2.3.1    Scale of forest utilization: distribution of benefits to commercial, artisanal and subsistence users

Commercial logging mainly benefits large-scale commercial and artisanal timber producers and, until recently, these values were often the only ones considered in forest management. The beneficiaries from other forest products vary by country. In many developing countries, subsistence forest users benefit from non-marketed goods and services such as livestock grazing. These forest products may be critical to the livelihoods of local communities and thus have a high social value even when the economic value of such products is low relative to commercial timber.

Shackleton and Shackleton (2002) document the importance of NTFP to rural livelihoods in South Africa based on extensive surveys of the use of NTFP throughout rural South Africa. All rural households make use of NTFP to some extent. The authors distinguish two types of use: daily or regular subsistence use of NTFP which saves scarce cash to be used for other household needs, and the emergency safety net use of NTFP, which is the additional use or trade in NTFP in response to unexpected household difficulties (drought, illness, etc.). The authors find that in times of hardship the poorer rural households increasingly rely on collecting NTFP for sale in informal markets. Thus, NTFP are important not only for daily subsistence but also as part of a coping strategy to diversify livelihoods and reduce vulnerability to hardships.

Recreational services primarily benefit households. In developing countries, forest recreational services may be enjoyed mainly by foreign visitors but they also provide employment to local communities. In some instances, recreational services are provided free of charge or at a cost that does not reflect its value to the beneficiary. This often occurs for use of natural parks, protected areas and natural forests. Thus the recreational services of forests are undervalued and do not send an appropriate signal for sustainable forest management. In developing countries, this may result in transfer of benefits from the host country to relatively well-off foreigners. Services that provide indirect benefits of a public goods nature, such as watershed protection and carbon storage, accrue to multiple beneficiaries.

While no forest accounts have been compiled with the idea of representing these distributional aspects, some observations may be made on the basis of the Swedish and South African forest accounts (Table 2.7). In both sets of accounts, artisanal producers are not distinguished from large‑scale commercial operators, but it is possible to distinguish commercial operators, household forest users and multiple beneficiaries. In South Africa, poor rural households depend on forests for subsistence livelihoods and accounts for the single largest forest value. In Sweden, households are also the major beneficiaries, but mainly in the form of recreational services.

2.3.2    Distance from forest: distribution of local, regional and global benefits

When major benefits do not accrue to land owners/users, the incentive for sustainable forestry declines, even though the social benefits from sustainable forestry may outweigh the benefits from land use conversion. Beneficiaries may be categorized as local forest owners/users, regional beneficiaries and global beneficiaries. Two examples are shown from the forest accounts for Swaziland and Spain to illustrate the forms these divisions may take (Table 2.8); these accounts are discussed in greater detail in the next chapter.

Table 2.7:        Distribution of forest benefits by purpose or scale of forest use

 

		Sweden, 1999 (million euros)	South Africa (million rands)
Private commercial operators		2370 (41%)	2721 (29%)
	Commercial logging and forest products	2370	1935
	Commercial agriculture (pollination in South Africa)	NA	786
Households		2595 (45%)	3663 (55%)
	Non-market timber and other NTFP (Subsistence production in South Africa)	225	3634
	Recreation (value to visitors)	2370	29
Multiple beneficiaries		820 (14%)	255 (16%)
	Carbon storage	810	480
	Other environmental services including negative impacts	20	-225
Total		5795	6639

NA: not available

Source: Tables 4.3 and 4.4

Local users/owners in both countries receive the largest share of forest benefits: 55 percent and 47 percent in Swaziland and Spain, respectively. These include benefits from commercial logging and household harvest of non-market timber and NTFP. In Spain commercial logging is carried out by local companies, but in Swaziland most of the logging operations are foreign owned, outsourced to local operators. It is useful to distinguish the foreign operators as their interests may differ from local operators.  There are no non-local, regional benefits identified in Swaziland’s forest accounts. In Spain, the proximity of the forest to Madrid makes it an attractive recreational site for city‑dwellers. Recreational services accruing to regional beneficiaries account for 42 percent of the forest’s total economic value, but visitors typically do not pay for the use of forests. Finally, the global community benefits from forest services such as international tourism, carbon storage and biodiversity protection, which account for 32 percent and 12 percent of forest values in Swaziland and Spain, respectively.

