Economic and environmental upgrading after a policy reform The case of timber value chain in Myanmar

.


Introduction
A rapid rise of global trade in tropical timber and wood products in recent years has seen higher than ever participation of smallholders in global value chains (GVCs).It has also revealed a number of sustainability challenges in timber-producing countries where governments grapple with the dual aims of forest protection and increasing value added from forestry.Illegal timber logging and trade as some of the main causes of deforestation and forest degradation figure prominently in international sustainability discourse.Three major timber-importers, the EU, the US and Australia, set the foundation for the 'timber legality regime' by introducing legal acts that prohibit imports of illegally sourced timber and wood products (Leipold and Winkel, 2016).Consequently, curbing illegal logging and trade to meet international timber legality requirements has become a top priority of forest policies in several tropical countries (Dooley and Ozinga, 2011;Newton et al., 2013;Carodenuto and Cerutti, 2014;Rutt et al., 2018;Furumo and Lambin, 2020).Against this backdrop, we ask to what extent the international timber legality requirements adapted to local national policies enable or constrain the economic and environmental upgrading outcomes in the Myanmar timber value chain.
Broader literature on forest governance that spans several continents (e.g., Adams et al., 2020) provides various accounts of differential impacts of legality verification systems primarily for traditional loggers (Mejia et al., 2015;Waldhoff and Vidal, 2015;Cromberg et al., 2022) and for forest-dependent local communities (Grieg-Gran et al., 2005;de Jong et al., 2006;Ramcilovic-Suominen et al., 2010;Nurrochmat et al., 2016;Tsujino et al., 2016;Myers et al., 2017;Fatem et al., 2018).Overdevest and Zeitlin (2018) find that international forest legality initiatives contribute to promoting civil society activism and holding public authorities accountable for addressing environmental challenges.In addition, they offer opportunity for business expansion, increased earnings from exports and the valorisation of investments (Carodenuto and Cerutti, 2014;Thuy et al., 2021).Others focus on factors limiting legal compliance and discover asymmetries between value chain actors related to timber legality (Smith et al., 2006;McDermott et al., 2015;Susilawati et al., 2019;Acheampong and Maryudi, 2020;Maryudi et al., 2020Maryudi et al., , 2021)).While differential adaptation mechanisms and perceptions on legality verification are assessed, consequences for economic and environmental upgrading have not been considered in the above-outlined forest governance literature.We contribute to filling this gap by drawing on the global value chain (GVC) approach, which has so far not provided extensive evidence on the role of policies governing the conditions of management, exploitation and trade of natural resources.
GVC studies have until now focused mainly on industrialization, trade, investment and innovation policies (Nadvi et al., 2004;Dong and He, 2018;Behuria, 2020;Tessmann, 2020;Whitfield et al., 2020;Horner, 2021).Here, the main focus has been on trajectories of global buyers and lead firms, not addressing the perspective of local firms (Sako and Zylberberg, 2019).Even though valuable evidence on business strategies of local small-scale firms is starting to emerge (Hansen, 2023;Hansen et al., 2016;Kadarusman and Nadvi, 2013), more research is needed to improve our understanding of upgrading strategies of local firms in the Global South (De Marchi and Di Maria, 2019).Asymmetries between global buyers and small-scale suppliers in realising economic and environmental upgrading have been detected (Poulsen et al., 2018;Khan et al., 2020a;Ponte, 2020), but it is not explicitly analysed how government policies, such as those aimed at timber and wood products legality verification, contribute to the process.This requires making an active assessment of government activities and not, as previously, considering them as part of the institutional setting (Gibbon and Ponte, 2005;Neilson and Pritchard, 2009;Vermeulen and Kok, 2012;Smith, 2015;Horner, 2017).Baglioni and Campling (2017) pointed to this omission in the literature and called for research aimed at investigating how the state shapes value chain relations in particular related to natural resources.Moreover, Tham et al. (2020) note a shortage of studies focusing on the evolution of timber value chains and their governance and coordination mechanisms.
To address this knowledge gap, and to discuss current interaction between the existing regulatory institutions and upgrading prospects for private sector enterprises, we conducted fieldwork in the main timberproducing regions in Myanmar, which boasts the largest forest cover in the Asia-Pacific region.Around 45% of the land area in Myanmar is under forest cover (World Bank, 2017).Notably teak (Tectona grandis), a high-quality and high-value timber, comprises 38% of the total forest area (Ladrach, 2009;Kaung, 2016).However, Myanmar's forest cover is degrading rapidly under the influence of increasing global demand for timber and wood products, which are delivered using unsustainable logging practices (Treue et al., 2016;Potapov et al., 2017).The annual loss of intact forest was about 10% in 2002-14 (Bhagwat et al., 2017) with the most rapid loss in Bago Region (Potapov et al., 2017).The highest rate of deforestation is linked to the military regime that preceded the first democratically formed government, but deforestation has not ceased completely afterwards either, due to increased fuelwood demand, commercial agriculture, urbanization, and illegal logging (Linn and Liang, 2015). 1 To meet the requirements of international buyers and to improve forest management, the government of Myanmar initiated a series of policy changes in the forest sector starting around 2012, principally aiming to put in place a system that demonstrates legality of timber and wood products (Springate-Baginski et al., 2014).The key elements of the policy reform include: (i) prohibiting export of unprocessed logs, (ii) placing a nationwide ban on logging, (iii) transferring timber extraction rights solely to a single state-owned enterprise, and (iv) introducing a comprehensive system of inspections and on-site verifications (from the forest to the port of export) that provide legality assurance of timber and wood products.
Despite its clear economic and environmental importance, the timber industry in Myanmar has not been studied extensively.Banikoi et al. (2018Banikoi et al. ( , 2019) ) outlined the Myanmar's forest management system and described the teak value chain; Woods (2011) explored the link between Myanmar's state formation and timber trade; Springate-Baginski et al. (2014) analysed the potential of the EU timber legality verification system to advance democratic reforms in Myanmar; and Dong and He (2018) analysed the distribution of benefits from timber trade between Myanmar and China.Investigating current interaction between the existing regulatory institutions and upgrading prospects for private sector enterprises in Myanmar is important as country-specific differences in adopting principles and processes from an international forest policy determine the benefits of timber legality processes (Wiersum and Elands, 2013).In addition, our study draws attention to the importance of institutional actors and regulations for economic and environmental upgrading in GVCs and contributes to an emerging body of studies that have detected asymmetric outcomes related to upgrading in GVCs (Poulsen et al., 2018;Behuria, 2020;Khan et al., 2020a;Ponte, 2020;Tessmann, 2020).

