PROSPERITY OR CRISIS? EXAMINING THE EFFICACY OF BRITISH COLUMBIA’S FOREST POLICY CHOICES TO SUPPORT THE FOREST INDUSTRY by Tammy L. Baerg B.Sc. Hons, University of Northern British Columbia, 2005 THESIS SUBMITTED IN PARTIAL FULLFILMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF ARTS IN NATURAL RESOURCES AND ENVIRONMENTAL STUDIES UNIVERSITY OF NORTHERN BRITISH COLUMBIA September 2016 © Tammy L. Baerg, 2016 Abstract The Forest (Revitalization) Amendment Act 2003 significantly altered the conditions for holding major forest tenure agreements in British Columbia. The goal of this forest policy was to allow the forest industry to become globally competitive. This thesis asks whether or not this goal was successful and what trade-offs were incurred as a result. Using the Prince George Timber Supply Area as a case study, mixed methods of inquiry were employed. This study examines changes throughout the forest industry over a fifteen-year period. In this case, the forest industry became globally competitive. But, significant consolidation within the forest industry occurred. Moreover, while the timber harvest increased and processing became centralized, the forest industry’s overall contribution to the provincial economy declined significantly. This thesis demonstrates that the government remains committed to supporting the current forest industry. However, the aforementioned revitalization legislation may have lasting negative repercussions for the provincial forest economy. i Table of Contents Abstract .................................................................................................................................i Table of Contents ................................................................................................................ ii List of Tables ....................................................................................................................... v List of Figures .....................................................................................................................vi Dedication and Acknowledgements ................................................................................. vii Chapter 1: Introduction ....................................................................................................... 1 1.1. Introduction ............................................................................................................... 1 1.2. Purpose of the Research ............................................................................................ 2 1.3. Research Questions ................................................................................................... 4 Chapter 2: Background and Context................................................................................... 8 2.1. Introduction ............................................................................................................... 8 2.2. Background ............................................................................................................... 8 2.3. Context .................................................................................................................... 17 Chapter 3: Literature Review............................................................................................. 24 3.1. Introduction ............................................................................................................. 24 3.2. Forest Policy Change in British Columbia ............................................................. 25 3.3. Responses by the Forest Industry to Policy Change in British Columbia .............. 29 3.4. British Columbia Forest Policy Legacies ................................................................ 36 Chapter 4: Methodology and Methods .............................................................................. 43 ii 4.1. Introduction ............................................................................................................. 43 4.2. Ethical Considerations ............................................................................................ 45 4.3. Research Design ...................................................................................................... 47 4.3.1. The Case Study ................................................................................................. 48 4.4. Data Collection and Analyses ................................................................................. 51 4.4.1. Mill Closures .................................................................................................... 51 4.4.2. Employment ..................................................................................................... 53 4.4.3. Timber Harvest and Valuation ......................................................................... 53 4.4.4. Semi-Structured Interviews .............................................................................. 58 Chapter 5: Results and Analyses ....................................................................................... 67 5.1. Introduction ............................................................................................................. 67 5.2. Mill Closures ........................................................................................................... 67 5.3. Employment ............................................................................................................ 70 5.4. Timber Harvest and Valuation ................................................................................ 74 5.5. Semi-Structured Interviews..................................................................................... 89 5.6. Conclusion ............................................................................................................ 113 Chapter 6: Discussion ...................................................................................................... 114 6.1. Introduction ........................................................................................................... 114 6.2. Rationalization of the Forest Industry ................................................................... 114 6.3. Timber Harvest and Valuation .............................................................................. 118 iii 6.4. The Impact of Mountain Pine Beetle .................................................................... 121 6.5. Research Questions Revisited ............................................................................... 125 Chapter 7: Conclusion ..................................................................................................... 132 7.1. Introduction ........................................................................................................... 132 7.2. A Way Forward ..................................................................................................... 132 7.3. Limitations and Future Research .......................................................................... 133 References........................................................................................................................ 136 Appendix I – UNBC Research Ethics Board Approval................................................... 153 Appendix II – Event Time Line 1997 to 2011 ................................................................. 154 Appendix III – Interview Guide ...................................................................................... 155 Appendix IV – Information and Consent Form ............................................................... 159 Appendix V – Template Analysis Themes/Questions ..................................................... 161 Appendix VI – Content Analysis and Theme Development ........................................... 162 iv List of Tables Table 1: British Columbia Employment by Detailed Industry, Annual Averages…………. 70 Table 2: 2006 Direct Employments in Logging and Other Wood Manufacturing Sectors… 72 Table 3: 2006 Economic Dependency Indicators by Forest District…..………………...…. 73 Table 4: Approximate Direct Employment Loss Due to Mill Closures in the PGTSA…...... 73 Table 5: Interview Participant Statistics……………………………………………………. 90 Table 6: Theme 8 Subthemes and the Relative Frequency of Subtheme Indicators by Group …………………………………………………………………………………………...… 106 Table 7: Theme 9 Subthemes and the Relative Frequency of Subtheme Indicators by Group ……………………………………………………………………………………………... 107 Table 8: Theme 10 Subthemes and the Relative Frequency of Subtheme Indicators by Group ………………………………….…………………………..…………...……………….… 109 Table 9: Theme 11 Subthemes and the Relative Frequency of Subtheme Indicators by Group …………………………………..…………………………………..…………………..…. 111 v List of Figures Figure 1. Map of the Prince George Timber Supply Area……………………………….…. 50 Figure 2. Forest licence volume harvested and scaled in the Fort St. James District…....…. 75 Figure 3. Forest licence volume harvested and scaled in the Vanderhoof District…............. 76 Figure 4. Forest licence volume harvested and scaled in the Prince George District….....… 77 Figure 5. Volume harvested by forest licences and scaled in the Prince George TSA….….. 77 Figure 6. Forest licence harvest in the Prince George TSA versus revenue collected…....... 80 Figure 7. Lodgepole pine harvested by forest licences relative to all species…………........ 81 Figure 8. Lodgepole pine harvested by forest licences in the Fort St. James District...…..... 82 Figure 9. Lodgepole pine harvested by forest licences in the Vanderhoof District……..….. 82 Figure 10. Lodgepole pine harvested by forest licences in the Prince George District…..… 83 Figure 11. Log grade distribution of lodgepole pine harvested by forest licences……..……84 Figure 12. Annual average stumpage paid for lodgepole pine versus all other species......... 87 Figure 13. Log grade distribution of all other species harvested by forest licences ……...... 88 Figure 14. Distribution of the scale of coniferous non-pine harvested in 2010 and 2011….. 88 vi Dedication and Acknowledgements I dedicate this thesis to my mentor, Rainer Muenter, and my father, William Robert Baerg. Both men have profoundly influenced my life and have helped shape the person I am today. Thank you. I began my study of forestry at Selkirk College in 1993. It was there that I was exposed to the concept that forest management is both a science and an art as it involves the interaction between forest ecosystems and human systems. Both the internal and interconnected dynamics of forest ecosystems and human systems are extremely complex, and navigating these systems requires knowledge that is both deep and broad. This has been my journey, and after 20 years of practicing various aspects of forest management and broadening my educational foundation I have come to realize that if we are to understand the complex interactions among natural and human constructed systems, then we need to cooperate in the production and dissemination of knowledge, beyond the constructs of academic disciplines. I have so many people to thank for not only making this journey possible, but so rewarding. First, to all of the interview participants, thank you for your contribution to this research, it was invaluable and meant a lot to me. I would also like to express my deepest gratitude to my committee; to my supervisor, Dr. Tracy Summerville, for giving me the freedom to discover my research topic and the guidance I needed at every stage to complete it and to Dr. Greg Halseth and Dr. Gary Wilson for their guidance, instruction, and timely advice. I would also like to thank my external examiner, Dr. Kenneth Coates, for offering your perspective and insight. I want to acknowledge other members of the UNBC community who have had an impact on my journey. To Dr. Catherine Nolin, thank you for opening my eyes to the kind of influence that social research can have in the world. To my fellow graduate students and NRES cohort, nobody understands graduate students like other graduate students. Your encouragement and friendship have made the difference between quitting and finishing, so thank you. Special thanks to Jessica Blewett for your unwavering support, perspective, and insight, and to Jeannine Randall and Aija White, for their endless encouragement, great company, and sound advice. Finally, I want to thank the British Columbia Public Service for their financial support through the Pacific Leaders Scholarship, and recognize all my supervisors and managers over the past five years at the Prince George Natural Resource District, thank you for giving me the flexibility to manage my work load to accommodate doing my Master’s degree. In closing I want to acknowledge and thank my friends, family and colleagues for all of your support, patience and encouragement, there are not enough words to express my love and gratitude. The best is yet to come! vii Chapter 1: Introduction 1.1. Introduction I have been interested in British Columbia forest policy since 1994, when the province was on the cusp of introducing legislation that promised to revolutionize forest practices in British Columbia. This legislation was the Forest Practices Code Act 1995. It was an exciting time. Other experts, like biologists, hydrologists, ecologists, and geotechnical engineers, were making significant contributions to the development of best management practices for the complex forest ecosystems that make up British Columbia. Less than ten years later, this revolutionary, but costly forest practices code was scrapped for a resultsbased regime brought to life by the Forest and Range Practices Act 2004. Prior to the Forest and Range Practices Act, the government repealed numerous policies in the Forest Act 1996 through the Forest (Revitalization) Amendment Act 200, to support and revitalize a struggling forest industry. With these most recent legislative shifts in forest policy, beginning in 2003, it has been my observation that there is a pervasive silence surrounding the direction that forest policy has taken and continues to take, which I find fascinating. Although the current economic forecast for the British Columbia forest sector is for stability (Bennett, 2016), there is a timber supply crisis unfolding in areas that were hardest hit by the mountain pine beetle. This crisis, in turn, is prompting a new round of legislative changes (FLNRO, 2012a). As well, spruce bark beetle populations are on the rise in the Prince George Timber Supply Area and other parts of the province (Westfall & Ebata, 2014). It has been estimated that 108,472 hectares of spruce have already been attacked by the spruce bark beetle in the northern half of the Prince George District and lower third of the 1 Fort. St. James District. Given the location of the outbreak, it is likely that it falls within the timber harvesting land base1 (THLB) (Westfall & Ebata, 2014). This latest bark beetle outbreak comes in the wake of the mountain pine beetle epidemic. Spruce and species other than lodgepole pine are expected to be the major contributors to the mid-term timber supply (Snetsinger, 2011). Yet, the Ministry of Forests, Lands and Natural Resource Operations, in its most recent service plan, declared its support for the forest industry and revealed its strategy to provide jobs and economic growth, where the stated goal of government is to: “[…] maximize the short and mid-term timber supply to support forest sector employment and industry sustainability […]” (FLNRO, 2016a, p. 5). 1.2. Purpose of the Research In 2003, the Liberal government made legislative changes in order to accommodate the provincial government’s economic strategy for revitalizing the forest sector, and building prosperity for British Columbia (MOF, 2003a). Although these policies were intended to provide the forest industry with greater flexibility, which in turn was to benefit the citizens of British Columbia (MOF, 2003a), these benefits have yet to be realized. The changes to forest policies made in 2003 could not guard against the externally driven perturbations that influence the boom and bust cycles of the lumber markets (Hoberg, 2010; Zhang, 2007). Furthermore, these policies protected and benefited large corporations, which account for the bulk of the forest industry, more than they did forest-dependent communities. The aforementioned legislative changes were made to the Forest Act 1996 via the Forest (Revitalization) Amendment Act 2003. Among other things, Bill 29 repealed Section 14 (f) of the Forest Act 1996, which required holders of forest licences to maintain a timber 1 “The timber harvesting land base (THLB) is the area of productive forest land available for timber harvesting” (Snetsinger, 2011, p. 10). 2 processing facility in the region where timber was harvested (Bill 29, 2003). Presently, there is concern about a rapidly declining timber supply in areas that have been hard hit by the mountain pine beetle and what that will mean to both the forest sector and the economies that depend upon it (ABCFP, 2011). The removal of the appurtenancy clause has generated fear in forest-dependent communities that forest corporations, via their job-producing processing facilities, will no longer be tied to the regions from which they extract timber (Hoberg, 2010). The removal of appurtenancy was only one in a suite of legislative changes designed to revitalize a struggling forest industry (Hoberg, 2010). Other changes permitted forest corporations to consolidate and subdivide forest licences, and gave them the ability to transfer agreements (Bill 29, 2003). These changes further strengthened the property rights of forest corporations on British Columbia’s public lands and makes compensation for breach of these rights less likely on the part of the provincial government (Luckert, Haley & Hoberg, 2011). However, they did not trigger the mass dumping of assets, in the form of sawmills, as some observers worried. Rather, the new rules facilitated further restructuring within the sawmilling sector, allowing it to remain viable. Though change is a constant factor in the resource sector, the pace and scale with which it is occurring in north-central British Columbia is increasing. The alarming reality is that resource extraction and export is happening at an unprecedented pace (Burda & Gale, 1998) in the midst of externally driven, market related perturbations that are also increasing in both frequency and severity (Halseth, Markey, Reimer & Manson, 2010). In the Northern Interior forest region of British Columbia, the timber resource is becoming depleted (FLNRO, 2012a). It also appears that there has been a lack of diversification within forest dependent communities (Horne, 2009). Innis (1933) argued that Canada was locked into a staples 3 economy and that its pathway was determined by its absolute dependence upon its natural resources, which is still true for British Columbia (Bowles, Lytle & Paterson, 2002; Hayter, 2000; Markey, Halseth & Manson, 2008). Freudenburg (1992) explains that commodity driven economies are extremely vulnerable in a globalized economy. Nonetheless, the government of British Columbia continues to look to other resources to exploit and replace timber as the province’s primary commodity and revenue generator (EMNG, 2013). 