Table 2.8:        Distribution of forest benefits between land owners/users and others in Swaziland and Spain (percent of total forest value)

Swaziland (national forest accounts)		Spain (accounts for Guadarrama Forest)
Local beneficiaries   Subsistence household   (non-market timber and NTFP) Foreign-owned local beneficiaries   (commercial logging)	68%   55%     13%	Local beneficiaries (commercial logging and NTFP)	47%
		Non-local regional beneficiaries (recreation)	42%
Global beneficiaries (carbon storage, international tourism)	32%	Global beneficiaries (carbon storage, biodiversity protection)	12%

Source: Adapted from Mbuli, 2003 and Capparós, 2001. See Chapter 5 for more detailed figures and discussion of sources.

2.4 Is economic growth based on the depletion of forests and other renewable resources?

In the past, loss of cultivated forest was included in national accounts but loss of natural forest was not. 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 2.9. 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 rough 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 2.9:        Costs of forest depletion and degradation in selected countries

 

Country	Change in GDP/NDP
Indonesia, 1971-1984	-5.4% of GDP
Costa Rica, 1970-1989	-5.2% of GDP
Philippines, 1988-1992	-3.0% of GDP
Malaysia, 1970-1990	-0.3% of GDP
Sweden, 1998	-0.03% of NDP
Swaziland	-0.83 % of GDP

Sources: Indonesia: (Repetto et al. 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); Swaziland: Mbuli, 2003).

There is increasing interest in measures of change in total wealth (produced capital plus natural 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.

2.5 Forest valuation and trade-offs among competing uses of forests

Improved understanding of the value of forests can be useful in cost-benefit analyses to determine the optimal use of forests among competing users, often providing a strong economic argument for forest conservation, or at least lessening the incentive for deforestation. In one example, Shahwahid et al. (1999) analyzed the trade-offs among three alternative uses of forestland 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 forestland might not have included even restricted-impact logging. A similar study of alternative uses of Tongass National Forest, an old growth, temperate rainforest in Alaska, compared the economic values of logging, tourism and protective services for the fishing industry (maintaining the water quality of rivers used as spawning grounds by fish). Studies showed that the value of forest services to recreation and fishing exceeded the timber value of forests (Alaska Rainforest Campaign, no date).

Although the Malaysian and Alaskan forest studies, and many other similar studies, did not use the SEEA forest accounting framework, they are examples of the kind of policy analysis that forest accounts can support. Forest accounts provide a framework for assessing the total value of forests, not just direct commercial value from extractive activities, but the goods and services (or loss of these services) provided to other industries as well. An additional example, based on the conservation forest reserves of Tanzania, is provided in Chapter 4 where the implementation of cost-benefit analysis using forest accounts is discussed.

The assessment of trade-offs among competing users can be estimated in a partial equilibrium cost‑benefit analysis, such as the study mentioned above, or in a larger, economy-wide general equilibrium modelling framework. The next section discusses such a modelling approach, which is used for capturing the full cross-sectoral impacts on forestry. But this modelling approach can also be used for evaluating alternative forestland use.

2.6 Modelling the economy-wide impact of non-forestry policies

Assessment of trade-offs in a partial equilibrium framework is a first step towards understanding the cross-sectoral policy impacts on forestry. But understanding the impact of broader changes, such as trade liberalization, population growth, agricultural policy, etc. often requires an economy‑wide environmental-economic model. Economic simulation models have been widely used to understand cross-sectoral impacts on forests. Forest accounts are quite useful in this type of analysis because they are an extension of, and consistent with, national economic accounts. 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.

2.6.1    Forestry multiplier and impact analysis

There is a long history in regional and forestry economics of applying input-output (IO) or social accounting matrix (SAM) multiplier analysis to evaluate the employment and income effects of forestry on a local economy[1]. 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 changes in forestland 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. 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 sawnwood by other wood processing sectors, the use of these wood products further downstream, etc. At each stage, upstream and downstream, employment and income are 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 SAM (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, the 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 forestry IO tables in conventional multiplier analysis, but such analyses usually include only the use and supply of wood products in physical units.

Forest accounts allow construction of a hybrid forestry IO table that includes non-market forest goods and services as well. For more extensive analysis, such as that described in the next section, extended forest-related accounts are required. Forest‑related accounts would include forest goods and services plus accounts for land and other environmental factors that may affect forests in a given area: energy, water, pollution, soil erosion, etc. The model thus includes physical and monetary data about all the forest-related resources needed for sustainable forestry management and for assessing cross-sectoral impacts on forestry. The framework for such a hybrid IO table is shown in Table 2.10.