Governance in global value chains
GVCs have been used conceptually to explain global production and distribution of various products.In a GVC context, governance is defined as the process of specifying, communicating, and enforcing compliance with key product and process parameters along the value chain (Humphrey and Schmitz, 2002).It is traditionally focused on relationships of relative power and coordination among firms, e.g.raw material suppliers, component manufacturers, assemblers, designers, distributors, retailers and recyclers (Kadarusman and Nadvi, 2013;Havice and Campling, 2017).The role of lead firms, which govern and co-ordinate product and value flows is of critical interest, as these firms influence the possibilities for actors along the value chain to innovate and upgrade (Gereffi et al., 2005;Ponte and Sturgeon, 2014).A merger with global production networks (GPNs) scholarship enabled a move away from focusing on linear and vertical governance structures to allow for a multi-actor analysis outside of the production process and consideration of how various actors play a part in the governance structure, how GVCs face multilevel regulatory systems, such as international standards, and how they operate under a political structure with the nation state as the main actor (Henderson et al., 2002;Coe et al., 2008).Attention is also paid to the distinctive contribution of market-based or third-party regulation, private-sector environmental commitments and investments, collective action, trade unions or NGOs and states in GVC governance (Bair and Palpacuer, 2015;Havice and Campling, 2017;Horner, 2017).
The state can be involved in GVCs in different ways.Horner (2017) settles on four different characterizations of the role of state, these being facilitator, regulator, producer, and buyer.As a facilitator, the state's role can be to facilitate and promote participation in GVCs (of both domestic and foreign actors), to impose requirements for joint ventures with nationals (Gereffi, 1994), to regulate foreign and domestic private investment, or to provide access to infrastructure, credit and tax facilities (Horner, 2017).The state's role as a regulator emerges from the need to limit economic activities to protect certain interests, such as those of citizens, businesses, consumers, or workers (Horner, 2017).States can thus actively support development and formation of value chains, regulate their functioning, and determine distribution of the value generated along and across value chains (Vilakazi and Ponte, 2022).They can also participate in value chains actively as producers (e. g. state-owned enterprises (SOEs)), or buyers engaged in large-scale purchases needed to offer particular social services or supply SOEs (Horner, 2017).
The intersection of the four roles of the state (facilitator, regulator, producer, and buyer) may give way to potential conflicting interests in functions and responsibilities, as well as to inefficient and/or unequal management of resources and distribution of benefits.It can, we argue, ultimately support or restrict prospects for local producers to enhance 1 The exact volumes and value of illegal timber logging and trade are difficult to ascertain.Two accounts from 2013 to 2015 mention that illegally traded timber from Myanmar is valued at USD200-600 million per year (Wadley, 2015;Forest Trends, 2021).capabilities and upgrade.

Upgrading in global value chains
Upgrading in GVCs is, according to Gereffi (1999), defined as a process of improving the ability of a firm or an economy to move to more profitable or technologically sophisticated capital and skill-intensive economic niches.Humphrey and Schmitz (2002) distinguish between four types of economic upgrading: product, process, functional, and inter-chain upgrading.These could be, for instance, increasing the efficiency of internal operations, enhancing inter-firm linkages, introducing new products or changing the mix of activities conducted within the firm (Kaplinsky, 2000).For Pietrobelli and Saliola (2008), economic upgrading can entail entering higher unit value market niches, entering new sectors, undertaking new productive functions, and increasing technological capabilities.The focus can sometimes be on improving just one specific aspect of firms' operations and sometimes it can be on improving several functions at the same time (Trienekens, 2011).More often than not, it is required to engage in multiple upgrading pathways at once (Hansen et al., 2023).
Environmental upgrading is of high practical relevance due to its focus on reducing the use of limited resources (Hart, 1995) and its cost-saving potential in manufacturing processes (Dauvergne and Lister, 2012).It entails 'the process by which economic actors move towards a production system that avoids or reduces environmental damage from their products, processes, or managerial systems' (De Marchi et al., 2019, p. 312).This type of upgrading can be achieved through process improvements (e.g., reorganizing production systems or using a superior technology to improve eco-efficiency); product improvements (e.g., developing more environmentally friendly product lines); or organizational improvements (e.g., improving overall firm functioning, often related with meeting international standards' requirements) (De Marchi et al., 2019).In the forest sector and wood industry, some of the possible paths to environmental upgrading include improving energy efficiencies and developing technological capabilities to reduce carbon emissions, e. g., through implementation of sustainable forest management and management of wood residues (Wan et al., 2015).
The relationship between economic and environmental upgrading is a complex one.Traditionally, the literature proposed that reducing environmental impacts can contribute to firms' economic benefits and competitiveness (De Marchi et al., 2013).The benefits can come from transforming firms' internal processes towards lower operational costs, reducing waste and the use of resources and energy, meeting legal environmental requirements or implementing third-party certifiable environmental standards.However, much of the value created by environmental upgrading comes with costs that are not obvious to the consumers and the general public (Ponte, 2020).More often than not, the cost of sustainability compliance and related risks are passed by the lead firms upstream towards producers (Havice and Campling, 2017;Khan et al., 2020a;Ponte, 2020).In a number of case studies, environmental upgrading was a prerequisite for value chain participation achieved at the expense of developing country suppliers.For example, garment factories in Sri Lanka did not obtain higher prices from buyers after environmental upgrading (Khattak et al., 2015), while Pakistani apparel suppliers sacrificed profitability to meet the sustainability requirements of global buyers (Khan et al., 2020b).Finally, the initial gains from environmental upgrading accruing to early adopters can decline when competing firms in the value chain adopt the same strategy (Goger, 2013;Khan et al., 2020b).