1.3. Research Questions The Northern Interior forest region of British Columbia is facing a timber supply crisis. Though this crisis was inevitable, its arrival was accelerated by the devastation caused by the mountain pine beetle epidemic (Snetsinger, 2011). Proper mitigation of this timber supply crisis is vital given that the area hardest hit exists in some of the most forest dependent regions in British Columbia’s interior. This crisis provides an opportunity and context to demonstrate how forest corporations have adapted their business, in terms of where they process the timber they cut, as a result of the relaxed regulations brought in by the Forest (Revitalization) Amendment Act 2003 and to critically evaluate long standing harvesting, revenue generation, and timber allocation decisions that continue to dominate British Columbia forest policy. This thesis will argue that while the government’s relaxation of restrictive legislation and promotion of a good business environment has assisted forest corporations and facilitated the temporary survival of the forest industry, the adoption of neoliberal economic theory into the context of the British Columbia forest industry has not strengthened British Columbia and brought prosperity. Rather, the application of this theory has weakened the forest economy as a whole by perpetuating the status quo of the timber supply being 4 controlled by forest corporations. Further, though the assistance to strengthen the forest industry has resulted in a more stable forest economy, this forest economy contributes a much smaller share than it once did in the provincial economy. At the same time, the people of British Columbia have not received fair compensation for the publically owned timber that drives this forest economy. Finally, this thesis will argue that the long term viability of a forest industry in British Columbia will likely require a more radical shift in forest policy, which could include a different arrangement of forest tenures. My research examines the policy changes of the Forest (Revitalization) Amendment Act 2003, invoked by the Liberal government, that were aimed at revitalizing the forest industry. Although little has changed over time with regards to British Columbia’s core forest policies (Haley & Nelson, 2007; Pearse, 2001; Rayner, Howlett, Wilson, Cashore & Hoberg, 2001), I suspect that the context and driving forces which shape forest policies have changed. It is the complexity of forest policy that warranted a holistic approach to this research. Such an approach not only examines the internal and external forces that drive government policy, but uncovers the interconnection between policy choices made and the impacts of those policies on both the intended and unintended targets. For these policy changes in question, the intended target is the forest industry and the unintended targets are the environment and human communities. My data collection was guided by my research questions, which are as follows: 1) What was the context that motivated government to make the 2003 forest policy changes?  Were these forest policies already in place in other jurisdictions, or were they unique to British Columbia?  What was occurring globally at the time that may have influenced this specific direction? 5  Were potential negative outcomes for the environment and/or human communities considered? If so, what were the mitigation strategies?  Were these changes related to the recent change in government, from the NDP (political left) to the Liberals (political right)? 2) How have forest corporations responded to the 2003 policy changes over time?  Has the number and distributions of sawmills and fibre flow changed over time?  Did these changes in forest policy help industry? If so, in what ways?  Did the forest industry participate in the formulation of the policies? To answer these questions regarding the forest industry, I collected and analysed data regarding timber harvesting, sawmill distribution, lumber markets, as well as employment and economic trends. The data that I collected was restricted in both time and space. The time frame in which data was collected was limited to the period between 1997 and 2011. The focal point on the British Columbia landscape was the Prince George Timber Supply Area. Next, I conducted semi-structured interviews with key informants who were involved in the forest industry within the Prince George Timber Supply Area between 1997 and 2011, and who had specific knowledge regarding the various stages of the forestry policy formation that came into being under the Forest (Revitalization) Amendment Act 2003. The purpose of the interviews was to solicit their explanations as to the findings of my secondary research of official records and documents. The remaining structure of this thesis begins with Chapter 2, where I lay a foundation of historical background of forest policy and provide context by discussing how forest dependent communities in northern British Columbia and neoliberal ideology are related to forest policy. This discussion will be followed by a review of the literature in Chapter 3 where I situate my research by discussing the changing dynamics of forest policy and the forest industry in British Columbia. Next, in Chapter 4, I present the methodology 6 and methods I used to answer my research questions and discuss the ethical considerations I had in undertaking this research, my choice of using a case study as my research design, my use of a mixed methods approach to collect data, and my methods of data analysis. In Chapter 5, I present the results of my research which provide evidence as to how the rationalization of the forest industry unfolded and its impact on employment, where timber was harvested and processed, how much revenue the Crown received for the timber, and what people connected to the forest industry had to say about these topics and the events surrounding the legislative changes to forest policy in 2003. In Chapter 6, I discuss the findings of my research and how they answer the research questions. Finally, I conclude the thesis in Chapter 7 by suggesting possible forest policy pathways for British Columbia, discussing the limitations of my research, and making recommendations for future research. 7 Chapter 2: Background and Context 2.1. Introduction To understand the present state of the forest industry in the Northern Interior forest region of British Columbia it is important to discuss aspects of British Columbia forest policy, forest dependence in relationship to the communities situated there (Bowles et al., 2002; Markey et al., 2008; Power, 2006; Waggener, 1977), and the relatively recent shift to a neoliberal political ideology in British Columbia (Summerville, 2010; Young, 2008). All of these topics underpin both the historical and lived realities of communities located within the Northern Interior forest region and are useful for understanding the significance of the shifts that have occurred within this forest industry. Moreover, to understand the changes made to legislation in 2003 by the Forest (Revitalization ) Amendment Act 2003 (Bill 29) it is helpful to provide the context in which the forest industry was working in British Columbia prior to 2003 and the conditions which may have led the government to make these policy choices. It is also useful to discuss the events that followed the changes brought in by Bill 29 because they provide the context for this research. 2.2. Background British Columbia has been described as a unique forest jurisdiction in that 95% of its land base is public land (Luckert et al., 2011). Thus, because the majority of the land base is public land, when the expansion of the forest industry and the conservation of the resource are juxtaposed, the governments of British Columbia have historically, with a few exceptions, ensured that ownership of land remained with the Crown (Clark 1985; Marchak, 1983). Given that the majority of timber harvested in the province is from public land, and that rights to Crown timber are allocated through a variety of timber tenures, discussion of forest 8 policy regarding how timber has been allocated historically is warranted. While it is evident that the evolutions of several key forest policies surrounding Crown timber, its disposition, its regulation, and its price, have strongly influenced the development of British Columbia’s forest economy (Hayter, 2000; Marchak, 1983), changes made to forest policy have emerged from the midst of conflict as private firms benefit from utilizing a public resource (Hayter, 2000; Hoberg, 2001; Howlett, 2001; Marchak, 1983). For better or worse, changes made to forest policy have served the purpose of the time (Clark, 1985). The formation of a timber tenure system in British Columbia was undertaken as a means of timber allocation, but the underlying goal was to create favourable economic conditions that would promote and support a forest industry (Howlett, 2001). The forest industry was supported by making the timber supply available to investors (Howlett, 2001). Though the number and types of timber tenures have expanded over time, the initial goal remains unchanged (Howlett, 2001). British Columbia was successful in attracting many multinational forestry corporations, with large amounts of external capital (Drushka, 1999); these players obtained the largest forms of timber tenures available (Drushka, 1999; Hoberg, 2000), the forest licence and the tree farm licence (Howlett, 2001). In 1996, 82% of the provincial allowable annual cut (AAC) came from tree farm licences and forest licences (Hoberg, 2000). The control over the provincial timber supply is not a recent development; rather, its roots go back to the beginning, when the alliance between government and private sector firms in the forest industry was forged (Howlett, 2001; Marchak, 1983). From a forest management standpoint, the history in British Columbia can be divided broadly into the unregulated (1858-1945) and regulated (1945 to present) periods. That is not to say that there were no regulations pertaining to forest management in during 9 the unregulated period, rather, that regulations were sparse and, more importantly, that timber rights were less secure. Between 1858 and 1865, the only way to access timber in British Columbia was to purchase the land on which it grew (Clark, 1985). In 1865, the government made changes in policy which enabled it to lease the rights to Crown timber (Clark, 1985). Leasing rights allowed the government to retain ownership of the land while simultaneously encouraging a timber manufacturing sector (Clark, 1985). Interestingly, in order to obtain these rights the lessee had, “to own and operate a sawmill” (Clark, 1985, p. 5). By 1905, under the leadership of Premier Richard McBride, British Columbia was desperate for revenue and began to sell off timber rights for a pittance (Clark, 1985). Coincidently, the broad relaxation/removal of regulation occurred at the same time that the conservation movement was gaining significant momentum in the United States (Howlett, 2001). An unprecedented wave of timber speculation followed and subsequent concentration of amassed holdings (Howlett, 2001) because these special licences were renewable and transferable (Clark, 1985). The era of timber speculation ended in December of 1907 and sparked the first Royal Commission, led by F.J. Fulton (Clark, 1985). Though this policy had generated substantial revenue for the province, Clark (1985) contends that it also set the tone for the management of the forest resource. Forest policy in British Columbia further developed through the recommendations and findings of several Royal Commissions, which were used by government as a means of both engaging and educating the public (Clark, 1985). The first Royal Commission was headed up by F.J. Fulton in 1910 and can be credited for the formation of the Forest Act of 1912 (Howlett, 2001). The templates for volume and area based timber tenures stem from the Royal Commission on Forestry led by Chief Justice Sloan in 1945 (Howlett, 2001). One of 10 the outcomes of the Sloan Royal Commission of 1945 was the creation of the Private and Public Working Circles (Hayter, 2000). The Private and Public Working Circles, essentially management units, were the foundational structures for timber administration that were intended to bring both public and privately held timber rights under sustained yield forest management (Pearse, 1976). In order to bring privately owned rights to timber into the sustained yield model, compensation had to be given by the provincial government to the holders of these rights. Membership into the Private Working Circle was offered to those who held Crown land grants and timber leases (Pearse, 1976). In exchange for agreeing to regulate the harvest over the privately held area, the government offered additional Crown land which would be amalgamated with the existing private holdings into one area based licence (Clark, 1985). So, the Private Working Circle consisted of these area based licences, which are now known as tree farm licences (TFL) (Clark, 1985). The Public Working Circle consisted of the remaining Crown timber land base and was administered by government (Pearse, 1976). From the Public Working Circle management unit, timber was allocated through volume based licences (Clark, 1985), the largest of which is known today a forest licence (FL). The Public Working Circle was later divided up into Public Sustained Yield Units (PSYU), which then became the Timber Supply Areas (TSA) (Clark, 1985). The Private Working Circle licenses afforded their holders tremendous property rights, which they could leverage as assets (Howlett, 2001). The government then gave preferential treatment in awarding the volume based licences in the Public Working Circles to those who had mills or secondary manufacturing, which enabled many of the players in the Private Working Circles to gain a foothold in the Public Working Circles (Clark, 1985). Thus, 11 those who controlled the Private Working Circles quickly gained control of the Public Working Circles (Clark, 1985). The management regime of sustained yield, introduced by the Sloan Royal Commission, was seen as a means of managing the timber supply (Marchak, 1983; Prudham, 2007). Sustained yield was also instituted to provide government with revenue, industry with timber, and communities with jobs (Prudham, 2007; Rayner, 2001). The normal forest is the underlying concept of sustained yield, which is the systematic conversion of primary forests, which are uneven-aged old growth, to even-aged ones (Zhang & Pearse, 2011). The strategy is to divide up the known timber supply, either by volume or area, over a projected time span of growth (at least 80 years in British Columbia) and harvest at a rate that liquidates the virgin forests over the allotted time (Zhang & Pearse, 2011). In the case of British Columbia, the timber supply is given out by volume and tracked on an annual basis, and the annual allotment of it is known as the allowable annual cut (AAC) (Rayner, 2001). One of the downsides of the strategy of sustained yield is that the total volume available in the harvest of virgin forests is far greater than what will be available in the subsequent harvests from regenerated forests (Zhang & Pearse, 2011). The volume difference between the initial and subsequent harvest is so large, it has been given the term ‘fall down’ (Zhang & Pearse, 2011). Another problem is that by employing a sustained rate of harvest there will also be a gap in available timber supply between the period when the old growth supply is exhausted, and the subsequent growth becomes available (Zhang & Pearse, 2011). The mid-term timber supply refers to the time period that follows once the old growth (natural and original timber) has been liquidated, but before plantations (second growth) are mature. Subsequently, the amount of timber available during this period is scarce. In British 12 Columbia, we have been approaching the fall down since the mid-1990s (Rayner, 2001). Although the recent mountain pine beetle epidemic has pushed the central interior of the province to this point (MOF, 2003b), the magnitude of the drop is much larger than anyone anticipated and there is not much chance of mitigating the impacts to the timber supply given almost all of the mature lodgepole pine, a significant contributor to the timber supply, is now dead (Westfall & Ebata, 2010). The development and expansion of British Columbia in the post war period was in large part due to the social-economic beliefs of the government led by W. A. C. Bennett (Bowels, Lytle & Paterson, 2002; Markey et al., 2008; Summerville, 2010; Wedley, 1990). The focus then was on creating an inviting investment environment for industries, which in turn would both establish and bring stability to communities (Bowles et al., 2002). W. A. C. Bennett was able to draw investment to the province by creating infrastructure, generating hydro-power, and establishing appurtenancy as part of the Forest Act (Markey et al., 2008; Summerville, 2010). The appurtenancy clause was a formal agreement between the Provincial government and forest companies where, in exchange for timber rights, forest companies holding major volume based forest tenures were to maintain timber processing facilities near to the origin of timber harvesting ; further, it is considered by many as a social contract, designed to bring stability and growth to forest resource dependent communities in British Columbia (Haley & Nelson, 2007; Luckert et al., 2011; Jackson & Curry, 2002). In northern British Columbia the majority of communities are small, and many of them are largely dependent upon, what Harold Innis (1933) described as staples economies. Sustained yield and the appurtenancy clause were established during the era of province building (Wedley, 1990) and understood then as a critical factor in promoting community 13 stability, and subsequently increased dependence on the forest resource (Power, 2006; Waggener, 1977). That these economies are becoming increasingly threatened in the context of the global economy makes the stability issue extremely important for people who live and work in northern British Columbia (Markey et al., 2008). The economic dependence of northern British Columbia communities upon natural resources is a well described and established phenomenon and is seen as a barrier to economic diversification (Bowles et al., 2002). Additional barriers include lack of finances, lack of infrastructure, conflict within the community, capacity within the community, lack of human capital, social cohesion and inclusion (Ryser & Halseth, 2010). A function of the sustained yield timber supply regime was that it was supposed to insulate forest dependent communities from the instability of the forest product market business cycle over the long run (Zhang & Pearse, 2011). However, attempting to control a market that is driven by demand by controlling supply does not work (Zhang & Pearse, 2011). For example, between 1997 and 2008, a number of events occurred that had a negative impact on British Columbia’s lumber market. They were: the Asian economic crisis (Hoberg, 2010); the rise in value of the Canadian dollar (Hoberg, 2010); the expiry of the Softwood Lumber Agreement with the United States (Zhang, 2007); the downturn, and subsequent crisis in the United States housing market (Hoberg, 2010); and, the general instability of national economies globally (Luckert et al., 2011). Further, volatile fluctuations in lumber markets have been shown to have negative impacts on forest sector employment, sometimes within as little time as two months (Byron, 1978). Add to these market issues the increasing global nature of the markets and the influencing sway of neoliberal ideology, and you have a seedbed for change in forest policy. 