In order to compare and combine data from different sources, it is important to use statistics that are collected using comparable terminology, definitions and measurement units. The following text describes the definition of the forestry sector used in this study, explains some important issues related to the measurement of employment, value-added and trade and presents the regions used in the study.
There is no commonly agreed definition of the forestry sector. Ideally, the sector should be defined to include all economic activities that mostly depend on the production of goods and services from forests. This would include commercial activities that are dependent on the production of wood fibre (i.e. production of industrial roundwood, woodfuel and charcoal; sawnwood and wood based panels; pulp and paper; and wooden furniture). It would also include activities such as the commercial production and processing of non-wood forest products and the subsistence use of forest products. It could even include economic activities related to production of forest services (although it would be difficult to determine exactly which activities are really dependent on forest services).
The text above gives a broad and inclusive definition of the forestry sector, but a narrower and less precise definition of the sector had to be used in this study. This was due to limitations such as the availability of published statistics and the definitions already used in the compilation of those statistics.
Employment and value-added. In most countries, statistics on employment and value-added are two important components of a broad range of macroeconomic statistics that are collected to monitor trends in the economy. These statistics are usually collected by national statistical offices (rather than sectoral agencies such as forestry administrations), so they are classified using classification systems that are structured according to the type of economic activity rather than the type of product produced by each sector.
Some countries use their own macroeconomic classification systems (see below), but many use the International Standard Industrial Classification of all Economic Activities (ISIC). This classification system is maintained by the United Nations (UN) and is periodically revised by the UN with the agreement of member nations.
The version of the ISIC used for most of the period 1990 to 2000 (ISIC Revision 3.0) includes three sub-sectors (called Divisions in the ISIC) that clearly fall within the definition of the forestry sector given above. These are:
• ISIC Division 02 (forestry, logging and related service activities);
• ISIC Division 20 (manufacture of wood and of products of wood and cork, except furniture; manufacture of articles of straw and plaiting materials); and
• ISIC Division 21 (manufacture of paper and paper products).
Box 1 The location of forest dependent activities in the ISIC Revision 3.0
The ISIC Revision 3.0 is divided into four levels of detail, called; Sections; Divisions; Groups; and Classes. The fundamental level of detail in ISIC Revision 3.0 is the Division, which is represented by a 2-digit code. Sections are used to group together similar activities (defined at the level of the Division) and are identified by a letter. They can be used to present information at broader level of detail. Below the Division level, ISIC Revision 3.0 is further sub-divided into Groups (3 digit code) and Classes (4-digit code). Forestry appears as a separate economic activity at the Division level in ISIC Revision 3.0 and forms part of Section A (agriculture, hunting and forestry). However, different forestry activities are not identified separately at higher digit levels (i.e. Groups and Classes). In other words, the forestry Division contains only one Group and only one Class. Forest processing activities are located under Section D (manufacturing), which contains a number of manufacturing sectors. Division 20 includes the production of sawnwood and wood based panels, plus the production of a number of further processed solid wood products. Production of pulp and paper is covered by Division 21, which also includes some further processing activities. Both of these Divisions are sub-divided into Groups and/or Classes. In addition, some charcoal production is part of Class 2411 and the production of wooden furniture is part of Class 3610. The location of all forest dependent activities within ISIC Revision 3.0 is shown below. (Note: this only shows part of the ISIC structure; there are many other activities not displayed under each of the Sections, Divisions and Groups shown here). SECTION A: AGRICULTURE, HUNTING AND FORESTRY SECTION D: MANUFACTURING |
Source: UN (2003a)
In addition to these three Divisions forest dependent activities also appear as part of other sub-sectors (i.e. Classes) in the ISIC (see Box 1). However, it is not possible to distinguish between forest dependent activities and other activities within these sub-sectors. Therefore, for the statistics on employment and value-added presented later on in this report, the forestry sector was simply defined as the sum of ISIC Divisions 02, 20 and 21.3 (For brevity, these three sub-sectors or Divisions are referred to as “forestry”, the “wood industry” and the “pulp and paper industry” hereafter, while the combination of all three sub-sectors is referred to as the “forestry sector”).
Box 2 The definition of “forestry, logging and related service activities” given in ISIC Revision 3.0
Class 0200 (forestry, logging and related service activities) includes the following activities: |
Source: UN (2003a)
In addition to the ISIC, some countries also use their own national (or regional) classifications of economic activities (e.g. the North American Industry Classification System or NAICS and the Statistical Classification of Economic Activities in the European Community or NACE). These classification systems are slightly different to the ISIC in terms of their structure and their definitions of each sector (for an example of one of the definitions used in ISIC, see Box 2). However, the differences are generally quite small and protocols exist for translating statistics from these classification systems to the ISIC structure. Furthermore, comparison of the different classifications showed that there is very little difference in the definitions of forestry sector activities.
