Executive Summary
Reports that a substantial proportion of wood raw materials, used by Chinese and Vietnamese manufacturers, are from illegal sources have drawn concern from major consumer countries who recently implemented timber legality regulations. These regulations, which include the Japanese ‘Goho-wood’ policy, the U.S. Lacey Act, the EU Timber Regulation and the Australian Illegal Logging Prohibition Act restrict the import of illegally harvested wood and are expected to have a direct impact on major wood processing countries, such as China and Vietnam.
 
This study focused specifically on the wood products industry and business practices in these two processing countries, targeting individual furniture and flooring manufacturers and wood products traders, as a way to clarify and evaluate the effects of timber legality regulations.
 
Surveys were conducted at trade shows in Ho Chi Minh, Shanghai and Guangzhou in 2013 and 2014 to assess how these regulations influence attitudes and perceptions regarding regulations, firms’ use of chain of custody certification, and impacts on the material sourcing and export market decisions of industry managers. Survey responses were evaluated using descriptive statistics, regression analyses, cluster analysis, non-metric multidimensional scaling and analysis of similarity.
 
The analysis showed that as firms increase in size they reduce domestic sales and show increased awareness and support for regulation, and that firms’ awareness of timber legality regulations plays a significant role in whether a firm decides to obtain certification. Analyses showed that Vietnamese firms have lower awareness of regulations while being more supportive of regulations. Chinese firms have higher awareness while having a more negative attitude towards regulations. The findings also highlighted a split between firms with a domestic focus and firms which export to foreign markets suggesting a split in the market which may reduce the impact of regulations. This segmenting of the Chinese market and to a lesser extent the Vietnamese market supports the idea that regulatory leakage is taking place, wherein sales of wood products from suspicious sources are shifting away from regulated markets and towards unregulated markets which are experiencing rapid increases in demand for wood products.
 

Executive Summary
The Washington State Department of Natural Resources (DNR) manages 2.1 million acres of timberland. Timber harvested from these lands has averaged 12 percent of Washington 's total annual timber harvest level from 1989 through 2003. Management plans have incorporated more ecosystems goals such as promoting ecosystem health, habitat conservation, and increasing the structural complexity of the timberland. The management plan adhered to by the DNR will likely result in more heterogeneous stands of timber and thus timber sales of more heterogeneous saw timber.
The goal of the study was to examine how the diversity of saw timber in a given timber sale impacted the final sale value in lump sum sales of western Washington . Saw timber diversity was defined by the distribution of volume among different species and saw log grades in a given sale. Prior research predominately located in the
U.S. South had shown a reduction in timber sale value as sales included a greater level of saw timber diversity. This result is based on the theory that increasing transaction costs associated with the processing of more heterogeneous stands led to a less desirable timber sale characteristic per sale. These processing costs include harvesting, sorting and the reselling of undesirable timber.
The study focused on lump sum state timber sales in western Washington . Lump sum sales require prospective buyers to bid on the right to harvest an entire timber sale. A parcel is first advertised and then auctioned as one unit, establishing a fixed payment for the timber sale by the winning bidder. Since this method of sale requires the buyer to purchase and harvest all timber advertised in a sale, it can result in forcing buyers to purchase timber species and saw log grades that they have little or no interest in. This rationale led to the hypothesis that a negative relationship would be observed between increasing saw timber diversity and the final sale value of lump sum timber sales in the data. The detailed inventory information provided by the DNR enabled the calculation of an index value for saw timber diversity that could be used as a metric in empirical examination of the relationship between diversity and the final sale value.
A diversity variable was created to facilitate examination of the impact increased heterogeneity of saw timber in a tract has on the final sale value of lump sum timber sales. This required the calculation of a diversity variable that would account for the species and grade characteristics of each individual timber sale. To this end the Shannon- Wiener index was selected as the best method for calculating this variable. It was an appropriate choice for this data because it could be calculated using the detailed inventory information. In this research stand diversity or heterogeneity applied only to the species of trees and log grades that were included in the timber sale data.
Wildlife and other facets of a timber stand were not included in the calculation of the diversity index.
