Chapter 4
Monitoring and measuring cross-sectoral impacts
with environmental accounts

Glenn-Marie Lange

INTRODUCTION

Sustainable forest management has been hampered by a lack of information about the multiple economic contributions of forests and how the goods and services provided by forests are linked to the rest of the economy. While there is information about the economic value of commercial timber, many other contributions of forests are often missing from national accounts, our primary source of information about the economy. Examples of these contributions include non-marketed forest goods and services that contribute to the livelihoods of rural populations, the use of forests for tourism and recreation and ecosystem protection services such as watershed protection for agriculture, fisheries and municipal water supply, or carbon storage.

To provide the information to build broad-based, cross-sectoral forestry policy, information is needed about the value of forest ecosystem benefits and integrating environmental and economic information. The United Nations, along with other major international organizations (OECD, Eurostat, World Bank, IMF), has prepared a handbook for environmental accounts, the System of Environmental and Economic Accounts (SEEA) (UN, 1993; 2001). Environmental accounts are widely recognized as providing such a framework: they bring together economic and environmental information in a common framework to measure the contribution of the environment to the economy and the impact of the economy on the environment. The accounts provide both indicators for monitoring sustainability as well as detailed statistics for use as a planning tool. Detailed statistics enable governments to set priorities, monitor the environmental impact of economic policies more precisely, enact more effective environmental regulations and resource management strategies and design more efficient market instruments for environmental policies.

The purpose of this report is to describe the potential use of SEEA for measuring the cross-sectoral benefits of forestry to other sectors of the economy and identifying impacts on forestry from non-forestry sector policies. This report will identify the relevant components of SEEA and develop a framework that can be used for economic analysis of the impact of cross-sectoral policy linkages and, hence, contribute to better sectoral and macroeconomic policy-making by governments and, ultimately, to more effective forest conservation.

OVERVIEW OF ENVIRONMENTAL ACCOUNTS

Over the past few decades, most countries have come to embrace the notion of sustainable development, popularly expressed by the Brundtland Commission Report, Our common future, as ‘…development that meets the needs of the present without compromising the ability of future generations to meet their own needs’ (World Commission on Environment and Development, 1987). The search for ways to operationalize this notion has focused, in part, on national economic accounts: incorporating the role of the environment and natural capital more fully into the conventional system of national accounts (SNA) through a system of satellite accounts for the environment. SNA (UN, 1993) is particularly important because it constitutes the primary source of information about the economy and is widely used for analysis and decision-making in all countries. However, SNA has had a number of well-known shortcomings regarding the treatment of the environment.

With regard to forestry, SNA has treated cultivated forests and natural forests quite differently. For cultivated forests, SNA records both production and changes in the forest stock so that the consequences of depletion or afforestation are accounted for. For natural forests, however, SNA records only the income from logging, but not changes in natural forest stocks. This can result in quite misleading economic signals about changes in a natural forest: income from over-exploitation would be recorded as part of GDP, but the corresponding depletion of the forest stocks would not be recorded. Similarly, the benefits from afforestation would not be recorded. The 1993 revision of SNA addresses some of these problems, notably by expanding the asset boundary to include a broader range of natural assets such as natural forests. Even with this expanded coverage, significant gaps remain. Non-marketed products are critical to rural livelihoods in developing countries, yet they are often not included in the national accounts. In principle, SNA includes such products, but measurement difficulties have limited implementation in many countries.

Many of the non-market services from forests are wrongly attributed to other sectors of the economy or omitted. The value of forest services provided as intermediate inputs to other sectors, such as livestock grazing or tourism, are attributed to the using sector, not to forestry, thereby underestimating the economic value of forests. Ecosystem services such as watershed protection and carbon storage may not be represented at all. Land and land use are not represented in SNA in the detail necessary for effective policy analysis. SEEA was developed as a set of satellite accounts to SNA to address these gaps (UN, 1993; 2001).

Environmental and natural resource accounts have evolved since the 1970s through the efforts of individual countries and practitioners, each developing their own frameworks and methodologies to represent their environmental priorities. Since the late 1980s, a concerted effort has been under way through the United Nations Statistics Division, Eurostat, OECD, the World Bank, national statistical offices and other organizations to standardize the framework and methodologies. The United Nations published an interim handbook on environmental accounting in 1993 (UN 1993b), Handbook of integrated economic and environmental accounting; this handbook is currently under revision.

Structure of SEEA

As satellite accounts, SEEA has a similar structure to SNA. SEEA consists of stocks and flows of environmental goods and services. It provides a set of aggregate indicators to monitor environmental-economic performance at the sectoral and macroeconomic levels, as well as a detailed set of statistics to guide resource managers towards policy decisions that will improve environmental-economic performance in the future. The definition of environmental goods and services in SEEA is much broader than SNA, attempting in principle to measure total economic value, not just market values. SEEA has major four components:

• Asset accounts, which record stocks and changes in stocks of natural resources.

Forestry asset accounts typically include stocks of standing timber, forest balance accounts, forest land accounts, land use accounts and carbon storage in forests.

• Flow or production accounts for materials, energy and pollution which provide information at the industry level about the use of energy and materials as inputs to production and final demand and the generation of pollutants and solid waste.

Forest flow accounts include supply and use accounts for detailed forest products (wood and non-wood, marketed and non-marketed) by sector, which are linked to the input-output (IO) tables and social accounting matrices (SAMs) used in economic models. Forest flow accounts include measures of forest ecosystem services. The forestry accounts also include flow accounts for other resources that may be relevant to forestry management, such as energy accounts, pollution (e.g. acid rain affecting forests) and environmental degradation (e.g. soil erosion from logging) accounts.

• Environmental protection and resource management expenditure accounts, which identify expenditures in the conventional SNA. For forestry, these accounts include forest management expenditures by government, environmental protection expenditures by public and private sectors, as well as user fees and taxes paid by forest users to the government.

• Environmentally-adjusted macroeconomic aggregates, which include indicators of sustainability such as environmentally-adjusted gross domestic product (GDP), net domestic product (NDP), national savings or national wealth. For this component of SEEA, forestry accounts provide the addition to GDP of unvalued forest goods and services, the subtraction from NDP of the economic cost of deforestation or loss of forest services due to a change in management and the contribution of forest assets to national wealth.

SEEA includes both physical and, to the extent possible, monetary accounts. When dealing with non-market goods and services, however, valuation can become difficult. However, there are a number of widely-used economic valuation techniques that may be applied.

Advantages of SEEA

There are two features that distinguish SEEA from other databases about the environment: integration of environmental data with economic accounts and comprehensive treatment of all important natural resources, linking them with the economic sectors that rely on them, directly and indirectly, and those sectors that affect them. In contrast to other environmental databases, the purpose of SEEA is to directly link environmental data to economic accounts. SEEA achieves this by sharing structure, definitions and classifications with SNA. The advantage of SEEA is that this kind of database provides a tool to overcome the tendency to divide issues along disciplinary lines, in which analyses of economic issues and of environmental issues are carried out independently of one another.