GVC governance and upgrading
Moving to a higher-value economic activity or a more environmentally friendly production process requires a supportive institutional context in which firms operate (Manning et al., 2012;Neilson et al., 2014;Eckhardt and Poletti, 2018).GVC scholarship has been highly focused on how private governance arrangements such as international standards shape outcomes of global and local value chain actors.Examples come from various value chains (Morris and Dunne, 2004;Hatanaka et al., 2005;Islam, 2008;Nadvi, 2008;Neilson, 2008;Tran et al., 2013;Schleifer and Sun, 2020).The active role of the state in shaping GVC governance and outcomes in value chains has not received much analytical attention (Smith, 2015;Horner, 2017Horner, , 2021)).The few existing studies offer polarized evaluations.
State-led GVC governance is perceived positively in a couple of studies.For example, Lebdioui (2022) argues that power asymmetries alongside GVCs justify the need for state interventions in developing countries.A successful state intervention in Malaysia achieved a positive value addition using coherent industrial policies and diverting rents towards inter-sectoral productivity improvement and capability development (Lebdioui, 2022).In addition to collective learning and capacity building, the key factor of successful state-led upgrading in Uzbek horticulture value chain governance includes the establishment of financial and political partnerships with international actors (Lombardozzi, 2021).Empirical results in Vietnam show advantages of high-capacity SOEs in the textiles and garments industry able to integrate into high-quality value chains and provide gains through employment while smaller private firms remain in regional markets (Nadvi et al., 2004).However, a combination of weak local manufacturing tradition global and buyer pressure can reduce efficiency of industrial policies to develop competitive local export firms (Whitfield et al., 2020).Behuria (2020) describes how rents from upgrading in the coffee value chain were captured by state-and military-owned exporters, politically affiliated private businesspeople and only the most efficient producers in Rwanda.Tessmann (2020) explains that, focusing on market access and market participation as the key to economic development, policy makers in the Ivory Coast reinforced the dominant position of foreign buyers at the expense of local processors.
Several studies conclude that various policies to formalise forest management and value chain participation do not address smallholder needs (Eba'a Atyi et al., 2013;Waldhoff and Vidal, 2015;Acheampong and Maryudi, 2020;Cromberg et al., 2022).Small enterprises face technical and financial burdens leading to, in some cases, extensive engagement in illegal practices to sustain their businesses (Obidzinski et al., 2014;Setyowati and McDermott, 2017;Acheampong and Maryudi, 2020;Maryudi et al., 2021).In contrast, larger manufacturers more easily fulfil legality requirements and realize financial gains from them (Maryudi and Myers, 2018).
Among the GVC studies until now, the evidence on the role of policies governing natural resource value chains is not extensive (Baglioni and Campling, 2017;Dong and He, 2018;Vilakazi and Ponte, 2022).In addition, there is need for increased focus on local firms as the GVC literature extensively covers outcomes of lead firms and global buyers (De Marchi and Di Maria, 2019;Sako and Zylberberg, 2019) and the forest governance literature does not provide insights on economic and environmental upgrading of local firms.

Materials and methods
To investigate the outcomes of the forest policy reform in Myanmar set in motion by international timber legality requirements, we apply a multi-level analysis and use a combination of qualitative and quantitative methods to generate a unique portfolio of information for providing empirically grounded results.To achieve this, different methods of data collection are applied at different levels of the timber value chain to obtain rich insights into the forest policy reform from different perspectives.
We use two kinds of primary data and several sources of secondary data in this study.First, we use data from a quantitative nationally representative enterprise survey of small and medium-sized manufacturing enterprises in Myanmar.The survey was conducted in 2017, 2019 and 2020 within the Myanmar Enterprise Monitoring System project (UNU-WIDER, 2017).The sample includes about 2500 private enterprises from all 15 regions and states in Myanmar.Out of these, about 200 enterprises operate in the wood industry. 2The survey data include information on enterprise characteristics and practices, such as the number and structure of workforce, technology and innovation, revenues and costs, main customer type, owner characteristics, and economic constraints.These data were used to produce descriptive statistics on the performance of private enterprises in the wood industry.Fig. 1 shows the location of timber enterprises from our survey data.
Second, we use data from in-depth face-to-face interviews with enterprise owners and other value chain actors conducted across Myanmar in 2018.The interviews were conducted over the course of six weeks from March to September 2018 in 14 townships from eight regions/ states in Myanmar.Table 1 presents an overview of interviews by state/ region.In total, 40 interviews were conducted with owners or managers of wood-processing enterprises and six were conducted with representatives of supporting institutions, such as officers in local branches of the Forest Department (FD) and Environmental Conservation Department.The highest number of interviews with wood enterprises was in Ayeyarwady Region and the lowest was in Kachin State, reflecting the prevailing distribution of various value chain actors in different parts of the country.
We selected the respondents for qualitative interviews from the quantitative survey database using a combination of purposeful and random sampling.The sampling categories were established to maximize the breadth and relevance of information based on firm size and location.In terms of location, we chose townships with the largest number of firms in the wood industry and then sampled our respondents randomly, taking into account firm size, such that we can obtain information from enterprises in different size categories: micro, small, and medium.The interviewed wood-processing enterprises were predominantly from the micro-size category, with one to nine employees (82.5% of those interviewed, equal to 33 enterprises).A further five enterprises were small, with 10-49 employees (12.5%) and two were medium-sized, with 50-300 employees (5%).The average number of employees was four in the interviewed micro firms, 15 in small firms, and 100 in medium-sized firms.
The interviews were arranged and conducted in co-operation with the Central Statistical Organization (CSO) of Myanmar.The list of enterprises selected for interviews was sent to CSO central office staff, who transferred the list to CSO regional offices.All interviews were scheduled by CSO regional office staff and staff from local administrative (village or ward) offices.CSO staff from central and regional offices were present during the interviews.The interviews proceeded with two-way interpretation between English and Myanmar language.In a few cases, additional interpretation from local minority languages was required.
The interviews took the form of semi-structured conversation.The key topics prepared in advance included production characteristics, supply chain, employment conditions, access to finance, formalization, business associations, and business environment.Additional and clarifying questions were added on a needs basis.All questions were openended.As a rule, the interviews started with questions about production characteristics, while sensitive questions, such as the ones about informal payments and challenges, came at the end.The order of other questions varied from one respondent to another to address fatigue issues.The interviews were recorded with consent and the research team also took handwritten notes.The interview material underwent thematic and interpretative analysis.The respondents were promised confidentiality, so our study does not refer to any persons, businesses, or brands by name.To increase the reliability of our analysis, comparisons  Source: Authors' qualitative interviews.
and contrasts were made between the survey and qualitative interview data, as well as with material from yearbooks, site visits, and conversations with administration officers and experts from relevant ministries.
Finally, we use secondary data on aggregate trade values and hardwood production at the union and state/region level from statistical databases maintained by FAO, UN Comtrade and CSO.As some of these data go back to 1960, we are able to obtain a valued historical perspective on the key developments in the sector.These data were analysed quantitatively to present descriptive statistics of trade and production of wood products across different regions.
Taking that global buyers and the international sustainability discourse have stimulated a reform of domestic institutions in Myanmar, namely the forest policy reform, we focus on the impact of these recent policy changes on upgrading prospects in the Myanmar timber value chain.Given the country context and information availability, we assess economic upgrading against several dimensions, including changes in production volumes, changes in product type and quality, export expansion, transaction and operating costs, and raw material access.Various instruments can decrease negative environmental impacts of imprudent forest management, so following suggestions from the literature (e.g., Tham et al., 2020), we evaluate environmental upgrading against the following measures: reforestation and plantation forestry programs, sustainability certification, international agreements, national legislation and transformation of firms' internal processes towards reducing emissions, waste, and the use of energy.