14 In British Columbia, the pressure of globalization and the dominant tenets of neoliberal ideology have been influencing the direction of government policies since the 1980s (Summerville, 2010). In its most basic definition, neoliberalism is an ideology that is based in free market economics (Summerville, 2010), which is a stark departure from the Keynesian economic model that dominated the post-war era in British Columbia (Hayter, 2000). One of the primary tenets of neoliberalism is that market freedom and greater prosperity are constrained needlessly by government subsidies, regulations, social policies, and general bureaucratic interference (Harvey, 2005; Summerville, 2010). The rise of neoliberal ideology is a phenomenon that can be seen globally (Harvey, 2005; Humphreys, 2009). Notable leaders who helped usher in neoliberal ideology are Margaret Thatcher in the United Kingdom in 1979, Ronald Reagan in the United States in 1980 (Harvey, 2005; Savoie, 1994), David Lange in New Zealand in 1984 (Roche, 1990), and Brian Mulroney in Canada in 1984 (Graefe, 2002; Savoie, 1994). While neoliberal ideology guided the dominant politics in the aforementioned jurisdictions, exactly how neoliberal agendas played out and to what extent varied dramatically in each country (Harvey, 2005) and even within Provincial jurisdictions of Canada (Graefe, 2002). The advancement of neoliberalism in Canada can be described as gradual, through the neoliberal agenda of fiscalization (Graefe, 2002). The term fiscalization: “refers to periods when financial concerns, especially considerations of expenditure restraint and deficit reduction, dominate deliberations on setting public policy priorities and contemplating social reforms” (Rice & Prince in Graefe 2002, p. 20). The neoliberal agenda was advanced amidst the pressure of globalisation and an economic recession via the empowerment of the Department of Finance in Canada’s provinces (Graefe, 2002). He also notes that there is a 15 lack of distinction between political parties, as fiscalization was applied in different provinces under the leadership of differing political parties. For example, these fiscalization policies were applied in Saskatchewan by Grant Devine (Conservative 1982-1991), in Ontario by Bob Rae (NDP 1990-1995), in Quebec by Robert Bourassa (Liberal 1985-1994), and in New Brunswick by Frank McKenna (Liberal 1987-1997) (Graefe 2002). In his study of provincial jurisdictions in Canada, Graefe (2002) notes that neoliberalism has been applied unevenly because the form and the extent to which different governments have adopted the ideology has varied. Harvey’s (2005) work on neoliberalism is based on the premise that neoliberalization is a “political project to re-establish the conditions for capital accumulation and to restore the power of economic elites” (p. 19). With respect to class, Harvey (2005) explains that, “[w]hile neoliberalization may have been about the restoration of class power, it has not necessarily meant the restoration of economic power to the same people” (p. 31). Harvey (2005) uncovers its origin to be with the Mont Pelerin Society, whose ideology had its roots within liberal ideals, such as personal freedom, but with “adherence to those free market principals of neoclassical economics” (p. 20). Members of the Mont Pelerin Society included some of the most renowned economists of the 20th century including, Ludwig von Mises, Friedrich Hayek, and Milton Friedman (Harvey, 2005). However, as Harvey (2005) points out, contradictions within the neoliberal position are many. In fact, neoclassical economics and liberal ideals are frequently diametrically opposed (Harvey, 2005). Therefore, he warns to take heed of, “the tension between the theory of neoliberalism and the actual pragmatics of neoliberalization” (Harvey, 2005, p. 21). Harvey (2005) says that while 16 neoliberalization has not been very effective in increasing global capital accumulation, it has been widely effective in either restoring or creating economic elite power. 2.3. Context In 1997 the forest industry was struggling under the complex and expensive environmental regulatory regime of the Forest Practices Code of British Columbia Act 1995, a quota system that regulated the flow of softwood lumber exports into the United States, and the unprecedented high level of stumpage charged for timber, known as super-stumpage that was rendered to the Crown (Grafton, Lynch & Nelson, 1998; Hayter, 2000). Earlier, in 1987, a new means for calculating stumpage was adopted by the province to appease the authors of a memorandum of understanding with the United States. This new method was known as comparative value pricing (CVP), which was essentially an ad valorem tax2 which was calculated using the market selling price of lumber and did not account for logging expenses (Grafton et al., 1998). Later in 1994, under an NDP government, the system was revised to increase the ad valorem at a higher rate as the market selling price of lumber increased, which was known as super-stumpage (Grafton et al., 1998). Meanwhile, the policies and regulatory regime created by the Forest Practices Code of British Columbia Act 1995 significantly increased the cost to harvest timber and over time eroded the ability of companies to respond effectively to changes in market conditions (Hayter, 2000; Pearse, 2001). However, by 1997 the Asian economic crisis had negatively impacted companies like MacMillan Bloedel that primarily exported lumber to Japan. These companies suddenly found themselves without a market and insufficient quota to ship their product into the United States (Pearse, 2001; Zhang, 2007). 2 An ad valorem tax is a tax that is based on value. For example, municipal property taxes are based on real estate values (Klemperer, 1996, p.279). 17 There was recognition by the provincial government that the forest industry was in trouble and it began to institute changes to help streamline the costly regime created by the Forest Practices Code of British Columbia Act 1995 (MOF, 1998). However, these changes were deemed insufficient by the Council of Forest Industries (COFI), which published its own ideas for improving the forest industry. The central theme of this document was that the forest industry needed to be globally competitive (COFI, 1999). The NDP, who were in government at the time, took note of the suggestions made by the forest industry and in 1999/2000 began the shift towards a more results-based model of forest practices than the prescriptive regime of the Forest Practices Code of British Columbia Act 1995. The new model was to be tested using five pilot projects (MOF, 2000). Further, the government was also beginning to explore a market-based pricing system that would generate stumpage values, knowing that stumpage values would be a critical factor for negotiating a new softwood lumber agreement with the United States (MOF, 2000). Despite the government’s efforts to relieve forest companies of the excessive and costly regulatory requirements, it would not alter its course in the way it collected stumpage (Pearse, 2001). Hence, companies struggled to balance the high cost of production with their softwood lumber quotas amid declining lumber prices in the United States (Pearse, 2001). All the while, the British Columbia forest industry faced an overall increase in competition from other timber producing regions (MOF, 2000). Market conditions for the forest sector continued to worsen through 2000 and into 2001, and to make matters worse, the Softwood Lumber Agreement expired on March 31, 2001 (MOF, 2001b). By April, the Coalition for Fair Lumber Imports in the United States had filed a petition with the Department of Commerce to investigate countervailing duties 18 and anti-dumping claiming that British Columbia lumber manufacturers received subsidies through stumpage collection and the prevention of log exports, and that lumber was dumped onto the market in the United States below cost (MOF, 2001b). The situation with the United States had become critical for the British Columbia forest industry, and while the provincial government tried to assist the federal government with negotiations, several forest corporations engaged in litigation with the United States (Zhang, 2007). In May of 2001 the BC Liberal Party was elected with a resounding majority. British Columbia was poised to embark into a “New Era,” in which the road to prosperity was guided by three pillars: fiscal responsibility; free enterprise; and, equality of opportunity and responsibility (BC Liberal Party, 2001). Meanwhile, in the lodgepole pine forests of British Columbia the largest mountain pine beetle (Dendroctonus ponderosae) epidemic in history (Westfall, 2001) began to seriously threaten the timber supply and the economies dependent upon it (MOF, 2003b). As the beetle infestation began to expand across much of the Northern Interior Plateau, government, the forest industry, and communities worked together to form strategies for managing the spread and damage caused by the epidemic (FLNRO, 2012b). As the epidemic progressed, these strategies shifted from mitigation of damage and spread to capturing losses (FLNRO, 2012b). The new government set to work on its mandate to assist the forest industry by giving it the flexibility to transform itself and become more competitive (MOF, 2002b). The government focused on changing the regulatory environment, such as moving to a marketbased pricing system for calculating stumpage and changing forest practices to a resultsbased, rather than a prescriptive-based regime (MOF, 2002b). However, these changes took time so the forest industry and the government had to endure the market restrictions and 19 retaliations imposed by the United States (Hamilton, 2001). Still, the focus of the government remained steadfast: to reduce government influence over the business decisions made by the forest industry. Reducing government influence over the business decisions of the forest industry was the driving ideological force behind the impetus for legislative change (Hayter & Barnes, 2012; Hoberg, 2010). Amidst the first strategies for managing the epidemic were concerns about the regulatory and operational constraints that would prevent effective containment of the beetle spread (MOF, 2001a). Government paid considerable attention to policies that would enable the forest industry to maximize the harvest of beetle infested lodgepole pine by temporarily increasing the allowable annual cuts (AACs), repealing constrictive regulations, and subsequently directing the harvest to timber supply blocks considered to be the most heavily infested (MOF, 2003b). Increasing the availability of timber for harvest was one challenge, having the incentive to harvest was quite another. There were perceived limitations for processing, which included the milling capacity of industry (FLNRO, 2012b), adverse market conditions, and the constraints of the Softwood Lumber Agreement with the United States (FLNRO, 2012b; Zhang, 2007). One response by government was to change forest policy so that the industry could be more flexible in its response to market signals (MOF, 2003a). The most recent wave of changes to forest policy began in 2003 with the enactment of the Forest (Revitalization) Amendment Act, where major changes to the Forest Act 1996 were made in order to accommodate the provincial government’s economic strategy for revitalization of the forest sector (MOF, 2003a). These changes were part of the Liberal government’s quest for prosperity; for example, Bill 29 – 2003, the Forest (Revitalization) Amendment Act 2003 repealed Section 14 (f) of the Forest Act 1982, which required holders 20 of forest licences to maintain a timber processing facility in the region where timber was harvested. Legislative changes continued in 2004 with the Forests and Range Practices Act (FRPA). Considered by industry as overly regulated and unduly prescriptive, the Forest Practices Code of British Columbia Act 1995 was slowly phased out in favour of the new results-based forest and range practices regime (MOF, 2002a). Despite the changes made to forest policy, by 2008 a severe downturn in the forest economy was experienced throughout British Columbia (BC). During the fervor of an accelerated harvest of beetle killed pine in the northern and central interior, the BC lumber manufacturing industry suffered severe blows to its lumber markets via the ongoing Softwood Lumber dispute with the United States, from 2001 to 2004 (Zhang, 2007), a drastic drop in housing starts in the United States in 2005, the rise in value of the Canadian dollar in 2007, and the housing market crisis in the United States in 2008 (Hoberg, 2010). By 2008, sawmills throughout British Columbia began to close, some temporarily and some permanently (Hamilton, 2008), which threatened the very existence of many communities (Martin, 2013). The temporary loss of BC’s largest lumber importer prompted the provincial government to pursue Chinese lumber markets more aggressively (MFR, 2008a). These inroads made by government into China assisted the forest industry by providing markets so that mills remained viable (Bowles & MacPhail, 2016; FLNRO, 2012b). In February of 2012 a report was released by the Auditor General of British Columbia criticizing the lack of timber stewardship by the Ministry of Forests, Lands and Natural Resource Operations (AGBC, 2012). In May, a Special Committee on Timber Supply was formed and they toured the communities in the Northern Interior forest region of the province seeking input from citizens and stakeholders alike. This committee released its 21 report in August, in which it made numerous recommendations in relation to the mid-term timber supply crisis (SCTS, 2012). This report was followed in October by a response from the Ministry of Forests, Lands and Natural Resource Operations (FLNRO) that addressed a number of the Special Committee’s recommendations to the Legislative Assembly. The FLNRO responded by saying:  that they will conduct a review of sensitive areas in terms of potentially expanding the timber harvesting land base;  that they will investigate the use of intensive silviculture practices as a means of growing more fibre;  and, that they will create legislation that would enable the conversion of volume based forest licences into area based tree farm licences (FLNRO, 2012a). Provincially, it is estimated that approximately 800,000 hectares of timber was damaged by the mountain pine beetle in 2001; the current cumulative estimate, observed in 2011, is a staggering 18.1 million hectares of timber damaged by the mountain pine beetle, which is still expanding north (FLNRO, 2012b). In the wake of the devastation caused by the mountain pine beetle, and the large scale salvage harvesting that has followed, is a landscape that has changed. The patchwork of clearcut areas, plantations, and mature timber has been replaced by extensive clearcut areas and remnant stands of rapidly decaying lodgepole pine (FLNRO, 2012a). Throughout this landscape are numerous communities, that are a part of this geography and which have been, and will continue to be, impacted by these agents of change (FLNRO, 2012a; Parkins & MacKendrick, 2007; Patriquin, Wellstead & White, 2007). Unlike older communities located in the southern reaches of British Columbia, these communities were established during the province building era and, as such, were purposefully linked to the forest economy (Bowles et al., 2002; Markey et al., 2008). Consequently, the majority of communities in the northern interior of British Columbia 22 remain highly dependent upon the forest economy (Halseth et al., 2010; Horne, 2009). The forested landscape has changed and all of the things that depend on it, from people and communities to flora and fauna, have been forced to adapt. 23 Chapter 3: Literature Review 3.1. Introduction The forest industry has long been an economically important sector for British Columbia (Hayter, 2000; Marchak, 1993). Not only is the industrialized forest sector important, but so are the forest policies that govern it, as they have the capacity to impact aspects of the province’s natural environment and its human communities (Burda & Gale, 1998; Markey et al., 2008). It is understandable then why major forest policy developments of recent decades, such as the Forest Practices Code of British Columbia Act 1995, the Forest and Range Practices Act 2004, and to some degree the Forest (Revitalization) Amendment Act 2003, have drawn research interest from a diverse group of scholars, from disciplines ranging from biology to sociology. While the Forest Practices Code of British Columbia Act 1995, and the Forest and Range Practices Act 2004 were/are about the regulation of forest operational and management practices, the Forest (Revitalization) Amendment Act 2003 brought about changes to the conditions by which private firms must abide in order to acquire and/or maintain long term rights to harvest public timber. The focus of my research concerns how the forest industry responded to the specific forest policy changes brought about by the Forest (Revitalization) Amendment Act 2003. I approached my research from the understanding that forest management involves complex interactions between natural ecosystems and human systems and as such requires insight from the disciplines of both the natural and social sciences. 