Papua New Guinea - The forestry sector : a tropical forestry action plan review

Repod No. 8031-PNG Papua New Guinea The Forestry Sector: A Tropical Forestry Action Plan Review - 1, 1990 Agriculture Operations Division Country Department V Asia Regional Office FOR OFFICIAL USE ONLY Docmen of the World Bak This document has a restricted distribution and may be used by recipients only in the performance of their official duties. Its contents may not otherwise be disclosed without World Bank authorization.  CURRENCY EQUIVALENTS Annual Averages 1981 K 1.00 = US$1.49 1982 1.36 1983 1.20 1984 1.12 1985 1.00 1986 1.03 1987 1.10 1988 1.15 August 1989 1.15 ABBREVIATIONS AND ACRONYMS ADB - Asian Development Bank AIDAB - Australian International Development Assistance Bureau m3 - Cubic meters CSIRO Australian Commonwealth Scientific and Industrial Research Organization DAL - Department of Agriculture and Livestock dbh - Diameter at Breast Height DEC - Department of Environment and Conservation DFP - Department of Finance and Planning DLPP - Department of Lands and Physical Planning DOF - Department of Forests FAO - Food and Agriculture Organization of the United Nations FIA - Forest Industries Association of PNG FIC - Forest Industries Council FRI - Forest Research Institute GOPNG - Government of Papua New Guinea LMP - Land Mobilization Prog-am (or Project) mai - Mean Annual Increment MEP - Minimum Export Price NEC - National Executive Council NFDP - National Forestry Development Progrum NPEP - National Public Expenditure Program PIP - Public Investment Program PNGRIS - PNG Resources Information System SPO - State Purchasing Option TA - Timber Authority TFAP - Tropical Forestry Action Plan UNDP - United Nations Development P:ogramme UNITECH - PNG University of Technology  _ ,1 AL PAPUA NEW GUINEA THE FORESTRY SECTOR: A TROPICAL FORESTRY ACTION PLAN REVIEW Table of Contents Page No. PREFACE ........*... ........... iv EXECUTIVE SUMAiARY . . v I. INTRODUCTION ............................................... 1 II. FORESTRY AND THE PNG ECONOMY . ... 5 A. The Papua New Guinea Economy: Directions and Policy Imperatives .. 5 B. Forestry in the Economy ................................. 6 L. The National Forestry Development Program ............... 11 D. The Institutional Framework .. 13 III. THE FOREST RESOURCE AND SUSTAINABLE MANAGEMENT ............. 15 A. The Natural Forest Resource of PNG . . 15 B. The Sustainable Management of the Resource . .5......... i C. Plantations ... . 22 D. Resource Assessments The Need . . . 27 IV. MAXIMIZING RETURNS TO THE FORESTRY SECTOR .................. 29 A. Government Revenues, Royalty Policy and Rents . . 29 B. Methods of Sale of Wood ... 35 C. Log Exports and the Transfer Pricing Issue ... 36 D. Forest Industries Development .......................... . 42 V. CONSERVATION AND LAND USE STRATEGY . . 51 A. Environmental Controls in Logging . . 51 B. Land Registration, Negotiation and Compensation ......... 52 C. Landowner Assistance and Education .................... #. 52 D. National Land Use Planning Issues ....................... 53 E. The Need for a Conservation Strategy ................... 55 F. World Heritage Possibilities and Sustainable Yield ...... 56 G. A Suggested Role for Non-Government Organizations ....... 57 This document has a restricted distribution and may be used by recipients only in the performance of their official duties. Its contents may not otherwise be disclosed without World Bank authorization.  - ii - VI. MANAGING THE FOREST RESOURCE. .............................. . . .... 59 A. The Present Problem ........................ ... 59 B. Major Constraints to Management . .................. 61 The Policy and Legal Frameworka................ 61 Management Personnelr..o 4 ne. . * * ............ . 63 Intergovernmental Relations. . ...... 64 Landowner-Government Relations. . ...... 66 C. A Strategy for More Effective Forest Management 8...... A N3w Act for Forestry in PNG. . ... 68 Institutional Requirementsu . ..... 69 A New Forest Service. ... .. ... .... ...... .... 71 Re3earch and Development .................. . ........ 73 Human Resources Development.......... ........ 75 VII. AN ACTION PROGRAO G R AM.. .. . ...... 77 A. Major Progrem Componenm. .. . ..... 77 Government Policy Statement. 77 New Legislationg....n .. ..78 Resources and Sustainable Yield . .78 Conservation and Land Use........ . 79 Institutional and Human Resource Development. . 81 Short Term Institutional Support Measures. 83 Further Studies ................ .................. 85 B. Implementing the Action Program ... .86 Interim Management Structure ....... . ... ............ 86 External Assistance--The Case for a Twi .. .87 Phasing of Implementation .......... . .... 88 C. Implications for Donors ........ ..... 89 ANNEXES 1. International and Coastal Freight Rates for PNG 2. Major Timber Species Exported from PNG - 1988 3. Log Export Data: PNG and Major S.E. Asian Suppliers 4. Calculation of Transfer Prices Using Ex Log Pond Japan Prices 5. A Proposed Forestry Actt Discussion and Drafting Instructions 6. Suggested Areas for Conservation/Preservation 7. Terms of Reference and TFAP Mission Members 8. Major Forms of Timber Sale Agreement in PNG 9. Summary of the Minister's Drafting Instructions for a New Forestry Act 10. Suggested Project Proposals BIBLIOGRAPHY  - iii LIST OF TABLES Pare No. 1. Country Data - Papua New Guines..... ................... 4 2. Forestry in the PNG Economy, 1988 .......................... 7 3. PIG Value of Exports by Type of Product .................... 8 4. Japanese Imports of Logs from SE Asia and Pacific Islands, 1965-88 ........... ...0................... . ................. 9 5. PNG: Volume of Major Exports by Type of Product, 1981-88 ... 9 6. Destination of PNG Forest Product Exports, 1984-88 ........ 10 7. National Forestry Development Programt Logging Concessions, 1987-91 ................... 12 8. NFDP and Original Programs. ... . .............. ........ . 12 9. Planting Targets by Year ................................... 13 10. PNG Forested Area, 1985 ..... ..... ..0... .......... 15 11. PNG Natural Forest Area and Grow ing Stock . ...15..* .......... is 12. Plantation Areas, 1979, 1985, 1988 ......................... 23 13. Age Classes of Plantations, 1986 ......8 6..................... 24 14. Mean Annual Increment Data, Major Plantation Species ....... 24 15. Plantation Development Strategy, 1986-2000 ................. 27 16. Sources of PNG Government Forestry Revenue ................. 30 17. PNG Log Export Volumes, Values, Prices .......... ...... 36 18. FOB Log Prices, Sabah, Sarawak and PNG, 1985-89....... 38 19. Estimated Losses from Transfer Pricing, 1985-87, 39 20. Estimated Losses from Transfer Pricing, 1982-85 39 21. Estimated Losses from Transfer Pricing, 1985-88 40 22. Processed Forest Produce Exports, 195 1-88 43 23. Management Personnel vs Operations, 1972173-87 64 Chart 1: Organization of Dep&rtment of Forests .......... 14 Cinart 2: Forest Service Proposed Structure ..... ............... .... 72 Map  - iv - PAPtA NEW GUINEA THE FORESTRY SECTOR: A TROPICAL FORESTRY ACTION PLAN REVIEW Preface 1. The Tropical Forestry Action Plan is an internation.z coordinating mechanism for halting the destruction of tropical forests and promoting their sustainable use for the economic and social benefit of people. There are compelling reasons for TFAP support for action in Papua New Guinea. First, of PNG's total land area of some 46 million ha., about 36 million ha (772) is forested and the economy will have to depend upon the development of forestry and its other renewable resources for sustained economic activity in future. 2. A secontd reason for TFAP support goes beyond the country's own econo- mic interests. PNG's forests are well known for their outstanding biodiver- sity. The country's unusual geology, with spectacular mountain landforms on the main island and volcanic features on the smaller islands, has produced a rich diversity of Malay and Gonwanaland flora and fauna. There are more than 11,000 species of flora, including 2,000 ferns alone, while the lowland rain forests comprise more than 1,200 tree species. Many species of flora and fauna are unique to PNG, and some of these are now threatened. 