A more significant problem identified in the data was changes in the statistics reported by countries after the change in the ISIC (from Revision 2.0 to Revision 3.0) in the early-1990s.4 This appeared to result in inconsistency in the data series for several countries. For example, in Belgium and Italy, employment and value-added statistics for the wood industry changed by more than 50 percent in one year, possibly due to the change from ISIC Revision 2.0 to ISIC Revision 3.0. Minor problems were also noted with some of the revisions to national classification systems.
Wherever possible, this study presents statistics that have been published using the ISIC Revision 3.0. However, in some cases, statistics collected using the ISIC Revision 3.0 were only available for a few years, compared to a complete set of statistics (i.e. for the whole of the period 1990 to 2000) collected using a national classification system. In such cases (e.g. Canada), the latter statistics have been presented to ensure consistency and comparability over time. As noted above, these national classification systems are generally very similar to the ISIC, so it is believed that these statistics are broadly comparable with the rest of the dataset.
While collecting all of these statistics, another problem that was encountered was the question of if (and if so, where) subsistence production is placed in the ISIC. The most recent revision to the ISIC (ISIC Revision 3.1 - issued in 2004) places the gathering of woodfuel and non-wood forest products for own use under a new Division 96 (Undifferentiated goods-producing activities of private households for own use). However, the previous version of the ISIC (Revision 3.0) is not clear about whether subsistence production should be included in the activities listed under other Divisions. It seems likely that ISIC Divisions 02, 20 and 21 do not include subsistence production and, therefore, that employment and value-added for the sector as a whole is underestimated in statistics published using the ISIC system.
A similar problem concerns employment and value-added along the marketing chain. For example, in many developing countries with a large informal sector (e.g. countries in Africa), it is quite common for large numbers of traders to be involved in buying, transporting and re-selling woodfuel and non-wood forest products. Indeed, this is often where most of the value-added is obtained in the informal sector. These activities could be considered as forest dependent activities but, strictly speaking, transportation and trade should be covered in other sections of the ISIC.
The previous text has highlighted some of the complexity of the issues concerning the definition of the forestry sector and the difficulties of identifying the sector within the most common classification systems currently used to monitor economic activities. The approach taken in this study was to use a fairly simple and restrictive definition of the sector. Therefore, it should be noted that the figures presented later on are probably a minimum estimate of the importance of the sector, particularly in countries where informal forestry sector activities are significant.
Trade. The forest products statistics compiled by FAO include the following: roundwood and charcoal; solid wood products (sawnwood and wood based panels); and pulp and paper products. Statistics are available for the value of international trade (imports and exports) in these products for every country and territory in the World over the period 1990 and 2000. In addition, FAO already publishes the total value of trade in these products in a separate table (Forest Products Trade) in the Forest Products Yearbook (FAO, 2003c). This total was used as the measure of the total value of forest products trade in this study.
It should be noted that the industries that produce these products do not exactly match the definitions used in the ISIC. Firstly, ISIC Divisions 20 and 21 include the production of further processed wood and paper products (e.g. builders’ joinery and cardboard boxes), but these are not included in the FAOSTAT trade statistics. Secondly, as noted above, charcoal production (outside of the forest) is one of a number of activities under Class 2411 in ISIC.5 The overwhelming majority of global charcoal production occurs in developing countries, where it is nearly all produced in the forest, but in industrialised countries a significant proportion of charcoal may be produced in factories (and, therefore, would fall under ISIC Class 2411). A third difference arises because the FAOSTAT trade statistics include trade in recovered paper, which is not included as an activity of the pulp and paper industry (ISIC Division 21).
These differences between the definition of the forestry sector used in the employment and value-added statistics and the definition used for forest products trade are quite small, but it should be noted that the trade statistics use a generally more restrictive definition of the sector (i.e. they exclude the value of trade in further processed products).
Employment. Official forestry sector employment statistics rarely define the measurement units used. There is a lot of part-time employment and seasonal employment in forests. In addition, informal employment and subsistence activities are significant in the forestry sector in developing countries. The latter is often missing from official employment statistics and some of the published statistics collected for this study gave figures that seemed very high (suggesting that some of this employment was part-time).
Problems such as these make it very difficult to compare between countries. Therefore, wherever possible, employment statistics measured in full-time equivalents were collected (or published statistics were converted to full-time equivalents if that was possible). The problem of employment in the informal and subsistence sectors could not be addressed in the study.
The employment figures presented in this report should be treated as figures for the “visible” or “formal” employment in the sector. For developed countries, this is likely to include all of the activities included in ISIC Divisions 02, 20 and 21. For developing countries, it is likely to include activities in ISIC Divisions 20 and 21 (wood industry and the pulp and paper industry) plus employment in industrial roundwood production (i.e. only part of ISIC Division 02, excluding woodfuel and charcoal production activities that are probably not captured in official employment statistics).