24 In my broad survey of the literature regarding forest policy and the forest industry in British Columbia, I focused specifically on the literature that spoke to change dynamics in forest policy and the forest industry. This literature draws primarily from four disciplines: political science; economics; sociology; and geography. Scholars from these disciplines have sought to understand the dynamics of forest policy change as well as change dynamics within the forest industry using a variety of theories, which are rooted in their specific disciplinary ontologies. Despite the disciplinary focus, at least two separate but related paradoxical observations continued to surface in this literature, in one form or another. The first ambiguity described in the literature is that despite various changes in forest policy, business carries on as usual. The second is that certain forest policies persist though they have been shown not to be achieving their intended goals. It is the pursuit to understand these contradictions surrounding British Columbia forest policy that makes this literature important to my research. The literature of change dynamics in forest policy and the forest industry lends itself to understanding the unique nature of the British Columbia forest sector as well as identifying potential factors that have influenced the dynamics of change within it over the past several decades. I will begin by discussing what scholars have looked at in terms of forest policy change. This discussion will be followed by a review of the various responses of the forest industry to changes in forest policies in. Finally, I will discuss what scholars have identified as enduring forest policy legacies. 3.2. Forest Policy Change in British Columbia Forest policy in British Columbia has traditionally been studied by scholars within the social science disciplines of political science and economics (Lindquist & Wellstead, 25 2001). Policy as it is used in this research is defined as, “the regulation of how, when, and where” (Hessing, Howlett & Summerville, 2014, p. 3). Broadly speaking, policies can be developed by any organization or group, however, for this inquiry the regulations are developed by the provincial government (Hessing et al., 2014). Lindquist and Wellstead (2001) contend that prior to the late 1980s, scholars studying forest policy change in the province assumed a rather simplistic state of policy making between the provincial government and the forest industry, and as such, little attention was paid to the actual development of forest policy in British Columbia. Rather, the efforts of political scientists were focused on how a policy was implemented and economists evaluated the effectiveness of the outcomes (Lindquist & Wellstead, 2001). This section will show how the approach to analyzing forest policy has changed over the past few decades. Until the 1990s, the drivers of forest policy change in British Columbia remained relatively consistent due to the strong relationship between the forest industry, government, and labour (Hagerman, Dowlatabadi & Satterfield, 2010; Prudham, 2007). The era of implied simplicity came to a close in the latter part of the 1980s in British Columbia when the policy arena became more contested (Lindquist &Wellstead, 2001), triggered by the coordinated effort of environmental non-governmental organizations and First Nations towards a common interest in the environment (Lertzman, Rayner & Wilson, 1996). As the complexity in forest policy dynamics increased in British Columbia, traditional approaches to policy analysis became problematic so political scientists began to apply more robust theoretical frameworks such as policy networks/communities and agenda setting theories to better understand the forest policy dynamics occurring in British Columbia (Lindquist & Wellstead, 2001). 26 Early investigations of theoretical frameworks to explain forest policy change in British Columbia began to emerge in the mid-1990s. What began as a controversy between the roles of ideas versus interests in policy dynamics (Howlett & Rayner, 1995; Lertzman et al., 1996; Hoberg, 1996) was later understood as an integration of ideas and interests (Howlett & Ramesh, 1998). Similarly, Hoberg and Morawski (1997) discovered that using policy network/sector theories to explain the dynamics occurring between the government and First Nations regarding forest policy in Clayoquot Sound were inadequate. Moreover, in his investigation of various agenda setting theories used to explain particular forest policies in British Columbia, Kamieniecki (2000) found that these theories could not account for the complexity of forest policy dynamics occurring in British Columbia. To overcome these deficiencies in understanding the complexity of forest policy dynamics, political scientists began to broaden their lens of inquiry, reaching beyond the specificity of their own discipline. This research informed the broader context in which British Columbia forest policy was taking place. In their comparative study of forest jurisdictions, Cashore and Vertinsky (2000) found that a firm’s response to an external pressure depends on the type of regulatory structure it is under as well as the type of relationship it has with the government, demonstrating the important role of institutions. Similarly, Bernstein and Cashore (2000) show how the globalization of economies and communication and the internationalization of trade have provided international actors and institutions with direct influence over domestic forest policies, citing the ability of the United States to directly influence forest policy through the Softwood Lumber Agreement. In the field of economics, forest economists draw from rationale choice theory and the policy instrument framework to guide their analyses of specific forest policies to 27 determine their economic efficiency (Lindquist & Wellstead, 2001). For example, in their assessment of British Columbia’s policy shift that resulted in British Columbia Timber Sales (BCTS) replacing the Small Business Forest Enterprise Program (SBFEP), Niquidet, Nelson and Vertinsky (2007) modeled characteristics of the sales that were thought to impact bid prices. They found that the new program of timber tenure delivery through BCTS to be more efficient than its predecessor, the SBFEP, though they suspected that resource rents may have been transferred from rural communities to the provincial government (Niquidet et al., 2007). Rational choice theory assumes that: “people’s preferences are static, society is a mathematical aggregation of homogeneous rational agents, public inputs are through market signals, and there is no role for any institution other than the market” (Kant, 2003, p. 40). As such, reasoning that is based on rational choice theory can be problematic as it involves minimal input parameters that results in an explanation that is over simplified (Kant, 2003). Oversimplification in the analysis of forest policy instruments is starting to be recognized and evidence demonstrating its limits can be found. For example, a long standing assumption by forest economists was that by strengthening timber tenure rights in British Columbia firms would make longer term investments in growing trees (Luckert, 2007; Zhang & Pearse, 1997). However, work by Arnot, Luckert and Boxall (2011) found that because there is no firm definition of tenure security, tenure security cannot be linked to economic behaviour. Further, Bogle and van Kooten (2013) have shown that government and firms have conflicting agendas when it comes to growing trees because firms view the task as a liability and constraint as opposed to an investment. While the above noted scholars have all made valuable contributions to the discourse regarding British Columbia forest policy, perhaps the most significant contribution, 28 in terms of understanding the dynamics of forest policy change in British Columbia, was made by Rayner et al. (2001). They argue that while the forest sector is made up of many sub-sectors, not all of them can be given equal scrutiny as some sub-sectors can be determined to be critical in the sense that they exhibit substantial influence over the dynamics occurring within the entire sector (Rayner et al., 2001). In their analyses of British Columbia forest policy changes in the 1990s, they identified three factors that affect policy change in the forest sector. First, that subsectors are organized around a small number of issues which attract specific policy networks, as policy networks are interest based. Second, that subsectors can be strongly linked together to support a policy goal, such as the policy of sustained yield. Third, those critical subsectors occupy a position of privilege, in that they “constrain policy development” (Rayner, et al., 2001, p. 329). They identified the tenure and timber supply subsectors as critical because of their direct link to the wellbeing of the forest industry (Rayner et al., 2001). In their research, Thielmann and Tollefson (2009) also found that numerous attempts to change land use policy were thwarted by the commitment of successive governments to uphold the forest industry and forestry jobs. 3.3. Responses by the Forest Industry to Policy Change in British Columbia The historical relationship between the government of British Columbia and the forest industry has been well documented (Marchak, 1983), which had become firmly established following the Royal Commission on Forest Policy led by Sloan in 1945 (Prudham, 2007). But, this relationship began to be tested by the late 1980s when the environmental movement and First Nations began to effectively challenge the historic relationship between government and the forest industry (Cashore & Vertinsky, 2000). The following section will show how the forest industry in British Columbia has adapted to forest policy change in order 29 to survive. This section will also describe the internal and external contexts of these adaptations as the forest industry was not only responding to British Columbia forest policy, but to the forest industry’s place within the global economy. In their model to predict how firms will respond to external pressures, Cashore and Vertinsky (2000) employ both policy network and regulatory typologies; they classify British Columbia’s regulatory style as non-legal/discretionary and its forest policy network structure as a clientele-pluralist network. In this model, the regulatory style refers to the role of the courts, where the clientele-pluralist network is described as a network in which: business interests are policy participants within the sub-government; state agencies rely on business interests for advice; business interests essentially have a veto over policy change; and, other organized interests are relegated to roles as policy advocates (Cashore & Vertinsky, 2000, p. 5). To demonstrate the progressive responses of firms to environmental pressure they use the example of Canfor, a British Columbia forest company. At the beginning, in the mid-1980s, Canfor used tactics of defiance, avoidance, and manipulation to thwart the pressure from environmental groups who were targeting old growth logging and pulp mill pollution (Cashore & Vertinsky, 2000). Here, Canfor outright dismissed the claims of the environmental groups while leveraging their connections to government agencies to avoid compliance with regulations (Cashore & Vertinsky, 2000). It was not until environmental groups became more organized, threatening to bring their fight to the market place, that Canfor responded in an accommodating and proactive manner, voluntarily pledging to abate the pollution levels by building a modern pulping facility (Cashore & Vertinsky, 2000). In this case, Cashore and Vertinsky (2000) argue that by taking a proactive stance Canfor was able: “to maintain the clientele-pluralist network and 30 discretionary legal setting under which they operated” (p. 15). Citing Canfor’s proactive approach in advocating for better forest practices and in obtaining sustainable forestry certification, Cashore and Vertinsky (2000) speculate that Canfor may not have prevailed within the market place or under a new government, formed by the NDP, had it not learned to adapt its responses. Thus, they contend that Canfor was able to effectively portray an environmentally friendly company to their customers while demonstrating to government that companies can self-regulate (Cashore & Vertinsky, 2000). Prior to the recession of the early 1980s, Edenhoffer and Hayter (2013a) describe Canfor as a dominant integrated forest company operating primarily in the coastal region of British Columbia. However, in response to the recession, Canfor moved the majority of its operations from the coast to the interior of the province, and in 1983 became a publically traded company (Edenhoffer & Hayter, 2013a). Using the investment dollars it received, Canfor embarked on its corporate restructuring plan, which involved consolidation through the acquisition of other firms, and rationalization through a number of cost saving measures (Edenhoffer & Hayter, 2013a). Canfor survived the 1980s recession, and as such, Edenhoffer and Hayter (2013a) postulate that at least parts of Canfor’s survival strategy were prevalent throughout the industry. A key aspect to this strategy of survival was the choice to pursue the mass production of low cost commodities rather than value-added commodities (Edenhoffer & Hayter, 2013a). They argue that not only did the pursuit of mass production lock in the industry’s evolutionary trajectory, but it reinforced the business cycle of booms and busts, which became increasingly more extreme and created greater instability (Edenhoffer & Hayter, 2013a). Further, they contend that the uncertainty surrounding timber supply, 31 markets, and land use in British Columbia have in some ways forced forest corporations towards making these kinds of choices (Edenhoffer & Hayter, 2013a). They note that another key to Canfor’s survival was moving to the interior when it did. At the time, the timber supply in the interior was more robust and there was less contention with First Nations than there was on the coast (Edenhoffer & Hayter, 2013a). Between 1980 and 2010 Canfor tripled in size in British Columbia while reducing its employment by half, and although Canfor made substantial capital investments in the interior of British Columbia in response to the mountain pine beetle, it has since made investments in the southern United States, using the money refunded from the illegal duties collected between 2001 and 2005 as part of the Softwood Lumber Agreement with the United States in 2006 (Edenhoffer & Hayter, 2013a). In the broader context, Edenhoffer and Hayter (2013b) explain the evolutionary trajectory of the British Columbia forest industry by combining the industry life cycle model with the resource cycles model, as both models have similar phases, into what they call the resource industry life cycle. This life cycle consists of four phases. The first phase of the industry life cycle/resource cycle is that of pioneering/discovery, where the industry is dominated by competitive entrepreneurs and resource exploitation is focused on that which is in closest proximity and commands the highest value in the market. The second phase is the growth/rapid expansion, where entrepreneurs are replaced by larger vertically and horizontally integrated companies and extraction costs increase as the distance to the resource increases. The third phase is maturity/plateau, where the innovation and technology drive firms to gain competitive advantage and exploitation mentality depletes and degrades the resource. The final phase is decline revival/decline and abandonment, where the decline can be gradual until, in the case of industrialized forestry, the supply of timber collapses, but 32 until such time efficiencies continue to be sought and investments continue to be made (Edenhoffer & Hayter, 2013b). Using the tenets of the resource industry life cycle model, Edenhoffer and Hayter (2013b) sketch out the development and evolution of the forest industry in British Columbia, noting that its prolonged boom period not only produced large integrated firms, but also helped to create what Freudenburg (1992) calls addictive resource economies. Although the plateau stage cannot be precisely pin pointed in time, Edenhoffer and Hayter (2013b) claim that the forest industry in British Columbia arrived on the plateau just as the recession of the 1980s began. They describe the plateau stage in British Columbia as one in which the industry is dominated by large vertically and horizontally integrated firms, stating that as resource exploitation expands and moves farther away from distribution hubs companies must leverage economies of scale in order to keep costs down, which is accomplished through mass production (Edenhoffer & Hayter, 2013b). The pursuit of mass production is still predominant in forest industry as Nelson, Cohen, and Nikolakis (2009) found that mills in British Columbia favour maximizing production over maximizing value. Just like the plateau phase, the decline phase is not a precise point in time, rather, it can be tumultuous and drawn out, filled with short lived revivals (Edenhoffer & Hayter, 2013b). Edenhoffer and Hayter (2013b) suggest that since the 1980s, restructuring in the British Columbia forest industry has been a matter of survival, stating that: “resource exploitation contains an internal dynamic of self-destruction and, as growth levels out in the plateau stage, decline and closure become possibilities” (p. 141). Despite the mountain pine beetle epidemic, Edenhoffer and Hayter (2013b) say that production in the forest industry peaked around 1987, which is indicative of the industry leveling out. They argue that the 33 recession of the 1980s was the catalyst of change for the British Columbia forest industry, as it was the crisis that triggered the softwood lumber trade dispute with the United States, the rise of societal protest against industrial forestry, and the subsequent re-regulation of the industry (Edenhoffer & Hayter, 2013b). In terms of the re-regulation of the industry, Binkley and Zhang (1998) measured the response of stock holders when the NDP government in British Columbia raised stumpage rates in 