Timber value chain and state regimes
A simplified graph showing key actors in the GVC for timber in Myanmar (Fig. 2) illustrates many successive stages related to production, extraction, processing and distribution of timber and wood products, as well as supporting products and services. 3The chain includes input and service suppliers, loggers, sawmills operating at various scales, selling timber and wood products to local traders and processors.Apart from selling locally, some sawmills and processors are also exporters with clients in other countries.Other actors include domestic government agencies which regulate and support the industry and issue licenses to sawmills, processors and traders, as well as foreign governmental and non-governmental agencies, wholesalers and retailers in importing countries.
The functioning and the performance of the forest sector and the wood industry in Myanmar are inextricably linked to the state regimes.Myanmar gained independence from British rule in 1948, when the government of the Anti-Fascist People's Freedom League (AFPFL) introduced the private and foreign sectors in addition to setting up stateowned, mostly large-scale industries (Aung, 2008).After 1962, a socialist economic system was established with the nationalization of large private enterprises.A military takeover by the State Law and Order Restoration Council (SLORC) in 1988 moved the country towards a more market-led economy, reversing all socialist policies but failing to mimic the high growth of other authoritarian regimes of the same period (ICG, 2012).Numerous political, ethnic and territorial tensions especially in the borderlands gave rise to protracted conflicts, which were fuelled by revenue from gemstone and timber sales (Jones, 2014;Christensen et al., 2019).Due to political and human right concerns, Myanmar was suspended from bilateral and multilateral agreements in 1997.Both the US and the EU imposed sanctions and trade restrictions, which were lifted in 2013, three years after a democratically elected government was formed (Bernhardt, 2017).The new government led by the National League for Democracy (NLD) inherited low levels of industrialization, a large informal sector, a dysfunctional financial sector, and unresolved conflicts with ethnic minorities (ICG, 2012).
In all these state regimes, the regulatory role of the state in the timber value chain was performed by the Forest Department (FD) that operates under the Ministry of Natural Resources and Environmental Conservation together with the Myanma Timber Enterprise (MTE), which is in charge of timber extraction and its placement to the market through public auctions.The FD has traditionally been in charge of implementing the Myanmar Selection System (MSS), a scientific method of forest management.The FD aims to assure forest conservation by determining ' … where, when, how, or even if the forests are to be exploited' (Bryant, 1997, p. 21), which is achieved by defining the annual allowable cut (AAC).The legal framework that the FD follows includes the 1995 Forest Policy and the Forest Law enacted in 2018, which set the foundations for forest conservation and exploitation.The FD plays a crucial role in assuring timber legality along the value chain as it oversees logging, trade, transport, sawmilling and processing of timber, and issues operating licenses to private sawmills that purchase timber from MTE auctions.
The timber industry represents a considerable share of the manufacturing sector, but it remains undeveloped in terms of product variety and sophistication (Berkel et al., 2018).The industry predominantly comprises small-scale sawmills and a small number of plywood and other wood processing factories that specialise in rough sawn timber and semi-finished products.The capacity to process timber further to produce higher value products, such as furniture, wood-based panels, moulding and joinery, is limited (Rand et al., 2019a), except in large companies partly owned by the state.An example is the Forest Products Joint Venture Corporation, which is a public-private corporation between MTE, FD and private investors.This corporation is the largest processor and exporter of teak and other hardwoods from Myanmar, which illustrates that regulatory and commercial roles of the state co-exist, leading to a concern that the incentives to follow harvest rules can easily lean towards extraction rather than sustainable forest management.

Forest policy reform
The emergence of the global timber 'legality verification regime' (Leipold et al., 2016) has strongly influenced the forest policy in Myanmar.In 2003, the EU adopted the Forest Law Enforcement, Governance and Trade (FLEGT) Action Plan, which stipulates that only timber and wood products that meet the European Union Timber Regulation requirements can be imported in the EU.The plan specifies the definition of legal timber, and incorporates a timber tracking system and a licensing scheme designed to ensure that only legal timber products are exported to the EU (Rutt et al., 2018).In 2008, the US introduced the Legal Timber Protection Act (LTPA) and in 2012, Australia issued the Illegal Logging Prohibition Act (ILPA), which prohibit the import of illegally logged or traded timber.
After the military regime dissolved in 2011, the new democratic government in Myanmar prioritized demonstrating its forest conservation objectives.As indicated in the Environment Conservation Law from 2012, the government started to voice commitments for a more sustainable forest sector giving increased attention to ' … conservation, management, beneficial use, sustainable use and enhancement of … forest resources'.Myanmar became active in the REDD + process in 2013 with the aim of reducing emissions from deforestation and forest degradation, and improving conservation, sustainable management of forests, and enhancement of forest carbon stocks.The National Forest Monitoring System and Reference Emission Levels were put in place, but a criticism has been raised that further political commitment is needed to address direct and indirect drivers of deforestation (Oo et al., 2020).In an attempt to assure that more value is captured before export, the government placed a ban on exporting unprocessed logs in 2014.In 2015, Myanmar started negotiating a Voluntary Partnership Agreement (VPA) within the EU's FLEGT Action Plan.Then, it issued a national logging ban in 2016 (until March 2017), as well as a regional 10-year logging ban in Bago Region (EIA, 2016).Similar instruments have been applied in other Asian countries, including China, Indonesia, Lao PDR, Malaysia, the Philippines and Vietnam, which all have active bans on exporting timber logs (Tham et al., 2020).
The new Forest Law issued in 2018 (Pyidaungsu Hluttaw Law No. 29/2018) sets the frame for forest exploitation, identifies legal sources of timber (state-owned MTE auctions, concessions, land conversion, plantations, and community forests), and establishes fines and penalties for illegal operations.Compared to earlier versions of the law, it prescribes stricter penalties for illegal actions, including stricter punishment for taking part in illegal logging, trade, possession and processing of illegal logs.Further steps to meet external verification requirements include establishing the Community Monitoring and Reporting System, the Timber Legality Assurance System and a third-party certification system, the Myanmar Forest Certification Committee (MFCC, 2018).Nevertheless, Myanmar received a negative assessment of the accountability and transparency of the timber and wood product exports' chain of custody.Teak imports have been shown to violate the EUTR requirements, such that in 2016 and 2017, Sweden and Denmark banned teak imports from Myanmar (EIA, 2019).Notably, current verification systems in Myanmar are also insufficient to ensure compliance with the US LTPA and Australia's ILPA.
In a coup d'état on February 1, 2021, the Myanmar military took power from NLD.The implications for the forest sector are still to be gauged, but it seems that the new regime will revert to the revenuedriven uncontrolled forest extraction (EIA, 2021).A recent review of satellite images revealed increased deforestation in the central parts of the country after the coup (Heubl, 2021).The US and the EU imposed sanctions against state-owned enterprises responsible for timber exports from Myanmar in April and June 2021, respectively (European Council, 2021; U.S. Department of the Treasury, 2021), so timber will mainly be exported to 'neighbouring states that fail to question its legitimacy' (Heubl, 2021).