3. Finally, TFAP support is especially appropriate because forest sector management in PNG is in a state of disarray. Despite the importance of the sector, the country does not now have the means to assess the full extent of its forest resource or its sustainable yield, to control existing logging operations, to reap the full benefits from the present level of proe Lction, or to determine future policies for conservation and industrial development. Yet the official medium term development strategy provides for even heavier logging in the future, with few measures suggested for improved monitoring and supervision, and for seemingly flawed policies to promote local processing. 4. Recognizing both the importance of the forest resource and the gravity of the management problem it faces, the Government of Papua New Guinea (GOPNG) requested the TFAP to review the forest sector and propose an Action Program for the future. In response, in May 1989, the TFAP mounted an inter- agency mission to PNG under the leadership of the World Bank, with participa- tion of the Asian Development Bank, Food and Agriculture Organization, United Nations Development Programme, the International Institute of Environment and Development, the International Union of Conservation and Nature, and the bilateral development assistance agencies of Australia, New Zealand and the Federal Republic of Germany. The mission actively involved government and non-government organizations in its work and was able to benefit greatly from the work of a judicial Commission of Inquiry which was drafting its final report during the mission's visit to PNG. The Prime Minister of PNG has since appointed a task force to report directly to him on the implementation of recommendations on forestry arising from the Commission of Inquiry and the TFAP Review.  v PAPUA NEW GUINEA THE FORESTRY SECTOR: A TROPICAL FORESTRY ACTION PLUN REVIEW Executive Summary 1. This report provides the TFAP's assessment of the status of PNG forest sector and a proposed Action Program for consideration by Government and TFAP member agencies. The report concentrates on six problem areas that warrant the most urgent attention: Maior Problem Areas and Priorities for Action (i) Resource Assessment 2. PNG lacks adequate information on the growth, yield or even the extent of its forest resourre. Thus, while the Constitution of PNG accepts the concept of achieving a sustained yield management syatem, there is not the data base to assess at what level this should be set. The urgent requirement, therefore, is to quickly obtain sufficient information on the resource to allow reasonable prescriptions for sustainable management to be drawn up, with adequate cross referencing to other major land use plans and options included. 3. Proposed Actions. Ultimately PNG will require a complete inventory of its forest resource. However, important decisions on resource use and related policies are required well before a full inventory could be completed. Therefore, the following actions are suggested: (a) A Rapid Resource Appraisal. The report advocates that a preliminary resource assessment should be undertaken and completed in 12 months. It would utilize an existing geographical information system data base, logging and DOF records, field measurement, and possibly some remote sensing imagery, to estimate regeneration. This will require externally funded contracts for updating and amending the data bases and for a long term consultancy in resource estimuation. (b) A Re-inventory of the Resource. The report outlines a four to five year program of activity to complete an inventory. It would rely for stratification and some baseline mapping on output from the rapid resource appraisal, and would therefore be scheduled to commence after that output becomes available. External funding of such an undertaking will be necessary. (ii) Managing the Resource 4. A major breakdown in effective management of the forest resource has occurred due to the lack of an effective decision-making framework to link the Central and Provincial Government and landholders in PNG. It is essential that landholders be adequately compensated and involved in forestry activi:y: it is neither practical, nor appropriate, to locate forestry activity only on alienated land in PNG. The TFAP Review, identified some major constraints to the effective management of PNG's forest resources:  - vi - (a) Some 982 of land in PNG is owned under custom law by clans. There are serious issues of land tenure and rights which need to be addressed: the adequacy of compensation and continuity of such payments; effective registration of tenure; assistance in the area of enhancing landowners' capabilities to negotiate and participate effectively in land use decision making. (b) The very rapid increases in output of logs (from 910,000 m3in 179 to 2,000,000 m- in 1987, and projected in GOPNG plans to reaches 6,000,000 m3 by 1990) have not been matched by increases in the capacity of the central and provincial government forest agencies to monitor and supervise operations. (c) Serious environmental consequences have already occurred in some areas of PNG as a result of uncontrolled logging programs, and more can be expected if no steps are taken to improve control. (d) The problem is more than simply one of staffing levels or funding, although these are clearly important. Essentially, the system of forestry administration in PNG has not evolved successfully following the devolution of central government powers to provinces. (e) At the provincial level, a lack of qualified personnel and equipment is prevalent. Provincial officers, and to a large extent the governments they work for, are placed in the position of having to apply policy set at a different level of government (the nationa, government), and serious problems of lack of coordination, misunderstanding and in many cases direct resentment are present. (f) At the national level, the Department of Forests is operating on a standard line structure, even though the field activity that is usually the object of such a structure is now for the most part in Provincial hands. Some of the worst exploitive practices have occurred when landowners have been allowed to deal directly with logging operators. On the other hand, when Government has been involved, implementation has often been slow and there has been insufficient consultation with landowners. In these cases, environmental damage has also occurred through lack of adequate planning and supervision. 5. Proposed Actions. The problems noted abovf, are sufficiently serious to wazrant a completely new approach to forest administration in PNG. The report advocates a new structure with five major compoLentst (a) A National Forestry Board that would formulate national forestry policies and plans, and advise the Minister on all major policy and project developments. (b) The Board would be serviced by specialized Standing Committees, in such areas as conservation, land use, research, marketing and economics. These Committees would have a funding base determined by levies on forest operations, to ensure adequate coverage of their responsibilities. They would have power to employ and co-opt assistance as required.  - vii - (c) Four Regional Forestry Boards would administer forest development plans, issue timber licenses at the provincial level, and provide the link to landholders through provincial government and non-government organizations membership. The regional boards were seen as a compromise between centralized control of forestry, which is politically unacceptable in PNG, and setting up 19 provincial boards, which would be an administratively unwieldy and probably unfeasibie solution. (d) A new Forest Service, which would service the national and regional boards. Existing provincial forestry officers would be transferred to the service, thus restoring a career structure, improved morale and effective communication of informatior. and instructions. The Service would be required to deal directly with regional boards, thus allowing provinces a measure of control over Forest Service operations. (e) A new financial framework that will create a number of special purpose funds and accounts with guaranteed sources of income *rrom related operations. 