These figures are likely to be a vast underestimate of the true levels of employment in the forestry sector. For example, according to the International Labour Organization (ILO, 2001), formal employment only accounts for about one-third of the total employment in the forestry sector.
This study also presents figures about the importance of the forestry sector to total employment. This is measured as the proportion (in percent) of the total economically active population employed in the forestry sector. The total economically active population6 refers to the number of all employed and unemployed persons (including those seeking work for the first time). It covers employers, self-employed workers, salaried employees, wage earners, unpaid workers assisting in a family farm or business operation, members of producers' cooperatives and members of the armed forces (FAO, 2003a).
Value-added. Value-added is a measure of the contribution to GDP made by an individual producer, industry or sector (UN, 2002a). The sum of value-added across all sectors is roughly the same as total GDP (indeed, calculating and adding together the value-added in all individual sectors is one methodology used to calculate total GDP - see Box 3).
Box 3 The calculation of value-added and national income accounting
The System of National Accounts (SNA) consists of a coherent, consistent and integrated set of macroeconomic accounts, balance sheets and tables. It is based on a set of internationally agreed concepts, definitions, classifications and accounting rules. The system also defines some major statistics that are widely used as indicators of economic activity (e.g. Gross Domestic Product or GDP). Because of the multitude of different statistical practices used in countries, it is difficult to describe precisely how countries produce their accounts. However, in general, the methodologies used by countries can be grouped into the following three broad categories:
The simplest approaches to national accounting aim to estimate total GDP and the disaggregation of GDP by:
Most often, GDP is estimated by the first approach above - the production approach. With some minor adjustments, total GDP equals the sum of the value-added generated by all producers. Value-added is calculated for each sector and then added together to give the total. This method normally uses the first digit level of the ISIC - including categories such as agriculture, mining, manufacturing, trade, transport. The second approach - the expenditure approach - works from the perspective of the final users of a country's outputs. GDP is then calculated as the total of private and government consumption, capital formation and net exports. GDP can also be calculated from the point of view of the costs incurred by producers - the cost or income approach. With this approach, GDP is equal to the sum of employee compensation (e.g. wages and salaries), taxes on production and imports, consumption of fixed capital and the operating surplus. Countries following these simple approaches do not necessarily produce GDP estimates using all three methods, but often estimate GDP using only one or two approaches. In most cases, the income approach is missing as it is generally regarded to be the most difficult to implement. |
Source: UN (1999 and 2003a).
At the simplest level, the value-added in a sector is defined as the value of output less the value of intermediate consumption. Alternatively, this can also be expressed as the total value of sales from that sector less the total value of purchases of that sector from other sectors.7 However, the calculation of value-added is slightly more complicated than suggested above, because value-added can be measured in a variety of different ways.
First, there are differences between gross and net value-added (depending on how the consumption or depreciation of capital assets is included in the calculation). Secondly, value-added can be measured at factor costs, basic prices, producer prices or market prices. The differences between these different measures depend on whether taxes and subsidies are included in the calculations.
In many countries, GDP is estimated using the production approach (see Box 3), so it was quite easy to obtain statistics for value-added in the forestry sector for countries that had published national income accounts. However, for a number of countries (especially developing countries) the availability of national income accounts was limited or the only statistics available were quite old. In such cases, a variety of methods were used to produce estimates to complete the dataset (see below).
Most national income accounts presented statistics for each sector that showed the gross value-added at basic prices or factor cost. Value-added at basic prices or factor cost are quite similar, so data collection focused on obtaining statistics for value-added calculated by these methods rather than statistics calculated using alternative measures of value-added. In a few cases, statistics were only available for value-added calculated in a different way, so there is some inconsistency in the dataset. A more significant problem was the large number of countries that presented value-added statistics without any mention of whether they were calculated at factor cost, basic prices, market prices, etc. This is another (possibly significant) source of inconsistency in the dataset.
The study also presents figures that show the contribution of the forestry sector to national economies. The measure used to show this was the gross value-added in the forestry sector, expressed as a proportion (in percent) of the total gross value-added at factor cost (i.e. the gross value-added of all sectors added together). Total gross value-added is roughly the same as total GDP, but it should be noted that published GDP statistics are often reported at market prices.8
Trade. Internationally agreed definitions of forest products already exist and are presented in the Forest Products Yearbook (for further details, see: FAO, 2003c). FAO statistics record the value of imports and exports of the following products and product aggregates: roundwood; wood charcoal; wood chips and particles; wood residues; sawnwood; wood-based panels; wood pulp; other fibre pulp; paper and paperboard; and recovered paper. Total trade of the forestry sector was defined as total trade in all of these products.
At the level of the three sub-sectors, products from the forestry sub-sector (i.e. ISIC Division 02) were defined as roundwood and charcoal; wood industry products were defined as wood chips and particles, wood residues, sawnwood and wood-based panels; and pulp and paper industry products were defined as wood pulp, other fibre pulp, paper and paperboard and recovered paper.