Regarding the second distinguishing feature, SEEA includes all the important natural resources, linking them with the economic sectors that rely on them, directly and indirectly, as well as those sectors that affect them, a feature that makes it ideal for addressing cross-sectoral issues such as forestry management. SEEA includes forest accounts as well as all other critical environmental stocks and flows related to forestry, such as land and ecosystem accounts, energy accounts, pollution and material flow accounts, etc.

For issues such as forest management, the advantage of the SEEA approach is clear. It is not possible to promote sustainable forestry purely from the narrow perspective of managing forests; rather, an economy-wide approach is needed that can identify benefits that forests provide to other sectors and threats to forests from non-forestry policies. SEEA makes possible joint analysis of economic policies and their impact on all relevant environmental variables.

Policy uses of forestry accounts

For all resources, policy analysis and decision-making take place on three relatively distinct levels: the local or company level, the sectoral or industry level and the macroeconomic (national) or regional level. The contribution of SEEA to policy analysis has been primarily at the sectoral and macroeconomic levels. At the macroeconomic level, SEEA is useful as a planning tool to coordinate policies across different line ministries and assess cross-sectoral impacts, weighing alternatives and trade-offs among sectors.

SEEA accounts for the value of non-marketed goods and ecosystem services of forests, which show how many sectors outside traditional forestry benefit from forest ecosystems. This information is critical in developing cross-sectoral policy-making institutions necessary for sustainable development. Such institutions would involve stakeholders from all sectors that benefit from forestry, such as rural development, tourism, agriculture, fisheries, municipal water supply and others. Even though agriculture is a main competitor with forests for land use in many countries, agriculture also relies on forest ecosystems in ways that are not often included in policy analysis, such as watershed protection.

The applications of SEEA will be discussed in greater detail below. However, a short summary is provided here. The contributions of SEEA fall into two categories: improved measurement of the value of forest ecosystems to all sectors of the economy and improved management of forestry by modelling economy-wide linkages to forestry.

1) Improved measures of the economic benefits from forest ecosystems to all sectors of the economy:

• Total economic value of forestry and forest land, including the non-market values of forest goods and services missing from national accounts.

• Cost of deforestation and land conversion, including losses of ecosystem benefits affecting non-forestry sectors.

• Distribution of benefits from forestry among different groups in society, e.g. commercial, artisanal and subsistence users of forests, as well as other users such as tourists, regional interests (local and regional environmental protection services) and global interests (carbon storage, biodiversity protection).

• Economic linkages, upstream and downstream, between the forestry sector and other sectors of the economy

2) Improved macroeconomic management that takes into account impacts on forestry:

• Measurement of the links between economic activities and demand for resources, including forest goods and services, land and energy.

• Impact of non-forestry sector policies on forestry.

• Impact of macroeconomic policies on the forestry sector, both direct and indirect effects.

• Measurement of environmental externalities caused by forestry and their impact on other sectors, as well as externalities generated elsewhere in the economy that affect the capacity of forests to provide critical goods and services.

USE OF SEEA FOR FORESTRY POLICY ANALYSIS

This section discusses the experiences of countries implementing SEEA and reviews the policy applications of the forestry accounts. Policy use so far has often been limited to an assessment of timber asset value and some valuation of non-timber benefits, but there are a growing number of studies that attempt to account for a wider range of non-market forest goods and ecosystem services on which other sectors depend. In many developing countries, the original motivation for forest accounts was to estimate the cost of deforestation and to assess whether rapid economic growth in countries like Indonesia was achieved by the liquidation of natural capital. A new interest in forest accounts has emerged from international efforts to reduce greenhouse gas emissions. One greenhouse gas mitigation measure is the creation of a market for carbon sinks in tropical forests. Some countries already receive payments for this service and the market may grow once international agreement is reached.

There are some cases where the forest accounts have been used for analysis of macroeconomic policies or national development strategies, but this is still a relatively new area for SEEA. Several examples will be described in detail. There is also a long tradition in forestry economics of input-output-based impact analysis to assess cross-sectoral linkages. While these studies are typically not based on SEEA, they can benefit from the extended data that SEEA provides. This approach, and how it can be improved with SEEA forestry accounts, is described.

Country experiences with forest resource accounts

Environmental accounts have been constructed for forest resources more often than for most other resources. The earliest set of forest accounts was constructed by Norway in the late 1970s. At that time only physical asset accounts for standing timber were constructed (Alfsen et al., 1987). Fuelwood was included in the supply and use table¹ for energy, which is widely used in Norway’s multi-sector macroeconomic planning model. Norway also constructed land accounts, which include information about forested land and land use by different sectors. Since that time, many other countries have constructed forest accounts and these have expanded to include monetary asset accounts for standing timber as well as non-timber goods and services.

Table 4.1 shows countries that have constructed forest accounts and the type of amenities included in the accounts. This table is limited to those countries with formal accounting programmes sponsored by government or non-governmental agencies in cooperation with governments. Forestry accounts are more common in developed than in developing countries. Since 1995, Eurostat has had an ongoing programme to develop forest resource accounts and many of the participating countries have developed extensive accounts. There are also many additional academic studies and one-off studies by governments or international agencies that are not shown here; some of these will be discussed later in the section on policy analysis. For an exhaustive review of all forest accounting efforts through 1997, see Vincent and Hartwick, 1997.

Table 4.1 - Forest accounts constructed by selected countries

Note: Countries included here have on-going accounting programmes by government, or non-governmental agencies in cooperation with governments. There have been many additional academic studies and one-time studies by governments or international agencies. See Vincent and Hartwick (1997) for a review of these studies.

The forest accounts for all countries include timber asset accounts in physical and monetary terms. Forests are disaggregated in different ways depending on the policy issues and characteristics of forests in each country. Virtually all forest accounts distinguish cultivated and natural forests and disaggregate forests by major tree species. Many developing countries limit timber accounts to commercial timber production, but are beginning to add non-commercial timber production and use of non-wood products. None of the developing countries regularly constructs detailed supply and use tables for timber and wood products. The non-timber benefit most commonly included in forest resource accounts is carbon storage. Virtually all developed countries include carbon storage accounts, but this practice is less widespread in developing countries. In most instances forest accounts have been constructed as part of a broader environmental accounting effort that includes other natural resources. In the next part of this section forest accounts will be examined in greater detail for two countries: Sweden and South Africa.