6. The first step in implementirg such a reform will be to enact new legislation. The judicial Commission of Inquiry identified inadequacies in the existing legislation as a major cause of over-exploitation and malfeasance in the forestry sector. GOPNG has accepted that existing legislation is inadequate, and at the time of the TFAP mission had dzafted instructions for a ne-w Act to go to Parliament. TFAP supports the intention of GOPNG to produce new legislation quickly, and has offered a number of suggested changes including the proposed new structura, which are discussed at length in this report. 7. The report also suggests that the donor community can assist in the process of re-structuring and improving supervision and monitoring by providing: (a) a minor input to assist with forest policy and legislation; (b) a coordinating input to assist with structuring donors' contributions during the reorganization phase; (c) external experts to join together with national forestry staff in monitoring and management task forces to reorganize harvesting and indigenous forest management, and supervision of logging operations. Involvement of landowners in the management process would be an important element of this activity. (iii) Maximizina Returns from Losaing 8. PNG is presently sustaining serious losses to revenue from poor control over logging and marketing activities, and an inadequate resource policy base. Whatever the level of sustainable yield that is determined for PNG, the country needs to reap the maximum benefits, in terms of foreign exchange and government revenues, from the annual cut. This does not happen  - viii - now. his review confirmed that extensive transfer pricing of exported logs has occurred, and measures adopted in the past to raise export log prices (such as the minimum export price system) have not worked. Nor are some of the new measures being propoaed to raise the export tax level likely to be effective. Equally serious is the fact that losses occur further back in the marketing c in, due to low efficiency in logging (related to the field monitoring and supetvision problem referred to earlier) and to unnecessarily large losses from degrade of logs due to exposure to thi elements, mis- identification of species, poor log yard and loading control, and other problems of an essentially supervisory nature. The report also argues in favor of a royalty system which can capture the full rental for higher valued species (which will not occur under presently proposed legislation), combined with an export tax system based on FOB prices, to take account of variability in average selling prices over time. Further, the report suggests the practice of including indirect levies (such as obligations to construct roads, bridges, schools, etc.) in logging contracts may be inefficient. 9. Proposed Actions. Some of the marketing problems identified will be included under the terms of references for the monitoring task force referred to below. In addition, the following are suggested% (a) GOPNG should consider a comprehensive review of its royalty, export tax and other revenue measures as applied to the log export trade. (b) the State Purchasing Option (a GOPNG acquisition scheme which allows it to sell up to 25Z of an exporter's log shipment if it can improve upon the exporter's prico, and to retain the difference) should be extended so that a greater impact on private sector marketing is possible. The SPO should also be made more autonomous and covmercial, and be given sufficient financial powers to become self- funding. (c) external assistance will be required to bring in a competent marketing firm (on a profit-share contract basis) to raise the level of activity and profitability of the SPO. Funding for equipment and training will also be required. (iv) Industrial Develooment Prospects 10. The forest processing sector in PNG is presently internationally uncompetitive. GOPNG has adopted the policy of encouraging the domestic processing of an increased proportion of its forest products. The report endorses this policy as an obvious means of increasing the benefits of forest resource management to the nation. However, there has been a dramatic decline in local processing since the export of logs was liberalized in 1979, and the report concludes that the savmilling and veneer manufacturing sectors in PNG are uncompetitive internationally. High costs, low productivity and the diverse nature of the resource are the main reasons for this. The future direction of industrial development, and the appropriate incentives framework for encouraging such development, therefore need careful study. 11. The key issue at present is how to encourage viable processing industries. GOPNG has already imposed a ban on the export ef logs o.1 some premium species, and has announced new measures to protect t1he domestic  - ix - market. The Government is also considering introducing a complete ban on log exports for all new concessions and, in time, for the whole sector. Given the real constraints on successful industrial development, the report contends that such a ban would not achieve its objectives and could lead to a serious misallocation of resources. The report argues in favor of supply guarantees, standardized grading and training enhancement, and certain taxation measures, to encourage domestic processing. The report cautions against the adoption of further trade protection measures to favor domestic processing. 12. Proposed Actions. The report cattions against hasty and possibly imprudent actions to stimulate industrial development, and suggests the follovings (i) A careful review of the costs and benefits of introducing a log export ban at this stage. The report suggests a number of alternative measures that could be combined into a suitable incentives package for promoting industry. (ii) Three major industry feasibility studies to determine the viability of (a) an efficiently-scaled integrated sawmill/board manufacturing plant; (b) woodchipping; and (c) a pulp mill. (v) Conservation and Land Use 13. Serious environmental problems as a result oZ uncontrolled logging activity have occurred in PNG. Wider issues of environmental management in the forested areas of PNG are also of concern. Many of the specific environmental problems which have arisen from logging operations in PNG will be better managed under the new legislation and administrative structure proposed in the report. Similarly, some of the critical land use and tenure issues will also be addressed. However, conservation and land use are much broader subjects that need to be tackled on a cross-sectoral basis. It is therefore necessary to devise an action program that goee beyond forest policy and management. 14. This review will deal with the critical land issue by cross referencing into some current activities in the land tenure area in PNG, and by suggesting some specific inputs in the area of landowner training, assessment of the feasibility of establishment of a land use research council, and assessment of mechanisms for titling and compensation. 15. Proposed Actions. This report proposes actions to address conservation and land use issues: (a) A World Heritage Proposal to take in a large part of the upland terrain and some island and coastal areas; (b) An early endeavor to formulate a national conservation strategy, with wide participation of comunuity, government and lndustry interests; (c) A program to rehabilitate existing nation$ctl parks; (d) A program to improve ecological survey and monitoring;  (e) A program to provide training to local leaders in negotiation skills and land use management; and (f) Funds to support the activities of NGOs engaged in forestry and conservation activities. (g) A feasibility study for establishment of a land use research council. 