As noted above, the definition of the forestry sector used with respect to the trade statistics is slightly different to definition used for the statistics for employment and value-added in the sector (e.g. the trade statistics do not include further processed products manufactured by the wood industry and pulp and paper industry and included under ISIC Classes 2022, 2023, 2029, 2109 - see Box 1). In addition, the trade statistics do not include trade in wooden furniture (part of ISIC Class 3610) or trade in non-wood forest products (but they do include trade in recovered paper). They are, therefore, an underestimate of the total value of international trade in products manufactured in the forestry sector.
In the report, the contribution of the forestry sector to trade is measured as the share of forest products trade in total merchandise trade (exports and imports). Merchandise trade includes the value of trade in all goods provided (exports) or received (imports) by a country from the rest of the world (World Bank, 2002). It should be noted that this excludes trade in services.
Measurement of value. The value-added and trade statistics are all reported in United States Dollars (USD) at the price levels and exchange rates prevailing in the year 2000. Most of the original statistics were reported in a variety of currency units and were not adjusted for inflation. Therefore, they were converted to this common measurement unit in three stages.
First, where necessary, statistics were converted to local currency units at the exchange rates prevailing in each year. Secondly, the statistics for each country were then converted to the price levels prevailing in each country in the year 2000 (i.e. they were adjusted for inflation). GDP deflators for each individual country were used for this purpose. Finally, the statistics for each country (measured in local currency at year 2000 price levels) were converted to USD using the exchange rate in 2000.
The structure and importance of the forestry sector is likely to be very different in different parts of the World. Therefore, the following analysis presents figures at the regional level in addition to figures for individual countries and the World as a whole. For this purpose, the World was divided into the following eight geographical regions:
• Africa (all African countries);
• West and Central Asia (Afghanistan, Armenia, Azerbaijan, Bahrain, Cyprus, Gaza Strip, Georgia, Iran, Iraq, Israel, Jordan, Kazakhstan, Kuwait, Kyrgyzstan, Lebanon, Oman, Qatar, Saudi Arabia, Syria, Tajikistan, Turkey, Turkmenistan, United Arab Emirates, Uzbekistan, West Bank and Yemen,);
• Developed Asia-Pacific (Australia, Japan and New Zealand);
• Developing Asia-Pacific (all countries in Asia and Oceania, other than those listed above);
• Western Europe (West European countries);
• Eastern Europe (Eastern European countries, including: Belarus; Moldova; Russian Federation; and Ukraine);
• North America (Bermuda, Canada and United States of America); and
• Latin America and the Caribbean.
Data were obtained from a variety of sources, including information from international organisations (e.g. UN, FAO, World Bank, United Nations Industrial Development Organization (UNIDO) and the ILO), plus information from more than 200 national statistical reports and other literature (see Annex 2).
Forestry employment statistics were collected from national statistical yearbooks, other national reports, Poschen (1997) and a recent ILO report for the European Forest Sector Outlook Study (ILO, 2003). Employment statistics for the wood industry and pulp and paper industry were taken from the UNIDO Industrial Statistics Database (UNIDO, 2003b), national statistics and other reports. Statistics for the total labour force were taken from the FAOSTAT database (FAO, 2003a).
Value-added statistics for forestry (ISIC Division 02) were obtained from the UN national accounts yearbook (UN, 2002a and previous). Statistics for value-added in the wood industry and pulp and paper industry were mostly obtained from the UNIDO database and supporting publications. All of this information was supplemented with statistics from various reports published by national statistical services.
All of the forestry sector trade data was taken from the FAOSTAT database, where the total value of exports and imports are presented in current USD (i.e. the figures are not adjusted for inflation).
Macroeconomic statistics for each country were obtained from an unpublished FAO database, which includes total GDP, exchange rates and GDP deflators for every country in the World from 1970 to 2000. This database was originally derived from the World Bank STARS database, which was updated as part of this exercise with more recent figures from the World Bank (2002), UN (2003a) and International Monetary Fund (IMF, 2003). Statistics for the total value of merchandise trade were obtained from the World Bank website.
The following analysis contains other measures derived from these basic statistics (e.g. value-added per cubic metre of production) and forestry statistics were also used to estimate missing data. Most of the forest and forest products statistics used in this analysis were taken from FAO databases. For example, statistics for the production, trade and apparent consumption of forest products were taken from the FAOSTAT statistical database and forest cover statistics were taken from the main report of the Global Forest Resource Assessment (FAO, 2001). Information on forest product prices was used in some places and this was obtained from FAO (2001, 1995) and various national publications.