Use of forest accounts for policy analysis

Forest accounts have been used by policy-makers to provide a better estimate of the total economic value of forests and, to a lesser extent, to understand the impact on forestry of other sectors of the economy. The major concerns addressed by forest accounts include the following:

Concerns about the “true” economic value of forests

1. What is the full economic value of forests, including non-market values, and who are the beneficiaries?

2. Is economic growth based on the depletion of forests? What is the cost of deforestation?

3. Are policy decisions based on the total economic value of forests, taking into account linkages throughout the economy and impacts on all stakeholders?

Concerns about the impact on forests of non-forestry policies

4. What are the economic trade-offs among competing sectors or optimizing forest ecosystems that produce benefits to multiple stakeholders?

5. How will economic growth, macroeconomic and other policies affect forests?

The use of forest accounts to address each of these issues is discussed below.

What is the full economic value of forests?: Forest management and decisions about forest land conversion, especially in developing countries, are often based on a limited range of economic, mostly timber, values. Better understanding of the full range of goods and services supplied by forests to many other sectors is essential for the optimal utilization of forests and may provide an economic rationale for conserving forests. This was one of the motivations for the South African forestry study described below. The estimated values for Sweden and South Africa indicate that non-timber values to other sectors can be greater than the value of commercial timber harvest. Sweden’s forestry accounts are part of a well-established, comprehensive programme of environmental accounting that includes detailed land and ecosystem, energy and pollution accounts. The forestry accounts in South Africa are part of a new environmental accounting initiative by Statistics South Africa and the Department of Environment and Tourism. Much work is also being done on water accounts, which is critical for South Africa. Land and ecosystem accounts and energy and pollution accounts have not been constructed, although there are extensive databases which could be used to construct such accounts.

The Swedish and South African accounts are fairly similar in some ways: both focus on timber accounts, non-timber goods, carbon storage and recreational benefits. But the accounts differ in terms of forest classification and some of the non-timber services of forests (Tables 4.2 and 4.3). A short time series of forest values is available for Sweden and it shows remarkable stability over a six-year period. Only one year’s values are available for South Africa, so it is not possible to assess what may be happening to these forests over time.

The Swedish forest accounts include stocks of standing timber and area of forest land disaggregated by several characteristics, such as major tree species, availability for wood supply and natural or planted forest. Detailed supply and use tables for timber are also constructed. Non-timber goods include wild foods (berries, mushrooms, game, etc.), some of which are marketed and some consumed directly by households, but which account for less than 5 percent of the total value of forests. Non-timber services include carbon storage, recreation and several protective functions (biodiversity, soil protection and noise abatement), as well as two indicators of overall health: defoliation of trees and changes in forest ecosystem land area and productive capacity. Recreational use of Swedish forests is the single most important forest value, greater than the value of timber harvest. The value of carbon storage is roughly half the value of timber². The forest protective services for soils and noise abatement are negligible. Of course, there are some forest services that could not be valued, but these accounts provide a reasonable estimate of the magnitude of non-timber forest values. (For a discussion of valuation methodologies, see Norman et al., 2001.)

Table 4.2 - Value of forest goods and services in Sweden, 1993 to 1999 (millions of euros)

Source: Norman et al. (2001).

South African forests are classified by three major types: cultivated forest plantations that provide most of the country’s commercial timber and tree products, natural forests and woodlands that are used by rural communities, and fynbos woodlands, which is a unique biome in South Africa, the Cape Floral Kingdom (Hassan, 2002). Forests in national parks and protected areas have not yet been included in the accounts. Stock accounts for standing timber have only been constructed for cultivated forests. The flow accounts include production, but not detailed supply and use accounts. In South Africa, commercial timber harvest accounts for well under a third of forest value. The largest single forest value is non-market goods from natural forests, which are used mainly by traditional rural communities.

Non-timber services include carbon storage of cultivated forests, livestock grazing in natural forests and fynbos, recreational services of fynbos woodlands and the pollination service provided to agriculture by wild bees in fynbos woodlands. Pollination and livestock grazing are examples of inputs provided to agriculture at no charge and whose value is embodied in the value of agricultural output, rather than woodlands. Combined with livestock grazing, the goods and services in natural forests account for over half of the total forest value. In contrast to Sweden, recreational use of forests is very small and limited to fynbos woodlands; tourism in cultivated forests and natural woodlands is negligible. No doubt this picture would change if the recreational value of forests in national parks and protected areas, which are major domestic and international tourism sites, were included in the forestry accounts.

The forestry accounts measure an important environmental externality: the cost of water abstraction by cultivated forests. Cultivated forest plantations of exotic species (mainly pines and gums) absorb much more rainfall than native species, thus reducing run-off. South Africa is a water-scarce country so this excess water abstraction has a cost in terms of foregone use of water by downstream users, amounting to about 12 percent of the value of commercial timber harvest. In South Africa, this externality is being treated quite seriously. The new South African water policy has proposed charging forest plantations for this water abstraction externality.

Table 4.3 - Value of forest goods and services in South Africa, 1998 (millions of rands)

NA: not applicable
NAV: not available
Source: (Hassan, 2002)

Is economic growth based on the depletion of forests and other renewable resources?: In the past, loss of natural forest was not included in national accounts. Forest accounts were constructed to adjust the commonly used measures of macroeconomic performance - GDP and NDP - for depletion of natural forests and it was hoped that these environmentally adjusted measures of GDP and NDP would provide more accurate indicators of sustainable development. This type of application was typical of early work in developing countries and some of the results are shown in Table 4.4. In some instances, Indonesia and Costa Rica, the cost of deforestation was quite high. In Sweden, this value is quite small.

The World Bank includes a crude estimate of forest depletion (timber value only) in its indicator of sustainable development, Genuine Savings (Kunte et al., 1998). Genuine Savings attempts to adjust conventional net domestic savings for environmental depletion and for investment in human capital. It subtracts from net domestic savings an estimate of depletion of forest and minerals, adds expenditures on education (viewed as investment in human capital) and subtracts a notional damage charge for carbon emissions. In the World Bank estimates, forest depletion reduced net domestic savings by 20 percent in low-income countries, mostly in Asia (Hamilton, 2001).

Table 4.4 - Costs of forest depletion and degradation in selected countries

Sources: Indonesia: (Repetto, 1987); Costa Rica: (Repetto et al., 1989); Philippines: (NSCB, 1998; Delos Angelos and Peskin, 1998; Domingo, 1998); Malaysia: estimated from (Vincent, 1997); Sweden: (Ahlroth, 2000a).

There is increasing interest in measures of changes in total wealth (produced capital plus natural capital and human capital) as an indicator of sustainable development (see, for example, Dasgupta and Maler, 2000). Some countries, such as Australia and Canada, are beginning to publish figures for total national wealth that include non-produced assets such as natural forests. In Australia and Canada, the total economic value of natural capital has been quite small and the share of natural forests, valued for timber only, was extremely small (Lange, 2001a, 2001b). However, in some developing countries, such as Malaysia (Vincent, 1997) and the Philippines (NSCB, 1998; Lange, 2000), the asset value of forests can be significant.

