Modern Economy, 2011, 2, 90-106
doi:10.4236/me.2011.22014 Published Online May 2011 (http://www.SciRP.org/journal/me)
A Spatial Equilibrium Analysis of Policy for the Forestry
and Wood-Pro cessing Industries in Northern Vietnam
Phan Sy Hieu1, Steve Harrison2, David Lamb3
1The Informatics and Statistics Centre for Agric ul t u r e a n d Rural Development, The Ministry of Agriculture
and Rural Development, Hanoi, Vietnam
2The School of Eco no mi c s , The University of Queensland, Brisbane, Australia
3Rainforest Cooperative Research Center and School of Integrative Biology, University of Queensland,
Brisbane, Australia
Received January 28, 2011; revised March 11, 2011; accepted April 15, 2011
Abstract
National objectives for forestry and wood-processing industries have not usually been achieved to the extent
that the Government of Vietnam planned. A spatial equilibrium model with linear supply and demand func-
tions for the forestry and wood-processing industries was developed to analyse policies for northern Vietnam.
Four policy scenarios were established, in each of which optimal levels of prices, supply and demand quanti-
ties were estimated, and solutions were subjected to sensitivity analysis with respect to the elasticity estimates.
The analysis indicates that government objectives cannot be achieved in the 2010-2015 period because plans if
implemented would force the main stakeholders to produce unprofitable quantities. For greater financial viabil-
ity and sustainability, the number of planned new wood-processing factories should be reduced and the design
production capacity of each new factory should be increased to achieve economies of size and international
market competitiveness. Also, the area of planted production forest in northern Vietnam will increase in all four
policy scenarios; however, the growth should be lower than the levels planned for 2015.
Keywords: Five-Year Governmental Plan, Policy Scenario, Supply Target, Engineered Wood Products, Asia,
Vietnam
1. Introduction
Globally the area of natural forest has gradually declined
and as a result many countries have sought to increase
their forest plantation area. The increase has contributed
to biodiversity, environment protection and economic
growth in many countries. A consequence is that devel-
opment of wood-processing industries has been under-
going change, with greater use of plantation timber.
National objectives can be presented in qualitative and
quantitative targets for a given period of time, for in-
stance one year and five years. Some countries regularly
devise five-year plans for forestry and wood-processing
industries. Qualitative and quantitative objectives are usu-
ally specified for the final year of each five-year period,
for example 2000, 2005, 2010, 2015 and 2020 as in
Vietnam [1] and 2002, 2007, 2012 and 2017 as in India
[2].
Qualitative objectives are usually presented as the in-
crease in the participation of private sectors in timber
plantation establishment, the expansion of the wood-
processing industry, and the increase in incomes of peo-
ple including the poor who manage plantation forest in
upland and remote regions. In Vietnam, the qualitative
objective of the national government has been that Viet-
nam will become self-sufficient in terms of domestic
production of wood and wood-processing products. In
the plan, the domestic forestry industry will supply ap-
proximately enough input for the wood-processing in-
dustry, and the domestic production of the wood-proc-
essing industry will not only meet the domestic demand
but also leave some surplus for export [3].
Key and measurable qualitative objectives are usually
quantified in forestry plans. For example, the Indian go-
vernment targeted that the national forest cover would
increase from 25% in 2007 to 33% in 2012 [4]. In Bhu-
tan, one of national targets is to maintain the current for-
est cover of 60% [5]. In Vietnam, a forest cover of 43%

P. S. HIEU ET AL. 91
was set in 1998 for 2010 [6]. More specifically, the
Government of Vietnam [GoV) set supply targets to be
achieved by 2010, 2015 and 2020 for wood and wood-
processing products. For example, the national supply
quantity planned is 5 million tonnes (Mt) for paper
products and 6 Mt of sawn-timber in 2020 [3,7]. These
national quantitative supply targets are divided between
provinces. The GoV assumed that when the set of supply
targets are met, the problems existing in the wood and
wood-processing industries will generally be solved; for
instance unprofitable farm-gate prices, high production
cost of domestic pulp and paper products and shortage of
industrial jobs in rural areas will be overcome [3].
To achieve national forestry objectives, policies in
three broad areas are usually considered for implementa-
tion by governments. The first policy group relates to the
private ownership of plantation forest land, with gov-
ernments proposing the transfer of more forest land to
famers and local communities rather than to state organi-
zations, as in Philippines [7] and Vietnam [8]. The second
group is policies of timber exploitation from natural for-
ests. The exploitation is usually banned (as in Vietnam
during 1992-1998), or highly limited as in Vietnam [8],
the Philippines [7] and Australia [9]. The third policy
group considers how to manage timber markets effec-
tively or how wood markets work competitively, as in
Philippines [7] and Bhutan [5]. Besides policy groups,
changes of governmental institutions are usually planned
in many countries, including Cambodia [10], Vietnam
[8,11], the Philippines [7] and Mongolia [12]. These in-
stitutional changes aim to make government forestry
policies func- tion more effectively.
In the case of Vietnam, the GoV has developed poli-
cies in two main areas, namely to increase investment in
expanding the design production capacity of existing
large-scale state-owned enterprises and to build new large-
scale wood-processing enterprises. To increase the sup-
ply of logs, the GoV implemented the Five Million Hec-
tare Reforestation Program (5MHRP) over 1998-2010
[13].
Globally, national qualitative and quantitative object-
tives for forestry and wood-processing industries are
often unachievable, for instance in Bhutan [5], the Phil-
ippines [7] and India [4]. For example, in India, accord-
ing to Joshi et al. (2010) [4] only three states in the total
of 28 states and 7 union territories will meet the planned
forest cover of 33%. A common reasons reported for
failures in these countries is weak policy and institution
frameworks that make markets work inefficiently, for in-
stance the high marketing cost in the Philippines [7].
Other reasons relate to insufficient funds from govern-
ments, for example in Mongolia [12], and high rates of
poverty in upland regions, e.g. greater than 90% in Viet-
nam [8] and in the Philippines [7]. In these regions, local
people have focused on short-term enterprises and par-
ticularly annual crops rather than long-term income
sources including planting timber trees.
In Vietnam, the forestry and wood-processing Indus-
tries have made major advances in the last 20 years. The
area of plantation forest increased from only about 0.2 M
ha in 1992 to 1.7 M ha in 2006 [14] and to 2.77 M ha in
2008 [15]. The wood-processing industry expanded rap-
idly in the first decade of the 2000s. The average growth
of export revenue was more than 30% annually during
2001-2007. Total export revenue was 61 million USD in
1996, 1.55 billion USD in 2005, 2.2 billion USD in 2006
and 3.2 billion USD in 2008. However, as in other de-
veloping countries, these achievements were much lower
than what the national government had planned. In 2005,
only two new wood-processing enterprises were built in
northern Vietnam, in comparison to the planned 14 new
enterprises. The plan of building a state-owned paper
factory in Kon Tum province was abandoned in 2003
due to the lack of raw materials [16]. The annual growth
of the area of planted production forest in Vietnam was
only 4%, compared with the planned growth of 6% in the
1998-2008 periods. In 2006, 80% of the wood require-
ment for the furniture sub-industry was imported. The
GoV had to adjust the planting area target of 5 M ha to
only 3 M ha in 2010 [17].
Spatial equilibrium modeling provides an appropriate
framework for analysis of national forestry policies. A
theoretical spatial equilibrium model with linear supply
and demand functions was developed by Takayama and
Judge, initially in 1964 and fully in 1973 [18]. The theo-
retical model inspired many studies for various products
and policies, for example Adams et al. [19] for timber in
north America, Jae [20] for southern pine lumber in the
USA, Goletti et al. [21] for the rice sector in Vietnam,
Kawaguchi et al. [22] for milk markets in the Japanese
dairy industry, Yoshimoto et al. [23] for Japanese timber
markets, Agrifood Consulting International for the lives-
tock sector in Vietnam [24], Stennes et al. [25] for
lumber in the USA and Canada, and Devadoss et al. [26]
for softwood lumber in the USA, Canada, Mexico, China,
Japan, New Zealand, Australia and European Union.
per spatial equilibrium modelling (SEM) is used to
explain why there is a gap between plans and observed
growth for the forestry and wood products industries in
northern Vietnam and to draw policy implications from
SEM applications. The next sections describe notable
factors—that were not considered by the GoV—affect-
ing the supply targets, as well as the structure of the spa-
tial equilibrium model developed. The method of data
collection and contents of four policy scenarios are pre-
sented, and sensitivity testing relation to elasticity esti-

