Patrick Durst and Thomas Enters
Introduction
Making appropriate and timely decisions is important in managing any resource. To make the right decision requires inputs and information—this is where the problem starts. Frequently, the information that we have is not the information we want. This is exacerbated by the fact that more often than not the information we want is not the information we really need. Finally, by the time we have figured out what we really need to know, the necessary information is not available. As a result, many decisions are based on inaccurate, incomplete or outdated information, and the decisions do not lead to the desired results.
Although the problem described above is common in many sectors, it is of particular concern for forestry because poor decision making can have long-term and wide-ranging effects. Consider reforestation with commercial species in remote (i.e. the wrong) locations, for which at the end of the rotation there is no market. Poor species selection may only become evident several years after planting, when growth rates—for some people although not for all—dramatically decline. As a consequence, investments in wood-processing facilities become subject to risk when the supplies of raw materials were overestimated. The list of examples could be extended easily.
The negative impacts of poor decisions in forestry are not only long-lasting but they also affect an increasing number of people, directly or indirectly. Concerns about climate change and biodiversity loss have made forestry a global issue, although actions are still taken at the local level. Demands for information about particular aspects of forestry can come from any corner of the world, and it is not sufficient anymore to apologize that requested information is not available!
We have known for many years that forestry statistics and information in Asia are inadequate and many efforts have been made at national and international levels to improve the situation. For example, the International Tropical Timber Organization has held more than 10 training workshops on tropical forestry and timber trade statistics since 1990. FAO has devoted time and resources to the issue, of which the EC-FAO Partnership Programme on Information Analysis for Sustainable Forest Management is only one example. The Forest Survey of India assesses the country’s forest resources every two years and the Forestry Department of Peninsular Malaysia has used a Sawmill Information System (SIS) since 1999, which has increased considerably the speed with which data can be validated, analysed and presented according to specific questions.
Certainly, progress has been achieved in some countries. Information is more readily accessible and some forest departments even make it available on their websites, thus accelerating the speed of information flow tremendously. Yet, the suspicion remains that we could have achieved far more, if we had looked more closely at the direct and the underlying causes for poor information on Asian forestry and forests. Too often policy makers, planners and donors assume that the lack of funds, infrastructure and skills are the main barriers to improving forestry information. While it is safe to assume that investments, particularly in capacity building and equipment can make a difference, it is useful at this point to review comprehensively the reasons for the poor state of forestry statistics and information in Asia.
(1) Lack of funds
Data collection, analysis, storage and dissemination are costly. In forestry, data collection is particularly expensive. Forests stretch over large areas; many natural forests are still located in areas that can be classified as remote or inaccessible and the number of operators in the forests and the wood-processing industries in many countries is staggering. Hence, it is not surprising that many forest inventories are as old as 10 years (or more) such as in the Philippines where the last inventory was conducted in 1989. Although aerial photography and remote sensing make it possible to assess forest cover from above it can be quite costly as especially satellite-based inventories require ground-truthing. Also, bird’s-eye-views provide little information on forest operations and wood processing.
Infrastructure for processing, storage and retrieval of data such as geographic information systems (GIS) or global position systems (GPS) have been hailed as breakthroughs for data collection and analysis, but they also come at a high price. Additionally, office computerization needs and costs climb out of reach for many countries.
(2) Inadequate skills and capacities
The advent of information technology in forestry requires new skills that are not always easy to acquire, assuming that the existing staff shows an interest in training and upgrading their skills (see below). In most countries, there are very few people who can manage GIS effectively, and training is no guarantee for strengthening capacities. More often than not, experienced and skilled people are promoted, relocated or leave their jobs altogether in search of higher incomes (e.g. in the private sector).
There is also a lack of suitable trainers conversant in local languages and many forest agencies are unable to provide financial resources for upgrading skills. At the field level, the problem is compounded by low literacy levels in some countries and a lack of understanding as to why data have to be collected and questionnaires have to be completed.