Who benefits from the goods and services provided by forests?: The question of who benefits from forests is increasingly important for development policy. This issue has two dimensions, an inter-generational one and an intra-generational one. Inter-generational equity concerns the forest wealth left to future generations - whether society is liquidating its natural capital to pay for current consumption or using it sustainably. This has already been discussed above.

Intra-generational equity concerns the distribution of benefits among different social groups in the present generation. For example, commercial timber mainly benefits large- and small-scale commercial timber producers. In developing countries non-marketed goods and services may be critical to rural livelihoods, even when the economic value of such products may be low relative to commercial timber. The services of forests may benefit local or regional communities (e.g. livestock grazing, flood protection, prevention of soil erosion) and even national and international communities (carbon storage, biodiversity protection). In the South African study, the non-marketed value of natural woodlands makes a contribution to rural livelihoods that is greater than the value of commercial forest plantations.

Forest accounts have not been used to address systematically the issues of equity and poverty, but this use of the accounts is likely to become important in future work (e.g. Lange and Hassan, 2002). Some observations may already be made on the basis of the Swedish and South African forest accounts. The forest accounts indicate that households are major direct beneficiaries, rather than commercial operators. In Sweden, one cannot determine from the information available which social groups benefit most. In South Africa, the poor rural households depend on forests for subsistence livelihoods.

Forest valuation and trade-offs among competing uses of forests: The majority of forest accounting initiatives aim to improve the understanding of the value of forests. There is a broad literature of cost-benefit analyses to determine the optimal use of forests among competing users, often providing a strong economic argument for forest conservation. Although these cost-benefit studies do not typically make use of SEEA, they are examples of the kind of policy analysis that SEEA can support, as the example for Malaysia, discussed below, indicates.

Shahwahid et al. (1999) analyzed the trade-offs among three alternative uses of forest land in the four catchments that make up Hulu Langat Forest Reserve in Malaysia. The Forest Reserve is currently used for catchment protection, providing soil protection and water to a dam for hydroelectric power and water regulation downstream. The alternative uses are two different methods of logging: conventional logging, which provides the most timber but results in high levels of soil erosion that reduce dam capacity, and restricted-impact logging, which provides less timber than conventional logging but also less soil disturbance. The study found that the economic returns to timber alone, under either logging method, were not as great as the economic value of forests from catchment protection. Further analysis showed that a combination of restricted-impact logging and reduced catchment protection provided the greatest economic value. The relatively small reduction of forest catchment protection services from logging was compensated for by the timber value of logging, as long as the restricted-impact method was used.

Some important additional forest benefits were omitted from the analysis - recreation and tourism, biodiversity, non-timber forest products and other protective services for downstream activities. The provision of these additional benefits is compatible with catchment protection, but would be reduced by logging; if they had been included, the optimal use of forest land might not have included even restricted-impact logging. The SEEA forest accounts would provide this kind of information for local development planning.

Influence of macroeconomic and non-forestry sector policies on forests: Two different methods have been used for analyzing cross-sectoral impacts on forestry: regression analysis and simulation modelling. SEEA has not been used specifically for regression analysis, although it can provide information about forest and land use often used in these analyses. Section 4.5 describes the relationship between SEEA and forestry indicators that may be used for regression analysis. Economic simulation models provide another approach to understanding cross-sectoral impacts and it is perhaps here that SEEA can make a significant contribution. Several versions of simulation modelling are reviewed, from relatively simple forestry multiplier analysis to more complex general equilibrium analysis based on hybrid forestry IO/SAM models. This section provides an overview of forestry multiplier and impact analysis, then describes work done using forestry accounts to assess the potential for carbon storage both from the point of view of the potential provider of forest carbon storage and the potential purchaser of carbon storage services. Finally, studies in the Philippines and Indonesia that explicitly combined environmental accounts with economic models to address cross-sectoral policy linkages to forestry are reviewed (Boxes 4.1 and 4.2).

There is a long history in regional and forestry economics of applying input-output multiplier analysis to evaluate the employment and income effects of forestry on a local economy³. For example, the US Forest Service has developed an IO multiplier model that can be applied for every county in the country (Alward and Palmer, 1983; Loomis, 1993). This method is used to analyze the dependence of a local economy on forestry and to answer questions, such as, how will a change in forest land management affect the local economy? Will the loss of jobs in one sector (e.g. logging, saw milling) be offset by job gains in other sectors (e.g. tourism)? What are the effects on employment and income in other sectors of the economy?

IO models represent the transactions among all sectors of the economy in a double-entry bookkeeping framework, where each transaction is recorded simultaneously as a sale and a purchase between two sectors (Table 4.5). This allows the calculation of ‘upstream’ and ‘downstream’ linkages from one sector to all others in the economy. The upstream linkages for logging, for example, include the direct inputs purchased by the logging sector such as fuel and materials, plus the indirect inputs needed to produce the direct inputs to logging. One can trace the impacts of logging on the economy by travelling downstream as well: the use of timber as input to sawmills, the use of sawn wood by other wood processing sectors, the use of these wood products further downstream, etc. At each stage, upstream and downstream, employment and income is generated. A small change in logging creates multiplier effects throughout the economy, affecting upstream and downstream industries and the employment and income associated with them.

Virtually all industrialized countries use these IO multiplier models, or more complex general equilibrium models based on a social accounting matrix (an IO table extended to trace the flows of income), for forestry impact assessments (e.g. Ashton and Pickens, 1992; British Columbia Ministry of Forestry, 1999; Macaulay Land Use Research Institute, 1999). Multiplier analysis is also used in developing countries where IO tables are constructed, such as China, India, Indonesia, Philippines, Korea, Mexico, South Africa, etc. Simple forestry impact models are derived from national accounts and usually represent only the monetary transactions in an economy. Analysis has traditionally focused on income and employment impacts of forestry or changes in forest management, but not on the broader environmental impacts, or the impact of non-forestry policies on forests. To include impacts on the environment, hybrid IO tables have been constructed, which extend the standard IO tables for environmental data represented in physical units. ‘Hybrid’ refers to the mix of monetary and physical units in the table. Hybrid accounts have been used extensively for energy analysis (e.g. Miller and Blair, 1985; Pearson, 1989; UN, 1999). There has been some use of partial forestry IO models in conventional multiplier analysis, but such analyses usually include only the use and supply of wood products in physical units.

SEEA allows for the construction of a full hybrid forestry IO model that extends the IO tables constructed from SNA with SEEA satellite accounts for non-market goods and services related to forestry. Hybrid forestry IO models include the use of forest products (detailed supply and use table in monetary and physical units), use of non-timber products, land use and other environmental factors that may affect forests in a given area: energy, pollution, soil erosion, etc. The model thus includes physical and monetary data about all the forestry-related resources needed for sustainable forestry management and for assessing cross-sectoral impacts on forestry.