16. The most ambitious of these proposals is for a World Heritage listing. The proposal is that GOPNG might offer a composite package, consisting of a World Heritage declaration over a significant area of its forests, together with an undertaking to manage its other forest resources at a sustainable rate of exploitation. In return, major donors might compensate the country for the revenues lost through adopting a lower production target until PNG's own expanding mineral revenues take up the slack. (vi) Institutional and Human Resource Deve'opment 17. At present, PNG does not possess the appropriate institutional structure, nor an adequate, trained cadre of managers, to design or implement forest management policies. Ultimately, whether or not PNG succeeds in reaping the maximum benefits from its forests, while maintaining the resource for future generations, will depend upon how fast the country can build up its capacity to manage the resource. The report concludes that PNG's present capacity is quite inadequate to provide even minimm supervision of present activities, let alone tackle the complex tasks thE.t lie ahead. 18. Proposed Actions. The report concludes that strengthening its mAnagement capacity will be the most important single challenge facing the sector, which will requIre considerable external assistance. The report suggests the outline of an organiz^tion structure for effective umnagement, but no structure will be effective unless it can be staffed with qualified people. The Action Program therefore proposes three areas for urgent attentions (a) The detailed design and implementation of the proposed new structure of national and regional boards and a single forestry service. (b) The design of a major human resource development program, comprising a phased staffing plan, an indicative training program for national staff, and a complementary strategy for securing external assistance to fill posts for which qualified nationals are not available. This is a major, long term task, which PNG might seek external assistance to belp plan, but which will require indigenous capacity to implement on an ongoing basis. (c) Short-term institutional support in the form of contract technical inputs while indigenous management capacity is being strengthened, in the key areas of forest management, the monitoring of logging and marketing. In each case, the report suggests that external expertise will be needed to supplement national staff to bring about urgently- needed improvements.  - zi - Imnlementina the Proaram 19. Even though the TFAP Review has tried to limit the proposed Action Program to only essential items, it still amounts to a very challenging set of policy and structural changes. The report therefore suggests three mechanisms for managing the process of changes (a) the establishment of a special, interim policy/planning structure within GOPNG; (b) external assistance to support this interim structure; and (c) a careful phasing of implementation. 20. The report suggests a two-tiered structure should be adopted as an interim mechanism for managing the Action Program: (a) A high level Steering Committee, reporting directly to the Minister, that could perform some of the intended functions of the National Board, pending the Board's establishment. It is possible that these functions could be performed by the task force recently set up by the Prime Minister. (b) A small Planning Group within the Department of Forests, composed of senior officials, which would be responsible for preparing and overseeing the implementation of the program. 21. To assist the planning group, the report suggests that GOPNG should seek external assistance for a small, expert team of planners for a minimum of two years. It is suggested that since the task of restructuring and strengthening the forest sector will be a long process, the best solution would be to seek a 'twinning8 relationship with a suitable external technical agency, which could not only field the planning team but also provide advisory support to GOPNG over an extended period. 22. The report has suggested that the program might be divided into ten principal tasks, with the initial implementation split into three phases over the next two years: Phase I - The mission suggested three actions to be implemented before the end of 1989, all of which are now under preparation: 1. Issue a GOPNG policy statement, in response to the TFAP Review, as a basis for the proposed donor meeting; 2. Set up steering committee and planning group in the Department of Forestry as an interim structures 3. Mobilize external assistance for (a) a planning support team in DOF to help manage the process of change and (b) if necessary, for the final drafting of the new Forests Act. Phase II - to be implemented before the end of 1990: 4. Enact the Forests Act; 5. Set up the new organizational structure; 6. Mobilize short term assistance/outside contracting to strengthen forest management, monitoring of logging and marketing services; 7. Undertake rapid resource appraisal; 8. Launch review of human resource development needs.  - xii - Phase III - to be implemented before the end of 1991: 9. Commence forest re-inventory. 10. Undertake studies of (a) revenue policy and (b) viability of industrial development alternatives. Implications for Donors 23. There is a need for substantial and prolonged assistance from the donor comunity, in both fineancial and more importantly technical assistance. The Action Program offers the donor community a challenging opportunity to help conserve PNG's unique environment and biological diversity. The report offers three unusual suggestions for the role of donors: (a) Support for a Worlc Heritage/Sustainable Yield Declaration. Donor support for such a declaration would be a pathbreaking initiative in the environmental field, and would call for new mechanisms for donor coordination. One major advantage of this proposal would be that it might mobilize support from bilateral donors and environmental NGOs that are not now actively engaged in PNG. (b) A Twinning Arrangement for Institutional Support. TVe proposal to find a twin agency to provide GOPNG with technical support over an extended period of time would require a bilateral aid agency to make a long term commitment to the PNG forestry sector, but with the prospect of making a major impact on its performance and on the sustainable development of the economy as a whole. 5C) The Role for the World Bank. The Government has requested the Bank to act as lead' agency in the forestry sector, not only in mounting the TPAP review, but also thereafter in helping to coordinate donors in supporting the implementation of the Action Program. The Bank takes this assignment seriously, and will make resources available over the next several years to provide support to the Government.  - xiii - BROAD FORESTRY1 SECTOR STATISTICS w!'. Forested Area 1.000 ha Total land area PNG 46.200 Natural broadleaved forest 35,623 Natural coniferous forest 520 Plantations 36 TOTAL 36.179 Operations 1979 1988 Log output (m3) 910,000 2,000,000 Volume exported (m3) 473,000 1,341,000 Value of exports (Millions Kina) 20.76 91.19 Cumulative plantation area (ha) 23,850 35,900 Numbers employed in sector 5,000 (est) Forest management personnel 567 445  I. INTRODUCTION 1.1 The Tropical Forestry Action Plan (TFAP) is an international coordinating mechanism for halting the destruction of tropical forests, and promoting their sustainable use for the economic and social benefit of people. The TFAk seeks to achieve these goals by bringing together national and international agencies to produce systematic and coordinated plans for forest use and development in countries which have requested TIAP assistance. At present, a large number of government and non-government organizations have joined the original sponsors of the TFAP--FAO, UNDP, The World Bank and the World Resources Institute--in promoting and planning the TFAP program. 