The diversity index created had a range of 0 for a completely homogeneous timber sale to 3.689 for a completely heterogeneous timber sale. Slightly less than 800 sales had values from 1.4 and 1.7. Another 700 sales included values from 0.7 and 1.3. About 500 sales had values from 1.8 to 2.2. The remaining sales were distributed above and below these ranges. A total of 2194 sales were included in the study.
Other variables examined in the model to explain the final sold value of the timber sale included the total number of bidders on a timber sale, the total acreage of the timber sale, the contract length of the timber sale, the total number of miles of required road reconstruction, the Douglas fir volume of the saw log grades P, 2P, 3P, SM, 1S, the Douglas fir volume of the saw log grade 2S, the Douglas fir volume of the saw log grade 3S, the Western Hemlock volume of the saw log grades P, 2P, 3P, SM, 1S, the Western Hemlock volume of the saw log grade 2S,

the Western Hemlock volume of the saw log grade 3S, all other volume included in the timber sale, and the WWPA lumber index price of Douglas fir. During the process of model fitting a diversity measure of the timber sale accounting for only the distribution among eight possible tree species and a diversity measure of the timber sale accounting for only the distribution among the five possible saw log grades were used in alternative model comparisons.
The final sold value of a timber sale in U.S. dollars represented the winning bid of a timber sale. By including the volume found in the three highest grades of both dominant and co-dominant species, as well as the other sale volume on the right hand side of the equation, the problem of scale with respect to the dependant variable was alleviated. In other words, the existence of large bid values skewing estimates simply due to large volumes was eliminated.
Seven models were estimated with alternative sets of independent variables. There was consistent evidence that the final value of DNR timber sales located in western Washington were negatively influenced by increases in the level of saw timber heterogeneity over the period of study. Heterogeneity among tree species was found to impact final sale value more than heterogeneity among saw log grades. A possible reason for this result is that commodity producers generally focus on a tree species or a certain range of grade classes. For instance a sawmill may be best geared to mill #2 and #3 saw logs, or perhaps a commodity producer uses only Douglas fir in the manufacture of its products. Increasing heterogeneity of saw timber in a lump sum framework forces these bidders to bid on greater volumes that they are not interested in and may in fact have to resell. This is believed to be viewed negatively by bidders as an additional cost of doing business. Alternative theories are likely to exist as well that can increase the cost of harvesting and marketing logs.
The impacts of saw timber heterogeneity are not well serviced by the lump sum method of timber sale. Timber sales in which greater levels of saw timber diversity are observed may return greater revenues to the DNR if another method of sale is instituted. Additional empirical work on heterogeneous timber sales focusing on how the DNR can create bundles of timber from these sales attractive to different bidders would be pertinent.
In addition to the effect of saw timber diversity, this study found significant evidence that an increased pool of bidders and therefore increased competition for a timber sale had a positive impact on the market value. However, the data indicates that the level of competitiveness declined over the period of study. The existence of a competitive framework among bidding firms is a key to achieving a final timber sale value at or near its true market value. Declines in the average number of bidders on timber sales in the data set may be caused by a number of factors. Regardless, these declines may be cause for concern and further research into why they are occurring and what can be done to alleviate the impacts would be relevant.
The total acreage of timber sales in the study region displayed diseconomies of scale. While this result was not predicted, it is not uncommon in the literature. Munn and Rucker (1995) and Boltz et al. (2002) both found significant evidence that parcel size negatively impacts final sale value. However, this variable presents a clear focus for future study to explore why increasing parcel size results in reduced final sale value in western Washington . In understanding the implications of this variable it is important to consider who purchases DNR timber and industry shifts over the period of study.
Research regarding how the DNR could increase the number of bids offered on its timber sales and alter its methods of sale would also be valuable. While increasing the competition among bidders is a good way of increasing the timber revenues annually generated by the DNR's timber sale program, mill consolidation in the state of Washington suggests that there may not be a lot of room for this to occur. Additionally, the DNR would not want to adversely impact business relationships it has developed with large commodity producers. They represent a steady demand for the states timber as well as important sources of employment. Future economic research is needed to determine the feasibility and impact of attempts to improve competitiveness and marketing of DNR timber sales.

Executive Summary
The study produced an assessment of the lumber manufacturing sector. It analyzed the changes in this sector and its relation to timber harvest levels. The study's objective was to determine whether Washington 's sawmilling sector can expand or not, given a projection of future timber harvest levels.