The aim of this report was to cover all of the 225 countries and territories in the World. For the macroeconomic statistics required to show the contribution of the sector to national economies (e.g. total labour force, total gross value-added, total merchandise trade), published official statistics could be obtained for almost every one of these countries and territories. However, in a few cases (12 countries and territories) it was necessary to estimate figures or use unofficial sources of information.9
At the level of the forestry sector and the three sub-sectors (forestry, wood industry and pulp and paper industry), the availability of statistics was much more limited. FAOSTAT contains trade statistics for every country and territory in the World, but published statistics about employment and value-added in the forestry sector were not available for a large number of countries.
Apart from countries where no statistics were available, the main problem encountered was that many countries do not publish their statistics in sufficient detail. Therefore, for example, statistics about value-added are often published at the level of the ISIC Section (e.g. Section A: Agriculture, hunting and forestry) and are not shown for individual Divisions (e.g. Division 02). In other cases, some countries publish statistics for non-standard combinations of sub-sectors (e.g. giving value-added in the paper and printing industries together). In the case of value-added statistics, another problem was that some countries only publish statistics that show the value of gross output (although this can be used as a basis for estimating the value-added in the sector).
The following text presents an overview of the availability of published statistics on employment and value-added, which can be used to judge the reliability of the global and regional totals presented later on.
The availability of official employment statistics for the forestry sector is presented in Table 1 and Table 2. Table 1 clearly shows that the availability of statistics is most limited for the forestry sub-sector (ISIC Division 02), where published statistics were available for only 41 countries each year (on average) out of the total of 225 countries and territories included in the study.
One of the main reasons for the small number of statistics for the forestry sub-sector is that there is no international database containing all of these statistics. The ILO has published a few studies on this topic, but much of the information presented here was collected from national statistical reports or reports about the forestry sector commissioned by FAO.
Table 1 Total number of official employment statistics collected for each year
|
Year |
Forestry sector |
Furniture industry | ||
Forestry |
Wood industry |
Pulp and paper industry | ||
1990 |
39 |
117 |
104 |
97 |
1991 |
27 |
111 |
102 |
98 |
1992 |
28 |
111 |
102 |
95 |
1993 |
33 |
114 |
101 |
100 |
1994 |
55 |
111 |
100 |
97 |
1995 |
54 |
105 |
95 |
90 |
1996 |
40 |
112 |
101 |
83 |
1997 |
42 |
106 |
95 |
84 |
1998 |
42 |
106 |
100 |
89 |
1999 |
42 |
98 |
93 |
81 |
2000 |
52 |
83 |
78 |
65 |
Average (1990-2000) |
41 |
107 |
97 |
89 |
Note: the figures represent the number of countries covered by published official statistics for each sub-sector in each year. For example, the table shows that official forestry employment statistics for the year 1990 were obtained for 39 countries.
The availability of employment statistics for the wood industry and pulp and paper industry was much better, with published statistics for employment in the wood industry for 107 countries each year (on average) and statistics for pulp and paper industry employment for 97 countries.
The greater amount of employment statistics for these two sub-sectors is due to the existence of the UNIDO industrial database and supporting publications. These references were the source of about 90 percent of the published statistics (although it should be noted that some of the UNIDO statistics - for about 10 to 20 countries each year - are UNIDO estimates rather than figures reported by countries).
The availability of official employment statistics remained quite stable over time. For the forestry sub-sector, the number of figures increased slightly in the late-1990s, due to recent FAO support for improved data collection in countries. In contrast, the availability of statistics for the other two sub-sectors declined slightly over the period. This is, perhaps, due to problems associated with the changes in classification systems from ISIC Revision 2 to ISIC Revision 3 during the 1990s.
Table 2 Average number of employment statistics collected for each year (over the period 1990 to 2000) by region
|
Region |
Forestry sector |
Furniture industry | ||
Forestry |
Wood industry |
Pulp and paper industry | ||
Africa |
4.1 |
19.1 |
12.1 |
15.3 |
West and Central Asia |
3.6 |
15.0 |
14.2 |
10.0 |
Developed Asia-Pacific |
1.5 |
3.0 |
2.5 |
3.0 |
Developing Asia-Pacific |
2.7 |
15.7 |
14.6 |
13.8 |
Western Europe |
12.3 |
16.9 |
16.9 |
15.9 |
Eastern Europe |
12.5 |
16.1 |
17.2 |
12.5 |
North America |
1.8 |
2.9 |
1.9 |
2.9 |
Latin America and the Caribbean |
2.7 |
18.0 |
18.0 |
15.5 |
World |
41.3 |
106.7 |
97.4 |
89.0 |
Note: the figures represent the average number of countries covered by published official statistics for each sub-sector in each region. For example, the table shows that official forestry employment statistics were obtained for 4.1 countries (on average) in Africa over the whole of the period 1990 to 2000.
The availability of employment statistics was also very different by region (see Table 2). As might be expected, most published statistics were available for developed regions (North America, Western Europe and the Developed Asia-Pacific region) and parts of Eastern Europe. It was much more difficult to obtain published statistics for many countries in Africa, Latin America and the Caribbean and the rest of Asia.