In addition, due to the changes that forestry has experienced and increasing complexity, sustainable forest management (SFM) requires expertise from many disciplines including ecology, silviculture, rural sociology, economics, soil science and engineering (Omoluabi 1998). However, multi-disciplinary expertise is not all that is required. Information needs have also escalated and it is obvious that many forest departments or related institutions have still not replied to the challenge that multi-stakeholder and multi-disciplinary forestry poses.
(3) Resistance to change
Historically, the main objective of forest management was timber production. Although forestry policies and forest management objectives diversified and expanded long before the United Nations Conference on Environment and Development (UNCED), since 1992 forestry has become even more multi-dimensional. Local stakeholders and the international community especially have voiced their concerns about the narrow definition of forestry. They view forests as a source for sustaining local livelihoods and providing local environmental services, as a reservoir of global biodiversity and as carbon sinks that need to be maintained to minimize the perceived negative effects of climate change.
Foresters have been slow to adapt to the expanding demands. Forestry statistics have been even slower in coping with the changes. The main focus has remained on timber, timber products, the wood-processing industries and forest resource assessments in terms of aerial cover. Scant attention is paid to collecting data on forest conditions, the extent of private plantations, non-wood forest products (NWFPs), trees outside forests, fuelwood use, biodiversity, the numbers of forest dwellers and forest-dependent people, degrees of dependencies, recreational visitors to forest areas and the like.
To some extent, this shortcoming can be explained by the lack of funds and capacities (see above). Forest departments have been slow to employ social scientists and ecologists whose skills are required to collect many new data. For example, botanists and ecologists are required to unlock the mysteries of biodiversity, especially related to changes over time. More important than the lack of resources is probably the human tendency to resist change, which Woon et al. (1999, p. 158) also experienced during the introduction of the Peninsular Malaysia Sawmill Information System:
"The human resistance to change issue is often overlooked in many major computerization projects and accounts for the high failure rate of such projects. Thus, the management team overseeing such projects should ensure that adequate on-the-job training be provided."
This apprehension explains why, although the rhetoric on forestry has changed and demands for information have increased, data collection and forestry statistics have remained almost unchanged over the decades.
(4) Inconsistent definitions and methodologies
What is the difference between a reserved forest and a forest? The first is the legal description of the forest area, whether it is covered by trees or not, the second depicts the actual forest area. While this is obvious to some, others are not aware of the difference, which accounts for many misunderstandings.
Solberg et al. (1996, quoted in Brown 2000) noted that:
"Estimating the area of forest plantations presents some challenges. The term ‘plantation’ has varied meanings, and even where a precise definition is available, it is not universally applicable."
Brown (2000, p. 5) concluded that "in many instances, because there is an extensive range of silvicultural practices applied in intensive forest management, the difference between a semi-natural forest and forest plantation is essentially arbitrary" and that many definitions include several ambiguities. As a result, countries such as Finland, Germany and Canada report no forest plantations, while neighbouring countries with seemingly similar forest practices and philosophies report significant plantation areas. What is a "rubber (Hevea brasiliensis) estate" in some countries, is a "forest plantation" in Malaysia. Why not? Malaysian forestry officials argue. There is not much of a difference between an Acacia mangium plantation and a rubber plantation and both produce timber for the furniture industry. What is a Pinus radiata plantation to most people is often referred to as "planted forest" in New Zealand. The list of examples could be extended to other forest-related issues including the use of different measurement units such as cubic meters, tons, or hoppus tons.