Production dynamics
Fig. 3 shows the annual teak and hardwood production in Myanmar for the period 1960-2019.We observe a pattern of incessant production growth with an average annual growth rate of 2%.The record level in hardwood production of 8.5 million m 3 was reached in 2013, four times larger than in 1960.Current production is stable at around 6 million m 3 .Timber extraction mainly takes place in Sagaing, Bago, and Mandalay regions.Consistent with the logging ban, extraction in Bago Region declined in 2016 compared to 1985, but it increased in Tanintharyi Region and Kayah State by 3-5% (Rand et al., 2019b), indicating a shift towards timber extraction in borderland areas, which supports the argument of the importance of timber for prolonged internal conflicts (Jones, 2014).
The limited product range shown in Fig. 4 indicates a low diversity of processed timber products.While sawlogs still dominate, their production has been declining since 2013 in favour of more roundwood and sawnwood.Production of higher-value products such as veneer sheets and plywood is about 20 times lower than the production of low value added products such as rough sawn timber and semi-finished products.Most enterprises focus on generic products with low value added, such as planks and boards of various size, so in theory, there is a wide range of options for economic upgrading by changing the product variety or increasing processing efficiency.While there is great interest in producing higher-value products, our respondents voiced concerns about their capability for doing so, which is inextricably linked with a lack of suitable machinery and equipment.In addition, the ability to change the Fig. 2. Timber value chain in Myanmar.Source: Authors' illustration product portfolio is very limited.One of the respondents stated that he could profit more from selling thinner floorboards, but the equipment he owned was not suitable for making such products.Technology used in the sawmilling sector is old with limited capacity for utilization of wood.In 2017, 61% of timber enterprises from our survey reported that the machinery and equipment used for processing is older than five years.Financial constraints prevent technical modernisation and automation, also observed in other South-East Asian countries that are losing international competitiveness (Ng and Thiruchelvam, 2012).
None of our interviewees was aware of government support to the private sector, such as offering training to increase enterprise capabilities or supporting contacts with foreign buyers, which could facilitate foreign market access.This seems like a missed opportunity given the experience of neighbouring countries, such as Indonesia, where the government put extensive efforts into supporting partnership formation between timber processors and buyers to achieve upgrading of the timber value chain (Tham et al., 2020).According to several interviewees, enterprises from the wood industry do not qualify for preferential loans backed by the government given the concerns over the financial viability of the entire industry.

Export dynamics
A commonly used measure of economic upgrading in GVCs is export expansion.Fig. 5 shows that the value of exports of timber and wood products from Myanmar peaked in 2013 at a value of USD1.46 billion for roundwood, after which it remained stable at around USD100 million.In 2017, sawnwood became the most valuable exported wood product, reflecting the initial impact of the forest policy reform towards forest preservation and higher-value exports.According to UN Comtrade (2018) data, about 60% of all timber and wood products from Myanmar was directed to India in the period 2010-17.China is the second most important destination for raw logs and sawnlogs, with around 9% of total export value.Direct export to more distant markets such as the EU, the US, and Japan is far less common.This trade pattern indicates that Myanmar primarily focuses on exporting to neighbouring countries, which are probably not end markets but serve as re-exporting hubs.
Unauthorised and unrecorded timber exports were estimated at USD5.7 billion in the period 2000-2013(EIA, 2014)).This amounts to 67% of the legally recorded industrial roundwood and sawnwood export value.Most commonly cited causes of illicit logging include increased demand for timber and timber products, especially in China, globally increasing prices of timber products, corruption, concessions to military regional commanders and ethnic leaders in conflict areas (Brunner et al., 1998;EIA, 2014;Prescott et al., 2017;EITI, 2020).
Fulfilling the timber legality requirements is expected to result in higher exports to the EU, as well as to other higher-value markets that require timber legality assurance from their suppliers.However, we are doubtful about the immediate potential of extended foreign market access to improve circumstances of the small-scale private sawmills and processors for two reasons.
First, the initial engagement of the private enterprise sector with foreign markets is very low, as recorded by our survey.While this holds for both timber and non-timber enterprises, the former are at a slight disadvantage.Fig. 6 shows that timber enterprises trade at closer proximity than non-timber enterprises (timber enterprises sell 74%, while non-timber enterprises sell 63% of their products in the same township), which is most likely due to a high reliance on local and international traders.A strong dependency on selling in local markets indicates that it is not certain that timber enterprises can benefit from the improved EU market access without receiving substantial support for improving their productive and marketing capabilities.In Malaysia and   Indonesia, the state provided incentives for innovation and technological upgrading, facilitated creation of business associations and disseminated technical advice through business associations and industrial clusters as part of a strategy to upgrade timber value chains (Ng and Thiruchelvam, 2012;Tham et al., 2020).
Second, exporting is more common among larger enterprises.Whereas micro enterprises sell 77% of their products in the township in which they are located, larger timber enterprises sell 77% of their products abroad.Even if the forest policy reform is successfully implemented in Myanmar, it is unlikely that the benefits of improved market access will be equally distributed along the value chain.This is because large firms have an advantage in participating in GVCs, which entails meeting legal and customer requirements in terms of product quality, as well as environmental and social conditions of conducting the production process.Small firms are at a technological, organizational, and skill disadvantage.Here, government efforts are crucial for addressing learning and power asymmetries in the value chain (Pietrobelli and Staritz, 2018).