P. S. HIEU ET AL. 92
mates is described. Finally the interpretation of optimal
solutions generated between the four policy scenarios is
examined and policy implications are drawn.
2. Factors Affecting the Government of
Vietnam’s Supply Targets for Wood and
Wood-Processing Products
Many factors that were not taken into account by the
GoV in 2006 can affect supply targets for the forestry
sector. Three main factors are clearly identified by re-
search conducted by Katja et al. [27], Barney [29] and
GTZ [28], namely the change of import taxes, the rela-
tionship between design production capacities of paper
factories and average production cost, and distances be-
tween wood-processing factories and unit transportation
costs.
2.1. Declining Protection for Domestic Wood
Industries
Vietnam’s trade policy has changed greatly since 1994
when the USA discontinued their embargo on Vietnam’s
international trade. In 1995, Vietnam became an official
member of the Association of South East Asia Nations
(ASEAN) and joined the ASEAN Free Trade Associa-
tion (AFTA). By 2006, Vietnam had implemented provi-
sions of AFTA of reducing tariffs on imported products
typically ranging from 30% to 100%, to less than 5%,
including the import taxes on paper and furniture prod-
ucts from ASEAN. In 1998, Vietnam became an official
member of the Asia Pacific Economic Cooperation
(APEC). In 2000, the USA normalised its relationship
with Vietnam in terms of bilateral trade policies. In De-
cember 2006, Vietnam was accepted as an official
member of the World Trade Organization (WTO). All
these changes have made Vietnam’s trade policies more
consistent with world trade policies. Import and export
taxes have been reduced, and are planned to be gradually
further reduced to from 0% to 5% by 2015. Import taxes
on paper in 2008 (40%) and furniture products in 2008
(33%) from non-ASEAN countries were also reduced to
only 5% in 2010 [29,30]. The declining prices of these
products have gradually affected the supply quantities of
most domestic factories, for instance reducing the do-
mestic supply quantities of paper products.
2.2. High Average Production Costs and Low
Design Production Capacities of Domestic
Paper Factories
Wood-processing factories in Vietnam have relatively
low production scales in terms of capital investment,
number of employees and design production capacity.
The design production capacities of paper factories vary
greatly, from only 1000 tonnes up to 300,000 tonnes per
year. For example, although Vietnam has about 300 pa-
per enterprises which could produce in total about 950
000 tonnes of paper in 2006, the three largest state-
owned paper enterprises, namely Bai Bang, Tan Mai and
Dong Nai, accounted for more than 50% of the total pa-
per production in Vietnam in 2000 [29], and 40% in 2006.
The supply quantities of the above three largest paper
factories were about 200,000, 140,000 and 50,000 tonnes
respectively in 2006 [6].
According to Roda et al. [31], only 20 paper factories
had design production capacities exceeding 10,000
tonnes a year. Therefore, the domestic production cost of
paper varied greatly, from 764 to 877 USD per tonne,
while the average production cost of paper products in
ASEAN countries was only 685 USD per tonne in 2006
[32]. The Cost Insurance and Freight (CIF) prices of pa-
per products imported from Indonesia and Thailand
through Vietnam seaports ranged from 822 to 842 USD
per tonne in 2006 [32]. Therefore, domestic paper prod-
ucts can be competitive only if taxes on imports are suf-
ficiently high; otherwise, only large-scale pulp and paper
factories are competitive. For example, the production
cost of Bai Bang state-owned paper enterprise (BA-
PACO) in 2006 was 764 USD per tonne. According to
Katja et al. [27] and GTZ (2007) [28], the design pro-
duction capacities of paper factories should be at least
300,000 tonnes a year to achieve sufficient economies of
scale for international price competitiveness.
2.3. Spatial Distribution of Wood-Processing
Factories and Proximity to Timber
Resources
The spatial locations of wood-processing factories and
their proximities to timber resources affect the supply
quantities of timber and wood-processing products and
their prices. For example, if other factors are constant,
the more factories are built in one area of timber resource
the greater the demand for timber and the higher the
timber prices will be. If the number of wood-processing
factories is too great and the supply of timber is limited,
the prices of timber will increase and the timber quantity
purchased by each factory will fall. Consequently, the
supply quantity and price of a wood-processing product
from each factory declines while its average production
cost increases. If the price is lower than the average pro-
duction cost for a new factory, the new factory should
not be built. The GoV’s plan to build wood-processing
factories did not take into sufficient consideration the
distance between factories and wood plantation size.

P. S. HIEU ET AL. 93
According to ADB [32] and Barney [29], a factory gen-
erally could operate profitably if located within a dis-
tance of 200 km from a main raw material supply area.
No data about the distances between wood-processing
factories are published in Vietnam, although Phan et al.
[6] estimated that the average distance between provinc-
es having existing and planned new large-scale
wood-processing factories is only 142 km. In addition,
some provinces plan to have more than one large-scale
wood-processing factory, including Phu Tho, Hoa Binh,
Bac Kan, Lang Son and Quang Ninh provinces. There-
fore, the average distance between these wood-process-
ing factories may become much less than 142 km. The
short distances between these factories can lead to the
above situation where there will be insufficient resource
for the factories to be all large.
3. Research Method
The research method includes three main activities: iden-
tifying the components of the industry and stakeholders
in the supply chain of wood and wood-processing prod-
ucts; formulating the spatial equilibrium model for for-
estry and wood-processing industries; and collecting data
to populate the SEM.
3.1. The Classification of Wood and
Wood-Processing Sub-Industries and
Their Products
Phan et al. [11] classified the wood-processing industry
in northern Vietnam into six sub-industries with seven
main input and output products, namely woodchip, sawn
wood, engineered wood, pulp, paper, engineered- wood
furniture and solid wood furniture, as in Figure 1.
The selection of regions of the spatial equilibrium
model was restricted by availability of provincial data in
northern Vietnam (defined as from Ninh Binh province
to the northern border of Vietnam). For example, the
transportation costs of timber are only available between
some provinces only, as reported in ICARD ([33,34]),
Nguyen et al. [36] and Lamb et al. [37]. Distances
Construction
Engineered-wood
furniture
Fuel-wood
and others
Pillar
Sugar
Solid-wood
furniture
Paper
En gineered
wood
MDF
Particleboard
Fib r eboar d
Plywood
Others
Sawn-wood
Shavings and
sawdust
Shavings and
sawdust
Residue of
sugarcane
High quality log
Low quality logWood chip
Wood
residue
Sugarcane
Sugar processing
factories
Harvested
trees
Pulp
Chemical
Mechanical
Recycled
Waste paper
Figure 1. Inputs and outputs for manufacture of intermediate and final wood-processing products. Source: ICARD [34],
ICARD [35], Barney [29], Nguyen et al. [35], Nguyen et al. [36], Lamb et al. [37] and Phan et al. [6].

P. S. HIEU ET AL. 94
between provinces in northern Vietnam have been re-
ported by Phan et al. [6], but are not available for southern
Vietnam (defined as from Thanh Hoa province to the
southern border of Vietnam).
Table 1 lists nine products representing the main inputs
and main outputs purchased and produced by seven main
stakeholders in 26 regions (i.e. provinces and cities) in
northern Vietnam and three regions outside northern
Vietnam. The 29 regions include Hanoi, Vinh Phuc, Bac
Ninh, Ha Tay, Hai Duong, Hai Phong, Hung Yen, Thai
Binh, Ha Nam, Nam Dinh, Ninh Binh, Ha Giang, Cao
Bang, Bac Kan, Tuyen Quang, Lao Cai, Yen Bai, Thai
Nguyen, Lang Son, Quang Ninh, Bac Giang, Phu Tho,
Dien Bien, Lai Chau, Son La, Hoa Binh. The three other
regions are the rest of Vietnam, ASEAN countries and
the rest of the world.
The seven main stakeholders are planters (i.e. farmers,
state forestry enterprises and other planters), operators of
woodchip factories, pulp factories, paper factories and
engineered-wood factories, owners of sawmills, furniture
producers and consumers. The nine products include low
quality log, high quality log, sawn timber, woodchip,
engineered wood, pulp, paper, solid-wood furniture and
engineered-wood furniture.
3.2. Formulation of the Spatial Equilibrium
Model1
A spatial equilibrium model was developed which has
three groups of linear functions, namely supply functions,
demand functions and an objective junction of the total-
economic surplus after subtracting the total transporta-
tion cost between regions. This model may be summarized
as:
Demand functions:
,, ,,,
1
K
kikikk ik i
k
pD 