Plantation area estimates provide another interesting example. The estimates of plantation areas are conducted in quite different ways. Some countries use seedling production or distribution as a proxy indicator for the area established. Needless to say, some seedlings die before they reach the planting area. Others are planted but do not survive. Gap planting can fix the latter problem, but leads to an artificial inflation of the planting area because the number of seedlings is used as an area indicator. Other countries report the annual planting area without considering mortality or the fact that some former plantation areas are replanted after harvesting. In the latter case, this is no actual expansion of the plantation area. In the first case, it is difficult to assess the difference between reported area and actual or net area. To do this, Pandey (1995) introduced a correction factor, which describes the success rate. In extreme cases, the success rate was as low as 15 percent. This reduced the farm forestry area reported by a social forestry project in Bihar (India) from 104 000 ha to 15 600 ha. The large-scale application of correction factors has been criticized but alternative and better approaches have not been applied widely. As a result, we know very little about forest plantations in Asia, which makes it next to impossible to calculate future wood supplies.
(5) Preponderance of illegal activities
Forestry statistics suffer greatly from widespread illegal activities in the sector. Hence it is not surprising that drastically different forest product data sets exist for some countries. For example, while FAO (2000) indicated that 1997 industrial roundwood production in Cambodia was 1.04 million m3, data from four different sources for the same year ranged from 212 000 to 4.32 million m3 (Castrén, 1999a). For Myanmar, FAO (2000) reported production of 3.44 million m3, while Castrén’s (1999b) estimates were less than 2 million m3. Gintings and Roliadi (in Enters 2001) estimated log production in Indonesia at about 28 million m3. Other official figures ranged from 29.15 to slightly above 40 million m3 (FLB 2000). Barr (2000) calculated production at 55 million m3, and referred to a 1999 study by Scotland and others which estimated production at 82.3 million m3, or nearly three times higher than the official figures. For China, Chen estimated log production in 1997 of around 64 million m3 (in Enters 2001) while FAO (2000) reported a figure of 109 million m3 for the same year. Finally, Thailand’s industrial roundwood production was approximately 2.9 million m3, as published by FAO (2000), compared with only 54 800 m3 (including confiscated timber but apparently excluding wood sourced from plantations), as published by the Royal Forest Department (RFD 2000).
Similar discrepancies can be found in other areas, in particular in trade flow statistics. Frequently import data (from country A to B) do not match export data (from country B to A). While some of the differences are due to errors, others clearly relate to illegal trade. This is not only the case for logs and wood products but also for NWFPs, about which in general very little in known.
(6) Vested interests, impacts on performance appraisals and embarrassing truths
While most of us would like to obtain an accurate picture of forestry in a particular location, there are some people who have a vested interest in concealing the truth or they are—frequently for understandable reasons—too embarrassed to report the truth.
Concealing the truth is in part related to illegal activities such as smuggling, tax evasion and transfer pricing. However, the problem goes much further. Hard evidence is difficult to come by but anecdotes reveal that investors in forest plantations tend to inflate planting areas for which they receive subsidies. In many countries there are no compelling legal instruments that institutionalize proper monitoring. Where they are in place, it is still possible to bribe forestry officials who then accept reported plantation areas, which may be a far cry from the actually planted area. The problem is similar in timber production. Data are transmitted usually by companies to local forest officers who then forward the same to the respective district or provincial forestry officers. Data verification is non-existent or weak.
In the Philippines, as well as many other countries, the performance rating of field officers is based mainly on their reported physical accomplishments versus targets. Therefore it should not be surprising that field officers hesitate to report failures, problems, forest destruction (e.g. forest fires or encroachment) and low "survival" rates. In fact, they are almost "forced" to withhold such information as it would affect their performance appraisal (Austria et al. 1997).
In a similar vein, policy makers, planners and donors are sometimes too embarrassed to face the truth revealing policy, program, project or institutional failures. Deforestation rates are deflated and the success rates of new programs are inflated in the hope of attracting more funds for initiatives that should actually be critically reviewed, modified or stopped. A favourite means of indicating a country’s success in biodiversity conservation is to raise the size of areas protected without any additional budget allocations. While the creation of "paper parks" prevents embarrassment in the short term, in international discussions, ultimately it leads to suspicion on any number.