Product quality and price
The forest policy reform has had another negative upgrading effect for private small-scale sawmillsby decreasing raw material availability, logging bans led to quality downgrading.Several interviewed enterprises raised the issue of timber quality, stating that they had to switch to working with timber in a lower quality grade than before.Higher quality grades of timber became unaffordable or the only type of legal timber they could procure was seized timber, which is assessed as having inferior quality.Searching for affordable timber, some sawmills occasionally use illegally harvested wood, which often entails going down the quality ladder.A recent study documented that illegally harvested logs in Bago Region tend to come from younger trees, to have smaller diameter and to be of lower quality (Saung et al., 2021).The viability of competition based on low price and quality is questionable, when global markets have moved towards competition based on quality, design, and image (Loebis and Schmitz, 2005).
The shortage of unprocessed logs in the market caused by the forest policy reform has predictably increased auction and finished product prices.While auction prices increased by about 20% (Forest Trends, 2021), timber enterprises from our survey reported an increase in the sales price of their main product of 49% (MMK725,427 per ton in 2017 and MMK1,079,101 per ton in 2019). 4In both 2017 and 2019, final product prices charged by medium and larger firms have been three times higher than the prices charged by micro enterprises.This indicates that larger firms either consistently work with logs in the higher quality grade or that they have a higher capacity to absorb market shocks in the form of higher input prices, which they convert to increased benefits by charging higher final product prices.

Environmental upgrading
Continuous effort aimed at assuring legality verification of timber and wood products could in principle result in an improved forest condition.However, the process is not likely to succeed if not accompanied by management practices that contribute to sustainable protection and regeneration of forests.If properly applied, the MSS of forest management has high sustainability and legality measures (Woods, 2013), but it has been abandoned for several decades in favour of revenue-driven exploitation.A return to strict implementation of the MSS should enable environmental upgrading if the extraction of forest proceeds such that the AAC levels are respected.Current AAC in Myanmar is about 2 million m 3 for hardwood and about 100,000 m 3 for teak.Treue et al. (2016) show that these levels started to be exceeded in 1978.Following global recommendations and examples of neighbouring countries, the government of Myanmar has also encouraged the expansion of plantation forests.However, the rate at which the plantation forests are established is nowhere near the targets specified in the nationally determined contribution (Myint et al., 2021).The government has set a target of 286,000 ha for plantations aimed at commercial timber production to be achieved by 2027, but as of October 2019, only 40,400 ha had been planted (World Bank, 2020).This mirrors the experience from China where timber logging quotas may have halted deforestation and forest degradation, but have failed to promote forest expansion (Liu and Xia, 2021).
Where policy reform has not secured conditions for sustainable forest use, voluntary forest certification has been increasingly adopted as an alternative way of improving forest management (Ulybina and Fennell, 2013).Obtaining environmental certifications, such as Forest Stewardship Council (FSC) or ISO 14001, has become a widely used tool for demonstrating commitments towards sustainable forest management and environment-friendly production among developing country suppliers.We therefore looked at the ownership of environmental certificates in the timber value chain, and found that none of the enterprises we surveyed owned environmental certificates.This is because both FSC and ISO 14001 require commitment to applicable regulations and legislation, which is not yet feasible in Myanmar given the difficulties of assuring compliance with the existing national regulatory framework.The FSC conducted a national risk assessment for Myanmar in 2018 and assigned a high-risk score to almost all of its 33 indicators of sustainable forest management (FSC, 2018).Besides the lack of implementation of relevant rules and regulations, forest management in Myanmar suffers from a high risk of corruption, illegal logging, and harvesting from areas with poor forest management, including disregard for protection of biodiversity, wildlife, and watershed areas (FSC, 2018).Drawing from the past experiences in China described by Chen and Innes (2013), improving the adoption of sustainable forest management practices and environmental certification would require overcoming constraints posed by inconsistent forest policies, limited finance, poorly developed infrastructure and transport systems, and inadequate knowledge and technical transfer.
Timber industry is associated with several kinds of adverse environmental impacts (Adhikari and Ozarska, 2018) and the activities of the middle segment of the value chain can to a great extent support environmental upgrading.Timber extraction and processing release particulate environmental matter during tree felling, log debarking, and sawing into boards (Adhikari and Ozarska, 2018).Further concerns include emission of carbon and air pollutants, discharge of used chemicals, and wood wastage.Nevertheless, some of the key actors in the value chain, such as the MTE management, see the wood industry as environment-friendly with the only harmful emissions coming from plywood industry that uses glues and polishing chemicals (Castrén, 1999).
In the wood industry, the type of technology determines productivity, the rate of wood utilization and the amount of waste created during processing, and as much as 45-50% of the total volume of sawlog input can go to waste (Dionco-Adetayo, 2001;Ramasamy, 2015).A natural way to increase the lumber value would be to optimize the performance of cutting operations by reducing materials use and minimizing waste.However, most of the surveyed sawmills work with old machinery and equipment purchased as used.Investing in new technology is rare among manufacturing enterprises, and more so among timber than other enterprises.Whereas 7% of non-timber enterprises invested in new technology in the two years prior to our 2019 survey, only 4% of timber enterprises did the same.This echoes experiences from neighbouring countries where environmental upgrading in the forest sector was constrained by limited finance and inadequate knowledge and technology transfer (Chen and Innes, 2013).Our interview data show that the raw materials shortage has inspired some sawmills to increase the use of reclaimed wood, but only a negligible fraction of interviewed sawmills adopted this strategy.
Sawmilling is energy-intensive.The electrical energy consumed during sawmilling is generated off-site at hydropower plants or power stations that burn fossil fuels, both with substantial greenhouse gas emissions (Ocko and Hamburg, 2019).The combustion of fossil fuels for energy discharges additional pollutants, such as methane, nitrous oxide, mono-nitrogen oxides, carbon monoxide, sulphur dioxide, and non-methane volatile organic compounds, which have a negative impact on air quality (Ramasamy, 2015).As one-quarter of sawmills in our survey does not have access to public electricity grid and as electrical energy supply in Myanmar can be unpredictable, almost all sawmills use own diesel-powered generators that emit various air pollutants, some of which are highly toxic.Improving the environmental footprint of the sawmilling operations in Myanmar would require increasing the rate of using renewable energy to power machinery.
Even though wood industry can be associated with adverse environmental impacts, a comparatively smaller contribution towards upgrading can at present be made by small-scale sawmills and processors.Instead, a major contribution to the environmental upgrading of the timber value chain should come from increased efforts to assure sustainable forest management and legality verification.Apart from assuring timber and wood products legality, additional environmental performance requirements are not placed on sawmills and processors neither from the state in the form of specific regulations nor from global buyers.In this regard, the Myanmar timber value chain resembles a typical producer-driven value chain, which can rarely achieve environmental upgrading as the power is distributed between a couple or several actors competing for control along the value chain (Ponte and Sturgeon, 2014;Ponte, 2019Ponte, , 2020)).