1994 in an effort to appease the Americans over the claim that the government of British Columbia was subsidizing industry. Using an event study of stock prices, they found that the stock prices of publically traded companies fell on the day that the government announced the changes to stumpage rates. Binkley and Zhang (1998) interpreted as a negative reaction by stock holders to the news of increasing stumpage rates. A similar study was done by Niquidet (2008) to measure the reaction in the stock market to the announcement of the Forest Revitalization Plan and the intended expropriation of 20% of the volume from major licensees. In his measure of thirteen forestry firms in British Columbia, only a few showed a significant negative response (Niquidet, 2008). Further inquiry prompted Niquidet (2008) to ascertain that the negative reaction had to do with a perceived loss of property rights. These findings are interesting, because the re-regulation of 1994 was quite different from the re-regulation of 2003. The 1994 stumpage rates were increased by $12/m3 on average, which does not seem like a substantial amount, but based on lumber prices at the time the increase came into effect on May 1, 1994, it represented an 81.1% increase for the Interior and 64.4% increase on the coast (Binkley & Zhang, 1998). The study by Binkley and Zhang (1998) shows the reaction of shareholders to the arbitrary increase made by 34 government to the price of timber. There is evidence to support that the increased stumpage in 1994 was viewed negatively by shareholders as it significantly increased the cost of supply, whereas in 2003, the timber expropriated was compensated for. Niquidet (2008) suggests that shareholders perceived a loss of rights and hypothesizes that the 20% take back of timber volume represented not only a loss of security, in the form of property rights, but would result in less control of the timber supply for major licensees. Niquidet’s (2008) hypothesis lacks supportive evidence given that expected compensation paid by the government to the licensees was to be $200 million, equating to approximately $24/m3. Further, Pinkerton, Heaslip, Silver, and Furman (2008) assert that the changes that accompanied the expropriation, specifically those changes brought in by Bill 29, allowed the major licensees to consolidate their control of the allowable annual cut (AAC) which further solidified their oligopsony. In their comparison of three forest jurisdictions, British Columbia, New Zealand, and Tasmania, Hayter and Barnes (2012) explore how neoliberal ideology has been expressed in these three different geographies. Although, neoliberal ideology began to influence the government regulation of the forest industry in British Columbia beginning in the 1980s and 1990s, Hayter and Barnes (2012) contend that neoliberalism was not directly applied to British Columbia forest policy until the 2000s. Lindquist, Langford and Vakil (2010) agree as they consider British Columbia behind the times in terms of neoliberal influence. Young and Matthews (2007) state that the BC Liberals, elected in 2001, directly ushered neoliberalism into the forest economy in three ways, by liberalizing licensee rights through Bill 29, by liberalizing the market through Bill 29, and by liberalizing space through various amendments to the Forest Act 1996 and through the Forest and Range Practices Act. 35 Similarly, Humphreys (2009) states that neoliberal mechanisms of marketization, public/private partnerships, and deregulation are found throughout international forest management. The BC Liberal Party may have ushered in neoliberal ideology in British Columbia, but according to Humphreys (2009): “[n]eoliberalism is very much the hegemonic ideology of our age” (p. 320). 3.4. British Columbia Forest Policy Legacies The British Columbia forest policies that have been identified as legacies in the literature are policies regarding timber tenures, timber supply, and timber pricing. All of these policies support the larger underlying forest policy of regulating the forest harvest using the sustained yield model (Rayner et al., 2001). Despite mounting evidence that the policy does not work (Prudham, 2007), and that public expectation for sustainable forestry is rising (Rayner et al., 2001), the sustained yield forest regulation policy remains steadfastly in place, with the current provincial allowable annual cut (AAC) set at 75.8 million cubic metres per year (FLNRO, 2016b). The questions being raised in the literature are asking why these forest policies persist (Burda & Gale, 1998; Fréchette & Lewis, 2011; Prudham, 2007; Rayner et al., 2001). Following the Royal Commission on Forestry led by Sloan in 1945, changes were made to the Forest Act in 1947 which enshrined sustained yield forest regulation as the doctrine to liquidate old growth and create in its place an even-aged predicable flow of timber, revenue, and stability (Prudham, 2007). Rayner et al. (2001) agree, noting that in theory, sustained yield forest regulation would uphold the forest industry which in turn would provide revenue to the Crown and employment stability for communities. Indeed, it could be argued that the policy of sustained yield enabled the forest industry to expand into the 36 Interior of the province during W. A. C. Bennett’s province building era (Marchak, 1983). However, although the sustained yield forest regulation model was institutionalized in British Columbia in 1947, it does not mean that this model has been effective (Prudham, 2007). Prudham (2007) argues that the ideology underpinning sustained yield forest policy: “allow[s] forests to circulate as financial capital” (p. 259). During the post-World War II era, the time period at which sustained yield forest policy was introduced, Prudham (2007) notes that: “ideas of nature and work were very much linked in BC” (p. 259); meaning that the forest policy fit easily with the context of the time and place. He notes that some of the unique aspects of this forest regulation model to British Columbia were that it granted private capital exclusive rights to harvest timber, was managed by scientific forestry and linked community economic well-being directly to the exploitation of the forest resources (Prudham, 2007). Thus, he contends that support for sustained yield forest regulation was garnered because the forest industry, the state, and organized labour all stood to benefit from this model (Prudham, 2007). Challenging the policy goal that sustained yield forest regulation to provide stability to communities, Byron (1978) examined unemployment in Quesnel and Prince George from the period of 1967 to 1975. In this research, he compared the degree of instability of employment in occupational groups for both cities. In the study, to allow reasonable comparison, Byron (1978) looked at the relative instability and absolute variability for each group. The data support the postulation that the volatility of lumber markets causes fluctuations in forest sector employment, in which there is a lag of one to two months between the cause and effect (Byron, 1978). Byron (1978) further argues that employment instability in small forest-dependent economies is a social issue which cannot be 37 solved using sustained yield forest policy concluding, “the permanence or survival of forest industry centres is neither assured by nor solely dependent upon the perpetual maintenance of nearby forests at or near sustained yield levels” (p. 61). Despite these findings by Byron (1978), sustained yield forest regulation has prevailed and as a result, the province’s allowable annual cut (AAC) has tripled since the 1950s (Burda & Gale, 1998). Burda and Gale (1998) argue that the mass volume production of commodity products that dominates the British Columbia forest industry benefits private firms and does not maximize jobs, provincial revenue, or community stability. They identify that decreasing timber supply, increasing costs to extract timber, and increasing barriers to the market in the United States all pose significant threats to this timber manufacturing choice (Burda & Gale, 1998). Taking the aforementioned threats to the industry into consideration, Burda and Gale (1998) contend that a commitment to high volume depletion via sustained yield forest regulation has the potential to have crippling long term economic consequences for the province in the future. This policy of sustained yield has also survived the legislative changes of recent decades which govern the operational practices of the forest industry. While examining the various policy changes brought about by the NDP government in British Columbia in the 1990s, Rayner et al. (2001) were interested to find that while the Forest Practices Code Act substantially changed the operational practices of industrial forestry to be conducted at a higher environmental standard, other policies, such as the allocation of timber through the forest tenure system and the allowable annual cut (AAC) remained virtually unchanged (Rayner et al., 2001). Although conflict between the environmental movement and the industrial forestry sector in British Columbia had been on the rise since the 1980s, Rayner et 38 al. (2001) note that by the 1990s, the environmental movement had gained considerable support and subsequent political influence. However, they also recognize that supporting the forest industry had itself become a policy, and a key mandate of the ministry responsible for forests (Rayner et al., 2001). Hence, because the forest industry remained an important economic feature in British Columbia, they argue that it continued to have substantial influence over forest policy stating that: “[…] authority over policy formulation and implementation remained in the hands of the same community of experts who, by and large, remained quite supportive of the status quo” (p. 325). In the 2000s, the Liberal government replaced the Forest Practices Code Act 1995 with a results-based version, the Forest and Range Practices Act 2004, which is essentially a performance-based framework (Hoberg & Malkinson, 2013). However, unlike other forest jurisdictions, it is the licensee that sets the result or strategy to achieve the eleven values established in the Forest and Range Practices Act by government (Hoberg & Malkinson, 2013). Peterson and Stuart (2014) contend that the traditional reliance by successive British Columbia governments upon the forest industry to provide jobs and stability via the extraction of timber has ultimately led to the erosion of the government’s regulatory power. Citing the mountain pine beetle epidemic, they claim that because of the Forest and Range Practices Act the government could not direct the licensees to harvest the mountain pine beetle infested stands (Peterson & Stuart, 2014). As a result, licensees harvested in the green timber longer than they should have, sacrificing long term viability for short term gain (Peterson & Stuart, 2014). While Peterson and Stuart (2014) argue that the forest industry continues to act in its own best interest and that the government has little ability to change the situation, they also note that the government appears to be committed to business as usual. 39 Nelson (2007) agrees that the forest industry focused on maximizing its own interests by harvesting as much profitable timber as possible. But, he also says that the forest industry only changed its focus to harvesting as much green timber as possible because it was unsuccessful in lobbying the government to address the growing mountain pine beetle epidemic (Nelson, 2007). There are several elements noted earlier in this section that merit further discussion as they provide greater insight into why sustained yield forest regulation persists in British Columbia. The first is that the underlying goal of sustained yield forest regulation, the accelerated liquidation of old growth timber, which is founded on an ideology that views the forest as a commodity to be exploited (Prudham, 2007). The second is that the relationship between the forest industry and government and, to a lesser extent, labour is founded on agreement to support sustained yield forest regulation (Prudham, 2007). The third is that the relationship between the forest industry and government is not exclusively between the elected officials and the forest industry, but between the forest industry and the bureaucracy, namely the ministry responsible for forests (Rayner et al., 2001). What the first two things tell us is that not only is the commitment to this model strong, but that the model of sustained yield has an ideological foundation. What the third thing tells us is that this ideology of sustained yield also has an institutional context. To consider change dynamics in the governance of forests one must consider underlying ideologies (Humphreys, 2007) and institutions (Fréchette & Lewis, 2011). Humphreys (2009) states that an ideology is very different from knowledge, in that the former offers both a critique on how things are perceived and an alternative to that which is being critiqued, whereas the latter is defined as fact that is both verifiable and believed. 40 Given that an ideology is based on morals, any individual or group may subscribe to the ideology, however, this is not to say that acceptance of a given ideology is universal, rather, that it is relative (Humphreys, 2009). Fréchette and Lewis (2011) regard institutions as enduring fixtures in human society that are resistant to change. It is helpful at this point to define what institutions are. For the purpose of this investigation, institutions are defined as: “the formal rules, compliance procedures and standard operating procedures that structure the relationship between people in various units of the polity and economy” (Hall, 1986, p. 7). As such, Fréchette and Lewis (2011) argue that: “[o]nce produced, rules may become entrenched for a number of reasons” (p. 583). They elaborate by saying that the rules can become entrenched for the following reasons: that rules exist in complex inter-dependent arrangements both vertically within the hierarchy of rules and horizontally with other rules; that changing the rules would often be considered too costly; and, they have a tendency to be captured by the interests that stand to benefit the most (Fréchette & Lewis, 2011). Moreover, Fréchette and Lewis (2011) argue that because institutions tend to provide greater certainty for vested interests, any change that would threaten that certainty would be opposed. Finally, they stress the importance of identifying what the rules to change the rules are, and who has the power to change them. In Canada, the authority to manage provincial forests resides with the respective governments of each province (Fréchette & Lewis, 2011). In terms of the institutional context of sustained yield forest regulation, Prudham (2007) describes the British Columbia forest sector as a command and control system that is only accessible by those with elite knowledge, and that the professional forest management system is dominated by forest professionals, who occupy positions within the forest industry 41 and the bureaucracy. Sustained yield forest regulation was part of the scientific forestry model endorsed by Sloan as part of the recommendations in the 1945-1947 Royal Commission on Forestry in British Columbia (Prudham, 2007). That sustained yield is considered a component of scientific forestry means that it is part of the curriculum that is taught to aspiring forest professionals (Prudham, 2007). So, just as sustained yield forest regulation became entrenched within the ministry responsible for forests, becoming part of the institutional fabric, it also became a conduit for perpetuating the ideology through time because of its connection to the forestry profession (Prudham, 2007). In his discussion, Humphreys (2009) defines discourse as: “a set of linked understandings and ideas that structure how people think about, interpret and understand the world” (p. 319). He further argues that: “different discourses may achieve different degrees of political influence in social life […and] may gain considerable political influence while lacking a clear epistemic basis” (p. 320). What Humphreys (2009) is saying is that the power behind an ideological discourse is not in its ability to prove itself viable, but that it is morally accepted by those who adhere to it. Given that sustained yield forest regulation remains in place, despite its failure as a policy instrument to secure employment and stabilize the industry of the forest sector, indicates that the ideology supporting it is predicated on moral rather than epistemic relativism (Humphreys, 2009). In summary, it could be argued that the persistence of sustained yield forest regulation in British Columbia can be explained by its institutional framework that is informed by ideology. Institutional frameworks are important because they: “define the ends and shape the means by which interests are determined and pursued” (Scott, 1987, p. 508). 