We analyzed the Washington State Department of Natural Resources mill surveys from 1968 to 2002. The analysis first defined five timbersheds in western Washington and two timbersheds in eastern Washington . The study then focused on an analysis of capacity utilization by the sawmilling sector. We produced trends of utilization rates and discussed potential reasons why the rates have changed over time and among timbersheds. Periods of high prices were related with periods of high utilization rates, with exception. During the 1980's, there occurred a shutdown of capacity that improved the average utilization rate for the sector following a collapse of high lumber prices in 1979. Currently, the average utilization rate is at historically high levels as one might expect given the strong U.S. housing sector. Lumber prices have recovered from a short period of lower prices, and they are currently at near-high levels providing impetus to the high capacity utilization in the sawmilling sector. We also found differences in the utilization rates among the timbersheds, and they are presented in the body of the report.
We followed the capacity utilization analysis with an assessment of log consumption. Log use by the sawmilling sector within respective timbersheds was compared with the timber harvest level. Except for the South Coast, Southwest and eastern Washington timbersheds, sawmills were now, by far, the main consumer of the harvest level. Substantial amounts of saw logs continue to move from one timbershed to another. In 2002 approximately 600,000 mbf of timber was transported across timbershed boundaries to be used by sawmills in other timbersheds. Timber heading to Oregon continued to be significant, and logs imported from British Columbia were now occurring.
We examined the potential supply of timber for western Washington timbersheds. The complexity of projecting uneven-aged stands found in eastern Washington limited the analysis to western timbersheds. While the projection was considered preliminary, it was a useful first step to gauge the wood availability required to maintain or expand the sawmilling capacity in western Washington. Further sensitivity analysis is required but was beyond the scope of this phase of the research. The projections indicated that current harvest levels can be sustained, and in the South Coast timbershed, the harvest level can be increased over the next several decades. There did not appear to be any indication that the harvest level will fall below the current level of 2.8 billion board feet. Timber inventories in all timbersheds revealed a significant growth in volume in older age classes given the projected harvest level suggesting there exists the potential for a higher, future harvest level if these forested lands were made available for timber production.
Our projections assumed current harvesting conditions will continue to exist into the future. One conclusion we draw from the analysis is that the biological potential of the timber land itself will not likely be a constraining factor in future timber harvest levels. Rather, regulatory and land-use factors are more likely to impede a harvest level that coincides with the biological potential of the forested lands.
There are important policy implications from our findings. Lumber manufacturing in Washington has become the principal consumer of wood fiber in the state, and we projected, given their current high rates of capacity utilization by existing mills and the biological potential for increasing harvest levels, that lumber manufacturing can expand. While Washington 's forest products sector has changed substantially due to the significant decline in timber harvest level over the past decades, the sawmilling sector has maintained its level of use of the harvest during this time. There has been a substantial decline in the number of sawmills, and the volume of log that crossed timbershed boundaries continues to be significant, but still, lumber manufacturing, a sector that consumed a small percentage of the log harvest level 20 years ago, is now the predominant end-user of logs harvested in Washington, and it is in a position to grow. The change in the composition of the forest sector was not driven by a substantial growth in lumber manufacturing but rather the decline in timber harvest levels and its impacts of the other forest sectors in Washington, primarily log exports.
The lumber manufacturing sector has not been without its share of change however. There has been a consolidation of milling capacity, and with it, a reduction in the amount of labor employed by sawmills. Lumber mills in Washington have transitioned from a large number of smaller-sized mills capable of utilizing a wide range of log sizes to a consolidated sector that utilizes smaller logs with more capital and less labor. There is also much less mill-type diversity within the sector.
We conclude that harvest levels in the future are such that they should allow for lumber manufacturing to expand. We support this conclusion with the facts that capacity utilization rates have been high during the past decade and that timber harvest levels in Washington have adjusted to a lower level than in previous decades. The harvest level simulations suggested that current harvest levels are sustainable into the future over a couple of rotations. The simulations assumed conditions today will continue into the future. For this reason, further work is needed to assess changes in land-use patterns and regulatory constraints that may impact future timber harvest levels from Washington 's forested lands.