Operating costs
By increasing operating costs, legality verification does not harm larger as much as smaller enterprises due to the former's greater absorptive capacity.Small sawmill owners were particularly dissatisfied over the strictness of legality regulations, both because they limit access to unprocessed logs and because they increase costs of doing business.While the FD issues operating licences, which can cost MMK50,000-100,000 per year, licences from other institutions are also required to demonstrate business legality, and they range from MMK12,000 to MMK50,000.Apart from verifying that enterprises own all required licences, the FD inspects sawmills after every purchase of wood, and before and after the cutting of logs.This slows down operations as there is a time lag between informing the FD about the timber purchase and them inspecting the site.As recorded in our survey, enterprises in the wood industry are inspected on average 2.7 times per year, an average which masks high variation.It is most common to be inspected one time per year (observed in 52% of the enterprises), with some enterprises inspected more than 10 times (8% of the enterprises).
A comprehensive documentation follows timber, from forest, through log extraction, public auctions, sawmilling and transport, to final customers, as every transaction needs to be documented.The interviewees were critical about the number of procedures needed to operate in the industry, stating that 'government procedures are too complicated', that 'they need many permits', or that they have to go through 'many unnecessary procedures'.The impact of legality requirements on increasing costs of doing business in Myanmar provides a supporting case for the importance of governance for the functioning and distribution of gains in GVCs.This finding supports the argument that power imbalances affect the concentration of gains in GVCs.Large firms can better absorb the costs pushed on them by global buyers (Goger, 2013) and compared to small firms, receive much more of the value created in the value chain (Tham et al., 2020).

Raw materials access
Under current regulatory conditions, timber sourcing is one of the main challenges for small enterprises.Until 2016, MTE was subcontracting about 75% of extraction tasks to private firms, associated with poor law enforcement and non-transparent supply chains (FSC, 2018;EITI, 2020).By issuing the new Forest Law in 2018, the state transferred all responsibility for timber extraction to MTE.This move could potentially improve traceability in the value chain, which is a prerequisite for verifying timber legality, but it is also vesting even more power in a state-owned company to dictate conditions of participation in the value chain.
Even though they are the principal legal source of raw timber, MTE auctions are not organised in all states and regions, which makes attendance difficult for enterprises from some regions.Smaller enterprises are often unable to match bidding prices of large enterprises.For instance, one of the sawmill owners indicated that the bidding price increased by about MMK250,000 per ton in the past three years, yielding a progressive exclusion of small enterprises from the value chain.Foreign firms can purchase timber in separate foreign-only auctions, and also in domestic auctions, as domestic co-ownership is mandatory-again to the detriment of smaller domestic enterprises, which are unable to match their bidding prices.
Interviewees overwhelmingly agreed that the amount of available raw materials is much lower than in the past, reflecting the immediate consequences of the logging ban.Based on our survey data, Fig. 7 illustrates the most important constraints to growth perceived by enterprises.Compared with other industries, wood-processing enterprises are much more likely to report that limited access to raw materials poses a serious constraint.In addition, while wood-industry enterprises in all size categories face constraints to growth, the problem of access to raw materials predominantly affects the smaller ones, indicating the limited access to raw materials as a source of enterprise performance inequality.
Enterprises that are unable to attend or match auction prices resort to secondary sources, such as purchasing from auction winners or purchasing illegal wood seized by the FD.While the former entails purchasing raw logs at a higher price, the latter cannot be consistently relied on.Seized timber is sold in separate MTE auctions, and it is usually of inferior quality compared to timber sold in regular auctions.The quality deteriorates due to exposure to elements that logs endure while the legal process against those in possession of illegal timber is active.Some respondents mentioned that the price of seized wood is lower than that of freshly cut wood sold in auctions.However, the practice that each regional FD sets the price separately creates a regional variation that cannot be used to the advantage of small enterprises, because it is prohibited to transport seized logs between regions.Even though access to seized wood may seem appealing, its irregular availability and unpredictable quality and price illustrate that it cannot form a long-term supply strategy.

Smallholder marginalization
Fig. 8 compares performance, measured as revenue and profit per employee, of enterprises from the wood industry with enterprises from other manufacturing industries.Three main observations stand out.The wood-processing enterprises have on average lower revenue and profit per employee than enterprises from other industries, indicating lower overall performance.Not only was the performance of wood enterprises worse in the observed period, it also declined more rapidly than in other industries since 2018.While enterprises in other industries experienced a performance boost in 2017, the improvement in performance of woodindustry enterprises was almost imperceptible.
An additional illustration of the difficulties in the wood industry can be seen in Fig. 9, in which we contrast the proportion of enterprises in the wood and other industries that compared to 2017 no longer operated in 2020.The figure shows a higher share of temporarily and permanently closed enterprises in the wood compared to other industries.The closures happened exclusively among micro and small enterprises and not among the larger ones.A much higher proportion of enterprises in other industries that stopped manufacturing now provides services as a survival strategy.An example of such services is repacking and selling rice instead of milling.A similar option does not seem to be possible in the timber industry due to the shortage of raw materials.One of the interviewed respondents reported that the firm could survive only because it was registered as both a sawmill and a furniture-making enterprise, and due to using illegal wood in furniture making, where legality requirements are still not as strictly enforced as among the sawmills.
Even though illegal forest extraction and the inclusion of illegal timber in processing constrain the implementation of sustainable forest management programmes (Chen and Innes, 2013), the regulatory environment of the timber industry in Myanmar entails contradictory   consequences for the way in which sawmills and processing enterprises operate.On the one hand, strict regulations prevent the widespread use of illegal sources of timber through the application of penalties and continuous inspections.On the other hand, some enterprises are pushed to operate completely outside the legal framework due to the high cost of operating legally.Such enterprises exclusively use illegal raw materials and operate without the necessary licences.The temptation for some otherwise-legal enterprises to use occasionally illegal sources of raw materials is high.Paying bribes to avoid fines and to keep operating is inevitable.The number of illegally operating sawmills is impossible to ascertain, but interviewees mentioned that their number is substantial and that they represent unfair competition.
Such an ineffective governance setting in which regulations increase costs of doing business is likely to impair the aims of forest policy to ensure sustainable forest management and trade.The evidence from Myanmar mirrors the experience with FLEGT in Ghana and Indonesia, where a recent study concludes that illegality may be a legitimate response when the legality requirements do not take into account the local context, such as low legitimacy of public institutions and high cost of going through the due diligence process (Acheampong and Maryudi, 2020).Similarly, a study from Cameroon shows that limited options to change regulatory arrangements perpetuate unsustainable harvesting practices and tendencies of small-scale enterprises to operate illegally (Tieguhong et al., 2015b).Consequently, marginalization of small enterprises into illegality intensifies the race to the bottom and creates favourable conditions for greater concentration of the industry.