42 Chapter 4: Methodology and Methods 4.1. Introduction My research focuses on forest policy change in British Columbia, a subject matter that can be laden with contention and polarizing viewpoints (Howlett, Rayner & Tollefson, 2009). Knowing this, I wanted to undertake my research in a more holistic manner with the goal to gain a balanced understanding of the data. I also considered what the constraints and barriers to this research might be and how I would be situated in this research (Guba & Lincoln, 2004). Methodology involves more than the way in which we collect and measure data, it is the framework from which we approach research, how we decide what to investigate, and how we will go about studying it (Guba & Lincoln, 2004). The framework of research comes from our own ontology, or how we see reality, and our personal epistemology, or how we go about understanding reality (Guba & Lincoln, 2004; Jones & Gomez 2010). I used a mixed methods approach, which leveraged both qualitative and quantitative methods to conduct my research. The mixed methods approach can add breadth and depth to the research, expanding the possibilities of discovery while allowing for triangulation (Greene, Caracelli & Graham, 1989; Moran-Ellis, Alexander, Cronin, Dickinson, Fielding, Sleney & Thomas, 2006). The concept of triangulation comes from the navigational technique of using two known points to calculate the location of a third point, in the same way, data gathered from different sources and analysed in different ways converges in support of the findings (Jick, 1979; Tashakkori & Teddlie, 1998). My intention in using mixed methods was also to allow for the expansion of discovery, where the findings from 43 one source of data informed the collection of subsequent data (Greene et al., 1989), which also enforced the rigour of my research (Baxter & Elyes, 1997). As a result of working for twenty years in different capacities within the forest industry in British Columbia I have a good understanding of the operational aspects therein. Therefore, I had to be all the more diligent to ensure that I applied rigour throughout my research process so that the research findings could be assessed as being trustworthy (Baxter & Elyes, 1997). As a means of enforcing the rigour in my research, checking my subjectivity (Bailey, White & Pain, 1999; Madge, Raghuram, Skelton, Willis & Williams, 1997), and monitoring for power relationships (Bailey et al., 1994; Katz, 1994), I adopted the practice of critical self-reflexivity when I first began formulating my research topic. I have used the practice of critical self-reflexivity throughout this research project not only to evaluate my goals and approaches, but to think about the ethical aspects of my research (Dowling, 2010). Therefore, disclosing my position as a member of the forestry profession when I introduced myself as a researcher became an important aspect of my research because it has undoubtedly influenced the questions that I have asked, the way that I perceived the data, and the way that I present the findings (Behar, 1996; Clifford, 1986; England, 1994). My position within this research is that I am both an insider and an outsider (England, 1994; Kobayashi, 1994). I am an insider in that I live and work within the case study area and my career is in the forest sector. I am a Registered Professional Forester, and I am a public servant with the province of British Columbia, working at the Ministry of Forests, Lands and Natural Resource Operations. I am an outsider in that my inside position has not afforded me a privileged or complete understanding. Rather, I am, “situated as dweller and sojourner within my own fields of knowledge” (Clifford, 1997, p. 218). The duality of my 44 position as the researcher, being both an insider and an outsider, has been challenging to manage. While my experience working within the forest sector for many years and in many roles has given me insight to aspects of forest policy and management, it has not given me a total comprehension. As Clifford (1986) contends: “[c]ulture, and our views of “it”, [are] produced historically, and are actively contested. There is no whole picture that can be ‘filled in’, since the perception and filling of a gap leads to the awareness of other gaps” (p. 18). I agree with Clifford (1986); during my research journey I no sooner understood some aspect of the forest industry or answered a particular question, only to unearth more gaps in my understanding in the process. Challenging the assumptions upon which my knowledge is based has been a humbling experience. 4.2. Ethical Considerations My intention was to interview people as a means of collecting data for my research. It is common for academic institutions to require some form of ethical review when human subjects are involved in research (Dowling, 2010). Therefore, I sought approval from the University of Northern British Columbia Research Ethics Board and was granted approval on May 16, 2014. See Appendix 1 for a copy of the approval letter. Prior to embarking on the interview phase of the research I considered what some of the ethical challenges might be and decided that they would likely be: effectively managing power relationships; and, potentially crossing boundaries between my professional ethics and my own moral and ethical framework. Whether you are an insider or an outsider, as the researcher you must always consider the well-being of the participants that your research involves (England, 1994; Kobayashi, 1994). Honesty in the approach of a topic and critical evaluation of self in the 45 context of being the researcher are important practices that can help the researcher, protect the participants from harm, and safeguard from the potential of the researcher exploiting the participants (Dowling, 2010; England, 1994; Nast, 1994). As the researcher, it was important for me to remind myself that power, knowledge, and ethics are interwoven (Madge et al., 1997). Power relationships can be relatively equal in terms of what the researcher and participant stand to benefit or it can be unbalanced, either to the advantage, or disadvantage of the researcher (England, 1994). Power relationships can also be quite fluid, changing during the course of the research, or as a result of the research (Kobayashi 1994). There is no particular method that will eliminate the power relations inherent in qualitative research (Dowling, 2010; England, 1994; Gilbert, 1994, Katz, 1994; Kobayashi, 1994, Nast, 1994). The best way to keep them in check is through self-reflexivity (Bailey et al., 1999; England, 1994; Katz, 1994). Although I did not identify any tangible risks for those wishing to participate, I did fully disclose my identity as a knowledgeable practitioner of twenty years working in the forest sector and currently employed as a Registered Professional Forester with the Ministry of Forests, Lands and Natural Resource Operations. I assured the participants that the research I was conducting was my own and not that of my employer. I gave disclosure of my position to potential participants during the solicitation process which helped to establish a baseline of honest communication between them and myself. To further ensure the protection and well-being of the participants, I have kept their identities confidential. What I discovered, however, was that there was more to preserving anonymity of the participants than just assigning a unique alpha-numeric identifier. There were things that were revealed by participants during the interview process that I could not report in my thesis 46 because to do so would compromise the anonymity of certain participants. Furthermore, I purposely chose not to cite participants in the themes generated through content analysis because I felt there was a risk of participants being identified through the associated theme tables. I even scrutinized the use of direct quotes in this section to ensure that I was not inadvertently revealing identities. 4.3. Research Design I wanted to investigate how forest corporations have responded to the forest policy changes made by the government of British Columbia in 2003, which may have subsequently impacted the forest industry. My goal was to understand the context of these changes more comprehensively, and in so doing, perhaps uncover some linkages that may have eluded policy makers. Initially I was interested in grounded theory as a research technique. However, with a restricted time frame and resources this approach could not be employed and used properly. Instead, I decided to use a case study as a way of organizing my research strategically. Case studies have been used extensively in social science research and for a variety of purposes (Tight, 2010; Yin, 1994). Historically, ambiguity surrounds the use of case studies concerning the question of whether a case study is a method, a methodology, or merely an example of some specific phenomenon (Baxter, 2010; Tight, 2010; Yin, 1994). As such, many social researchers regard it as an inferior qualitative method; yet, it can provide robust explanations from complex social phenomenon (Baxter, 2010; Yin, 1994). The structure of the case study lends itself to the use of mixed methods, allowing the researcher to fully explore a given phenomenon within the context it is occurring (Baxter & Jack, 2008). The case study also provides a format in which the convergence of data can occur, which in 47 turn gives the research credibility (Baxter & Jack, 2008). Moreover, the construct of a case study research design permits the researcher to investigate ‘why’ and ‘how’ a phenomenon is occurring, which gives the findings of the research a degree of generalizability in that concepts uncovered can be transferred to other cases (Baxter & Jack, 2008). In this research, the case study was used as a research strategy to purposefully focus data collection and analysis (Yin, 1994). The topic of this research is complex, involving multiple variables whose relationships to the events in question are unclear. A case study is a research strategy that places boundaries around the topic in such a way that it focuses the data collection and analysis within the context of the real world (Yin, 1994). Though it appears very rigid because it places boundaries, it is in fact flexible and promotes the collection of evidence from multiple sources (Yin, 1994). To assist with the conceptualization of the case (Stake in Tight, 2010), I considered the following questions: 1) Has the number and distribution of sawmills changed over time? If so, what has driven this change? 2) Has the fibre flow changed? If so, how has it changed, and what does this change potentially mean for the forest industry? Asking these questions led me to decide to restrict the case study both in space and in time. I chose the Prince George Timber Supply Area as the geographic boundary and a fifteen year time period from January 1, 1997 to December 31, 2011. 4.3.1. The Case Study I chose the Prince George Timber Supply Area as the geographic region for this case study. It consists of three forest Districts: Prince George, Vanderhoof, and Fort St. James (see 48 Figure 1 for a map of the study area). I chose to make the case study within the geographic boundaries of a Timber Supply Area (TSA) for two reasons: 1) the TSA is the geographic unit in which the allowable annual cut (AAC) is determined; and, 2) TSAs contain administrative units known as Districts, which are geographic units in which harvesting and scaling data can traced. The Prince George Timber Supply Area (TSA) includes the municipalities of Prince George, Fort St. James, and Vanderhoof, as well as other smaller communities such as Hixon, Fraser Lake, Sinclair Mills, Bear Lake, and McLeod Lake (Snetsinger, 2011). In addition, 24 First Nations have asserted traditional territories within the TSA (Snetsinger, 2011). I chose the Prince George TSA because it is unique in the sense that it has varying timber types across the forest District landscapes (Snetsinger, 2011), and varying levels of forest dependence within the communities (Horne, 2009). As well, this TSA is the largest contributor of volume in the province, and accounts for more than 40% of the Northern Interior forest region’s TSA contribution (FLNRO “AACTSA”, n.d.). I chose the period of 1997 to 2011 primarily for two reasons. The first was that I wanted to have data that would reflect the period prior to the change in forest policy in 2003 and the period afterwards. The second was that I began extracting numerical data from a publicly accessible provincial database in 2012 and the public records only go back for a span of 15 years. For this case, 1997 was the oldest year of data I could collect using public access. 49 Figure 1. Map of the Prince George Timber Supply Area Source: Map retrieved from http://www.for.gov.bc.ca/hts/tsa/tsa24/map.gif 50 Finally, for the purpose of my research, I defined forest corporations as being major primary timber manufacturing companies with production capacities large enough to be classified as medium to large by the ministry responsible for forests3 in the province of British Columbia. In British Columbia, lumber manufacturing accounts for the majority of total primary manufacture of timber (FLNRO, 2011a), thus, my research focuses on lumber manufacturing companies within the Prince George Timber Supply Area. 4.4. Data Collection and Analyses I gathered data from both primary and secondary sources using a mixed methods approach, with the collection and analysis of the secondary data preceding and informing the collection of the primary data. The essence and complexity of the bigger picture within my research questions demanded that I investigate more than one source of data. In fact, I believe that the following sources of data I have used are interrelated in terms of the research questions and cannot be understood independently as such (Gaber & Gaber, 2007). A mixed method approach to the analysis of the data was used because the data were not collected in the same way and as such cannot be lumped together and analyzed as an aggregate (Brannen, 2005). All data collected, regardless of method, were within the scope of the case study parameters of both space and time. For clarity, I have chosen to present the methods of data collection and analysis by topic area the data were collected for. 4.4.1. Mill Closures I used three sources to collect secondary qualitative data: media outlets; corporate annual reports; and official state documents (Bryman, 2008). These data were first collected between October 2011 and March 2012. I conducted a media scan of the Prince George 3 I am using the generic term ‘ministry responsible for forests’ to avoid confusion because the ministry has undergone several name changes throughout its time. 51 Citizen newspaper using an electronic database subscribed to by the University of Northern British Columbia. I searched for news surrounding closures, start-ups, shut downs, and upgrades involving lumber manufacturing companies located within the Prince George Timber Supply Area. From the data that I collected, I extracted specific information about lumber manufacturing companies, which I tracked in a table that I constructed in a Word document. To better understand the context of sawmill closures and openings within the Prince George Timber Supply Area, other events linked to the forest industry and occurring between 1997 and 2011 were mapped out alongside the sawmill closures and openings to see if there was any correlation between them (see Appendix II for details). I used annual reports from forest corporations and the provincial government that were available to the public, as well as official documents published by the Government of British Columbia through the ministry responsible for forests to further inform my understanding of the events taking place within the Prince George Timber Supply Area. I analyzed these data using qualitative document analysis techniques because I was interested in themes and patterns of words to provide context rather than in the mere frequency of words (Altheide, Coyle, DeVriese & Schneider, 2008). I looked for themes that would help explain the trends identified in the numerical data and this inquiry yielded several themes which I added to the table tracking the closings, openings, and upgrading of sawmills. These themes were also used to help generate interview questions. During the examination of the results, I conducted a broader media scan regarding the closures of mills picked up in the initial media scan. The second media scan was used to uncover other details which might help to explain the particular circumstances surrounding each mill closure. In August of 2015 I looked for internet based media accounts of each mill 52 closure using the Google search engine as well as searching the Prince George Citizen and Vancouver Sun newspapers using an electronic database. 4.4.2. Employment In February 2012, employment statistics were collected from BC Statistics using the electronic database for BC Statistics. The employment statistics were extracted by industry and development region, as well as employment/unemployment rates by industry for the Cariboo and North Coast/Nechako regions. There was no secondary inspection of this data set. The data were examined to see if there were any obvious trends. During this data collection from BC Statistics, I discovered a special set of analyses that had been done by Horne (2009) for the provincial government where data collected for BC Stats and Statistics Canada were reconfigured to align with Forest District boundaries. Unofficial job loss numbers associated with specific sawmill closures in the Prince George Timber Supply Area were collected in the broader media scan of sawmill closures that was conducted in August, 2015. 