Table 4.5 shows the kinds of resource and environmental impacts that might be included in the forestry IO table. Each of the resources shown, such as non-wood products or land, would be further disaggregated by characteristics relevant to the economy and forest.

A new interest in forest and land accounts has emerged from international efforts to compensate for greenhouse gas emissions by creating carbon sinks in tropical forests. A growing number of studies have analyzed the potential value of forest land as a carbon sink compared to its value under alternative uses. Analyses of forest and land accounts in tropical countries have estimated reservation prices of farmers: the minimum payment farmers would be willing to accept for using forest land for carbon storage rather than other purposes. A study by Castro and Cordero (2001) estimated the reservation prices in eight regions of Costa Rica (which have different opportunity costs and carbon productivity) for 27 different agricultural activities. The reservation prices were lowest for livestock and rice and highest for export crops like coffee and pineapples.

The reservation price countries are willing to accept to store carbon in forests can be matched by the corresponding price that countries are willing to pay for carbon storage. There is extensive analysis in Europe, based in part on SEEA, of green taxes, especially carbon taxes. These models, usually multi-sectoral CGE models, use the energy and pollution accounts of SEEA to assess how high carbon taxes would have to be to achieve a target level of emissions. However, policy-makers can also consider other carbon mitigation measures, such as purchase of tradable carbon emission permits or carbon storage by forests. Tropical forests can offer attractive options for carbon storage.

Table 4.5 - Hybrid forestry input-output table

A. Inter-industry table (in monetary units)

A Swedish study (Nilsson and Huhtala, 2000) analyzed the advantages to Sweden of purchasing carbon-trading permits as an alternative to implementing measures to reduce domestic levels of carbon emissions in order to meet Sweden's carbon target under the Kyoto Protocol. The analysis estimated a ‘reservation price’ indicating the maximum amount a country would be willing to pay for carbon storage in tropical forests. As carbon trading becomes established, the forestry accounts can be useful to help assess both sides of a trade: willingness to accept payment for carbon storage and willingness to pay.

FRAMEWORK FOR ANALYSIS OF CROSS-SECTORAL POLICY LINKAGES

SEEA-based framework

Based on the experience with forestry accounts reviewed above, a generalized SEEA-based framework for analyzing cross-sectoral linkages can be proposed. Some policy issues can be addressed by a relatively narrow forestry accounting framework, while others require more comprehensive accounts for forests and related resources. This framework may be used to assess three major kinds of issue:

1. Impact of forest use on macroeconomic performance. (Is economic growth based on the depletion of forests? What is the cost of deforestation?).

2. Full economic value of forests, including non-market values, taking into account linkages throughout the economy and impacts on all stakeholders.

3. Economy-wide impacts of non-forestry policies on forest use.

Each application, and the components of SEEA on which it is based, is discussed below. Table 4.6 summarizes the components of SEEA required for each application.

Impact of forest use on macroeconomic performance: At the macroeconomic level, SEEA provides indicators of total forest value and the cost of forest depletion which may be used in macroeconomic planning, such as:

• GDP that includes the production of all forest benefits, market + non-market, not just those values included in SNA.

• NDP that includes the cost of forest depletion and environmental degradation.

• National savings and wealth adjusted to include the value of both cultivated and natural forests and the capitalized value of all forest goods and services, not just timber.

The indicators tell policy-makers, for example, how dependent the national economy is on forests, whether this dependence is increasing or the economy is becoming more diverse and the extent to which economic growth is sustainable or has been obtained by liquidating natural capital like forests. This macroeconomic indicator framework is used to integrate the results from more detailed analysis, such as simulation models, with measures of national economic performance.

Table 4.6 - Major applications of SEEA for sustainable forestry

The components of SEEA used for this include:

• Asset accounts, physical and monetary, for forests and land, which provide the cost of depletion and forest values for national wealth and savings.

• Flow accounts for production of forest goods and services as well as environmental degradation which provide figures for total forest production for GDP; detailed supply and use accounts are not required.

Full economic value of forest ecosystems, accounting for benefits to non-forestry sectors: Forest accounts address one of the often-cited indirect causes of pressure on forests: the absence of market prices for non-timber benefits. Forestry accounts measure how non-forestry sectors benefit from forests and what they have to lose from deforestation. As the forest accounts from Sweden and South Africa showed, non-timber and, in general, non-market benefits are a substantial - if not the major - forest value.

SEEA provides information for a cost-benefit approach that may be used for local/regional land use and development planning to determine optimal forest use, ensuring that all forest values are taken into account. This analysis compares the economic trade-offs among competing uses of forest, such as logging, subsistence use of forests, recreation and tourism, conversion to agriculture, etc. The example from Malaysia (page 123) shows how land use conversion and deforestation could result from a lack of information about all the goods and services provided by forests to other sectors of the economy and a lack of institutions or regulations to monetize these services so that forest owners are compensated for the services forests provide. The SEEA components used for this application include the flow accounts, especially the non-timber values such as:

• Subsistence livelihoods (non-market goods provided as well as services used as input to livestock production).

• Recreation and tourism industry, biodiversity (in its own right and as part of a tourism industry).

• Services used as input to agriculture (livestock grazing, pollination, etc.).

• Protective services to soil and water (affecting activities as diverse as hydroelectric power generation, agriculture or fishing) or global climate (carbon sequestration value).

In a more extensive analysis, the value of forests under an alternative mix of uses may be capitalized into SEEA forest and land asset values to measure how different forest use and land use policies affect national wealth. This use of forestry accounts can provide a strong basis for building the cross-sectoral alliances necessary for sustainable forestry because it identifies how other sectors depend on and benefit from forests.

Economy-wide impacts of non-forestry policies on forestry: Simulation analysis is an ambitious economy-wide assessment of the chain of effects from policies at the macroeconomic level, through the activities and policies of different economic sectors to decisions by individual agents about the use of forests. Consequently, simulation analysis requires a great deal of data about the economy, the use of natural resources and the impact on the environment of economic activities. The examples for the Philippines and Indonesia used environmental-economic models based on a combined database of an IO/SAM and SEEA can trace the interdependencies of the economy and the environment and the chain of effects of economic policies on the natural resource base.

The components of SEEA that would typically be required for analysis of the sort carried out for Indonesia and the Philippines are shown in Table 4.7. They include forest, land and other related accounts depending on a country’s specific environmental characteristics and policies. Forest asset accounts include timber and non-timber values, forest land and carbon sequestration. The asset accounts record the state of forests and their changes in a given year. Forest balance accounts are constructed from a combination of the forest asset and flow accounts. Forest flow accounts include the use of commercial and non-commercial wood products as well as non-wood goods and services throughout the economy. These accounts are included in order to trace the effect on forests of changing demands for forest products due to changes in policy. The demand and supply of commercial wood products are often included in analytical models; SEEA provides the same information for non-wood forest products and forest services. Environmental degradation that results from specific uses of forests and the impacts on other sectors of the economy would also be included.