Supply functions: ,, ,,,
1
K
kikikk ik i
k
rS 




Objective function:

,,,,,, ,,
11 1
,,
IK J
kikikikikij kij
ik j
fDSQpdDrdSC Q
 




 

The model has following six groups of constraint in-
equalities and equalities.
,,ki ki
p
r Supply price and demand price must be equal
by commodities (k) and regions (i)
,, ,
-
kik ji j
pp C The differences of demand prices of commodi-
ties (k) between regions (i) must be smaller
than the unit transportation costs (C).
,,
11
RR
ki ki
jj
SD



 Total supply of commodity k from all regions
must be equal to the total demand of commod-
ity k from all regions
,, ,
1
R
kij ki
j
QS



Total quantity transport of commodity k from a
region i to all regions j must be equal to total
supply of commodity k in region i.
,, ,
1
R
ki jk j
i
QD


 Total quantity transport of commodity k from
all regions i to a regions j must be equal to total
demand of commodity k in region j.
,, ,,
, ,0
ki kikij
prQAll variables (supply prices, demand prices and
quantity transported among regions) are
nonnegative.
The symbols in above functions, equalities and in-
equalities are as follows.
i Region i, 1, 2,,iR

 S Supply quantity
j Region j, 1,2,,jR

 p Demand price
k Product k, 1,2,,kK

 r Supply price
'kProduct 'k, '1,2, ,kK

 α Constants of demand
functions
Q Transportation quantity
between regions θ Constants of supply functions
C Unit transportation cost
between regions ω Price coefficients of supply functions
D Demand quantity β Price coefficients of
demand functions
In the spatial equilibrium model, supply and demand
prices2, supply quantities, demand quantities and traded
quantities are endogenous variables. By maximizing the
objective function subject to the above six groups of
constraint inequalities and equalities, optimal levels of
these endogenous (policy) variables are found. As an
example of a change in policy, suppose a new paper fac-
tory is to be built in a province. This will create a new
supply function for paper in that province, and lead to new
optimal levels of endogenous variables. The comparison
between old and new optimal solutions indicates the af-
fect of the policy on prices, supply quantities, demand
1For more detail see the journal article ‘A Review of the Formulation
and Application of the Spatial Equilibrium Models to Analyze Policy’
written by the author. The journal article was accepted to publish on 4
February 2011 by the Journal of Forestry Research [38]. The journal
article reviewed how the spatial equilibrium model is developed and
applied. The journal article also presented various methods that can be
applied to solve spatial equilibrium models. Finally, the journal article
described how to interpret the solutions generated when the model is
solved in the software of General Algebraic Modeling System (GAMS)
The author’s programming file written in GAMS to solve spatial equi-
librium models by linear programming, non-linear programming and
mixed complementary programming was also accepted to publish in the
website of GAMS on October 2010 [39].
2In economics theory, scholars usually want to separate the behaviou
r
of consumers and producers and then prove that if consumers and pro-
ducers try to maximize their own economic benefits then the interaction
b
etween them will lead to equality of supply price and demand price.
As a result, in theoretical mathematical-economic models, supply prices
are expressed in supply functions and demand prices are expressed in
demand functions.

P. S. HIEU ET AL. 95
Table 1. Main inputs and main outputs of stakeholders in the spatial equilibrium model for wood and wood-processing in-
dustries in northern Vietnam.
Stakeholder Major input (demand quantity) Major output (supply quantity)
Planters Area of planted production forest
Low quality logs
High quality logs
Operators of woodchip factories
Low quality log
Woodchip
High quality log
Operators of pulp factories Woodchip Pulp
Owners of sawmills High quality logs Sawn-timber
Operators of Paper factories Pulp Writing paper
Operators of Engineered-wood factories Woodchip Engineered wood
Furniture producers
Sawn timber Solid-wood furniture
Engineered wood Engineered-wood furniture
Consumers
Writing paper
Assumption: consumers do not produce any
product.
Solid-wood furniture
Engineered-wood furniture
quantities, traded quantities and the total economic sur-
plus.
3.3. Data Collection Method
The establishment of the spatial equilibrium model for the
forestry and wood-processing industries in northern Viet-
nam requires three groups of data: main policies affecting
the operation of the two industries; own-price, cross-price
and income elasticity of supply and demand functions;
and prices, supply quantities and demand quantities of the
nine wood and wood-processing products in the base year,
2008. The main policies—about trade and investment—
have been obtained from published reports of official state
organisations, particularly the Vietnam General Depart-
ment of Custom (GSO) and the Vietnam Ministry of Ag-
riculture and Rural Development (MARD).
For the data group of price and income elasticities,
according to Phan et al. [6], data published about wood
and wood-processing products usually are discon- ti-
nuous, inconsistent between major organisations and are
irregularly updated. As a result, data available in Viet-
nam is not adequate to estimate these elasticities, and
hence it has been necessary to rely on estimates reported
in international econometrics journals, referred to here as
external elasticities. For example, Hseu et al. [40] ap-
plied econometric methods to estimate own-price and
cross-price elasticities of input demand and output sup-
ply in the pulp and paper industry in the USA and Can-
ada. McCarthy et al. [41] reported estimates of income
elasticities and demand price elasticities of pulp and pa-
per products in Asia, Europe, North and South America.
Waggener et al. [42] estimated own- price elasticity of
supply of sawn-timber in Indonesia. Stennes et al. [25]
estimated own-price elasticity of demand for lumber in
America. Turner et al. [43] estimated own-price elastici-
ty of demand for furniture products. McCarthy et al. [41]
estimated own-price elasticity of supply for industrial
roundwood in New Zealand.
For the data group of prices, supply and demand quan-
tities of wood and wood-processing products in 2008, the
inconsistent data published meant that it was necessary to
standardize concepts and conversions rates. Four main
steps were conducted to establish the single dataset for
the model, as described schematically in Figure 2.
Step 1: A literature review was conducted to select the
most suitable data and data sources for the wood and
wood-processing products in northern Vietnam. Two main
indicators were applied for the selection, namely the clear
application of concepts and formulae and whether the data
are collected by official state organisations. Eight sources
of data were finally selected, namely ICARD ([36,44],
Barney [29], Lamb et al. [37], Nguyen et al. [35], MARD
[3], DoF [45] and GSO [14].
Step 2: The concepts in the data published from eight
sources were clarified, and reasons why the published
data differ were investigated. The concepts of enterprise,
factory and sub-industry were developed and defined to

P. S. HIEU ET AL. 96
Ye s 
Selected secondary data
-Concepts of factory, enterprise and sub-industry
-Conversion rates between measurement units
-Conversion rates between outputs and inputs
No
Adjust the
rates by data
collected in
field trips in
2008
No
Ye s 
Final dataset for the paper
Dataset for data review meetings
Are the data consistent
by regions and
p
roduct s?
1) Quantity supplied = quantity demanded + quantity
exported-quantity imported?
2) Total output quantity supplied = conversion rate x
total input quantity supplied?
Is the dataset
accepted?
Re-organised secondary data
Step 4
Step 3
Step 2
Step 1
In terms of concepts, are the data
presented clearly and officially?
No
The data are not
used
Collected secondary data
Ye s 
1) Understandable concepts?
2) Published by official organisations?
1) First decision given by participants
2) Final decision given by ICARD
Figure 2. Data collection method.
understand and re-organise published data. Conversion
rates between various quantity measurement units used in
Vietnam and between quantities of inputs and outputs in.
the wood-processing industry in northern Vietnam were
examined, and conversions made to standard units
Step 3: Programs were written in the Data Analysis and
Statistical Software (STATA) package to classify data
according to the new concepts developed and adjusted