​Executive Summary
The study examines where excess wood exists, how much can be imported to the US, and what opportunities exists for US investments overseas.  The study begins with an examination of global demand.  Data on global consumption of industrial roundwood reveals a structural break in consumption patterns during the early 1990’s.  Part of this break is the result of the collapse of the former Soviet Union.  The shut down of not only its consumption but also its production sectors has had a visible impact on global consumption.  Also efforts to produce timber in a sustainable fashion in tropical forests and environmental restrictions on softwood timber harvests significantly constrained timber supply in the 1990’s, leading to reduced global consumption of forest products.
Two projections of future consumption are made.  Using a growth rate of consumption observed prior to the 1990’s results in projected consumption of nearly 3 billion cubic meters by 2050.  Using a growth rate of consumption estimated during the 1990’s results in a projection of 2 billion cubic meters by 2050.  Near-term consumption is projected to increase from 300 million to 800 million cubic meters over the next 20 years.
The study also examines global timber supply projections using the ATLAS timber projection model, and their implication for excess supplies of wood fiber.  Excess supplies are defined as the volume over rotation age assuming no growth in current demand.  The model produces the biological potential for timber production for plantations established in New Zealand, Australia, Indonesia, Brazil, China, Argentina, Uruguay and Chile.  The projections are based on specific assumptions about growth rates, rotation ages and areas planted in 1995.  The projections indicate that in the short to medium term (from now to 20 years from now), there may be up to 400 million cubic meters (MMm3) of available wood in the Asian wood basket.  These wood resources are close to China and represent 3.5 times the timber consumed in China in 1997.  The Asian plantation wood is likely to compete with non-plantation wood particularly from Russia, and to some extent Scandinavia.  Over the longer term, it is likely to compete with Chinese plantations as well.
Several countries in Latin America have the potential to develop inventories above rotation age totaling over 500 MMm3.  Much of the additional short-term fiber from these plantations may fill European and North American markets, but are likely to come under competition from large non-plantation wood fiber sources in the northern hemisphere, such as Canada and Scandinavia, as well as the US fiber resource in the South.
The study also estimates economic supply for softwood logs using the CGTM.  We develop cost curves by ranking the quantity of sawlog supply available at a given price.  The cost curves assume no growth in demand over the projection period from 1993 to 2040.  An additional 200 MMm3 of sawlogs would be produced with an increase of $188/mbf (2000US$ or $40/m3 in 1980US$).  Finland produces the lowest cost sawlogs followed by New Zealand and then the US South.  The three regions provide the bulk of the first 100 MMm3 of addition sawlogs.  The interior region of western Canada, sourcing wood fiber from native forests, provides additional wood in the mid- to longer-term.  Regions such as the US West provide little or no additional wood supply because they are meeting current demand.
Supplying the least cost manufacturing capacity is modeled in a similar fashion as supplying least cost sawlogs.  The European region of Finland, Sweden and the western continental countries provide the majority of the lower cost manufacturing capacity.
The study concludes that while plantation wood may have a biological potential to produce nearly 1 billion cubic meters of wood fiber in the near-term term, there will be competition from wood fiber from non-plantation sources including Scandinavia, Canada and the US.  The sourcing of non-plantation wood fibers from these regions appears to remain more competitive with plantation fiber according to simulations with the CGTM. 
The 1 billion cubic meters of potential wood fiber is greater than the upper bound of 750 million cubic meters projected for the near-term demand for industrial roundwood.  However, the projection of biological supply is sensitive to plantation growth rates and rotation ages used in the study.  Timber inventory projections decline significantly with changes in these assumptions.  Future work will require better information on area planted, their growth and management intensities defining rotation ages.  

​Summary
The Pacific Northwest has traditionally been a strong participant in international forest products markets in the Pacific Rim.  The combination of a substantial resource base, processing capabilities to manufacture high quality timber products and export facilities have contributed to a comparative advantage in the export of wood and wood products.  The region has been able to maintain its comparative advantage, and hence its market share in the Pacific Rim at a time when other U.S. regions, notably the South, have successfully increased domestic market share.