Although the total number of countries covered by the statistics collected for this study seems quite small (e.g. less than 50 percent of all countries and territories in the World), it should be noted that official statistics were available for most of the countries with significant forest areas and significant levels of forest product production. For example, the countries where official employment statistics were obtained covered the following:
• Forestry sub-sector: official employment statistics were obtained from countries that account for 89 percent of the global forest area, 98 percent of global industrial roundwood production and 90 percent of global roundwood production (in 2000).
• Wood industry sub-sector: official employment statistics were obtained from countries that account for 99.7 percent of global sawnwood and wood-based panel production (in 2000).
• Pulp and paper industry sub-sector: official employment statistics were obtained from countries that account for 99.9 percent of global pulp and paper production (in 2000).
As the above figures show, the official employment statistics included in the dataset cover a significant proportion of global forestry sector activities (particularly in the pulp and paper industry). Furthermore, forestry sector employment is probably zero (or close to zero) in many of the countries where data could not be obtained (e.g. many of these countries do not report any production of forest products to FAO). Therefore, it is likely that the global and regional totals presented later on are quite reliable (particularly in the case of developed regions).
With respect to the quality of the official employment statistics, it was difficult to judge how accurate or reliable the published official statistics were, because very little information could be obtained about how they were collected. It was suspected that the main problem with official statistics might be the estimation and measurement of employment in the informal forestry sub-sector and in small-scale wood processing operations (where it is often difficult to obtain reliable information).
It was suspected that most countries do not include informal sector activities in their employment statistics. However, one problem that was encountered was where developing countries included estimates of employment in woodfuel collection in their employment statistics for the forestry sub-sector. In almost all cases, these figures were extremely high, suggesting that they had not been adjusted to full-time equivalents.10
A decision was taken to deliberately exclude woodfuel employment figures (wherever possible), in order to get statistics that were probably more comparable across countries (e.g. the woodfuel employment statistics were discarded where they were presented separately or, in other cases, very high figures for total employment were reduced by the proportion of woodfuel production in total roundwood production). Thus, the employment figures presented in this report are for employment in the formal forestry sector.
In general, the availability of value-added statistics for the forestry sub-sector (ISIC Division 02) is better than the availability of employment statistics. For example, statistics were obtained for 85 countries each year (on average) compared with only 41 countries producing statistics for employment in the forestry sub-sector.
A greater number of value-added statistics was collected because many countries produce national income accounts, which are either published as national reports or are included in the UN database of national income accounts. Thus, the availability of statistics was generally higher. However, it was still a problem that many countries do not provide sufficient detail in their statistics to identify forestry activities separately from the total for all activities in agriculture, hunting and forestry.
For the other two sub-sectors (wood industry and pulp and paper industry) the availability of value-added statistics was slightly lower than the availability of employment statistics. The main reason for this was that the UNIDO database and publications contained slightly fewer statistics for value-added compared with the statistics available for employment.
Table 3 Total number of official value-added statistics collected for each year
|
Year |
Forestry sector |
Furniture industry | ||
Forestry |
Wood industry |
Pulp and paper industry | ||
1990 |
72 |
97 |
90 |
80 |
1991 |
75 |
97 |
92 |
85 |
1992 |
83 |
93 |
85 |
79 |
1993 |
89 |
99 |
93 |
86 |
1994 |
93 |
100 |
97 |
85 |
1995 |
97 |
100 |
95 |
83 |
1996 |
98 |
104 |
97 |
84 |
1997 |
92 |
95 |
91 |
79 |
1998 |
86 |
93 |
89 |
79 |
1999 |
75 |
84 |
80 |
72 |
2000 |
75 |
71 |
66 |
58 |
Average (1990-2000) |
85 |
94 |
89 |
79 |
Note: the figures represent the number of countries covered by published official statistics for each sub-sector in each year. For example, the table shows that official statistics for value-added in the forestry sub-sector in the year 1990 were obtained for 72 countries.
The numbers of published official statistics for value-added in each sub-sector are shown by year in Table 3 and by region in Table 4. Table 3 shows that there is no trend in the availability of published statistics over time, although the availability of statistics for the year 2000 was generally much less than average. This is because of the significant amount of time it takes for countries to prepare and publish their national income accounts (i.e. many countries had not published their national income accounts for the year 2000 at the time that the statistics were collected for this study).