P. S. HIEU ET AL. 97
through the conversion rates. The outputs from running
these programs include supply quantities, demand quan-
tities and prices in 2008 for products by region in northern
Vietnam.
Step 4: Three Data Review Meetings were held in
Hanoi city and Quang Ninh province in November 2008
with forestry and statistics experts to obtain comments
and make adjustments to the data. A final dataset was
agreed by the experts who took part in the Data Review
Meetings (and was submitted to Informatics and Statistics
Centre for Agriculture and Rural Development (ICARD)
as an official dataset to publish).
4. Establishment of Policy Scenarios
The four policy scenarios according to four spatial equi-
librium models were established basing on three main
assumptions. One assumption is northern Vietnam is the
small economy; as a result it does not affect the world
supply and demand of log and wood-processing products.
This assumption was led by the fact that in terms of the
percentages of world supplies, the supplies of logs and
wood-processing products in Vietnam in general and in
northern Vietnam in particular have been very small. Ac-
cording to Brown and Durst [46], in the Asia- Pacific
region, the five countries of Japan, China, Indo- nesia,
India and New Zealand accounted for 88 percent of
wood pulp production. Similarly, the seven countries of
Japan, China, Republic of Korea, Indonesia, India,
Thailand and Australia accounted for 96 percent of the
region’s production. According to the estimation of Phan
et al. [6], no wood and wood-processing products pro-
duced in northern Vietnam could account for more than
3% of the world supplies in 2008.
Another assumption is the changes of external factors
including international demand and supply from 2008 to
2015 will create good opportunities for the wood-proc-
essing industry in northern Vietnam to grow fast. This
assumption was led by the fact that almost international
economic studies indicated that the growth of forestry
and wood-processing industries in Vietnam will be very
high by 2020. For example, De et al. [34] esti- mated
that the export value of wood and wood-process- ing
products in Vietnam will be from 7 - 9 billion USD by
2020 compared to 3.4 billion USD in 2008 [47].
The third assumption is domestic resources like land
and capita have not been used effectively in northern
Vietnam. This assumption is based on the fact that the
studies of ICARD ([33,44]), Barney [29], Karsenty [32]
and Dinh et al. [48] indicated that the domestic transac-
tion costs of wood and wood-processing products have
been high or the planned spatial locations of new
wood-processing factories should be adjusted.
Because of the above three assumptions, the future
growth of the wood-processing industry in northern
Vietnam will depend mainly on how these resources, for
instance the area of planted production forest and the
number and scale of wood-processing factories, will be
combined effectively together. Therefore, four policy
scenarios including internal factors have been developed
and examined. Each policy scenario is defined as a group
of policies (e.g. building new paper factories and adjust-
ing import taxes) and exogenous variables (e.g. the esti-
mated resident population). The content of each policy
scenario is presented in Table 2.
The base policy scenario is used to describe the wood
and wood-processing industries in the base year (2008),
as presented in Table 2. The differences between the
content of other following policy scenarios and that of
base policy scenario are mainly expressed by adding new
supply functions or shifting the old supply functions of
wood and wood-processing products to the right.
The planned policy scenario includes all policies that
were planned by the GoV for 1998-2015. For example,
in this scenario the construction of eight new pulp and
paper enterprises and five new engineered-wood enter-
prises by 2015 is planned. Import taxes on wood and
wood-processing products are 5%. The annual growth
rates of the area of planted production forest, the esti-
mated resident population and income per capita are 6%,
1.5% and 3% respectively from 2008 to 2015. The main
policies presented in this policy scenario are summarized
in Table 2.
The implemented policy scenario includes only policies
that were already implemented in the 1998-2008 period
and policies that are going to be implemented in the 2008 -
2015 period as reported in Phan et al. For example, the
paper enterprise being constructed in Tuyen Quang prov-
ince with the design production capacity of 300,000 tonnes
a year is included, as well as the engineered-wood enter-
prise being constructed in Yen Bai province with the de-
sign production capacity of 15,000 tonnes a year. Both of
these are expected to be completed by 2015. The annual
growth rate of the area of planted production forest is ex-
pected to be 4%/year in the 2008-2015 period. Other policy
variables and exogenous variables are equal to that in the
planned policy scenario (Table 2).
The recommended policy scenario is that established
in this study to achieve financial sustainability, and is
designed such that the building of any new wood-proc-
essing factory will guarantee that all stakeholders can
produce and sell their products at profitable quantities
and prices in 2015. The process to establish the recom-
mended policy scenario is described in Figure 3. A sim-
ple example about how to recommend the optimal design
production capacity of a paper factory in Tuyen Quang

P. S. HIEU ET AL. 98
Table 2. Main contents of the four policy scenarios.
Content Base policy scenario Planned policy
scenario
Implemented policy
scenario Recommended policy scenario
Description
Includes only policies
implemented in the
1998-2008 period
Includes all policies
planned to implement
in the 1998-2015
period.
Includes only policies
implemented by the GoV
in 1998-2008 period.
Includes only policies implemented by the GoV
in the 1998-2008 period.
Includes only policies being certainly
implemented between 2008 and 2015.
Includes only policies suggested by the
relationship between prices, production costs
and design production capacities of
wood-processing factories.
Includes only policies
being implemented
between 2008 and 2015.
New
enterprises
Eight new pulp and
paper enterprises will
be completed and
commence operation
in 2015.
Five new
engineeredwood
enterprises will be
completed and
commence operation
in 2015.
Tuyen Quang paper en-
terprise with the design
production capacity of
300,000 tonnes a year will
be completed and com-
mence operation in 2015.
One engineered-wood
enterprise with the design
production capacity of
15,000 tonnes a year in
Yen Bai province will be
completely and commence
operation in 2015.
Tuyen Quang paper enterprise with the design
production capacity of 350,000 tonnes a year
will be completed and commence operation in
2015.
One engineered-wood enterprise with the
design production capacity of 50,000 tonnes a
year in Yen Bai province will be completed and
commence operation in 2015.
One engineered-wood enterprise with the
design production capacity of 50,000 tonnes a
year in Bac Giang province will be completed
and commence operation in 2015.
Five woodchip factories with each design
production capacity of 10,000 tonnes will be
completed and commence operation in Lao Cai,
Lai Chau, Son La, Dien Bien and Ha Giang
provinces
Existing
large-scale
enterprises
Design production
capacity of Bai Bang
paper factory was
300,000 tonnes/year in
2008.
Design production
capacity of Bai Bang
pulp factory is 165,000
tonnes/year in 2008.
Design production
capacity of Bai Bang
paper enterprise will
still be 300,000
tonnes/year in 2010.
Design production
capacity of Bai Bang
pulp factory will be
185,000 tonnes/year
in 2010.
Design production
capacity of Bai Bang
paper factory will still be
300,000 tonnes/year in
2015.
Design production
capacity of Bai Bang pulp
factory will be 185,000
tonnes/year in 2010.
Design production capacity of Bai Bang paper
enterprise will be 350,000 tonnes/year in 2015.
The operation of existing small-scale paper
factories in Phu Tho and Hoa Binh provinces
will be ceased in 2015.
Trade
policies
applied for
non-ASEA
N countries
Import taxes on paper
products were 40% in
2008.
Import taxes on paper products will be 5% in 2010.
Import taxes on
furniture products were
33% in 2008.
Import taxes on furniture products will be 5% in 2010.
Area of
planted
production
forest
The base level in 2008
The annual growth
will be about 6% per
year in the 2008-2015
period.
The annual growth will be about 4% per year in the 2008-2015 period.
Population The base level in 2008 The annual growth will be 1.5% from 2008 to 2015
Income per
capita The base level in 2008 The annual growth will be 3% from 2008 to 2015.