(7) Complexity of agencies involved in forestry statistics
Usually it is assumed that forestry departments are the collectors of all forestry-related data and disseminators of information. In practically every country this is far from correct and Malaysia serves as an example. The main agencies collecting statistical information on forestry and timber trade are forest departments (forestry statistics) and the Malaysian Timber Industry Board (trade statistics). Others include the Forest Research Institute Malaysia, Ministry of Primary Industries, Ministry of Finance, Ministry of Trade and Industry, Ministry of Industrial Development Sarawak, Statistics Department, Land and Survey Department, Customs and Excise Department, Timber Trade Associations in Peninsular Malaysia, Sabah and Sarawak, Malaysia Centre of Remote Sensing (MACRES), Malaysian Industrial Development Authority (MIDA), Malaysian External Trade Centre (MATRADE), district offices, embassies etc.
This multitude of actors can create high transaction costs and lead to the duplication of efforts. There may be no established channel of communication between customs and forestry departments regarding import of timber and export/import of value-added products such as furniture, as is the case in Nepal. Some actors are reluctant to share information and raw data are protected as secrets. In the absence of mandatory data transfers, information kept at one level or one department cannot be accessed easily unless it is specifically and formally requested, a fr equently frustrating process.
(8) Proprietary business information
What used to be the domain of government departments has been transferred to the private sector to various degrees across the region. New Zealand is probably the most extreme example concerning the privatization of forestry (Clarke 2000). The increasing significance of the private sector translates automatically into a less significant role for public sector institutions. Areas that the public sector used to control, such as plantation development and management, are today in many countries the domain of private investors. As investors they are careful about making data about their operations available and view many issues increasingly as corporate or business secrets, as they feel that they need to maintain their competitive advantage.
As a result the public sector finds it more and more difficult to provide a complete overview of national forestry aspects including even research. Many enterprises refuse outright to complete and return questionnaires and thus developments in particular sectors, such as plantations or the domestication of NWFPs, have become blurred. This is particularly the case where government officials view their role as enforcers of the law and controllers but not as facilitators in development.
(9) Weak understanding of the objectives of data collection
The primary objectives of data collection, analysis and dissemination are to facilitate planning, management and the formulation of policies. Adequate, accurate, timely and relevant information also serves to direct scarce resources to areas of need, thereby minimizing risk and waste of resources. Data collection, their analysis, storage and dissemination are costly. Hence only relevant data should be collected and processed, with the detail and precision appropriate for decision making. Unfortunately, only few people understand the need to prioritize.
Data are often collected for the sake of data collection, or because that is what the job description specifies. Little time is spent on selecting the appropriate level of precision, choice of data-collection method and updating intervals. Survey forms are designed poorly and when they are returned, no one is assigned to code the data for further processing. All efforts are diverted to particular sectors of forestry and as a result other sectors, deemed less important, such as domestic timber markets, NWFPs or trees outside forests, are neglected totally by the institutions responsible for providing accurate and relevant information.
The types of statistics used in many government departments are mainly descriptive, in table form, graphics or pictures. Essentially data remain data and are not translated into information. This is like asking for the balance of a checking account and receiving only a printout of transfers. Making tables available on web pages also does not improve the situation. There is a substantial disregard for the needs of decision makers, planners and the general public, and as long as the translation of data into information is neglected, decisions will remain poor and people will continue to be misinformed.
(10) Data loss
Finally, data is also lost due to computer crashes, office relocations, changes in institutional arrangements, fires and occasional instability arising from survival and competition problems (Omoluabi 1998). While this is probably an infrequent problem it should not be discounted.
Summary
The discussion of the problems faced by forestry statistics in Asia is not intended to leave the reader with the impression that no matter what is tried, the situation cannot be improved. Instead, it is intended to serve as food for thought for further discussions and deliberations on the most appropriate ways to improve the current situation. Recognizing and understanding the underlying problems is often the first step in the right direction. Woon et al. (1999) responded to the challenge posed by resistance to change and actively assisted during the familiarization process when the SIS was introduced. Without the assistance from the program developers, scarce resources would have been wasted.