The components of SEEA used for simulation models also include those resources that are closely related to forest use and deforestation, most importantly land. Land accounts have been constructed using many different types of classification depending on the purpose of the land accounts. For forestry issues, land could be classified by ecological characteristics such as type of land cover, slope and soil and agricultural potential. Land accounts can also be classified by economic or institutional characteristics such as degree of forest protection, accessibility to settlers, economic user of the land (with detailed accounts for users like agriculture and infrastructure, which put the most pressure on forest land). Finally, other accounts, such as pollution, energy and water, may be useful depending on the causes of deforestation in a particular country.

Table 4.7 - Components of SEEA used for simulation models

Strengths and limitations of the SEEA framework for forestry

SEEA provides a powerful tool for analyzing policy impacts on forestry. Its major advantages, as discussed above, include the comprehensive treatment of all related environmental benefits and the integration of environmental information with national accounts. However, there are two kinds of limitation of SEEA that warrant some discussion. The first concerns the data requirements of SEEA and its applications, especially the simulation models. The second concerns the spatial disaggregation of SEEA; while official forestry accounts are usually compiled at the national level, local level accounts may be more useful for policy analysis.

Data requirements of SEEA and its policy applications: The first application of SEEA to sustainable forestry - assessing the contributions of forest goods and services to GDP and the cost of changes in forest land use - does not require a great deal of data. The basic forestry accounts are required and these can be constructed fairly easily from existing data. The second application requires more detailed information about the economic benefits to all stakeholders. Often the biophysical data may not be complete and economic valuation is challenging. But there is now a large body of literature to draw on for monetary valuation and a reasonable estimate of the value can often be obtained fairly easily.

Simulation models used to assess impacts of non-forestry policies on forests require a great deal of data. Even forestry multiplier models require input-output tables of the economy, which are not constructed by all developing countries and the data are not always very reliable. CGE models are designed to assess the response of households and firms to changes in market signals, such as the relative prices of products, labour or exports, so they are particularly well suited to addressing the cross-sectoral policy linkages affecting forestry. CGE models are based on SAMs, which represent the most detailed implementation of national accounts. The drawback of simulation modelling is the amount of data required. For countries that do not compile SAMs, IO tables may be available, which can be used for more limited simulation modelling.

Spatial characteristics of SEEA forest accounts: Researchers have noted that forest management is often a relatively localized phenomenon; a nation often has separate forests with different uses and economic values. At the national level, the value of forest ecosystem benefits may constitute a relatively small share of GDP, even when all the non-market benefits are accounted for. However, forest benefits may be very high for specific regions within a country. Moreover, SEEA is most often constructed at the national level, obscuring the regional and local importance of forests.

It is increasingly common to build national forest accounts from accounts for more detailed sub-national regions or for specific forests. For example, a collection of six case studies of forest accounts compiled for individual forests in Spain, USA and Costa Rica is provided in Campos (2001). For analysis of the impact of macroeconomic policies, a national-level model is necessary. In both the Philippine and Indonesian studies the disadvantage of national accounts was overcome by disaggregating forest and land accounts by geographic and ecological characteristics. The national-level framework of SEEA then became a strength rather than a weakness, because it provided a framework for consistent and comprehensive treatment of all land and forests, which in turn allowed aggregation of localized impacts to determine the cumulative impacts for the national economy.

SEEA AND SUSTAINABILITY INDICATORS FOR FORESTRY

In recent years, there have been a number of efforts to develop criteria and indicators for sustainable forestry, based on economic, social, ecological and institutional statistics. There is considerable overlap between work on sustainability indicators and SEEA, although these two efforts have proceeded, for the most part, independently of one another. One of the advantages of SEEA is that it produces both indicators as well as the detailed statistics needed for analysis. The relationship between SEEA and two sets of indicators of sustainability is described below. The first set, the Montreal Process indicators, provides a fairly well developed set of indicators for forests. The second set is based on the UN’s Driving Force-State-Response system for sustainability indicators.

SEEA and the Montreal Process indicators

The Montreal Process represents one such attempt to develop and implement internationally agreed criteria and indicators for the conservation and sustainable management of temperate and boreal forests. (See their website http://www.mpci.org for more information.) The Montreal Process has identified a set of criteria: categories of conditions or processes by which sustainable forest management may be assessed. Each criterion is characterized by a set of related indicators, quantitative or qualitative variables which can be measured or described and which, when observed periodically, demonstrate trends. Table 4.8 shows the relationship between the Montreal Process criteria and indicators and the information provided by SEEA; many of the Montreal Process indicators are provided by SEEA.

Table 4.8 – Correspondence between sustainability indicators and SEEA

SEEA and the Driving Force-State-Response indicators

Indicators for forestry that can be derived from SEEA may be grouped for convenience within the Driving Force-State-Response Framework (DSR) adopted in 1995 by the UN Commission on Sustainable Development. In DSR, driving force indicators represent human activities, processes and patterns that have an impact on sustainable development; they provide an indication of causes of changes both positive and negative in the state of sustainable development. Driving force indicators can pertain to various levels and issues such as individual companies, sectors and demographic and social trends. State indicators can pertain to qualitative and quantitative dimensions of sustainable development, including abundance of natural resources, level of education, average life span, etc. Response indicators provide an indication of the willingness and effectiveness of society in providing responses to sustainability issues. Conceptually, no unique attribution of sustainability indicators into one of these three categories appears warranted as one and the same indicator could serve as a driving-force indicator as well as a state or response indicator. The following list (Table 4.9) provides examples of indicators that may be derived from SEEA; many additional indicators could also be obtained from SEEA.

Table 4.9 - Examples of indicators provided by SEEA for the DSR framework of indicators

CONCLUSIONS

In most countries, forest accounts have mainly been used to assess forest asset values and the value of forest goods and services, providing a better indication of the benefits of forests to all sectors of the economy and what would be lost from deforestation. This information can be useful in cost-benefit analyses to assess the economic benefits and trade-offs from alternative uses of forests. Few countries have taken full advantage of the opportunities provided by forest accounts for analysis of the linkages between forestry and other sectors of the economy or macroeconomic developments. Part of the problem is one of information. Detailed information is needed about the flows of forest goods and services to each sector of the economy, as well as the use of land and other resources by each sector of the economy. As seen in Table 4.1, only developed countries compiled such detailed accounts on a regular basis. Two countries, Philippines and Indonesia, have used environmental accounts to examine cross-sectoral policy impacts on forestry. Although events have largely overtaken both these countries since the time of the studies, they illustrate the kind of analytical framework that may be developed from SEEA.