P. S. HIEU ET AL. 99
Ye s Recommended
scenario
- Are optimal prices
greater than average
production costs?
- Are optimal supply
quantities greater than
the minimum profitable
supply quantities?
No
Analysing and adjusting scenarios
CONOPT
solver
command in
GAMS
Demand constant
coefficients
Transportation
cost
Supply constant
coefficients
Supply own-price
and cross-price
coefficients
Demand
own-price and
cross-price and
income
coefficients
Optimal supply
prices
Optimal demand
prices
Optimal supply
quantities
Optimal demand
quantities
Figure 3. Process to examine the optimal levels and establish the content of building new wood-processing factories for the
recommended policy scenario.
province is conducted as in the following steps. A new
factory will create a new supply curve of paper in Tuyen
Quang province and its first design production capacity
(e.g. 200,000 tonnes a year) determines a constant coef-
ficient of the supply curve. The spatial equilibrium model
with the first new supply curve identifies the optimal
supply price and quantity of paper in Tuyen Quang prov-
ince. The optimal solution of paper price is compared to
the average production cost of 810 USD per tonne. The
optimal solution of the paper supply quantity is compared
to minimum profitable supply quantity of 130,000 tonnes
a year (200,000 × 65%)3. If the optimal price is lower
than the average production cost or the optimal supply
quantity is lower than the profitable supply quantity, the
first design production capacity is adjusted to the second
design production capacity (e.g. 300,000 tonnes a year).
The spatial equilibrium model with the second new
supply curve identifies the optimal supply price and
quantity of paper in Tuyen Quang province. The optimal
price is compared to the average production cost of 760
USD per tonne and the optimal solution of the supply
quantity is compared to the profitable supply quantity of
195,000 tonnes a year (300,000 × 65%). This process is
continued until the two indicators are satisfied and then
the optimal design production capacity of the paper fac-
tory in Tuyen Quang province is recommended.
A set of indicators has been devised which compares
optimal levels of endogenous variables of prices and
supply quantities generated with the national average
production costs and profitable supply quantities (as
presented in Table 3).
In the recommended policy scenario, the paper enter-
prise under construction in Tuyen Quang province will
have the design production capacity of 350,000 tonnes a
year in 2015. The engineered-wood enterprise being
constructed in Yen Bai province will have the design
production capacity of 50,000 tonnes a year in 2015. One
new engineered-wood enterprise with the design produc-
tion capacity of 50,000 tonnes a year will be built in Bac
Giang province in 2015. Five new woodchip factories
each with design production capacity of about 10,000
tonnes will be completed and commence operation in
2015 in the remote provinces of Lao Cai, Lai Chau, Son
La, Dien Bien and Ha Giang. The design production ca-
pacity of Bai Bang paper enterprise will be 350 000 ton-
nes/year in 2015. Existing paper factories with the design
production capacities of less than 100,000 tonnes a year
will cease supplying paper products in 2015. Other vari-
ables are equal to that in the implemented policy scenario.
The contents of the third policy scenario are summarised
in Table 2.
To examine the effect of errors in estimates of elastic-
ities on optimal solutions generated by solving these spa-
3The figure of 0.65 or 65% is the percentage of minimum profitable
supply quantities of wood-processing factories compared to the design
p
roduction capacities of those factories.

P. S. HIEU ET AL. 100
Table 3. Indicators to examine optimal levels of endogenous variables and establish the recommended policy scenario.
Indicator Product Value in 2008 Source Note
The average production
cost in 2008 Low quality log 12.12 USD per tonne [33,44]
Re-estimated by participants in
the three Data Review Meetings
organised by Phan and his
colleagues in Hanoi city and
Quang Ninh province in
November 2008
The average production
cost in 2006
Paper produced by Bai Bang
Paper Factory 764 USD per tonne [32]
Paper produced by factories with
design production capacities of
less than 300 000 tonnes a year
805 - 842 USD per
tonne [32]
The profitable supply
quantity Paper and engineered wood
65% of a factory’s
design production
capacity
[49]
tial equilibrium models, three sets of elasticity levels
have been tested for each policy scenario: the external
elasticities, all elasticities reduced by 30%, and all elas-
ticities increased by 30%.
5. The Optimal Solutions and Their
Interpretation
By solving the spatial equilibrium model for the four
policy scenarios, four optimal solutions have been gen-
erated. These optimal solutions are compared between
policy scenarios, supply targets and the two indicators of
profitable prices and profitable supply quantities. Two
products of low quality logs and paper are selected for
these comparisons because the supply quantities of these
two products are the most targeted by the GoV’s invest-
ment policy on building new large-scale state-owned
paper factories and expanding the area of planted pro-
duction forest via the 5MHRP. In addition, the data about
average production cost are only available for domestic
low quality logs as reported in ICARD ([33,44]) and
Phan et al. [6] and for domestic paper products as re-
ported in Karsenty [32].
5.1. The Optimal Levels of Prices and Average
Production Costs of Low Quality Logs and
Paper
In the planned policy scenario, policies planned by the
GoV will lead to two notable outcomes. First, optimal
prices of most products in 2015 will be much higher than
that in the base policy scenario in 2008, and higher than
average production costs, except the optimal price for
paper. Table 4 indicates that the selling price of low
quality logs will be more than 30 USD per tonne—much
higher than the average production cost of about 12 USD
per tonne in 2008. Optimal prices in 2015 for low quality
logs will be much higher in the provinces having new
wood-processing factories including Lai Chau and Son
La provinces. Second, eight in 10 paper factories—in-
cluding seven of the eight new paper factories—will re-
ceive unprofitable prices for their products because op-
timal prices for paper will be lower than the average
production costs of from 805 to 822 USD per tonne re-
spectively in 2006 as reported in Karsenty [32].
In the implemented policy scenario, compared to pric-
es in 2008, optimal prices in 2015 will be higher for
woodchip, sawn timber, engineered wood, paper, solid-
wood furniture and engineered-wood furniture in most
provinces in northern Vietnam. Table 5 indicates that the
optimal prices of paper products in 2015 will be at least
about 852 USD per tonne and that will be much higher
than the average production cost for paper products of
between 805 and 822 USD per tonne. However, imple-
mented policies will lead to optimal prices in 2015 of
low quality logs of not more than the average production
cost of about 12 USD per tonne in 2008 in seven prov-
inces in northern Vietnam, namely Son La, Dien Bien,
Lai Chau, Lao Cai, Cao Bang, Bac Kan and Ha Giang
provinces.
In the recommended policy scenario, compared to
prices in 2008, optimal prices in 2015 will be higher for
high and low quality logs, woodchip, sawn timber, engi-
neered wood and furniture. An example is optimal prices
of low quality logs presented in Table 4. In contrast,
optimal prices of pulp and paper in 2015 will be much
lower. Optimal prices of all products in 2015 will be
higher than the average production costs in all 26 prov-
inces in northern Vietnam. Examples are prices of low
quality logs and paper presented in Table 4.
5.2. Optimal Supply Levels and Factory Design
Production Capacities in 2015
In Table 5, the optimal supply levels in 2015 are com-
pared to factories’ design production capacities in per-
centages. In the planned policy scenario, planned policies
result in percentages in 2015 which are mostly lower
than the profitable figure of 65%. Planned new paper fac-
tories will produce at unprofitable quantities in six prov-

P. S. HIEU ET AL. 101
Table 4. Optimal prices of low quality logs and paper generated by scenarios in r elevant provinces (USD per tonne).
Province
Base policy
Scenario
Planned policy
scenario
Implemented
policy scenario
Recommended
policy scenario
Prices of low quality logs in 2008 Optimal prices of low quality logs in 2015
Ha Giang 14.3 41.9 6.4 34.5
Cao Bang 14.4 37.2 11.6 22.5
Bac Kan 20.9 31.2 12.2 28.8
Lao Cai 12.0 35.2 3.3 24.5
Dien Bien 11.9 39.5 3.3 33.0
Lai Chau 7.9 34.2 0.0 27.7
Son La 16.1 32.4 7.4 37.9
Prices of paper in 2008 Optimal prices of paper in 2015
Bac Kan 837.3 787.3 863.0 808.6
Tuyen Quang 834.6 786.2 861.9 807.4
Lao Cai 843.9 777.2 852.8 798.4
Yen Bai 835.0 785.6 861.2 806.8
Lang Son 840.7 791.0 866.6 812.2
Bac Giang 836.0 793.6 866.6 814.9
Phu Tho 830.5 789.8 869.3 811.0
Dien Bien 843.9 778.2 865.4 799.4
Lai Chau 846.9 775.6 853.8 800.7
Son La 840.9 781.4 855.1 802.6
Hoa Binh 834.1 791.6 857.1 812.8
Table 5. Optimal supply quantities as percentages of design production capacities by scenarios in relevant provinces.
Province or city
Planned policy scenario Implemented policy scenario Recommended policy scenario
Engineered wood Paper Engineered wood Paper Engineered wood Paper
Hanoi 58.37 0.00 62.37 0.00 63.87 0.00
Bac Kan 118.18 58.68 0.00 0.00 0.00 0.00
Tuyen Quang 0.00 66.11 0.00 73.02 0.00 81.14
Lao Cai 0.00 55.42 0.00 0.00 0.00 0.00
Yen Bai 214.69 58.04 93.87 0.00 66.39 0.00
Lang Son 63.59 63.21 66.21 0.00 67.88 0.00
Quang Ninh 62.74 0.00 65.05 0.00 67.55 0.00
Bac Giang 169.19 65.33 0.00 0.00 65.72 0.00
Phu Tho 57.66 86.80 62.66 93.34 64.03 84.27
Lai Chau 0.00 60.00 0.00 0.00 0.00 0.00
Son La 0.00 61.22 0.00 0.00 0.00 0.00
Hoa Binh 117.00 86.57 62.97 93.24 64.46 0.00