The scarcity of resources should also remind us that we need to prioritize and to determine the kind and precision of information required to support decision making. We will never be able to collect all the data that we want. Instead, we need to focus on what is required.
Investments in infrastructure should be minimized where data quality is a problem. Maps generated by a GIS might be more impressive than hand-drawn maps. However, if they are based on outdated and inaccurate inputs the quality of the final product is deceiving.
Some of the problems discussed above cannot be tackled directly, such as the preponderance of illegal activities. Other challenges can be dealt with more easily. The use of indicators will remain important. Hence they should be selected carefully. Assuming that seedling production can be translated into plantation area is erroneous. Performance appraisal procedures should also be reviewed. Those who successfully solve a problem but do not reach a predetermined target should not be penalized while those who report unverifiably that the target has been reached are promoted.
Finally, huge gains could probably be made if the one-way flow of communication was changed to a two-way flow. Data collectors have to feel that they are part of the information generation process. They need to feel ownership in the final products by being acknowledged and provided with an understanding of the final uses and value of the final product. Only then will their interest in collecting data increase.
References
Austria, I.V., S.A. Asuncion and M. Guitanguijo. 1997. Country report – Philippines. Final Report: ITTO training workshop on tropical forestry and timber trade statistics. 14 April to 2 May 1997 in Kuala Lumpur, Malaysia. International Tropical Timber Organization, Yokohama, and Malaysian Timber Industry Board, Kuala Lumpur.
Barr, C. 2000. Will HPH reform lead to sustainable forest management? Questioning the assumptions of the "sustainable logging" paradigm in Indonesia. Manuscript.
Brown, C. 2000. The global outlook for future wood supply from forest plantations. Working paper: GFPOS/WP/03. Rome, Food and Agriculture Organization of the United Nations.
Castrén, T. 1999a. Timber trade and wood flow-study Cambodia.
www.mekonginfo.org/mrc-en/doclib.nsf/0/6433AA461EE95499C7256831002D5/FULLTEXT.HTM
Castrén, T. 1999b. Timber trade and wood flow-study Myanmar.
www.mekonginfo.org/mrc-en/doclib.nsf/0/6590A35F0AD9B4B5C7256832000A …/FULLTEXT.HTM
Clarke, M. 2000. Devolving forestry ownership in New Zealand: processes, issues and outcomes. In T. Enters, P.B. Durst and M. Victor, eds. Decentralization and devolution of forest management in Asia and the Pacific. pp. 55-68. RAP Publication 2000/1. Bangkok, Food and Agriculture Organization of the United Nations.
Enters, T. 2001. Trash or treasure? Logging and mill residues in Asia and the Pacific. RAP Publication 2001/16. Bangkok, Food and Agriculture Organization of the United Nations.
FAO. 2000. FAO yearbook forest products (1994-1998). Rome, Food and Agriculture Organization of the United Nations.
FLB. 2000. Forest Liaison Office of the European Union in Jakarta.
www.eu.flb.or.id/statistiks/supply-demand.htm
Omoluabi, A.C. 1998. Methodological guidelines for improvement of forestry statistics in Africa. A report on the development of forestry statistics in Africa – based on AFWC-FAO initiatives.
http://www.fao.org/forestry/FON/FONS/outlook/Africa/Mutare/Table%20of%20Contents.htm
Pandey, D. 1995. Forest resources assessment 1990. Tropical forest plantation resources. FAO Forestry Paper 128. Rome, Food and Agriculture Organization of the United Nations.
RFD. 2000. Forestry statistics of Thailand, 1998. Bangkok, Data Center, Information Office, Royal Forest Department.
Woon, W.C., L.Y. Poh, Maizurah Ishak and Shahira Ishak, 1999. Development of the sawmill information system (SIS) for the forestry sector. The Malaysian Forester 62(3): 145-159.