Sustainable development requires that forestry policy be addressed in a cross-sectoral context. This can be achieved only when there is a clear incentive to build an alliance among stakeholders in different sectors. A powerful incentive is the information that SEEA forest accounts provide about the economic benefits of forest ecosystems to non-forestry sectors, including rural development agencies, agriculture, fisheries, tourism, municipal water supply and others. Potential stakeholders include government, business and civil organizations, as well as private citizens.

In addition to the political and institutional motivation, SEEA forestry accounts provide a useful technical framework for analysis of cross-sectoral impacts. SEEA provides a framework for assessing the total economic value of forests, as well as a framework for linking information about forestry to the use of other resources and to the broader economy, integrating forestry policy with national development and monitoring interactions and feedback across different industries. Forestry sector managers gain important information from SEEA about the total economic value of forest resources, especially the inputs provided to sectors that are not part of the traditional forestry sector. Perhaps more importantly, SEEA puts the information they normally produce and work with in the context of the national economy. SEEA provides them with a tool to identify and address threats to forest resources that originate outside the forestry sector, which can improve their ability to protect this resource.

Policy-makers and stakeholders outside the forestry sector benefit from SEEA in several ways. First of all, SEEA provides a measure of the economic value of forest ecosystems to non-forestry sectors and the distribution of these benefits among different stakeholders in society. SEEA also provides better indicators of sustainable development that include forestry and forest land resources, as well as a method to integrate forestry into macroeconomic policy and planning tools. With SEEA forest valuation is linked to the national economy. This makes it possible to evaluate local and regional land use decisions in terms of how they affect the national economy.

SEEA can provide a powerful informational tool to promote cross-sectoral policy-making but no country is taking full advantage of its potential at this time. To be more effective, forest accounting practices need to be improved, partly in terms of technical content and partly in terms of how it communicates technical information to stakeholders. The following are some suggestions:

Technical improvements:

Increase forest accounting at the regional or forest-level. Because forest utilization tends to be a local or regional issue, there is a need for more disaggregated forest accounts, rather than purely national forestry accounts. The increasing number of case studies that apply the SEEA framework to individual forests is a promising development.

Use existing information to begin constructing accounts for non-market goods and services. Because the importance of tropical forests has been recognized worldwide, there is a great deal of information that may be used to construct forest accounts, even in countries with relatively limited data collection resources. The extensive body of literature on valuation of non-market forest goods and services can provide a starting point for many countries.

Communicating information to stakeholders:

To be effective in building alliances across different sectors and stakeholders, forestry information needs to be presented in a clear and policy-relevant manner, not just in the dry technical format of formal accounts. There are two ways to achieve this:

Build forest accounts around policy priorities. An issue or set of issues such as carbon storage, rural livelihoods, tourism and biodiversity or watershed protection can provide a theme for communicating the results of the accounts. Clearly, the broader the group of stakeholders affected, the more effective the forestry accounts will be.

Develop easily-accessible, non-technical information for educating stakeholders. A good example of such an approach is the publication, Econnections, by Statistics Canada, which provides information from the environmental accounts of Canada. Econnections was designed as a tool to educate all stakeholders, government policy-makers, business groups, NGOs and even for use in schools. The first volume of the report presents the accounts for important environmental themes such as forests or energy use in terms of the policy issues they address. The second volume provides a more technical explanation of each account. An accompanying CD provides more detailed data so that users may perform their own analysis.

REFERENCES

Ahlroth, S. (2000a). Correcting NDP for SO2 and NOx emissions: implementation of a theoretical model in practice. Swedish National Institute for Economic Research.

__________ (2000b). GDP simulations in Sweden: simulations of tax policy for CO₂ reductions. Unpublished paper of the Swedish National Institute of Economic Research.

Alaska Rainforest Campaign. No date. See website www.akrain.org

Alward, G. & Palmer, C. (1983) IMPLAN: an input-output analysis system for forest service planning. In: IMPLAN training notebook. Land Management Planning, Rocky Mountain Forest and Range Experiment, US Forest Service, Fort Collins, Colorado.

Ashton, P. & Pickens, J. (1992). Community stability: natural resource based economic activities and relationships to national forest management on 10 western forests. Report for the US Forest Service.

Barbier, E. (2001). Agricultural expansion, resources booms and growth in Latin America. Paper presented at Conference on Natural Capital, Poverty and Development, University of Toronto, 5-8 September.

Bartelmus, P., Lutz, E. & Schweinfest, S. (1992). Integrated environmental and economic accounting: a case study for Papua-New Guinea. World Bank Environmental Working Paper No. 54.

British Columbia Ministry of Forestry. (1999). Soo timber supply area analysis report. Vancouver, British Columbia.

Broadhead, J. (2001). Cross-sector policy impacts in forestry: examples from within and outside FAO. Report to Forestry Department, FAO.

Bull, G., Mabee, W. & Sharpenberg, W. (1998). Global fibre supply model. Rome. FAO.

Campos, P. 2001. Forestlands new economic accounting: theories and applications. Special issue of Investigación Agraria.

Castro, R. Cordero, S. (2001). Tropical forests and the emerging CO₂ market. Investigación Agraria, No. 1, pp. 185-204.

Contreras Hermosilla, A. (2000). The underlying causes of forest decline. Occasional Paper No. 30. Bogor, Indonesia. CIFOR.

Cruz, W. & Repetto, R. (1992). The environmental effects of stabilization and structural adjustment programs: the Philippine case. Washington, DC. World Resources Institute.

Das Gupta, P. & Maler, K-G. (2000). Net national product, wealth and social well-being. Environment and Development Economics. Vol. 5, pp. 69-94.

Davis, G. & Moore, D. (2000). Valuing mineral stocks and depletion in green national income accounts. Environment and Development Economics. Vol. 5, parts 1 and 2, February and May, pp. 109-128.

Delos Angelos, M. & Peskin, H. (1998). Philippines: environmental accounting as an instrument of policy, In: P. Bartelmus and K. Uno, eds. Environmental Accounting in Theory and Practice. Kluwer Academic Press, Dordrecht, Netherlands.

Domingo, E. (1998). Philippines: adaptation of the United Nations system of environmental accounting. In: P. Bartelmus and K. Uno, eds. Environmental Accounting in Theory and Practice. Kluwer Academic Press, Dordrecht, Netherlands.

de Montalembert, M.-R. (1995). Cross-sectoral linkages and the influence of external policies on forest development. Unasylva. Vol. 46, no.182, pp 25-37

Eurostat (2001). Environmental accounts for forests: non-ESA/SNA functions. Eurostat: Luxembourg.