P. S. HIEU ET AL. 102
inces, namely Bac Kan, Lao Cai, Yen Bai, Lang Son, Lai
Chau and Son La provinces. Engineered-wood factories
will produce at unprofitable levels in four provinces,
namely Hanoi, Lang Son, Quang Ninh and Phu Tho pro-
vinces. In the two extreme cases, engineered-wood fac-
tories in Yen Bai and Bac Giang provinces will produce
at non-achievable levels, 214% and 169% of their design
production capacities respectively in 2015. Overall, al-
most all planned new wood-processing factories will
produce at the level outside the profitable range.
In the implemented policy scenario, nearly all the per-
centages of the design production capacities of engi-
neered wood factories are mostly lower than the figure of
65%, from about 62% to just over 65%, except the engi-
neered-wood factory in Yen Bai province with the per-
centage of 93% (Table 5). In this scenario, all paper fac-
tories will produce at profitable quantities because the
corresponding percentages of optimal factory supply
quantities compared to their design production capacities
will be more than 70% in 2015. Nearly all existing and
newly constructed wood-processing factories will pro-
duce at profitable quantities. Compared to the planned
policy scenario, many fewer factories are newly built in
the 1998-2015 period. For instance, only one new paper
factory is built in Tuyen Quang province while the
planned number is nine.
In the recommended policy scenario, the existing state-
owned engineered-wood factories in Hanoi, Phu Tho and
Hoa Binh provinces will produce at levels that are
slightly lower than the figure of 65% of their design
production capacities as indicated in Table 5. All rec-
ommended new engineered-wood factories in Yen Bai,
Lang Son, Quang Ninh and Bac Giang will produce at
profitable quantities (Table 5). Compared to the planned
policy scenario, the optimal numbers of recommended
new factories are lower and their design production ca-
pacities are higher. For example, Table 5 indicates that
the number of engineered-wood and paper factories in
the planned policy scenario are eight and 10 respectively,
compared with only seven and two respectively in the
recommended policy scenario. Compared to the imple-
mented policy scenario, the number of recommended
new paper factories is unchanged. There is only one new
factory in Tuyen Quang province. The number of new
engineered-wood factories is higher in the recommended
policy scenario—with two factories in Yen Bai and Bac
Giang province—than in the implemented policy sce- nario
with only one factory in Yen Bai province. Com- pared
to the implemented policy scenario, however, all recom-
mended new factories have much higher design produc-
tion capacities. For example, the design production ca-
pacity of Tuyen Quang paper factory is about 300 000 and
350 000 tonnes a year in the implemented policy scenario
and in the recommended policy scenario respectively.
5.3. Optimal Supply Levels and Supply Targets
Table 6 summarizes optimal supply levels in 2015 gen-
erated by solving the four policy scenarios and compares
these optimal levels with supply targets set by the GoV
in 2006 for 2015. In the planned policy scenario, the
planned policies will lead to optimal supply levels in
Table 6. Optimal supply levels and supply tar gets by sc enario.
Product and area of
p
lanted production fores
t
Unit Target in 2015
Quantities
supplied in
2008
Optimal supply quantities in 2015 Supply quantities in 2015 relative to 2015
target (%)
Planned
policy
scenario
Implemented
policy
scenario
Recommended
policy
scenario
Planned
policy
scenario
Implemented
policy
scenario
Recommended
policy
scenario
Paper Mt 645 324 1090 555 621 169 86 96
Pulp Mt 774 233 1230 529 951 159 68 123
Engineered-wood
furniture Mt 484 360 401 414 743 83 85 153
Solid-wood furniture Mt 962 246 278 276 668 29 29 69
Engineered wood Mt 532 425 659 477 525 124 90 99
Sawn-timber Mt 1480 184 206 206 887 14 14 60
Woodchip Mt 1591 848 1947 1596 1832 122 100 115
Low quality logs Mt 2865 1259 3364 1766 2792 117 62 97
High quality logs Mt 3289 410 815 729 778 25 22 24
Area of planted
production forest 000 ha 1200 334 697 416 595 58 35 50

P. S. HIEU ET AL. 103
2015 much higher than the planned quantities of paper
and engineered wood but much lower than the planned
quantities of sawn timber and solid-wood furniture. The
estimation of the optimal area of planted production for-
est depends on the average productivity of timber. How-
ever, the statistical data reported about the average pro-
ductivity per ha per year of timber trees planted in pro-
duction forest differs between major data sources in
Vietnam, for instance 8 - 10 m3 in 2000 [29], 20 - 25 m3
in some province [29], 8 m3 [33,44,45] and 4m3/ha/year
[14]. According to Phan et al. [6], the average productiv-
ities in 2008 were 6 m3 for Eucalyptus spp., 7 m3 for Aca-
cia spp ., 5 m3 for Pinus spp., 4 m3 for Bambusoidae spp.,
and 5 m3 for Styrax tonkinensis. If the average productiv-
ity of timber trees in planted production forest in north-
ern Vietnam is about 6 m3/ha/year, the area of planted
production forest in the planned policy scenario is pre-
dicted to be about 658 000 ha. This area equals to about
58% of the area planned by the GoV in 2006 for 2015.
In the implemented policy scenario, none of the supply
targets for 2015 will be achieved, other than for wood-
chip. The highest percentages of the supply targets will
be for paper (86%), engineered-wood furniture (85%),
engineered wood (90%) and woodchip (100%). The
quantities supplied in 2015 of high quality logs, solid-
wood furniture and sawn-timber will only equal to about
22%, 29% and 14% respectively of the supply targets. If
the average productivity of timber trees in planted pro-
duction forest in northern Vietnam is about 6 m3/ha/year,
the area of planted production forest is predicted to be
about 412 000 ha, equal to 35% of the area planned in
2006 for 2015.
As percentages of the supply targets, most of the op-
timal supply levels in 2015 generated by solving the spa-
tial equilibrium model for the recommended policy sce-
nario will be approximately equal to the supply targets,
for instance 96% for paper, 99% for engineered wood,
97% for low quality logs, 123% for pulp and 115% for
woodchip. The optimal supply level in 2015 for engi-
neered-wood furniture will be about 50% higher than the
supply target. The supply quantities planned by the GoV
are highly ambitious for high quality logs, sawn timber
and solid-wood furniture compared to the growth of these
industries in the 1998-2008 period. The supply quantities
in 2015 of these products will be still much lower than
that planned by the GoV in 1998 for 2015, for instance,
69% for solid-wood furniture, 60% for sawn timber, 24%
for high quality logs and 50% for the area of planted
production forest.
5.4. Sensitivity Analysis
Most of the above outcomes in each policy scenario hold
for the elasticity scenarios of ‘+30%’ and ‘–30%’. For
the comparisons between optimal prices and average
productions costs, most new paper factories will receive
unprofitable prices for their products in the planned pol-
icy scenario. This outcome will worsen if the average
production cost of paper in 2015 is 20% higher than that
in 2008 due to the assumed annual inflation rate of 3.5%
in Vietnam. In this situation, all existing and new paper
factories in northern Vietnam will receive unprofitable
prices for their products. Similarly, in the implemented
policy scenario, optimal prices of low quality logs in
2015 generated for the lower and higher elasticity coeffi-
cients will be lower than the average production cost of
about 12 USD per tonne in 2008 to produce them in
many provinces in northern Vietnam including Son La,
Dien Bien and Lai Chau provinces. In addition, if the
average production cost of low quality logs in 2015 is
20% higher than that in 2008 due to the assumed annual
small inflation rate of 3.5% in Vietnam, farmers in 12
provinces in the total of 26 provinces in northern Viet-
nam would receive unprofitable prices for their products.
The changes of elasticities also have little effect on
profitable supply quantities of new engineered-wood and
paper factories in northern Vietnam. In the planned and
implemented policy scenarios, the optimal levels of sup-
ply quantities of engineered wood and paper produced by
new factories are all lower than 65% of their design pro-
duction capacities.
The only case where the changes of elasticities only
have profound effect on optimal supply quantity of solid-
wood furniture in the planned policy scenario is where
30% higher elasticities increases the supply of solid-
wood furniture in northern Vietnam by 47%. This effect
is because the solid-wood furniture sub-industry has
higher supply price coefficients than those of other sub-
industries. Since 2000 about 50% of annual solid-wood
furniture outputs of northern Vietnam have been ex-
ported [6]. For example, the small changes of supply
elasticities in China or those of demand elasticities in
USA are predicted to affect profoundly the supply of the
solid-wood furniture in northern Vietnam.
6. Concluding Comments
The spatial equilibrium analysis indicates that the GoV’s
historical plans for the forestry and wood-processing
industries in northern Vietnam could not be achieved.
The conclusion was tested by elasticity changes in each
policy scenario. If all policies were implemented in the
2008-2015 period as planned in 2006, most planned new
paper factories would receive unprofitable prices for
their products in 2015. In the framework of currently
implemented policies, the farm-gate prices of logs will