Table 4 Average number of value-added statistics collected for each year (over the period 1990 to 2000) by region
|
Region |
Forestry sector |
Furniture industry | ||
Forestry |
Wood industry |
Pulp and paper industry | ||
Africa |
20.2 |
19.2 |
16.8 |
14.0 |
West and Central Asia |
9.5 |
9.8 |
8.9 |
7.4 |
Developed Asia-Pacific |
1.6 |
2.6 |
2.6 |
2.5 |
Developing Asia-Pacific |
14.4 |
15.5 |
14.5 |
14.0 |
Western Europe |
13.9 |
16.5 |
16.3 |
15.2 |
Eastern Europe |
8.7 |
11.3 |
10.6 |
8.8 |
North America |
1.5 |
2.0 |
2.0 |
2.0 |
Latin America and the Caribbean |
15.1 |
17.1 |
16.9 |
15.3 |
World |
84.9 |
93.9 |
88.6 |
79.1 |
Note: the figures represent the average number of countries covered by published official statistics for each sub-sector in each region. For example, the table shows that official statistics for value-added in the forestry sub-sector were obtained for 20.2 countries (on average) in Africa over the whole of the period 1990 to 2000.
Table 4 shows that the availability of published official statistics for value-added in the sector was again generally better for developed countries compared with developing countries. However, the number of developing countries publishing statistics for value-added in the forestry sector was much higher than the number that published employment statistics.
For example, official statistics for value-added in the forestry sub-sector were obtained for 20 countries in Africa (on average each year) compared with only four countries (on average each year) publishing employment statistics for this sub-sector. Similar increases in the availability of statistics were also noted for the other developing regions.
In contrast, the availability of value-added statistics for the forestry sub-sector in European countries was slightly worse than the availability of employment statistics. This was due to the quite large amount of information presented in the recent ILO publication on forestry employment in Europe (ILO, 2003) and the lack of detail presented in the national income accounts published by some countries in Europe.
Again, in order to give a more reasonable assessment of the significance of the countries that produced official statistics, the following should be noted:
• Forestry sub-sector: official value-added statistics were obtained from countries that account for 85 percent of global forest area, 97 percent of global industrial roundwood production and 89 of global roundwood production (in 2000)
• Wood industries sub-sector: official value-added statistics were obtained from countries that account for 99 percent of global sawnwood and wood-based panel production (in 2000).
• Pulp and paper sub-sector: official value-added statistics were obtained from countries that account for 99.9 percent of global pulp and paper production (in 2000).
It was not possible to assess the quality of the official statistics for value-added obtained as part of this study, although it can be expected that similar comments to those above (for employment statistics) would apply to some of the official statistics for value-added in the forestry sector.
As part of the analysis presented later on, the statistics on employment and value-added were converted to a number of indicators of productivity in the three sub-sectors (e.g. production volume per employee in cubic metres (CUM), value-added per CUM of production, etc.). These measures were also used to assess the validity of published official statistics, to make adjustments to these statistics (in some cases) and to estimate missing data.
For example, for the forestry sub-sector (ISIC Division 02), labour productivity was calculated as total roundwood production per employee in North America, Western Europe, Eastern Europe and the Developed Asia-Pacific region and as industrial roundwood production per employee in developing countries (all other regions).
As noted previously, in some cases (e.g. Kenya and Mali), the official forestry employment figures were reduced by the proportion of woodfuel production in total roundwood production, to convert the total employment statistics to figures for employment in formal forestry activities (i.e. industrial roundwood production). This conversion led to estimates of productivity that were comparable to similar countries in the same sub-region. In other cases, official statistics for employment in woodfuel production were deliberately not included in the statistics shown here for employment in the forestry sub-sector. These changes were made to improve comparability across countries, although they result in an underestimation of total employment in the forestry sub-sector in developing countries (i.e. because they exclude informal activities such as woodfuel production).
Similarly, productivity in the wood industry (sawnwood and wood based panel production per employee in CUM) and in the pulp and paper industry (pulp and paper production per employee in metric tonnes or MT) was also calculated to check and validate the published official statistics. The presence of informal activities is likely to be much less of a problem with the statistics in these two sub-sectors. Therefore, wherever possible, potential discrepancies or errors due to mis-classification were identified, checked and corrected in consultation with national experts.
As noted above, a large number of the 225 countries included in this study do not report any production of forest products to FAO. In these cases, it can be safely assumed that employment and value-added in the (formal) forestry sector is probably zero (or close to zero). Therefore, no further action was taken to try to estimate missing data for employment and value-added in the sector in these countries.
However, there were a number of countries with significant levels of production in the forestry sector, but no published official statistics for employment and value-added in the three sub-sectors. In such cases, estimates for employment and value-added were produced for the most important countries to complete the dataset across all countries and all years.
For the estimation of employment in the three sub-sectors, the following methodologies were used (in order of preference):11
E 1. For countries with some official forestry employment statistics (e.g. for a few years), estimates of employment were based on the reported production levels in the years where no employment statistics were available. Thus, for example, productivity per employee (roundwood or industrial roundwood production per employee) was calculated from the available official employment statistics. These figures were then multiplied by the reported production levels in the years without official employment statistics, to produce estimates of employment in those years.
The same process was used to estimate employment in the wood industry and pulp and paper industry sub-sectors in years where no official statistics were available. In these sub-sectors, the measures of productivity used to produce estimates were the amount of solid wood production per employee (i.e. sawnwood and wood based panel production in CUM per employee) or the amount of pulp and paper production per employee (in MT).