Recent environmental legislation has reduced the timber resource base substantially and will impact the management of additional areas.  A 44 percent decline in timber sales volume in 1989 will result in an annual 9.7 million cubic meters (2.0 billion board feet) reduction by 1995 in public timber harvest.  As a consequence, public timber revenues will deadline and there will be higher relative sawlog cost to both domestic and foreign purchasers of timber produced in the region.  The 75 percent log export restriction by the state of Washington will lower these costs for domestic mills, but increase them further log foreign purchasers.  The reduced timber harvest and trade restriction will also diminish the export value of logs and finished products originating from the region.  Hence, the region's comparative advantage and market share will decline as other regions will become more competitive with the Pacific Northwest in forest products trade.
Specific measures of these impacts associated with a 2 billion board feet reduction in public timber harvests are:
1)  The decline in timber sales volume from public lands reduces public timber receipts by$176 million (in 1988 dollars) in 1995 and reaches $188 million in 2000.  The total revenue decline for the public sector amounts to $1.9 billion d6llars for the 11 year period (1989-2000).
2)  The reduction in timber sales volume from public lands in the PNW region leads to increased log production from overseas producers.  Other producers offset nearly 90 percent of the timber reduction in the public sector.  Over half of the harvest reduction is made up overseas in only two years.  The private sector in the PNW region is unable to effectively increase the share of the log market.  Harvests from the private sector in the region begin to decline in 1993.  By 2000 timber harvest from the private sector in the PNW is only ten percent of the contribution by other regions.
3)  The impact on sawlog costs for the sawnwood and plywood sectors is substantial.  A 9.7 million cubic meter reduction in public timber harvest will increase annual log costs by $1 billion in 1995 and 2000 on a global basis.  Log costs in 1995 in the PNW region will increase $211 million compared to $63 million for the South and $34 million for Canada.  A similar pattern is observed in 2000.  In Japan, the impact of reduced timber supply increases log cost by $131 million in 1995 and $90 million in 2000.  Log cost impacts in 2000 are lower as Japan adjusts to the decrease in public timber harvests.  The price increase for sawlog will allow marginal softwood log producers to harvest more logs.  The substitution of hardwood logs for softwoods can also be expected in regions where technological and economic constraints are not binding.
4)  The impacts on profits for sawmills and plywood mills are also substantial and regionally distributed. Profits in the U.S. West decline in 1995 by $79 million.  There are only modest gains in profits in the South:  $7.4 million.  In 1995, Canada increases its profit in the sawmill industry by $84 million.  The decline in profits in 2000 for the U.S. West is $60 million, while the South's profit increase by $1 .3 million.  Canada, by 2000 increases its profits in the sawmill industry by $90 million.  Japan's sawmill industry sees a decline in profits of $77 million in 1995 and $32 million in 2000.
5)  In the export markets, the value of U.S. log exports declines by $152 million in 1995 and$168 million in 2000.  Globally, the value of log exports decreases by $75 and $83 million in 1995 and 2000 respectively. Nearly one quarter of 'the decline in U.S. log export value is recovered by Chile and New Zealand during this period.  In the sawnwood markets, the decline in value of coniferous sawnwood is matched by an increase in export value from Canada.  The value sawnwood exports by U.S. West is reduced $135 million and $246 million in 1995 and 2000, respectively.  Canada increases its export value of sawnwood by $152 and $205 million in 1995 and 2000 respectively.  In the plywood sector, the U.S. West experiences a decline in export value of $45 million in 1995 and 2000, while the U.S. South increases its export value by $57 and $58 million in 1995 and 2000 respectively.
6)  The diminished supply of timber in the region will reduce lumber and plywood production globally and increase the prices of finished product.  Globally the cost to consumers in increased lumber and plywood prices is $849 million in 1995 and $904 million in 2000.  For the U.S., the consumer costs are $161 million in 1995 and $198 million in 2000.  The price increase for lumber and plywood will promote greater non-wood substitution as end users will prefer new technologies that incorporate less wood inputs on the basis of factor cost.