P. S. HIEU ET AL. 104
be lower than the average production costs to produce
logs in many provinces in northern Vietnam in 2015. To
transit to optimal plans as a longer-run objective, the
GoV should adjust the policies planned in 2006 for 2015
and policies implemented in the 2008-2015 period. This
would require an increase in design production capacity
of existing larger-scale paper and engineered-wood fac-
tories. Planned new small-scale paper factories with de-
sign production capacities of less than 300,000 tonnes a
year and engineered-wood factories with design produc-
tion capacities of less than 50,000 tonnes a year should
not be built. Closure of existing small-scale paper facto-
ries of less than 50,000 tonnes by 2015 should be con-
sidered. Consequently, the planned area of planted pro-
duction forest in northern Vietnam should be less much
than 900,000 ha, considerably lower than the planned
level of about 1.2 million ha.
7. Acknowledgements
Funding for this research from a John Allright Fellow-
ship is gratefully acknowledged. I would like to thank
Drs Sharon Brown and Peter Dart, at The University of
Queensland for their advice and encouragement. I would
also like to thank my colleagues working at various state
organisations under MARD—including Dr. Trinh Duc
Huy, Ms. Nguyen Thanh Huong, Mr. Vu Dinh Xuan, Ms.
Vu Huong Thuy, Mr. Pham Duc Thuan, Mr. Le Thanh
Van, Mr. An Van Khanh and Mr. Tran Lam Dong—for
support in obtaining data and checking the consistency of
data presented in this paper.
8. References
[1] ICARD (Informatics and Statistics Center for Agriculture
and Rural Development), “Assessing the Changes of In-
stitution and Policy from Centrally Planning Economy to
Market-Oriented Economy in the past 20 Years,” Infor-
matics and Statistics Center for Agriculture and Rural De-
velopment (ICARD), Hanoi, 2004.
[2] T. Chakra, “Five Year Plans,” 2008.
http://www.tradechakra.com/indian-economy/five-year-pl
ans/index.html
[3] MARD (Ministry of Agriculture and Rural Development),
“Vietnam Forestry Development Strategy 2006-2020,”
Agricultural Publishing House, Hanoi, 2007.
[4] A. K. Joshi, P. Pant, P. Kumar, A. Giriraj and P. K. Joshi.
“National Forest Policy in India: Critique of Targets and
Implementation,” Small-Scale Forestry, Vol. 10, No. 1,
September 2010, pp. 83-96.
doi:10.1007/s11842-010-9133-z
[5] BMoA (Royal Government of Bhutan Ministry of Agri-
culture), “National Forest Policy,” 2008.
http://www.moa.gov.bt/moa/downloads/downloadFiles/M
oADownload3ea6992sr.pdf
[6] S. H. Phan, H. T. Vu, D. T. Pham and T. V. Le, “Col-
lected and Estimated Data for Wood and
Wood-Processing Products in Northern Vietnam,” Infor-
matics and Statistics Center for Agriculture and Rural
Development (ICARD), Hanoi, 2008.
[7] A .P. Carandang, “Forest Resource Assessment—Na-
tional Forest Assessment: Forestry Policy Analysis: Phil-
ippine,” Food and Agriculture Organization (FAO), 2005.
http://www.fao.org/forestry/15461-1-84.pdf.
[8] S. H. Phan, “The Changing Administration and the Role
of Forestry in the Economy of Vietnam,” Small-Scale
Forestry, Vol. 3, No 1, 2004, pp. 85-98.
[9] DAFF (Australian Department of Agriculture, Fisheries
and Forestry), “Regional Forest Agreements Home,”
Australian Government, 2008.
http://www.daff.gov.au/rfa/.
[10] CMoE (Cambodian Ministry of Environment), “Cambo-
dian National Environment Action Plan 1998 - 2002,”
1998.
http://www.camdev.org/_publications/NEAP%20text-En
glish.pdf.
[11] S. H. Phan, H. T. Vu, D. T. Pham and T. V. Le, “Main
Characteristics of Statistical Data and the Statistical Sys-
tem for Wood and Wood-Processing Products in Viet-
nam,” Small-Scale Forestry, Vol. 10, No. 2, September
2010, pp. 185-198.
[12] D. Munkhzorig, “Report on National Forest and Water
Policy of Mongolia,” Ministry of Nature and Environment,
2009.
http://www.rinya.maff.go.jp/faw2002/17%20D.%20Mun
khzorig%20(Mongolia).pdf.
[13] D. S. Do and Q. T. Le, “Forest Policy Trends in Viet-
nam,” The Forest Science Institute of Vietnam, Hanoi,
2001.
[14] GSO (General Statistical Office of Vietnam), “Statistical
Yearbook 2008,” General Statistical Office of Vietnam
(GSO), Hanoi, 2008.
[15] GSO (General Statistical Office of Vietnam), “Forest
Status Classified by Provinces on 31 December 2008,”
(Hiện trạng rừng có đến 31/12/2008 phân theo địa
phương), 2010.
http://www.gso.gov.vn/default.aspx?tabid=390&idmid=3
&ItemID=8752.
[16] C. Lang, “Vietnam: Construction of Kon Tum Pulp and
Paper Mill Suspended,” World Rainforest Movement,
Montevideo, Uruguay, 2003.
http://www.wrm.org.uy/bulletin/66/Vietnam.html.
[17] C. Lang, “Vietnam: What is Happening in the Pulp and
Paper Sector?” World Rainforest Movement, Montevi-
deo, Uruguay, 2006.
http://chrislang.org/2007/02/28/vietnam-whats-happening
-in-the-pulp-and-paper-sector/.
[18] B. A. McCarl and T. H. Spreen, “Applied Mathematical
Programming Using Algebraic System,” Department of
Agricultural Economics, Texas A&M University, College
Station, 2002.
http://agecon2.tamu.edu/people/faculty/mccarl-bruce/boo