4.4.3. Timber Harvest and Valuation In British Columbia, the scaling of timber calculates the volume and grade (quality) of logs, and the timber mark tracks all timber volume from its origin to its destination. (FLNRO, 2011b). The appraisal of timber is regulated by the provincial appraisal manuals and stumpage rates for timber are set by the timber mark of the cutting authority (FLNRO, 2011b). The Harvest Billing System (HBS), introduced in 2003, is the provincial electronic database that tracks and manages all of the timber that is harvested by managing the scaling data, which in turn in conjunction with stumpage rates are used to generate stumpage billing (FLNRO, 2011b). The Harvest Billing System (HBS) is an open database system containing 53 both current and historical scale and billing data. Historical data are limited depending on what is being queried. For this data collection, historical data were available up to 15 years prior to the year of inquiry. I used HBS to gather data on where timber in the Prince George Timber Supply Area was harvested and scaled, what species were scaled, and how much stumpage was collected for the timber scaled. To help answer the question of how industry was responding to the changes made to legislation I wanted to look at where timber was being harvested by forest licence holders within the Prince George Timber Supply Area and where it was being scaled. I extracted data from HBS in March of 2012. Initially I ran reports for every year from 1997 to 2011 setting the filter parameters to restrict the harvest to the District and by forest licences, and the report configuration parameters to be grouped by District scaled and scale site. I also included volume, value, species and grades within the report configuration parameters. I discovered cruise based scaling in the 2010 and 2011 data so I had to run additional reports so that I could sort out the data. A cruise based scale means that the timber is scaled on the stump, in the District in which it is harvested in, rather than at the mill (MFR, 2010a). To extract the destination information I ran reports on each District cruise based scale site by timber mark so that timber marks belonging to forest licences could be extracted and the destinations of the timber could be adjusted accordingly. I also ran reports on each District from 1997 to 2011 setting the filter parameters to restrict the harvest by District and by Crown Land, and the report configuration parameters to be grouped by management type. Management types are: community forest agreements (CFAs); woodlots; tree farm licences (TFLs); outside (not under management) and; Timber Supply Area (TSA). Using the data in these additional reports, I was able to do two things: 54 compare the total harvest within a District to that under TSA and also determine the proportion of the TSA harvest that was under forest licence; and cross check the data I gathered to ensure the accuracy of the queries I used for generating reports. In all, 32 reports were generated for each District of the three Districts for a total of 96 reports. I conducted secondary analyses on the data collected in 2012. These analyses are considered secondary because the data were collected by another organization for a different purpose (Bryman, 2008). My purpose of doing these analyses was to answer questions regarding timber harvested by forest licences in the Prince George Timber Supply Area. Namely, where was timber harvested and what species were included in these harvests, where was the harvested timber processed, and what was the revenue generated through stumpage for the Crown. For these analyses I used Excel spreadsheets to manage and analyze the data. Though the data collection occurred in 2012, the analyses were conducted at various times as new information was acquired, or new gaps in knowledge were identified, with the exception of the fifth analysis, where data were extracted in 2015. The 2015 data extraction was done to simplify the extraction and amalgamation of data. Secondary analyses were also conducted with these data. In the first analysis I sought to answer the question of how much timber harvested within a District was also processed there, and so, I compiled the data by year, location of harvest, and scale. Next, I summarized how much of the harvest within each District was scaled in that same District for each year from 1997 to 2011. In the second analysis, I looked at the total amount of timber harvested within the Prince George Timber Supply Area and how much revenue it generated for the province. I compiled the total harvest for all three Districts for each year from 1997 to 2011 and compared it to the total revenue that was 55 generated from each harvest year for the Crown. Next, to account for inflation I used an online inflation calculator found on the Bank of Canada website to calculate each year’s total value to 2011 dollars (Bank of Canada, n.d.). Based on the results of the analyses noted above, as well concerns that were raised by a number of people I interviewed, I conducted a third analysis on lodgepole pine. In this case I looked at the data from 2000 to 2011 for each District. I chose the starting year of 2000 for three reasons: 1) in 2000, the mountain pine beetle was active in all three Districts; 2) in 2001, the first mountain Pine Beetle Action Plan was released, and; 3) in 2002, the first uplift in the allowable annual cut (AAC) was put into effect to address mountain pine beetle epidemic. For the third analysis, to determine the relative proportion of lodgepole pine harvested within each District by forest licences, I compiled the harvest data collected in 2012 in a spreadsheet by District for the years 2000 to 2011. I sorted the data for each year by species where I was able to sum the volumes of lodgepole pine and the volumes of all other species. Using the total volumes of lodgepole pine and all other species, I was able to determine the harvest of lodgepole pine relative to all other species by year and by District. For the fourth analysis, I isolated the lodgepole pine data from all Districts and sorted it by year and then by log grade. Isolating the lodgepole pine gave me the ability to sum the log grades by year and see the distribution of log grades by volume proportion from 2000 to 2011. For this inquiry, because there were changes made to the Interior log grades in 2006 (FLNRO, 2011b), and that “Stand as a Whole Pricing” was introduced in 2010 (MFR, 2010c), I decided to amalgamate the log grades into two categories: off grade and sawlog grade. The off grade category included grades 3, 4, 5, 6, 7, and 8, whereas the sawlog category included grades (blank), 1, and 2. It is important to acknowledge that there is a 56 difference between a scaled grade and a cruise based billing in terms of the timber appraisal. The term “scale based” means the stumpage payable is based on the scale of the timber in accordance with Part 6 of the Forest Act (MFR, 2010c, p. 1-5). Conversely, “Stand as a Whole (SAAW) Pricing” (cruise based) means that one stumpage rate is determined for all of the net merchantable timber on the cutting authority area. Therefore, for a cruise based cutting authority, the single stumpage rate applies to all of the net merchantable volume identified in the cruise conducted in accordance with the Cruising Manual, (MFR, 2010c, p. 1-5). What I have termed grades, particularly grades 7 and 8, are actually billing codes in HBS (MFR, 2010a). So, for the purpose of this case, the term grade and billing code are considered to be the same. To better understand the decline in stumpage revenue from 2006 to 2011, I extracted data from HBS in December of 2015. As I was not interested in isolating the harvest by District for each year, a new data extraction would simplify the review. I ran a series of 12 reports, one for each year starting in 2000 and ending in 2011. In each year I set the filter parameters to restrict the harvest by forest licences and organise the data by management type so that I could extract the data for the Prince George Timber Supply Area (Timber Supply Area 24). I also included in the reports volume, value, and species and grade. For the fifth and final analysis, I extracted the data from the reports into an Excel spreadsheet, with a tab for each year. With the data now assembled by year, I was able to sort by species and associated grades as well as sum volumes and associated dollar values. In this probe, for each year I separated the lodgepole pine from all of the other species. I then tallied the volumes and associated values so as to calculate the average stumpage rate paid by forest licences for lodgepole pine and for all other species. For each 57 year I used mathematical formulas to calculate the average stumpage rates for the following categories: all lodgepole pine; lodgepole pine sawlog grade; lodgepole pine off grade; all other species combined; all other species sawlog grade; and all other species off grade. Also, for each year, I summed the sawlog grades and off grades of the non-pine species each year to determine percentages. Finally, to account for the cruise based scale that occurred in 2010 and 2011, I tallied volumes and associated values by sawlog grade (grade 1 and grade 2), off grade (grade 4 and grade 6) (MFR, 2010c), and cruise based grade (grade 7 and grade 8) (MFR, 2010a). I did these calculations for both 2010 and 2011 for all non-pine volume and then excluded the deciduous species and grades. 4.4.4. Semi-Structured Interviews To provide further context and depth to my research, I conducted semi-structured interviews. I chose a semi-structured interview technique because it did not limit me to asking questions that reflected my research questions. Rather, it gave me the flexibility to explore other questions that arose from my background research, or to respond to information revealed by the participants (Dunn, 2010). I developed an interview guide (see Appendix III) to use during the interviews. This guide contained: the interview questions in sequential order; the context for the questions in the form of a preamble; and prompts, which served as reminders for me while I was conducting the interview (King & Horrocks, 2010). In developing my interview guide I consulted the seasoned researchers, who are part of my advisory committee, for input on the design of my questions and on the construction of the interview process. This consultation was very helpful, especially because I received sound advice for opening and closing the interview. As well, prior to finalizing my interview guide 58 I conducted two mock interviews with people who understand forestry in British Columbia to test both the clarity of my wording of the questions and to time the length of the interview. I chose purposive sampling for this research because I wanted to gain specific information that is relevant to this case (Yin, 2016). I used two purposeful sampling strategies to identify potential participants: criterion and snowball sampling techniques (Bradshaw & Stratford, 2010). I chose this sampling strategy because I wanted to get informed opinions from people who are involved in the forest industry and who could bring a specific understanding of the issues surrounding forest policy. The criteria I used to qualify participants were the following:  they had to have worked in forestry from at least between 1997 to 2011;  they had to have lived within the case study area for at least ten years and at least partially during the period of 1997 to 2011;  and, they had to be associated predominantly with industry, with government, or at arm’s length from one or both of these groups. Initially I identified potential participants using my own knowledge (Berg, 1998). From there expanded the sample by using the snowball sampling technique, in which I asked interviewees who they thought I should interview (Babbie, 2007). I used the same criteria, noted above, to select these potential participants with the added qualifier that they had to have been suggested by at least two interviewees, unless the potential participant had a unique perspective not yet sampled. One of the advantages of using the criterion sampling technique is that it allows the researcher to investigate a topic by identifying potential recruits who have specific knowledge regarding the research topic (Van den Hoonaard, 2012). Potential recruits are then identified using set criteria (Van den Hoonaard, 2012). I used the snowball sampling 59 technique as a way of limiting my own bias while getting the most variation in the sample (Yin, 2016). Another consideration for using this technique is highlighted Van den Hoonaard (2012), who says that it may serve as a means of accessing populations that are difficult to reach and/or identify. Berg (1998) agrees saying that by asking participants to refer other people that they may know who would fit the criteria and who may be interested in participating is a way to expand the target population. A strength of the snowball sampling technique is that it leverages the social networks of participants, but, it is also viewed by many as a weakness, in that the technique has a, “perceived potential for bias” (Stehlik, 2004, p. 39). The potential for bias is based on the perception that the researcher is limiting the pool from which the sample is taken by relying on informant referrals, which will not necessarily be representative (Atkinson & Flint, 2001). For all of the aforementioned reasons, the purposive sample is not meant to be a representative sample (Yin, 2016). As I was not interested in representing a population, I was not concerned about my sample size as it relates to statistical analysis (Atkinson & Flint, 2001). But, I was interested in making sure that my sample was representative. I recognised that there was potential for bias in using the snowball sampling technique, both in the referrals received from informants, and as the researcher initiating the chain reference because I identified the initial informants. I was concerned that by relying on referrals that were provided by my informants I may be overly representing the opinions of those particular social networks (Atkinson & Flint, 2001). I was also worried that I might be barred access to certain social networks (Atkinson & Flint, 2001). Atkinson and Flint (2001) state that snowball sampling is in fact: “biased towards the inclusion of individuals with interrelationships, and therefore will over-emphasise cohesiveness in social networks and will 60 miss ‘isolates’ who are not connected to any network that the researcher has tapped into” (p. 96). I employed a number of strategies to minimize bias and to enhance the reliability of the interview data. First, to ensure that I was not selecting participants from one specific group, I decided to select people from different employment/career groups. Second, within each employment/career group, I selected people from a range of positions within the hierarchy of each group. For the actual selection of the initial participants I used my own knowledge of having worked and lived in the Prince George Timber Supply Area since 2006 to identify potential participants and also drew from the knowledge I had gained from studying background information about the case (Bradshaw & Stratford, 2010). For the snowball sampling, I applied the same qualifying criteria for referrals that I did to the original participants with the addition of the criterion that the person must receive at least two referrals or must bring a unique perspective, not yet sampled. I also used the snowball sampling as a method for cross checking my initial participant choices. I began the interview process in late October of 2014 by soliciting potential participants using a formal invitation via email, or by letter. Once potential participants contacted me by email expressing their interest in participating, I sent the information and consent form (see Appendix IV) for them to review and approve. Once they reviewed the information and consent form, arrangements were made to conduct the interview. Participants were also given the option to have the interview questions in advance of the interview. The interviews were conducted in November and December of 2014. I used a digital recording device to record each interview but also used the interview guide that I had 61 constructed to navigate the questions and to write notes during and after the interview (Dunn, 2010). Once an interview was complete, I downloaded the digital file from the recording device on to my computer and then erased the file from the recording device to ensure security. Given that I am learning to do research, I chose to do a full transcription of each interview myself for two reasons. First, doing a full transcription meant having a text version of the interview. Second, doing the work myself gave me the opportunity to re-visit the interview and become more familiar with the data (King & Horrocks, 2010). I used the transcription module from the Olympus Dictation Management System software package with the AS-2400 transcription kit which included the RS27 footswitch. I used this software and hardware package because it enabled me to control the speed of the digital track but also control the pause and play using the footswitch, making the task of transcription more efficient. Member checking is an important part of the interview process and enforces the rigour of the research (Baxter & Eyles, 1997). Upon completion of each interview transcript I then made a copy for editing in which I removed all of the ‘ums’, ‘ahs’, various slang, and otherwise distracting elements used in spoken language from the transcribed text. I then emailed this transcript to the participant in Word document format encouraging participants to read through the document and to make any changes that they wanted to what they said. The process from interview to return of the transcript to the participant for checking and input took between two and four weeks. I asked participants to email me back their approval or disapproval of the transcript and to also include a copy of the changed transcript should they have made any changes. Once a participant approved of the transcript, with or without 62 changes, that version of the transcript became the official record. Allowing participants to edit what they said not only ensures that you have captured what they said, but allows them to articulate what they meant to say or reconsider what they said. Asking participants to give their approval or disapproval of the transcript gave them every opportunity to withdraw their participation if they so wished. I used two approaches to analyzing the interview data: template and content analyses. I used template analysis for the specific research questions and the more conventional approach of content analysis to identify themes that occurred outside of the interview questions (King & Horrocks, 2010). Prior to conducting the analyses, I assigned a unique alpha-numeric identifier (P1-P15) to each interviewee and corresponding transcript to preserve the anonymity of each individual during the analyses. These alpha-numeric identifiers were also cross-referenced with the group number. Keeping the identity of the participants anonymous for the analyses served to: guard against my own interpretation or embellishment of the data based on my recollection of the interviews; and from being inadvertently influenced by knowing which transcript belonged to which participant (Van den Hoonaard, 2012). This strategy also served to reinforce the rigour of my research (Bradshaw & Stratford, 2010). Template analysis lends itself to a flexible coding structure where the researcher can predetermine the themes in advance of the investigation by constructing a hierarchical theme template (King & Horrocks, 2010). The majority of the interview questions that I asked were very specific because I was seeking clarification and greater meaning to themes I had already identified in the review of other data. I was also using this data as a means to check the validity of my initial conclusions from the other data (Greene et al., 1989). One of the 63 strengths in using template analysis is that it allows the researcher to look for trends among various participant perspectives (King & Horrocks, 2010). Therefore, because these particular interview questions were already established themes and the questions were asked across three different groups of people (King & Horrocks, 2010), I decided to use template analysis for the specific interview questions. For the specific interview questions I constructed a template using an Excel spreadsheet that I set up ahead of time with the alpha-numeric identifiers and corresponding group numbers for each question/theme (for question/themes used in this examination, see Appendix V). Before I began putting information into the template I re-read each transcript and the answers to the questions in a separate Word document. I then summarized each participant’s answers further and entered the data into the spreadsheet template. After entering all of the participant’s answers I was able to examine all of the answers given to each question and look for trends within the whole and among the groups. Not every question that I asked was part of