Table 2.10:     Hybrid input-output table for forest and forest-related resources

A. Inter-industry table (in monetary units)

Intermediate consumption by ISIC code

Final users

1

2

3

4

5

6

7

Consumption (public + private)

Imports

Exports

Capital formation

1. Agriculture

2. Forestry

3. Mining

4. Manuf. of wood products

5. Other manufacturing

6. Utilities and trade

7. Services

Value-added

Employment

B. Extension for hybrid forestry IO table (in physical units)

Wood products

Non-wood products

Land

Energy

Water

Pollution, soil erosion, other environmental impacts

 

2.6.2    Forestry simulation models and forest accounts

There are several examples of hybrid simulation models for forestry based on environmental accounts. The first two studies reviewed here are relatively limited in scope—a study of fuelwood use in Tanzania and the potential demand for forests as a carbon sink. The next two studies address the issue of deforestation in a much broader context—the impact of structural adjustment programmes in the Philippines and the impact of Indonesia’s second long‑term development plan on forests. The latter two studies are good examples of attempts to fully understand the complex linkages between macroeconomic policy and deforestation.

Peskin et al. (1992) undertook a study of factors influencing deforestation in Tanzania using environmental accounts for forestry and energy. They found that the use of fuelwood was a major contributing factor to deforestation; fuelwood is not only widely used by households, but is also widely used for processing agricultural products, notably tobacco curing. Peskin found that fuelwood use was strongly influenced by energy pricing and macroeconomic policies that affected the foreign exchange rate. Deterioration of the exchange rate created incentives to substitute fuelwood for imported commercial fuels. In addition, a decline in the exchange rate increased demand for products like tobacco, requiring more fuelwood and increasing pressure on forests. A more complete set of accounts would have included stock accounts for forests and land use, tying the demand for fuelwood direct to the supply, but this information was not available.

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 forestland as a carbon sink compared to its value under alternative uses. Peck and Descargues (1997) reviewed a range of energy policies that could be considered in Europe and their potential impact on forests. The authors found that energy policy would not, by itself, have a major impact on forests. However, when policies to mitigate carbon emissions from fossil fuel were considered, they found a positive impact on forests. This study did not make use of forest accounts, but represents the kind of study that could make use of such accounts.

One of the major applications of SEEA in Europe has been in analyzing green taxes—especially carbon taxes, but also taxes on other air pollutants. These models, usually multisectoral computable general equilibrium (CGE) models, use the energy and pollution accounts of SEEA. Typically, these models 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.

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. Analysis of forest and land accounts in tropical countries have estimated corresponding reservation prices—the minimum payment the country would be willing to accept to use forestland 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.

Environmental accounts have been used in developed countries, especially Europe, mainly to analyze issues related to pollution and environmental taxes. Economy-wide studies of forestry are largely restricted to traditional multiplier analyses that show the employment and income generated by forestry. Two studies in developing countries have explicitly combined environmental accounts with economic models to address cross-sectoral policy linkages to forestry, one for the Philippines and one for Indonesia.

Philippines: environmental-economic modelling and forestry

In the early 1980s the Philippines experienced a debt crisis and the World Bank and IMF stepped in with stabilization and structural adjustment programmes. Stabilization programmes are short term in order to address macroeconomic imbalances such as unmanageable balance of payments deficits. They usually reduce government expenditures considerably, shift resources into the production of internationally tradable goods and introduce measures to refinance debt. Structural adjustment programmes (SAP) have a longer-term objective of restoring sustainable economic development, often through the promotion of economic liberalization that targets exchange rate and trade policies, the size and composition of government expenditures and the extent of government control over the economy. The previous discussion of underlying causes of deforestation noted that such programmes might create incentives for more intensive, unsustainable exploitation of forests and other natural resources, which would be exported in order to pay off the debt or at least interest on the debt.

There have been many analyses of the economic impacts of stabilization and SAP, but a purely economic model cannot inform policy-makers about the impact on the environment. Similarly, there have been numerous studies of the changing condition of the Philippines forests, but they have not been linked to the impacts of macroeconomic policy changes throughout the economy.