1.2 In general, according to the TFAP, the consequences of not reducing rates of deforestation, will be severe; by 2000 AD: (a) half of the population of the developing world will be short of fuelwood; (b) forest dwellers will continue to be displaced, and their cultures destroyed; (c) 10-202 of the earth's plant and animal species will have become extinct; (d) further widespread degradation of watersheds, with consequent land and food losses, and natural disasters, will occur; (e) large quantities of potentially valuable timber will be lost, reducing economic benefit fron. its utilization. 1.3 PNG faces these problems, albeit in some cases to lesser degrees than in more heavily populated countries. The final item in the list - the economic consequences of loss of forests - may in fact be potentially more significant for PNG than for other countries, given the high ratio of forest cover per capita, and the extent to which PNG may have to depend on the resource for sustainable economic activity in the future. 1.4 A second compelling reason for TFAP involvement in PNG is the great ecological and biodiversity value of the PNG resource. Some 85? of the total area of PNG remains under forest, at the present time. These forests are well known for their outstanding biodiversity. PNG is at the leading edge of subduction of the ancient Gondwanaland tectonic plate, which has craated spectacular mountain land forms and volcanic features both on the main island and on the smaller islands. The main island of PNG is the largest and highest tropical island in the world, and it exhibits most kinds of environment (apart from desert) and vegetation types. The geology has produced a rich diversity of Malay and Gondwanaland flora and fauna. PNG has the world's largest and smallest parrots, the largest doves, the largest lizard, some of the smallest frogs, and the largest butterflies, moths, and stick insects. It has the tallest tropical trees, the largest rhododendron flowers and exceptionally rich mangrove and sea grass flora. In general, the PNG flora comprise more than 11,000 species, including 2,000 ferns. The  - 2 - lowland rain forests have more than 1,200 tree species. Rare and threatened species--especially among the fern and orchid families--are numerous. 1.5 A third, and more immediate reason for TFAP involvement in PNG comes from the findings of a judicial Commission of Inquiry into Aspects of the Forestry Industry. (referred to henceforth in this report as the Commiss.on of Inquiry) which has been releasing reports from its investigations for almost a year. This Inquiry, under the chairmanship of Mr. Justice Barnett of the Supreme Court of 1'NG, hai reported widespread malfeasance in the sector, and has concluded that the forestry sector is out of control, leading to widespread environmental and social damage and a lack of 'roper economic returns to either the State or landowners. 1.6 The Government of Papua New Guinea (GOPNG) recognizes the importance of its rich forest resource. In 1988, GOPNG requested the TFAP to review the forestry sector. Ir response, in May 1989 the TFAP mounted an inter-agency mission to PNG under the leadership of the World Bank. This mission actively involved government and non-government agencies in PNG at the national, provincial and local levels. The following external agencies also participated: (a) the Asian Development Bank. (b) the Australi&n International Development Assistance Bureau. ,c) the International Union for the Conservation of Nature. (d) the International Institute for Environment and Development. Se) the Food and Agriculture Organization of the United Nations. (f) the United Nations Development Programe. (g) the New Zealand Ministry of External Relations and Trade. (h) the technical cooperation agency (GTZ) of the Government of the Federal Republic of Germany. 1.7 The principal objectives of the review were to: (a) examine the role of forestry in land use and the economy of PNG, particularly with respect to the TFAP priority areas of: land use; industrial development; fuelwood/energy; conservation of tropical ecosystems; institutions. (b) to devise an action program to coordinate Government and donor agency inputs around a program of reform and restructuring of the sector. Annex 7 to this report contains expanded terms of reference, and lists the participants in the TFAP Mission.  - 3 - 1.8 Duriag the Mission, the following major issues were identifieds (a) the serious environmental consequences of the rapidly expanding logging program, and the inadequacy of existing legislation and institutional structures to manage it; (b) insufficient involvement of either landowners or Provincial Governments in forest land use decisions and management; (c) technical and economic inefficiencies in the operation of the market for logs; (d) insufficient consideration of the policy options for industrial development in the forestry sector; (e) a serious lack of resources data sufficient to allow effective sustainable management planning; (f) a lack of a coherent conservation policy.  Table l: COUNTRY DATA - PAPUA NOW GUINEA Area Population Doneit 461,700 eq km 8.c *ilIIon (1986) 7.6 per sq km (1986) Population Chsracteritticu He& Ith Crude birth rate (per 1,000) /a 37.0 Population per physician 1 11,700 Crude death rate (per 1,000) 7i 18.0 Population per hospital lih (1986) 247 Life Expectancy at Birth (yea r) (1982) 51.0 Infant Mortality Rate (per - 1,000 liv. births) (1080) 72.0 Acces to Safe Water (X of Population) Energy Consumption Per Capita Most (KG Coal Equivalent) 1978 Recent Total 20 1R 1970 149 Urban so 66 Rural 19 10 19S0 279 Nutrition La Educatton Per Capita Supply of: Adult literacy rate (1980) (M) 82.0 Calories (per day) 2,146 Primary school nroll _mnt (1985 (X) 64.4 Proteins (Grams per day) 40 Lower secondary enrol Im_nt (1986) (U) 14.5 ON? Per Capita In 1908: US8896 Cross Domestic Product in 1988 /o Annual Rate of Growth (X. 1981 prices) USS Min X 1981-86 1986-87 1987-88 /a CDP at market prices 8,481.9 .00.0 2.9 5.0 8.2Qross Domestic Investment 867.3 26.1 -8.3 6.6 16.7 Grovs national savings 506.6 14.6 Current account balance -861.7 -10.4 Exports of goods, NFS 1,547.4 44.7 4.9 6.9 8.7 Imports of goods, NfS 1,798.4 61.9 -1.6 -8.6 8.8 Output and Employment Overall Labor Force Value Added Labor Force 1984 1986 1980 (Forml Sector) USSUin X Min X '000 X Agricult-ure 898.5 85.6 1.28 85.0 48.4 24.9 Industry 684.7 27.1 0.05 8.6 84.8 17.7 Services 940.8 87.8 0.17 11.4 111.4 67.4 Total 2,524.2 100.0 1.60 100.0 194.2 100.0 Government Finance Central Qovornment K eIn X ofr Po LOST 1988 1 i985 1887 1988 /j Current Receipts 822 880 81.1 29.9 29.4 Current Expenditure 762 799 81.0 27.7 26.4 Current Surplus 80 00 0.1 2.2 8.0 Capital Expenditure 97 105 2.6 8.5 8.6 Overall Balance -87 -15 -2.5 -1.8 -0.6 I Moat resnt estimate. Source: World Bank, 1989  II. FORESTRY AND THE PNG ECONOMY A. The Papua New Guinea Economy: Directions and Policy Imperatives 2.1 The World Bank has recently published a report on the opportunities and challenges for accelerated development in Papua New Guinea (World Bank (1989)). This brief review draws largely upon that report, as well as an earlier economic report published in 1988, and a recent analysis of prospects for the aevelopment of the agriculture sector (World Bank (1988)). 2.2 Table 1 reproduces basic country data for Papua New Guinea. 2.3 In general, Papua New Guinea has a good record of financial management over recent years. Less success has been achieved in. the area of structural adjustment and investment in growth and employment creation - and these observations will be borne out in specific examples in the forestry sector throughout this report. 2.4 From the viewpoint of future policy options, a significant factor in PNG is the future of the mineral sector. 2.5 While the recent closure of the Bougainville copper and gold mine has caused a sharp drop in export earnings in the short terra, it is possible that future developments in the mineral sector will add significantly to the nation's financial strength, by the middle of the 19908. However, as is often the case, mining in PNG is an enclave activity: most investment expenditure is on imported capital goods, and forward linkages to the rest of the economy are also weak. In particular, the employment generating capacity of the mineral sector is small. Moreover, revenue flows from mining will be essentially temporary in nature; at some point commercially extractable reserves will decline. 