Discussion
Attempting to improve foreign market access and forest management, the government of Myanmar has reformed its forest policy to provide better assurance of the legality of timber and wood products.Based on the insights from the GVCs literature, legality verification would enable both environmental and economic upgrading, i.e., creation of added value for the participants in the legal segment of the chain.Highlighting the tensions at the core of forest policies aiming to assure a balance between forest protection and monetisation, our findings challenge the assertion that legality mandate is requisite for sustainable forest management.Instead, achieving economic and environmental upgrading in the value chain requires additional effort beyond demonstrating legality.
We established that the government has coupled its regulatory and productive roles, so the policy reform has not met its environmental upgrading objectives of putting in place systems of sustainable forest management and timber legality verification.The rules are set to protect the position of the state as the key producer and supplier of timber, which perpetuates inequalities in downstream segments of the chain.The GVC literature is traditionally concerned with inequalities between lead firms and smaller competitors, as well as between powerful foreign actors and retailers and local firms (Poulsen et al., 2018;Khan et al., 2020a;Ponte, 2020).The literature has only recently started to analyse the relevance of different roles of the state in GVCs (Horner, 2017(Horner, , 2021)).We juxtaposed the regulatory and the productive role of the Myanmar state and discover how competing interests in terms of environmental protection and timber extraction contrive to prevent realising economic and environmental upgrading of local firms.Our findings illustrate the need to assess different roles of the state, beyond the facilitator role traditionally assumed in the GVC literature, as the state in other roles can enter a value chain as a powerful actor, strongly influencing the distribution of opportunities and constraints in natural resource value chains.
GVC-oriented policies should support the creation of linkages of home country suppliers with lead firms and enable participation, value capture, inclusiveness and resilience of the value chain actors (Dallas et al., 2019;De Marchi and Alford, 2021;Pietrobelli et al., 2021).It is apparent from our analysis that, by increasing the transaction and operating costs, the forest policy reform in Myanmar has created favourable conditions for a race to the bottom and concentration in the timber value chain.This is in contrast to studies showing that the state engagement in GVCs contributes to value chain upgrading (Lombardozzi, 2021;Lebdioui, 2022) and in accordance with empirical findings from other contexts showing that GVC policies can contribute to increasing inequalities among value chain participants (Leipold et al., 2016;Behuria, 2020;Tessmann, 2020;Aguilar and Wen, 2021).
The new rules and regulations constrain micro and small enterprises more than they do larger enterprises.Consequently, there is a limited potential for economic and environmental upgrading of the smallholder sector given the prevailing system of restricted incentives, limited linkages with foreign markets, and insufficient public support for relieving financial constraints and promoting technology and knowledge transfer.The limited availability of raw materials through legal means, such as through public auctions, creates a highly competitive environment in which small enterprises become the last to access inputs and, even if they continue to operate, often end up working with lowerquality timber which they sell at lower profit margins.Downgrading can be a viable strategy for achieving economic advantages from volume sales of lower quality products (Gibbon, 2008;Ponte and Ewert, 2009), but combined with the restricted access to inputs, it represents more a survival than an upgrading strategy.This finding indicates that future analyses of upgrading in GVCs should not neglect the role of government policies, which, as we highlight, can strongly challenge upgrading opportunities of local firms.Forest policy will have to be adjusted in a comprehensive manner to address these issues for it to be both efficient and effective.

Conclusion
Our findings demonstrate that improving forest management sustainability and implementing an assurance system for legality of timber and wood products cannot go without considering more carefully the local impacts of the policy reform's effectiveness and efficiency.Furthering forest regeneration goals seems to require decoupling of the productive and regulatory functions of the state, as forest conservation goals tend to be easily side-lined in pursuit of the state revenue.Given the described implications of policy changes for multiple value chain actors, forest policies should not be viewed in isolation from other policies affecting a particular value chain.Where a threat to livelihoods and business viability is possible, supplementary policies opening alternative opportunities should be devised and implemented.For example, if access to raw materials is restricted by the logging ban, industrial policies for extended financing or capability development through retraining could be introduced.Reflecting on complementarities and remedies available in the broader policy landscape when discussing regulatory constraints within single sectors is advisable.
The timber value chain would benefit from reforestation and plantation forestry programs, enabling improved access to finance, better access to technology for producing higher value added products, and increased investment in capability development so a larger share of higher value added products can be produced.The state could take an active role in supporting partnership formation between timber processors and buyers to achieve upgrading of the timber value chain (Tham et al., 2020).Some deregulation for small-scale processors is a potential option to address concerns about their business viability.Similar proposals have been put forward in other countries where legality verification requirements have had mixed success.For instance, legality verification has been relaxed and replaced with a legality declaration for private smallholder timber production in Indonesia and similar measures have been suggested in the SME sector (Susilawati et al., 2019;Maryudi et al., 2021).It would, from the findings in this study, appear that similar initiatives merit consideration in Myanmar.
In combination, the above-suggested initiatives could help prevent concentration of gains among large companies and further marginalization of small-scale enterprises.Consequently, regulatory reforms for better forest sector and wood industry outcomes should take into consideration the adverse effects of high transaction and operational costs, concentration of the industry, relegation of smallholders into illegality, and disincentives to plant trees.Put directly, regulatory policies need to go hand in hand with industrial and development policies to promote participation and enable survival of smallholders in GVCs.

Fig. 1 .
Fig. 1.Location of timber enterprises in Myanmar.Source: Authors' illustration based on survey data.

Fig. 7 .
Fig. 7. Most important constraints to growth as perceived by enterprises.Source: Authors' illustration based on survey data.

Fig. 8 .
Fig. 8. Revenue and profit of SMEs in the wood industry.Source: Authors' illustration based on survey data.

Fig. 9 .
Fig. 9. Higher share of closed enterprises in the wood than in other industries.Source: Authors' illustration based on survey data.

Table 1
Location and number of qualitative interviews.