P. S. HIEU ET AL. 105
ks.htm.
[19] D. M. Adams and R. W. Haynes, “The 1980 Softwood
Timber Assessment Market Model: Structure, Projections,
and Policy Simulations,” Journal of Forest Science, Mo-
nograph 22 (Supplement to Number 3), 1980, pp.
a0001-z0001,
[20] K. S. Jae, “A Spatial Equilibrium Analysis of Southern
Pine Lumber Pricing and Allocation,” Journal of the An-
nals of Regional S c i e nce, Vol. 19, No. 1, 1984, pp. 61-76.
[21] F. Goletti, A. Raisuddin, N. Minot, P. Berry, B. Romeo,
V. H. Nguyen and T. B. Nguyen, “Rice Market Monitor-
ing and Policy Options Study,” International Food Policy
Research Institute (IFPRI), 1996.
[22] T. Kawaguchi, N. Suzuki and H. M. Kaiser, “A Spatial
Equilibrium Model for Imperfectly Competitive Milk
Markets,” American Journal of Agricultural Economics,
Vol. 79, No. 3, 1997, pp. 851-859. doi:10.2307/1244426
[23] A. Yoshimoto and K. Yukutake, “Japanese Forest Sector
Modeling through a Partial Equilibrium Model,” Forest
Resource Use vs. Environmental Policy & Timber Trade,
Vol. 7, No. 1, 2002, pp. 41-48.
[24] ACI (Agrifood Consulting International), “Vietnam Live-
stock Spatial Equilibrium Model (VILASEM),” the
Document for a Training Course at the Informatics and
Statistics Centre for Agriculture and Rural Development
(ICARD), Hanoi, 2002.
[25] B. Stennes and W. Bill, “An Analysis of Lumber Trade
Restrictions in North America: Application of Spatial
Equilibrium Model,” Canadian Journal of Forest Policy
and Economics, Vol. 7, No. 3, 2005, pp. 297-308.
doi:10.1016/S1389-9341(03)00067-4
[26] S. Devadoss, A. H. Aguiar., S. R. Shook and J. Araji, “A
Spatial Equilibrium Analysis of U.S.-Canadian Disputes
on the World Softwood Lumber Market,” Canadian
Journal of Agricultural Economics, Vol. 53, No. 2-3,
2005, pp. 177-192.
doi:10.1111/j.1744-7976.2005.04024.x
[27] S. Katja and S. Jayant, “India’s Pulp and Paper Industry:
Productivity and Energy Efficiency,” Ernest Orlando
Lawrence Berkeley National Library, 1999.
http://ies.lbl.gov/iespubs/41843.pdf.
[28] GTZ (German Technical Development Cooperation
Agency or Deutsche Gesellschaft für Technische Zusam-
menarbeit), “Enterprises Operating in the Vietnam’s For-
estry Industry: Current Characteristics and Challenges,”
(Doanh Nghiep Hoat Dong trong Nganh Lam Nghiep
Viet Nam – Hien Trang va Thach Thuc), Presentation at
the Forum of Vietnam forestry, 2007.
http://www.vietnamforestry.org.vn/PSC/VN/ForestryInve
stmentForum2007/2_GTZ-Forestry percent20Enterprise
percent20- percent20status percent20& percent20chal-
lenges percent205May07 percent20final percent20draft_
VN_1.ppt
[29] K. Barney, “Central Plans and Global Exports: Tracking
Vietnam’s Forestry Commodity Chains and Export Links
to China,” Forest Trends, Washington DC, 2005.
[30] DoC (General Department of Custom of Vietnam), “Im-
port and Export Taxes Applied in 2008,” (Bieu Thue
Xuat Nhap Khau Viet Nam nam 2008), General Depart-
ment of Custom of Vietnam (DoC), Hanoi, 2008.
[31] J. M. Roda and R. Santosh, “Feeding China’s Expanding
Demand for Wood Pulp: A Diagnostic Assessment of
Plantation Development, Fiber Supply, and Impacts on
Natural Forests in China and in the South East Asia Re-
gion, Vietnam Report,” Center for International Forestry
Research (CIFOR), 2006.
http://hal.cirad.fr/docs/00/19/35/45/PDF/Vietnam.pdf.
[32] A. Karsenty, “Study on the Impacts of Market and In-
vestment Liberalization on Vietnam’s Pulp and Planta-
tions Sector, with Particular Reference to China,” Viet-
nam Report, Center for International Forestry Research
(CIFOR), 2006.
http://www.cifor.cgiar.org/publications/pdf_files/research
/governance/foresttrade/Attachment4-CIRAD-VietnamRe
port-2.pdf.
[33] ICARD (Informatics and Statistics Centre for Agriculture
and Rural Development), “Policies Analysis Matrix for
Plantation Forest in Mountainous Regions in Vietnam,”
Agricultural Publishing House, Hanoi, 2002.
[34] L. De, H. Miyazono, R. Prasad and A. Gerrand, “Re-
gional Implications of National Changes—Some Major
Influences,” In: R. N. Leslie, Ed., The Future of Forests
in Asia and the Pacific: Outlook for 2020, Food Agricul-
ture Organisation (FAO), Chiang Mai, 2007.
[35] N. B. Nguyen, D. Q. Chu, V. T. Duong and H. A. Bui,
“Handbook for the Vietnam Forestry,” The Handbook
was written by the support of the Forest Sector Support
Partnership (FSSP) under the Ministry of Agriculture and
Rural Development (MARD), Equipment Investment and
Printing Joint Stock Company, Hanoi, 2006.
[36] N. B. Nguyen, V. T. Nguyen, C. N. Bui and Q. T. Trinh,
“Why do Farmers Choose to Harvest Small-Sized Timber?
A Survey in YenBai Province, Northern Vietnam,” The
Economy and Environment Program for Southeast Asia
(EEPSEA), 2006.
http://www.idrc.ca/eepsea/ev-99959-201-1-DO_TOPIC.h
tml.
[37] D. Lamb, D. N. Huynh, S. Brown and P. Dart, “Mixed
Species Plantations of High-Value Trees for Timber Pro-
duction and Enhanced Community Services in Vietnam
and Australia,” Forest Development and the Planting of
Native Trees in Three Mountainous Northern Provinces,
Project ID: FST/2000/003 Funded by the Australian Cen-
tre for International Agriculture Research (ACIAR),
2006.
[38] S. H. Phan and S. Harrison, “A Review of the Formula-
tion and Application of the Spatial Equilibrium Models to
Analyze Policy,” Journal of Forestry Research, 2011, in
Press.
[39] S. H. Phan, “spatequ.gms,” Spatial Equilibrium, Model
Library section in the Website of General Algebraic
Modeling System Development Corporation (GAMS),
2010. http://www.gams.com/modlib/libhtml/spatequ.htm.
[40] J. S. Hseu and B. Joseph, “Output Supply and Input De-
mand in the Pulp and Paper Industry: A Nonparametric

P. S. HIEU ET AL. 106
Model for the United States and Canada,” Journal of
Forest Science, Vol. 43, No. 1, 1997, pp. 297-308.
[41] P. McCarthy and L. Lei, “Regional Demands for Pulp
and Paper Products,” Working Paper, The Center for Pa-
per Business and Industry Studies (CPBIS), 2007.
http://www.cpbis.com/research/findings/070604%20app_
demand_draft.pdf.
[42] T. R. Waggener and L. Christine, “Asia-Pacific Forestry
Industry Outlook Study,” Working Paper No. APFSOS/
WP/02, International Food and Agriculture Organisation
(FAO), 1997.
http://www.fao.org/docrep/W4388E/w4388e00.HTM.
[43] J. Turner, K. Andres and B. Joseph, “Implications for the
New Zealand Wood Products Sector of Trade Distortions
due to Illegal Logging,” A Report Prepared for the Minis-
try of Agriculture and Forestry, New Zealand, 2007.
http://www.maf.govt.nz/forestry/illegal-logging/trade-dist
ortion-implications/executive-summary.htm.
[44] ICARD (Informatics and Statistics Centre for Agriculture
and Rural Development), “Estimated Economic Results
of Forest-Growing Households in Dai Tu District, Thai
Nguyen Province,” The Research Project Funded by Ford
Foundation, Hanoi, 2002.
[45] DoF (Department of Forestry), “Introduction to Five Mil-
lion ha Program,” 2008.
http://dof.mard.gov.vn/661/gioithieu/lydo.htm.
[46] C. Brown and P. Durst, “State of Forestry in Asia and the
Pacific—2003: Status, Change and Trends,” Food and
Agriculture Organisation (FAO), Regional Office for
Asia and the Pacific, Bangkok, 2003.
[47] ITTO (International Timber Trade Organisation), “An-
nual Review and Assessment of the World Timber Situa-
tion,” International Tropical Timbers Organisation, Yo-
kohama, 2009.
[48] H. K. Dinh and N. S. Vu, “The Plan of Raw Material
Area for Paper and Engineered-Wood Production in
Vietnam in the 2006-20 Period,” (Qui Hoach Cac Vung
Trong Nguyen Lieu Giay va Van Nhan Tao Vietnam
trong Giai Doan 2006-2020), Specialised Information
Bulletin, Informatics and Statistics Centre for Agriculture
and Rural Development (ICARD), Hanoi, 2008.
[49] N. Blower, M. H. Ngo, J. Allan, M. Adam, P. David, Q.
H. Pham, Q. Derek, Q. Nguyen, T. Mandy, T. B. Do, V.
H. Hoang, W. Bob and V. David, “Paper, Prices and Poli-
tics: An Evaluation of Swedish Support to the Bai Bang
Project in Vietnam,” A Swedish International
Co-operation Development Agency (Sida) Evaluation
Report, Centre for International Economics, 1999.