2.6 Thus the impact of the mineral's sector on the rest of the economy is uncertain. It will be determined very largely by the effectiveness with which increased financial flows from mining are invested in the remainder of the economy. The Government of PNG will need to look increasingly to opportunities for investment in sectors of the economy that are both more sustainable and more strongly linked into the economy, than is the mining sector. 2.7 Thus, while a mineral boom will provide opportunities for accelerated development in PNG, these will only be sustained if effective investment of the mineral revenues elsewhere in the economy takes place. This will call for a coherent overall strategy, and adoption of appropriate policies. Whatever is proposed for forestry in PNG must take these general imperatives into account. Also, some key constraints in the economy must be borne in mind: (-) shortage of skilled labor (b) limited capacity of public institutions (c) limited infrastructure (especially in the internal transport area)  - 6 - (d) the wide disparity in regional incomes, due to natural and historical development factors. 2.8 The 1989 Bank study puts forward a policy agenda, to achieve development over the medium term, in light of these major constraints. First: effective management of mineral resource development, primarily through the promotion of policies to enhance community relations and income distribution, and review of sector and fiscal policies that may inhibit exploration and development. Seconds continuing strong support for agriculture and other renewable resources and development of rural infrastructuret in particular, construction of rural roads is a priority. Thirds human resource development via an emphasis on primary education, restructuring educational institutions and improAing health care. Fourth: increased investment in the non-mineral sector, which implies izproved project preparation and implementation capacity; more efficient land mobilization; streamlining regulations and institutions dealing with foreign investment and development of small business; and improved macroeconomic policies, especially in the areas of wage and exchange rate flexibility. 2.9 Underlying all these objectives is the urgent need in PNG for institution building in the public sector, improved management of resources and accelerated human resource development. Without strong improvement in these areas, little of a sustainable nature will be achieved in the development program. 2.10 Development of agriculture, while an obvious priority, is unlikely to continue to be able to absorb 5 out of 7 entrants into the labor force, as it currently does. PNG must therefore look to its other major non-mineral resources for development: fisheries and forests. These sectors will also have important regional implications, in those areas of the country where agricultural development prospects are limited. Ultimately, PNG will also have to consider development of industry and services beyond their presently vestigial levelas forestry, in particular, is often seen as a suitable vehicle for traf.sformation from a solely resource oriented to a value added sector, given that certain types of forest processing can be technically efficient at relatively small scale and low capital intensity. 2.11 Many of the larger themes raised above will resonate through the specific forestry sector discussion which follows. In particular, frequent reference will be made to the need to reorganize and develop the capacity of institutions to effectively manage the forest resource, in the PNG real world of communal ownership of land and a system of government based on devolution of many responsibilities to Provincial Governments. B. Forestry in the Economy 2.12 Table 2 gives the basic forestry economic statistics for PNG. In 1988, PNG earned about US$111 million from forest product output, of which US$94 million was for logs. This amounted to about 8.5Z of the total value of exports from PNG. The share of forestry in total exports is varied (but about no particular trend over the current decade (Table 3)), whereas the share of agriculture and fish products has generally declined. The value of all exports has increased in real terms by 651 over the 1981-88 period and in the  case of forestry unit values have moved up, but a large part of the export gain has been due to volume increases: from 743,000 m3 in 1981, to 1.3 million m3 in 1988 (see Table 5). Volumes of logs exported from PNG began to rise significantly after 1979, when a new Forestry Act which was more lioeral than its predecessor -wth regard to log exports was promulgated. Table 2: FORESTRY IN THE PNG ECONOMY IN 1988 ($US m) PNG Forestry PNG Forestry GOPNG Forestry GDP share exports share resources share Sm Z Sm Z $m Z 3,461.9 159.2 4.' 1,312 111.3 8.5 680.4 21.9 3.2 2.13 Papua New Guinea remains a relatively insignificant supplier of export logs to major North East Asian markets, but its share of that market has increased dramatically, due to the combination of rising export~ from PNG, and declining log exports from previous major supplies such as Indonesia and the Philippines. Table 4 shows imports of logs from the region by Japan since 1965. As shown in Table 6, Japan, Korea and Taiwan import more than 90 percent of PNG log exports. 2.14 PNG imports relatively insignifIcant amount. processed wood products (which attract an import tax of 30 per cent ad valorem): the figure has varied between K1-K2 million in recent years. Imports of paper and paperboard have become significant, however, and recently exceeded K 20 million p.a. 2.15 Employment in the industrial forestry Sector of PNG has fluctuated between 4,900 and 5,500 since 1979. Of this figure, some 2,200 people are reckoned to be employed in sawnilling while most of the balance work in logging. This overall employment level is extremely low for a sector of this size. The predominance of logging as well as the relatively capital-intensive nature of wood processing in the sector, and the until now minor role played by the much more labor intensive plantation forestry and agro-forestry, have made PNG forestry a poor generator of employment. In this, however, the existing structure of the industry is more a reflection of certain economic fundamentals (high wages in the organized labor sector, high infrastructure and local freight rates, etc.) than any indication of the low labor absorptive capacity of the sector. The scope that PNG forestry offers for additional employment c early cannot be discussed separately from the wider economic policies which have shaped the industry's structure as well as determined its sheer size.  - 8 - Table 3: PNG VALUE OF EXPORTS BY TYPE OF PRODUCT. 