E 2. For countries with no official employment statistics for one or more of the sub-sectors over the whole of the period 1990 to 2000, employment was estimated in roughly the same way. However, in these cases, reported production in a country was multiplied by the average productivity levels (in each sub-sector) calculated from the official statistics for countries in the same ecological sub-region.12
For the estimation of value-added in the three sub-sectors, the following methodologies were used (in order of preference):
V 1. For countries with some official value-added statistics and a more complete set of official statistics for the value of gross output, estimates of value-added were based on the gross output statistics (if available) in the years where value-added statistics were not available. This was done by calculating the ratio of value-added to the value of gross output (in the years where both statistics were available) and applying this ratio to the gross output statistics in years where these statistics were available but value-added statistics were missing.
V 2. For countries with some official value-added statistics, but no gross output statistics for the years where value-added statistics were unavailable, the average value-added per unit of output was used with reported production statistics to estimate total value-added. This calculation was similar to that explained in E1 above. For example, the available official statistics were used to calculate value-added per CUM of roundwood or industrial roundwood production (for the forestry sub-sector) and these figures were then multiplied by reported roundwood or industrial roundwood production to estimate total value-added in the forestry sub-sector in the years where official statistics were not available. The same calculations were performed for the wood industry and pulp and paper industry sub-sectors, except that the productivity measures used were value-added per CUM of solid wood production (sawnwood and wood based panels) and value-added per MT of pulp and paper production.
V 3. For countries with no official value-added statistics, value-added in the three sub-sectors was estimated in the same way as above (V2), but using the average value-added per unit of output calculated from official statistics for countries in the same sub-region (i.e. in the same way that employment estimates were produced under method E2 above).
The above text gives a broad outline of the different methodologies used to check and estimate missing data. The application of these methodologies varied from country to country, depending on the amount of official published statistics that were available and the trends observed in the productivity measures calculated from these statistics.
For example, in some cases, average productivity levels (from all of the published statistics) were used to produce estimates for years where statistics were not available. In cases where statistics were not available for one year, the average of the levels of productivity in the preceding and following years was used to produce an estimate for that year. In other cases, trends in productivity levels were observed and were used to produce estimates.
A final point to note is that some countries in the dataset did not exist in the early-1990s (e.g. countries of the former-USSR, Czechoslovakia and Yugoslavia). In the tables presented later in this report, the figures for these countries in earlier years have been estimated by dividing the totals for the original larger countries according to each new country’s estimated share of total production in the larger country.
These figures are probably not very reliable at the country level. However, at the level of the regional analysis, this is not very important. Only the countries of the former-USSR are divided into two of the regions defined and used here and almost all of the forestry sector production in the countries of the former-USSR occurs in the Eastern Europe region.
3 This definition matches the definition of the forest sector approved by the Ministerial Conference on the Protection of Forests in Europe (MCPFE, 2002), which has been used to develop improved Pan-European indicators for sustainable forest management.
4 ISIC Revision 3.0 was issued in 1990, but many countries did not change their statistics until some time after 1990.
5 Charcoal production in the forest is included in ISIC Revision 3.0 under Class 0200 (see Box 2), but the definition of Class 2411 (Manufacture of basic chemicals, except fertilizers and nitrogen compounds) also includes manufacturing of charcoal from the distillation of wood.
6 This is also called the labour force.
7 Note that rents, wages, salaries, dividends and interest payments are not subtracted in the value-added calculation, as these represent the distribution of value-added to owners of land, labour and capital.
8 Total gross value-added at factor cost was used in this calculation so that comparable figures were used to calculate the importance of the sector. Readers may be more familiar with published GDP figures, which are often expressed at market prices. These figures are roughly 10 percent higher than GDP at factor cost, so readers should be aware of this discrepancy.
9 In almost all of these cases these were very small countries, where it is likely that the forestry sector is not very important. Therefore, the reliability of these estimates is unlikely to affect the reliability of the estimates presented at the sub-regional and global levels. This comment also applies to some of the other statistics required for this analysis (e.g. exchange rates and national GDP deflators).
10 For example, in many cases, the employment statistics suggested levels of roundwood production of only one or two cubic metres (CUM) per employee per year.
11 Note: these codes - E1, E2, V1, V2 and V3 - are used in the notes to the country tables presented at the end of this report.
12 Note that the sub-regions used for this purpose were much smaller than the regions presented earlier. Smaller sub-regions were used because countries in these smaller sub-regions are more likely to be similar to each other than the larger groups of countries in the broader regions presented earlier (e.g. Africa was divided into six smaller sub-regions). This was done to improve the validity of taking productivity figures from one country and using it to estimate employment in another country. The sub-regions used were those defined in Table 2 of the State of the World Forests 1999 (FAO, 1999).