Barbuda; Argentina; Armenia; Aruba; Australia; Azerbaijan; Bahamas; Bangladesh; Barbados; Belarus; Belgium; Belize; Benin; Bhutan; Bolivia; Bosnia and Herzegovina; Brazil; Brunei Darussalam; Bulgaria; Burundi; Cabo Verde; Cambodia; Cameroon; Canada; Central African Republic; Chad; China; Colombia; Comoros; Congo, Dem. Rep.; Congo, Rep.; Costa Rica; Croatia; Cuba; Cyprus; Czech Republic; Denmark; Dominica; Dominican Republic; Ecuador; Egypt, Arab Rep.; El Salvador; Equatorial Guinea; Estonia; Ethiopia; Fiji; Finland; France; Gabon; Gambia; Georgia; Germany; Ghana; Greece; Grenada; Guatemala; Guinea; Guinea-Bissau; Guyana; Haiti; Honduras; Hungary; India; Indonesia; Iran, Islamic Rep.; Ireland; Israel; Italy; Jamaica; Japan; Jordan; Kazakhstan; Kenya; Korea, Rep.; Kyrgyz Republic; Lao PDR; Latvia; Lebanon; Liberia; Libya; Lithuania; Luxembourg; Macedonia, FYR; Madagascar; Malawi; Malaysia; Maldives; Mali; Malta; Mauritius; Mexico; Moldova; Mongolia; Morocco; Mozambique; Namibia; Nepal; Netherlands; New Zealand; Nicaragua; Nigeria; Norway; Pakistan; Panama; Papua New Guinea; Paraguay; Peru; Philippines; Poland; Portugal; Romania; Russian Federation; Rwanda; Sao Tome and Principe; Senegal; Seychelles; Sierra Leone; Singapore; Slovak Republic; Slovenia; Solomon Islands; South Africa; Spain; Sri Lanka; St. Kitts and Nevis; St. Lucia; Sudan; Suriname; Swaziland; Sweden; Switzerland; Syrian Arab Republic; Tajikistan; Tanzania; Thailand; Togo; Trinidad and Tobago; Tunisia; Turkey; Turkmenistan; Uganda; Ukraine; United Kingdom; United States; Uruguay; Uzbekistan; Vanuatu; Venezuela; Vietnam; Zambia; Zimbabwe. 2).

Appendix 2. Bivariate correlation matrix for deforestation analysis (N = 155). (1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (1) Forest Loss, 2001–2014 (ln) 1.000 (2) Democracy Composition, 2000 −.167 1.000 (3) Multilateral Tree Treaties, 1993–2014 −.217 .201 1.000 (4) EINGOs, 2000 .132 .135 .289 1.000 (5) Agricultural Land Area, 2000 .162 −.033 −.109 .061 1.000 (6) Population Density, 2000 −.213 .055 −.024 .129 .319 1.000 (7) GDP per capita, 2000 −.252 .359 .113 .092 −.176 .020 (8) Forestry Exports (% of GDP), 2000 −.025 .012 .260 .171 −.089 .053 1.000 (9) Total Population Growth, 1990–2000 .367 −.363 −.184 .144 −.193 −.179 −.027 1.000 (10) Urban Population Growth, 1990–2000 .220 −.395 −.186 .159 −.046 .038 −.295 −.013 1.000 (11) Rural Population Growth, 1990–2000 .327 −.315 −.191 −.037 −.103 −.117 −.338 .092 .587 1.000 −.513 −.081 .740 .245 1.000