1981-88 La (In Millions of US$. F.O.B.) 1981 1982 1983 1984 1985 1986 1987 1988 Est. Mineral 447.0 409.6 447.4 365.5 490.1 580.6 780.3 924.4 Copper 200.2 166.5 193.0 151.5 164.3 160.6 289.1 395.8 Gold 236.6 232.0 240.9 205.0 318.8 413.0 479.7 515.6 Silver 10.6 10.2 13.4 8.9 7.0 7.0 11.5 12.9 Non-Mineral 366.0 331.2 354.1 528.4 4CU.7 428.0 418.9 387.8 Agricultural 260.0 236.2 276.9 425.8 330.2 341.5 284.8 268.7 Cocoa 51.8 43.2 49.7 75.0 62.4 57.9 62.4 58.0 Coffee 109.4 106.1 113.5 123.8 117.5 214.7 142.7 125.0 Tea 10.7 9.1 12.5 19.2 11.6 7.7 7.1 9.0 Copra 30.0 17.4 28.5 54.9 3;.4 10.3 16.8 19.2 Copra Oil 18.6 16.5 23.6 44.1 23.7 10.7 16.0 21.3 Palm Oil 21.1 29.5 28.5 84.7 61.5 29.1 37.3 51.4 Rubber 5.1 2.0 2.6 2.6 4.0 3.1 2.8 5.0 Other /c 13.4 12.5 18.1 21.5 16.1 8.0 0.0 9.6 Forest Products 65.2 83.7 66.3 91.4 67.3 78.4 122.0 111.3 Logs 46.8 67.3 52.5 78.2 58.4 71.4 116.2 99.3 Other id 18.4 16.4 13.8 13.2 8.9 7.1 7.4/b 8.4lb Fish Products 40.7 11.3 10.9 11.2 12.2 8.1 12.1 11.4 Tuna 29.7 1.9 0.3 0.6 5.1 - - - Other /e 11.0 9.4 10.6 10.6 7.1 8.1 12.1 11.4 MAJOR PRODUCTS 813.0 740.8 801.5 893.9 899.9 1.008.8 1.199.2 1.312.2 Other Exports /f 28.7 33.8 40.9 25.6 18.0 44.4 30.3 53.4 TOTAL EXPORTS 841.7 774.6 842.4 919.9 917.9 1.053.0 1.229.2 1.385.4 Forest Products aS X of Total Exports 7.7 10.8 '.9 9.9 7.3 7.4 9.9 8.0 Logs as 2 of Total Exports 5.6 8.7 6.2 8.5 6.4 6.8 9.5 7.1 la Based on customs data. lb Small discrepancies exist between DTI and Department of Forest data. Here, the latter are used. /c Includes a very small amount of manufactured agro-based goods. /d Includes sawntimber, plywood and woodchips le Includes prawns, barramundi and crayfish. If Includes re-exports. Sources: Department of Trade and Industry Department of Forests  Table 4: JAPANESE IMPORTS OF LOGS FROM SE ASIA AND PACIFIC ISLANDS, 1965-88 (1,000 m3) 1965 1970 1975 1980 1985 1968 (est) Philippines 5,632 7,542 2,853 1,073 559 8 Malaysia 3,471 6,019 6,659 8,580 11,402 10,637 Indonesia 99 ^.391 7,298 8,639 202 - Solomon Is 18 202 168 2_9 318 191 PNG. 62 339 304 438 734 819 Sourcest Japan Lumber Journals, Jan 1980 - April 1989. Timber Digest, Dec. 1988, PNG Dept. Forests. ; !H Table 5: PNG: VOLUME OF MAJOR EXPORTS BY TYPE OF PRODUCT, 1981-88 (Thousand Tonnes) La 1981 1982 1983 1984 1985 1986 1987 1988 Est. Ib Copper 164.0 172.2 181.1 163.3 168.7 178.7 211.5 222.0 Gold (Tonnes) 17.6 19.1 ;9.1 19.4 32.0 36.5 34.5 37.3 Coffee 46.4 41.2 52.5 49.4 40.6 53.1 64.3 77.4 Cocoa 27.8 28.6 26.3 34.1 30.9 30.9 34.7 43.2 Copra 102.9 74.8 78.7 93.5 103.5 91.9 85.5 113.1 Copra Oil 30.1 42.7 36.0 40.7 41.4 40.7 40.1 41.0 Fish Products 31.2 3.9 2.8 2.9 13.2 1.7 1.4 1.9 Palm Oil 34.1 88.9 77.9 129.9 123.8 122.3 123.4 160.7 Tea 7.0 6.1 6.8 7.3 6.6 5.7 5.7 7.0 Rubber 4.5 2.2 2.7 3.4 5.4 5.0 3.7 5.5 Logs (000 m3) 743.0 1,063.0 1,019.0 1,278.0 1,158.0 1,299.0 1,450.0 1,289.0 /c la Unless otherwise indicated. lb Prelimin;ry figures. /c Preliminary figures of DTI and DOF show a small discrepancy. Here, the latter are used. Sourcess Department of Trade and Industry. Department of Forests.  Table t: DESTINATION OF PNO FOREST PRIIUJCT EXPOMS, 1994-88 1984 1965 1906 1967 1906 Volu_e Value Volu Value Volume Volve Volu_e Value Volme Volvo (000 c ) (K el) (000 co v) (Ku 1i) (000 co a) (K el) (000 co m) (Ke *i) (000 NO ) ( ml 1) Total lam ugpogte 1288.9 70.8 l156.8 58.4 1814.0 70.8 1442.0 1OC.S 1209.0 91.2 of which sbare at Japan 49.0 49.0 60.1 69.4 66.6 S6.4 60.1 J0.9 6". 60.4 Korea 44.8 44.4 86.8 U.6 87.1 86.9 88.0 $1.9 81.9 29.0 Talan 6.8 5.0 8.8 8.2 8.6 2.9 8.8 2.6 2.8 2.6 OOhre 0.9 1.0 0.8 0.6 8.5 4.8 8.6 4.6 1.7 7.4 (100.0) (100.0) (100.0) (100.0) (100.0) (100.0) (100.0) (100.0) (100.0) (100.0) Total * antiinr eumorts 17.6 8.8 14.8 8.0 7.6 1.6 8.6 1a 2.7 0.7 of which ahare of Australia 40.6 54.0 52.0 52.0 42.8 42.6 52.6 66.8 44.8 89.8 New Zealand 28.4 26.5 25.6 27.9 45.1 45.4 86.7 39.6 81.4 34.9 Japan 2J.8 10.0 20.9 17.4 5.4 1.6 9.7 8.2 24.8 25.6 Othre 1.7 8.5 2.6 0.7 7.2 10.4 0.7 0.7 - _ (100.0) (100.0) (100.0) (100.0) (100.0) (100.0) (100.0) (100.0) (100.0) (100.0) Total woodehlip xports 102. /a 61.8 1 1.0 /I 65.7 /a 67.l La Japan 100.0 91.7 75.2 66.3 76.0 Taiwan - 6.8 24.8 22.5 24.0 Others - - 11.2 _ (100.0) (100.0) (100.0) (100.0) (100.0) la Unit: air dry tonne.  - 11 - C. The National Forestry Develorment Proaram 2.16 In September 1988 the Department of Forests published The Forestry Subsector in the Medlum Term Develovment Strateav. 1989-93. Key policy objec- tives put forward in that document ares (a) the guiding principle of management will be sustained yield; (b) resource replacement and regeneration will be provided by a balance of natural forest management and forestry plantation development; ; (c) forest development will be achieved by an equitable distribution of benefits from economic development throughout the country and to provide basic needs of the people; (d) a National Forest Development Plan will provide direction for orderly and rational forest development; (e) integrated land use planning will form the bas!.s of forestry sector planning; (f) increased attention will be paid to small scale forest based develop- ment to promote participation by customary forest owners; (g) on-shore processing will be a major requisite for all future forest development; (h) promotion of participation and involvement of PNG nationals including small-scaled industries; (i) trengthening of a central marketing authority in association with the State Purchase Option; (J) mandate is given to improve forest research to provide essential support to the operational aspects of the management and utilisation of forest resources so that they are conserved and maintained to provide a sustained supply of timber and other forest produce; and (k) forest training and education will be a priority to provide the moti- vation and skills necessary for the forestry and forest industry sectors. 2.17 The development strategy provides guidelines and a list of priorities which need addressing over the 1989-93 period to ensure implementation of these policies. The development strategy also revised logging levels in the 1987 National Forest Development Program (NFDP) upward to a level that the Department of Forests felt they could adequately handle. The revised NFDP is included in Table 7:  - 12 - Table 7s NATIONAL FORESTRY DEVELOPHENT PROGRAM: LOGGING CONCESSIONS, 1987-91 No. and Area (1,000 ha) 1987 1988 1989 1990 1991 Description No. Area No. Area No. Area No. Area No. Area New investment projects 5 11 828 14 1,625 14 841 16 321+ Renegotiated ongoing projects - 2 580 1 290 - - - - Defunct projects reviewed - 5 125 - - - - - Back-up areas 3 5 57 8 102 6 57 2 -+ Total 8 23 1.590 23 2.017 20 898 18 321+ + Project areas not yet measured. Source: Department of Forests. 2.18 A comparison between the revised NFDP and the original program is given in Table 8: Table 8: NPDP AND ORIGINAL PROGRAMS Annual Volumes (millions m3) 1987 1988 1989 1990 1991 NFDP review 3.0 4.2 5.4 6.0 6.0 Original ProKram Total harvest 2.7 2.9 3.0 3.0 3.0 Log export 1.8 1.8 1.6 1.3 1.0 + Timber export 0.1 0.2 0.3 0.4 0.5 + Timber domestic 0.34 0.36 0.40 0.44 0.50 + Sawn Timber Conversion Factor 5O0. Source: Department of Forests. The revised program has major implications for government and landowner revenues, Department of Forests monitoring manpower and budgets.  - 13 - 2.19 The NFDP review does not specify the tactical plans for indigenous forest management or reforestation. However, the 1987 Annual Report indicates a proposed reforestation program as detailed in Table 9: Table 9: PLANTING TARGETS BY YEAR Area (ha) Year Target (halyr) Target (ha) 1987 3,400 33,900 1988 3,400 1989 3,400 1990 3,400 48,200 1995 73,500 2000 100,000 Source: Department of Forests 1987 Annual Report D. The Institutional Framework 2.20 Since Independence, in 1975, there has been a continuous devolution of central government powers to the Provinces, and forestry has been no exception to this process. The result is that the central Department of Forests is now a residual of an original line department conceived when control over forestry was a national government matter. Chart 1 outlines the upper structure of the Department of Forests (DOF). The Department now has 164 management personnel (including those involved in research and training). The 19 provincial governments have a total of 281 management personnel, organized in varying ways, and a further 600 employees. 2.21 In 1989, the budgetary allocation to the national and provincial forestry agencies totalled K 6.4 million. Government revenues in the same year from forest operations totalled K 22.7 million.  - 14 - Chart 1: ORGANIZATION OF DEPARTMENT OF FORESTS I ii -~~~~~~~~~~~ 4I ! . 