Cruz and Repetto (1992) examined the impacts of structural adjustment in the Philippines using an environmental-economic model to simulate the impact of the actual policies of SAP and alternative policies that could have been undertaken by SAP. The authors constructed a multi-sectoral CGE model of the economy and combined it with environmental accounts and a population migration model. They point out the need to link the CGE model of the economy with environmental accounts in order to analyze how economic changes result in changes in forestry and land use, energy use, generation of pollution and demand for other natural resources. The forest and land accounts were disaggregated by geographic area as well as ecological characteristics such as type of forest and agricultural potential. This was one of the first attempts in developing countries to create a framework that made use of both economic accounts and the environmental accounts for policy analysis.  

Their analysis provided quite detailed results regarding the impact of SAP on the environment. Regarding forests, there was initially concern that SAP would encourage increased exploitation of forests; in fact, output from forests declined, partly due to the collapse of the domestic economy and domestic demand for forest products, but also in response to falling world market prices. Despite declining timber production, deforestation increased because of land clearing by impoverished households. While migration of poor people to forestlands as shifting cultivators trying to earn a subsistence livelihood was already occurring, the increased unemployment and poverty that resulted from SAP accelerated migration and the resulting deforestation. The environmental-economic model also showed that the negative impact of SAP could have been reduced if environmental concerns had been incorporated in SAP and safeguards put in place to protect forests and other resources. While their results may be disputed, the researchers did demonstrate the usefulness of such a model to understand this complex issue.

Indonesia: environmental-economic modelling and forestry

To assess the environmental implications of Indonesia's second long-term development plan (1994‑2018), an environmental-economic model was constructed by integrating environmental accounts (land, forests, water, energy, pollution) with a multi-sector, dynamic input‑output model (Hamilton, 1997; Lange, 1997). Land and forest accounts were disaggregated by geographic region and agricultural potential. Conflict over resource use and the deterioration of the environment required evaluations of trade-offs between economic growth and potentially serious degradation of the natural resource base, especially forests. The study assessed the demands of the country's development plans on the natural resource base and identified the kinds of technological and policy changes that might make it possible to achieve the development objectives given the environmental constraints.

In the late 1980s and early 1990s, much of the concern over deforestation in Indonesia had focused on excessive logging of natural forests for timber exports and, to a lesser extent, the clearing of forests by slash-and-burn cultivators. However, analysis revealed that a large and growing share of timber products was used domestically in manufacturing and construction. Promotion of rapid macroeconomic growth combined with plans to develop a large paper and pulp industry would increase demand for wood products and decimate Indonesia’s forests, even with strict controls over timber exports. At the same time, the plan to maintain food self‑sufficiency would require substantial increases in land for farming, which could further increase pressure on forests.

The analysis found that development objectives could be met only if there were substantial changes both in the forest sector and other sectors of the economy, as well as careful land use planning. The required changes included increased efficiency of timber harvesting and wood processing, increased efficiency of wood use in the construction industry, pricing policy reforms, but most importantly, an expansion of land under forest plantations to reduce pressure on natural forests. This last requirement brought the needs of sustainable forestry in conflict with agriculture. Detailed land accounts indicated that if forest plantations were to expand only in degraded areas not suitable for agriculture it would still be possible to meet many agricultural objectives.

2.7 Summary and comments

This section has reviewed how forest accounts have been implemented in different countries and how they have been used. 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 benefit of sustainable forestry and what would be lost from deforestation. In particular, the accounts were able to identify the forest values that are attributed to other sectors (like agriculture and tourism) or totally omitted from conventional national economic accounts. This information can be useful in cost-benefit analyses to assess the economic benefits and trade-offs from alternative uses of forests.

However, from the examples reviewed here, coverage of forest values is incomplete and varies widely among countries. The most comprehensive forest accounts have been constructed under the Eurostat pilot programme, but even these did not attempt to measure the value of forest environmental services, except for carbon storage. There is still a great deal of work to be done.

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 by macroeconomic policies. 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 2.1, only developed countries compiled such detailed accounts on a regular basis. The developed countries have used parts of their environmental accounts for policy modelling, and supply and use tables for timber and wood products are used in modelling. But the broader forest accounts supply and use tables have not been used much.

Two countries, Philippines and Indonesia, have used environmental accounts to examine cross‑sectoral policy impacts on forestry. Although events have largely overtaken both of these countries since the time of the studies, they illustrate the kind of analytical framework that can be developed from forest accounts and the broader accounts of SEEA.