________ (2000). The European framework for integrated environmental and economic accounting for forests--IEEAF. Eurostat: Luxembourg.

FAO (2001). State of the world’s forests 2001. Rome. FAO.

Førsund, Finn (1985). Input-output models, national economic models and the environment. In: Kneese, A.V. and Sweeney, J.L., eds. Handbook for natural resource and energy economics. Vol. 1, 325-341, Elsevier Publishing Co., New York.

Hamilton, C. (1997). The sustainability of logging in Indonesia’s tropical forests: a dynamic input-output analysis. Ecological Economics. Vol. 21, pp. 183-195.

Hamilton, K. & Clemens, M. (1999). Genuine savings rates in developing countries. World Bank Economic Review 13:2, 333-56.

Hassan, R. (2002). Forestry accounts: capturing the value of forest and woodland resources. In: G. Lange, R. Hassan and K. Hamilton, Environmental accounting in actions: case studies from Southern Africa. Edward Elgar Publishers, Cheltanham, UK.

Huitric, M., Folke, C. & Kautsy, N. (2002). Development and government policies of the shrimp farming industry in Thailand in relation to mangrove ecosystems. Ecological Economics 40 (3): 441-455.

Kaimowitz, D. & Angelsen, A. (1998). Economic models of tropical deforestation: a review. Bogor, Indonesia. CIFOR.

_________________ (1999). The World Bank and non-forest sector policies that affect forests. Bogor, Indonesia. CIFOR.

Kunte, A., Hamilton, K., Dixon, J. & Clemens, M. (1998). Estimating national wealth: methodology and results. Environment Department Papers, Environmental Economics Series No. 57, Washington, DC. World Bank.

Lange, G. (1997). Strategic planning for sustainable development in Indonesia using natural resource accounts. In: J. van den Bergh and J. van der Straaten, eds. Economy and Ecosystems in Change: Analytical and Historical Approaches, Aldershott, UK. Edward Elgar Publishing.

_____________ (2000). The use and policy applications of the Philippine System of Environmental and Natural Resource Accounts. Report for the Philippines National Statistical Coordinating Board. Manila.

_____________ (2001a). Applications and policy uses of the SEEA. Report to the London Group for preparation of the Revised Handbook on the System of Integrated Environmental and Economic Accounts. March.

_____________ (2001b). Environmental accounts: uses and policy applications. Working Paper of the Environment Department, World Bank.

Lange, G. & Hassan, R. (2002a). Using environmental accounts to link poverty, environment and development. Paper presented at the workshop on poverty reduction, environmental management and sustainable development: theory and practice. 24-26 May 2002, Durban, South Africa.

Loomis, J. (1993). Integrated public lands management. Columbia University Press, New York.

Macaulay Land Use Research Institute (1999). Scottish forestry: an input-output analysis. Report for the Forestry Commission.

Miller, R. & Blair, P. (1985). Input-output analysis: foundations and extensions. Prentice-Hall, Inc: Englewood Cliffs, New Jersey.

Nilsson, C. & Huhtala, A. (2000). Is CO₂ trading always beneficial? A CGE-model analysis on secondary environmental benefits. Paper of the NIER. NIER, Stockholm

Nordhaus, W. & Kokkelenberg, E. (1999). Nature's numbers. National Academy Press: Washington, DC.

Norman, L., Eriksson, H., Eriksson, M. & Skanberg, K. (2001). Environmental accounts for forests: non ESA/SNA functions. Report to Eurostat.

NSCB (National Statistical Coordination Board) (1998). The Philippine asset accounts: forest, fishery, minerals, land/soil and water. NSCB: Manila.

Pearce, D. (1996). Global environmental value and tropical forests: demonstration and capture. In: W.L. Adamowicz et al., eds. Forestry, economics and the environment. Wallingford. CAB International.

Pearce, D., Putz, F. & Vanclay, J. (1999). A sustainable forest future? Report to the Natural Resources Institute and the Department for International Development.

Pearson, P. (1989). Proactive energy-environment policy strategies: a role for input-output analysis? Environment and Planning A, 21, 1329-1348.

Repetto, R., Wells, M., Beer, C. & Rossini, F. (1987). Natural resource accounting for Indonesia. Washington, DC. World Resources Institute.

Repetto, R., Magrath, W., Wells, M., Beer, C. & Rossini, F., (1989). Wasting assets: natural resources in the national accounts. Washington, DC. World Resources Institute.

Schmithüsen, F., Bisang, K. & Zimmermann, W. (2001). Cross-sector linkages in forestry: review of available information and considerations on further research. Paper prepared for the Forestry Department, FAO.

Sedjo, R. & Simpson, R.D. (1999). Trade liberalization, wood flows and global forests. Discussion Paper 00-05. Washington, DC. Resources for the Future.

Shahwahid, M., Noor, A., Rahim, A., Zulkifli, Y. & Razani, U. (1999). Trade-offs on competing uses of a Malaysian forested catchment. Environment and Development Economics. Vol. 4, part 3, July pp. 279-312.

Statistics Canada. (1998). Econnections. Ottawa. Statistics Canada.

Statistics Finland (2000). Forest Accounting 1990-1998. Helsinki. Statistics Finland.

Statistics Sweden (2001). Environmental Accounts for Forests. Stockholm. Statistics Sweden.

United Nations (1993a). System of national accounts. New York: UN.

United Nations (1993b and 2001). Handbook of integrated environmental and economic accounts. New York. UN.

______________ (1995). Operational manual for the system of integrated environmental and economic accounts. New York. UN.

________________ (1999). Handbook of input-output table compilation and analysis. New York. UN.

van Tongeren, J., Schweinfest, S. and Lutz, E. (1991). Integrated environmental and economic accounting: a case study of Mexico. World Bank Environment Working Paper No. 50. Washington, DC. World Bank.

Verolme, H. & Moussa, J. (1999). Addressing the underlying causes of deforestation and forest degradation: case studies, analysis and policy recommendations. Biodiversity Action Network. Washington, DC.

Vincent, J. (1997). Resource depletion and economic sustainability in Malaysia. Environment and Development Economics. Vol.2, Part 1, pp. 19-38.

______________ (1999). Net accumulation of timber reserves. Review of income and wealth. Vol. 45, pp 251-262.

Vincent, J. & Hartwick, J. (1997). Forest resources and the national income accounts: concepts and experiences. Report to the FAO Forestry Department.

World Resources Institute (2001). World resources 2001. Washington, DC. World Resources Institute.

¹ The supply and use table is part of SNA and shows the supply of a product by each industry and the use of that product by each industry; it is the basis for the input-output table.
² Several methods were used to value carbon storage. The value reported here is one of the lower values and is the one favoured by the authors.
³ Analysis also uses, where available, social accounting matrix models, which are IO models expanded to include more detailed information about the generation and spending of incomes among different categories of household.