While policies that promote the conservation of forest resources for non-timber outputs are likely to be successful within the regions in which they are implemented, allowing the region to achieve a higher environmental standard, the impacts of reduced timber harvests will increase timber output pressure on forest resources in other regions and other forests within the region and allow non-wood resources potentially more damaging to the environment to substitute for the wood-based items.  The overseas regions that increase their timber outputs do not maintain their forest resources under a sustainable management regime, forest productivity and forest area will decline bringing into question the overall environmental gains achieved with conservation efforts in the Pacific Northwest.  In effect, imposing a single regional environmental policy only shifts the environmental benefits from one region to another -- old growth forest area will stabilize in the Pacific Northwest, but forest area and productivity in other regions will fall.  When evaluated globally, the environmental benefits of such a trade-off may be negative:  restricting the use of renewable forest resources in one region may produce a negative environmental impact globally.
The study demonstrates the globalization of specific Pacific Northwest harvest constraints and illustrates the need for a better understanding of the environmental benefits which the forest products industry can provide on a global level.  Allowing timber production to take place in regions where sustainable management is more likely to be successful will increase the comparative advantage of forest products versus non-wood substitutes, as well as maximizing environmental benefits within a global context.  The present study is a first step in demonstrating these benefits.  An extension of the present analysis will quantify the environmental effects with the appropriate linkages between forest products production and trade and the production of environmental by-products.
 

​Summary
Starting in late 1988, environmental litigation had a major impact on US Forest Service's timber sales from Westside Region 6 (the Pacific Northwest).  Currently, the prices for most of the timber sold but not yet cut are above the price level that can be expected to be competitive in the market for the foreseeable future.  In combination with a declining trend in softwood lumber demand, Pacific Northwest mills are facing a period of negative growth and restructuring.  The reduction in forest industry output and the negative regional economic impact will be considerably greater than that being caused by the reduced harvest from spotted owl conservation.  The quantity of uncompetitive timber is substantial.  Between 1.5 and 3 billion board feet of the US Forest Service (USFS) inventory from Westside Region 6 is currently uncompetitive.  Related problems exist on Bureau of Land Management (BLM) and state sales that will add to this volume.
Even though changes in contract terms were introduced after the timber buyout legislation in 1984, a setting for massive defaults on contracts has occurred again for the second time in just ten years.  The social and economic costs associated with these problems are substantial, including lost revenue for the public sellers, bankruptcy losses by the processors and customers of the public sellers, substantial litigation costs, regional tax revenue losses, increased public assistance costs, and substantial costs to final consumers.
The current mechanisms in the timber sale contracts--stumpage adjustment clauses and early harvest discounts--have not prevented harvest price speculation and the occurrence of uncompetitive timber sales.  Speculative or desperation bidding, as a result of the environmental litigation, has resulted in contracts for timber that will generate economic losses U the timber is harvested. Substantial increases in product prices would be needed to make these contracts competitive, which does not seem likely in the near future.  A quick resolution of the current problems can reduce the magnitude of these negative impacts.  More fundamentally, it needs to be recognized that the mechanics of current contracts are not conducive to making regional suppliers competitive or to allow stable marketing relationships with their customers, fundamental parameters in fostering a healthy and sustainable economic environment.  Policy changes are required both for relief of the current problem and for the long term benefit of public sellers, producers and consumers.

​Executive Summary
CINTRAFOR is engaged in the Pacific Rim Assessment (PRA), an ongoing analysis of forest products markets in the Pacific Rim.  The primary objective of this research is to assess the future outlook for forest products production, consumption, trade, and prices in the Pacific Rim.  A second objective is to develop analytical techniques and tools to study international forest products markets and policies.
The focus during the current phase of this project is on solid wood product markets, log/fiber markets, and related resource developments.  Although the emphasis is on Pacific Rim regions, especially those that affect the future of the Pacific Northwest, the research encompasses the entire world to maintain comprehensiveness and account for “third-party” trade interactions.
While this manuscript constitutes the final report of the current phase, it might best be viewed as a progress report.  A project like the Pacific Rim Assessment requires significant start-up costs.  At this point, the basic groundwork has been completed and CINTRAFOR is now poised to commence more in-depth analysis of international forest products trade.  We have developed a sound working model, assembled a fairly comprehensive data base, and conducted an extensive statistical analysis of market behavior in the Pacific Rim.  Still much remains to be done.  New data are always becoming available – more recent data and revised data, as well as new data sources.  New research continues to add insights into our understanding of Pacific Rim market behavior.  We hope that this report stimulates discussion and comments that will improve the quality of future analysis related to the Pacific Rim’s for