a theme that was analyzed using template analysis. These questions and other dialog were analyzed separately using content analysis. To identify themes within the remainder of the discussion that flowed out of the interviews I began with open coding and progressed to focused coding and finally to the formation of themes (Van den Hoonaard, 2012). I began by re-reading each transcript several times. Using open coding I took notes of words that seemed to occur often (Van den Hoonaard, 2012). I then sorted the interview transcripts by group and compiled all of the interview transcripts into one Word document per group, removing the text that belonged to me. 64 Creating these three master Word documents allowed me to maintain individual anonymity for the study, but retain the group identity. Using the word search navigation function of Microsoft Word, I was able to find and tally specific words. I conducted both manifest and latent content analyses (Dunn, 2010) using the words I had recorded earlier when reading through the transcripts as well as synonymous or associated words. Thus, I was able to tally the number of times a word was used by group and in what context. I used an Excel spreadsheet to track and tally the words searched by group. The next step was to use focused coding to sort the words into subthemes (Van de Hoonaard, 2012). Once the search words were sorted into subthemes I classified the words as subtheme identifiers (see Appendix VI). Next, I needed a method of separating sub-themes that did not have a specific context as well as the data that were already analyzed in the template analysis. So after forming subthemes I assigned each subtheme the value of either ‘general’ or ‘specific’, as it related to the context of the subtheme. Subthemes assigned ‘general’ were those that had a very general context whereas subthemes assigned the value ‘specific’ had specific context. In addition, all the subthemes were identified as to whether or not they were covered by the first analysis. These subthemes were assigned the value ‘direct’. Those subthemes not covered by the first analysis were assigned the value ‘indirect’. Subthemes that were both ‘specific’ and ‘indirect’ were then sorted into themes, which were topical and contained more than one subtheme. Those subthemes that did not form a theme can be seen in Appendix VI. The subthemes that had been identified as both ‘specific’ and ‘indirect’ were placed into an Excel spreadsheet where the subtheme identifiers could be tallied by subtheme and by theme. To identify which themes were important to the interviewees, I used the absolute sum 65 of subtheme identifiers to rank the themes (see Appendix VI for details). After re-reading the transcripts, I inferred that an absolute sum of 100 or greater meant that the theme was important to the interviewees (Gray & Densten, 1998). In other words, the transcripts affirmed that the themes scoring higher than 100 were important because they were easily found within the documents, whereas themes with scores less than 100 were more obscure and difficult to find within the documents. Using this technique was a way for me to limit the power I have as the researcher, deciding what to present and what not to present (Dunn, 2010). I revisited the participant transcripts to ensure that what came out of both analyses were in fact what the interviewees were communicating. The accuracy checking procedure involved re-reading each interview transcript several times from start to finish and subsequently checking specific transcripts for specific details. The procedure not only served to check the accuracy of both analyses, but also to check my own subjectivity. The themes that were deemed important are presented in the next chapter. For all the themes identified I also made tables that included the subthemes with the relative frequency of subtheme indicators by group. I made these theme tables to demonstrate how often a group talked about the subtheme relative to the other groups, and to identify if there was any solidarity within the groups themselves (see Appendix VI for details). 66 Chapter 5: Results and Analyses 5.1. Introduction My research data were acquired through a variety of sources and the scrutiny of these data were as complex as their acquisition. Therefore, for clarity, I have chosen to present the findings in the same format and order that I did in the Data Collection section of Chapter 4. I will begin with what I found out about mill closures from the media scans and document analysis which will be followed by a summary of employment. Next, I will reveal the results of my inquiries regarding the timber harvest and valuation information I obtained from British Columbia’s Harvest Billing System (HBS). Finally, I will demonstrate what I discovered in the breakdown of the semi-structured interviews that I conducted. 5.2. Mill Closures The media scan of the Prince George Citizen newspaper revealed that during the period between 1997 and 2011, within the Prince George Timber Supply Area, a total of eight lumber manufacturing facilities were closed, one was opened, and two received substantial upgrades. These mill closures were as follows: Northwood PG Wood, Canfor Netherlands, Canfor Upper Fraser, Winton Global Lumber, Canfor Rustads, and Canfor Clear Lake in Prince George District; Pope & Talbot Inc. and Stuart Lake Lumber Co. Ltd. in Fort St. James District. There was also the loss of a plywood plant in the Prince George District due to a fire. Six of the nine closures occurred between 2008 and 2011. During the same time period there was one sawmill opening within the Fort St. James District, one sawmill upgrade within the Vanderhoof District, and one sawmill upgrade within the Prince George District. 67 As there appeared to be no direct correlation between the mill closures and the changes to legislation in 2003, I needed to look for other explanations for the closures. To compare and contrast the sawmill closures and openings within the Prince George Timber Supply area to other events happening, I constructed an event timeline (see Appendix II). The types of events that I included were those having to do with the mountain pine beetle epidemic, lumber markets, forest policies, and legislation. Looking at all of the events within the context of the forest industry and time period, the event timeline provided me with clues as to why the mill closures occurred. On the surface, mill closures and openings seemed to coincide with events occurring within the lumber markets. An ancillary investigation of the internet for news regarding the mill closures identified above revealed that for some mills, the official closing date was years after the fact (Canfor, 2011). So, for the purpose of this inquiry, the closing date is the date when production ceased. Using the mill closure criteria listed above, the additional investigation revealed that the six closures within the Prince George Timber Supply Area occurred between 2007 and 2010. In 2007 Winton Global Lumber (Hoekstra, 2011), Stuart Lake Lumber Co. Ltd, and Pope & Talbot Inc. (Hamilton, 2008) were closed. In 2008 Canfor Corporation (Canfor) closed its Rustad mill (Hoekstra, 2010), and then its mill in Clear Lake in 2010 (Hoekstra, 2011). Meanwhile, the mills that remained open reduced their production capacity by reducing the number of shifts at various times (Canfor, 2008; Hamilton, 2008). As for the two defunct mills in Fort St. James, both were purchased by other companies in 2009. Pope & Talbot Inc. was purchased by Conifex Timber Inc., who re-opened the mill in 2009 (Tice, 2009). However, Dunkley Lumber Ltd. chose not to re-open the mill when they purchased Stuart Lake Lumber Co. Ltd. (Nielsen, 2011). Incidentally, Canfor did not rebuild its 68 plywood plant that was destroyed by a fire; rather, the insurance money was used to invest in upgrades for other facilities (Canfor, 2008). To understand some of the reasons why these mills were closed, it is useful to look at what was occurring elsewhere in the province. In fact, the Prince George Timber Supply Area was not the only location to experience sawmill shutdowns as there were massive shutdowns occurring throughout the province at that time, both in the Coast and the Interior forest regions (Hamilton, 2008). For example, Pope & Talbot Inc., a company based in the United States with numerous lumber and pulp operations in British Columbia, filed for bankruptcy protection in October of 2007 (VanderKlippe, 2007). For many forest companies like Pope & Talbot Inc., the decline in profitability began in late 2006 when lumber prices began to fall as a result of declining housing starts in the United States. Profitability was further confounded in 2008 when the housing market crisis in the United States began to unfold (Hoberg, 2010). Not only were lumber prices reaching historic lows, the value of the Canadian dollar was high (Hoberg, 2010). Furthermore, the United States was still British Columbia’s largest consumer of lumber products (BC Stats, n.d.) so these downturns in the lumber market would have had a significant negative impact on the lumber manufacturing sector. The housing crisis in the United States was not the sole cause of the problems facing Pope & Talbot Inc. (VanderKlippe, 2007) or other forest corporations (Hamilton, 2008). Most companies were already stretched financially due to the lengthy legal battle with the United States over softwood lumber and the ensuing high tariffs imposed by the US Department of Commerce in the form of countervailing and anti-dumping duties (Zhang, 2007). From 1997 to 2001 the 69 export of Canadian lumber into the United States was controlled by quotas established by the Softwood Lumber Agreement of 1996 (Zhang, 2007). 5.3. Employment BC Stats reveals that there has been a steady decline in employment in the forest industry since 1997, with an overall decline of 43% within the forest industry in British Columbia. However, employment in the forestry and logging sector has had a much greater decline (57%) than that of the wood manufacturing sector (33%) (see Table 1). What is interesting is that if you look at the annual change there are three relatively large negative declines in the forestry and logging sector, 2001 (10,700), 2004 (6,000), and 2008 (6,900). In contrast, the wood manufacturing sector shows only one significant negative decline, it occurred in 2008 with a loss of 10,100 jobs. Cumulative losses from 1997 to 2011 are 18,900 jobs in the forestry and logging sector and 14,700 jobs in the wood manufacturing sector. Table 1: British Columbia Employment by Detailed Industry, Annual Averages. NAICS* Sector Forestry ŧ 1000s Product § 97 98 99 00 01 02 03 04 05 06 07 08 09 10 11 32.9 30.2 29.9 35.5 24.8 25.2 27.1 21.1 21.3 21.4 24.2 17.3 13.9 16.1 14.0 Resource 60% 59% 62% 68% 61% 67% 61% 61% 58% 50% 52% 39% 35% 40% 35% Goods 8% 7% 8% 9% 7% 7% 7% 5% 5% 5% 5% 4% 3% 4% 3% All 2% 2% 2% 2% 1% 1% 1% 1% 1% 1% 1% - - - - 1000s 44.8 43.1 42.1 45.9 48.7 43.5 48.4 45.7 45.1 43.8 43.7 33.6 26.8 28.7 30.1 Manufacture 23% 22% 22% 23% 25% 22% 24% 22% 23% 23% 22% 18% 17% 17% 18% Goods 11% 11% 11% 11% 13% 11% 12% 11% 10% 10% 9% 7% 6% 7% 7% All 2% 2% 1% 1% 1% 2% 2% 2% 2% 2% 2% 2% 2% 2% 2% Source: Statistics Canada, Labour Force Survey (unpublished data). Prepared by BC Stats, February 2012.*NAICS – North American Industry Classification System; ŧ Forestry and Logging with support activities; § Wood Product Manufacturing. Percentages reflect the relative representation to the specific sector, the Goods-Producing Sector, and All Industries. 70 The significant loss of 10,100 jobs in the wood manufacturing sector in 2008 can definitely be linked to the large number of mill closures throughout the province that year (Hamilton, 2008). However, the significant losses experienced in the forestry sector in 2001, 2004, and 2008 require a more protracted explanation. As jobs in the forestry sector are linked to the timber side of the business rather than the manufacturing side, one needs to look at those factors which may have influenced the harvesting activities of firms. In 2001, the forestry sector experienced a loss of 10,700 jobs. The forest industry had been under the costly regulatory regime of the Forest Practices Code of British Columbia Act1995 since 1994 (Hayter, 2000; Pearse, 2001), and expiration of the Softwood Lumber Agreement in 2001 created unsavory business conditions (Zhang, 2007). However, it was likely a combination of the high countervailing duties placed on lumber and the high price of stumpage that had the greatest impact on forest sector jobs (Hamilton, 2001; Jackson & Curry, 2002). Given that the forest industry in the Coast forest region of the province was having much more financial difficulty, due in part that logging costs there superseded those elsewhere in the province (Grafton et al., 1998; Pearse, 2001), it is reasonable to assume that that most of these job losses likely occurred there. In 2004, the loss of 6,000 jobs in the forestry sector can likely be partially attributed to changes in technology (MOF, 2001b). Changes in logging and trucking technology allowed companies to reach higher levels of efficiency (Stirling, 2004), which meant greater production and fewer jobs. As for the 6,900 job losses in 2008, these can be linked to the slowdown in the economy triggered by the housing crisis in the United States (Hoberg, 2010). As a means of preservation, forest companies drastically curtailed the production of lumber and their harvesting operations (Canfor, 2008; Hamilton, 2008). 71 More specifically to the Prince George Timber Supply Area, an analysis conducted by Horne (2009) compared forest industry employment in 2006 between the Prince George, Vanderhoof, and Fort St. James Districts (see Table 2). Table 2: 2006 Direct Employments in Logging and Other Wood Manufacturing Sectors. District Logging Prince George 2443 Vanderhoof 752 Fort St. James 290 Source: Horne (2009). Percent of Total Direct Employment Other Wood Manufacturing 8% 18% 19% 2931 992 418 Percent of Employment Total Direct Estimate in Employment Forestry Sector 10% 26% 24% 45% 28% 49% Note that in terms of relative importance, the employment in the Vanderhoof and Fort St. James Districts is nearly double that of the Prince George District. Horne (2009) also produced a special report for the ministry responsible for forests in 2009 where he reconfigured the alignment of data collected by BC Stats and Statistics Canada in 2006 from Regional District to Forest District boundaries. This data reveal the economic dependency of each Forest District in the province, where it is described using indicators such as regional diversity, dominant basic income sources, dependence on forestry and wood processing, forest sector vulnerability, dependence on underground resources, and dependence on tourism. From this information it is clear that the Vanderhoof and Fort St. James Districts are extremely dependent upon forestry, (see Table 3 for details). 72 Table 3: 2006 Economic Dependencies by Indicator and Forest District Economic Prince George Dependency Forest District Indicator Regional Diversity 2 Dominant Basic Public Sector Income Source Dependence on 21-26% Forestry & Wood Processing Forest Sector 4 Vulnerability Source: Horne (2009). Vanderhoof Forest District Fort St. James Forest District 1 = Least Diversified Forestry &Wood Processing 27-70% 1 = Least Diversified Forestry & Wood Processing 27-70% 5 = Most Vulnerable 5 = Most Vulnerable In addition to the analyses done by Horne (2009), approximate job losses associated with the closing of mills between 2007 and 2010 in the Prince George Timber Supply Area were obtained from the ancillary media scan (see Table 4). For the three year period, there were an estimated 767 direct jobs lost in the wood manufacturing sector in the Prince George Timber Supply Area. The purchase of Pope & Talbot Inc. sawmill by Conifex Timber Inc. in 2009, which was reported to employ approximately 230 people (Tice, 2009), are included in the final total estimate of 767 direct jobs. Table 4: Approximate Direct Employments Loss Due to Mill Closures in the PGTSA. Mill Closure Closure Year Affected Employees District Winton Global 2007 ~ 220 Prince George Stuart Lake Lumber 2007 ~ 100 Fort St. James Pope & Talbot Inc. 2007 ~ 272 Fort St. James Canfor – Rustad 2008 ~ 220 Prince George Canfor – Clear Lake 2010 ~ 185 Prince George Source: Data for Canfor - Clear Lake (Hoekstra, 2010), all other data (Hamilton, 2008) Specific employment figures for the Prince George Timber Supply Area are not available to the extent that the provincial figures are. However, the similarities between the timing of mill closures locally and provincially indicate that the Prince George Timber 73 Supply Area likely experienced similar job losses in the forestry and wood manufacturing sectors. Also important to note is that the estimated 767 job loss in the wood manufacturing sector cannot be added to the work done by Horne (2009), as his analysis was done using official census data and by splitting jurisdictions. 5.4. Timber Harvest and Valuation For the inquiry using the Harvest Billing System (HBS) data, an assumption was made that logs are processed where they are scaled. Although trading does occur between timber processing facilities, it is assumed that the majority of logs are transported directly to where they will be processed because handling and transporting logs is expensive. Thus, it is assumed that for economic reasons, logs are sorted at the point of harvest so that they are delivered directly to where they will be processed. One particular limitation of using scale data from the Harvest Billing System is that the majority of the logs in the British Columbia interior are weigh scaled (FLNRO, 2011b). As defined by the Scaling Manual (FLNRO, 2011b), weigh scaling is a statistically valid method of scaling where only a sample of the population is measured. So ultimately, the method of weigh scaling results in a certain amount of averaging of the scale data. Hence, when a truckload of logs arrives at a mill site the truck is weighed and the logs it is carrying are placed into a stratum by the weigh scale operator. A stratum is a defined species mix ratio that is predetermined on an annual basis and are unique to the scale site (FLNRO, 2011b). Each stratum has its own sampling frequency, which determines how often loads are sampled (FLNRO, 2011b). So, after the truck load of logs is placed into a stratum and weighed, the logs are then unloaded and the truck is weighed again on the scale empty to calculate the net weight of the logs. The conversion factor for that stratum is then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