00~~~~~~ 0~~~~~~~~  - 15 - III. THE FOREST RESOURCE AND SUSTAINABLE MANAGEMENT 3.1 The total land area of Papua New Guinea is 46.2 million ha. In 1985, it was estimated that the forested area was 36.2 million ha. Table 11 simmarizes the categories of forest: Table 10: PNG FORESTED AREA, 1985 Forest Type 1.000 ha 1. Natural forest - Broadleaved 35,623 - Coniferous 520 Total Natural Forest 36,143 2. Forest plantation 35 Total Closed Forest Area 36.179 A. The Natural Forest Resource of PNG 3.2 FAO has collected a large amount of information on the natural forest resource of PNG, and the bibliography attached to this report lists the rele- vant publications. FAO (1987) estimates the operable (or commercial) forest area and total growing stock in PNG forests as follows: Table 11: PNG NATURAL FOREST AREA AND GROWING STOCK 1980 1985 1990 Broadleaved forest area (1,000 ha) - operable 14,035 15,000 15,000 - inoperable 21,696 20,623 20,583 Growing stock /a (millions m3) - heavy hardwoods 417 443 439 - other species 975 1,036 1,025 /a 23Z of the growing stock is >50 cm diameter at breast height. 3.3 In the same publication, FAO estimated that net annual deforestation in the natural forests averaged 22,000 ha p.a. over 1981-85. Logging opera- tions covered some 70,000 ha p.a. over the same period. Often, in PNG,  - 16 - various forms of agriculture follow logging (due to enhanced access). The potential role of logg'ng as an agent in deforestation, if not a direct cause, must therefore be borne in mind. The average volume achieved from commercial harvesting operations was 30 m3 per ha. 3.4 The estimates of utilizable forest are based on interpretation of aerial photographs taken by the Australian military during the Second World War, w:th some further photography in the 19609. These images were not intended for use in inventory work, and their accuracy is questionable. In any event, environmental, socio-demographic and agricultural parameters have altered significantly in subsequent decades. 3.5 During the 1960s and 1970s the DOF did carry out extensive reconnaissance ground surveys in proposed timber rights purchase areas to gauge species mix, piece size and indicative yields by way of transect plots with a view to assessing the value of the resource for royalty purposes. However, the yields derived have been subject to significant changes since the time of surveys due to review of environmental issues, marketability of species, harvesting technology, economics of operating and landowner aspirations. Moreover, inventory prior to logging has usually been done at less than 1Z sampling intensity, which is inadequate. 3.6 There is considerable doubt as to how much of the PNG natural forest is accessible to commercial logging, and as to what the zondition of the forest is. As long ago as 1970, when the common estimate of commercially utilizable forest was 19.4 million ha, McIntosh and Granger argued that the area of forestry in accessible areas of 100,000 ha or more, which was considered the minimum area necessary to justify commercial investment in logging, was only 2.6 million ha. 3.7 White (1974) presented some data to show that in one major forest area near Madang, only 25? of the forest area was carrying forest of appropriate size and species characteristics to be classified as commercial, and some part of this area was inaccessible to commercial operations. None of it, however, was immune to shifting cultivation, and a detailed study of one part of the area showed that some 36t of the area was man-induced regrowth cr grassland, and the remainder was disturbed in some way. 'White emphasizes that such figures should not be generalized to the whole resource. However, the figures do serve to shed considerable doubt on any current estimates of commercially accessible volume in the PNG natural forests. B. The Sustainable Hanagement of the Resource 3.8 Rainforest is the predominant feature of the PNG environment in the zone between 0-1,500 m altitude. These lowland forests have a large component of species which, ecologically, are secondary in nature: they are light demanders, which reach large sizes and live for long periods. The processes controlling rainforest dynamics are such that the further the disturbance pattern associated with forest management (e.g. timber harvesting) differs form the natural disturbance pattern, the greater the floristic and structural differences between unlogged and regenerating stands. The normal pattern of disturbance in nature is the death and toppling of large trees. Any silvi- cultural system which drastically reduces basal area and species composition  - 17 - will require considerable inputs to maintain forest cover, and even then output of other forest products more dependent on biodiversity are likely to be lost. 3.9 PNG therefore has a range of options in its na.ural forests, from minimum disturbance systems through to clearfelling and cor-qrsion to other land uses. If White's analysis cited earlier is any guide, she minimum disturbance option may be beneath the level of activity already practiced in the more populous areas, let alone that necessary in commercial logging. 3.10 Allowable cut, defined very broadly, calls for prescriptions which do not remove more volume than the resource is capable of regenerating. The objective of this form of management is basically to retain a given area under forest, and to practice some form of steady state of growth and output from that resource. The range of possibilities within this option is wide and complex. At one extreme is selective logging with the objectives of retaining quality trees below a certain size limit for subsequent logging, and of allowing natural regeneration to replace what is taken. Then there are a range of heavier logging options associated with timber stand improvement and/or enrichment planting, to produce a mosaic of larger remaining trees and an improved composition (from the point of view of wood values) of regrowth. At the other extreme is localized clearfelling and replanting. Allowable cut prescriptions, especially when defined in volume of log output terms, are highly fluids they can be affected by changes in market demand (especially in the mix of species required), by the methods of logging and roading employed (since the condition of the regrowth and remaining original trees is the major factor) and by the silvicultural techniques employed. Setting allowable cut targets too high, in a given set of circumstances, will obviously continuously reduce the size and quality of regrowth and residual Ftands to the point where it becomes an uneconomic proposition to retain what is left. Setting targets too low, on the other hand, undervalues the forest stand (and therefore makes conversion options appear more economically attractive) and, in the case where a significant proportion of the existing stand was overmature, actually physically reduces the vclume that comes from the stand. 3.11 Sustainable yield in tropical rainforest has attained the reputation of being difficult to achieve--especially in an environment where population pressure is intensifying traditional shift and burn agricultural practices, and is also increasing the general demand for permanent farming land (recently roaded and logged foiests represent a convenient basis for such expansion). Moreover, the economics of sustainable yield management of forests is frequently questioned. Leslie (1977) suummarizes the conventional views "Even the most ardent supporter of natural forest systems could not dcny that all the effort in tropical forestry has yielded very few reliable and reproducible methods for natural regeneration of moist trt'pical forests" 3.12 However, in PNG there are both political and economic reasons for retention of a large proportion of forest cover.  - 18 - 3.13 Turning first to the political factors virtually all forested land in PNG is owned by clan groups under custom law, rather than by the state. This can expose resources to serious over exploitation, because forests can be committed to logging without any technical or environmental supervision of what is done-and this has occurred in PPG, as will be discussed later in this report. On the other hand, it offers the opportunity for traditional landholders to resist pressures for logging in favor of retaining the resource for future generations, where state forestry bureaucrats might encounter problems in resisting powerful political pressures for exploitation. However, it is important that landholders be offered adequate alternatives. if this opportunity is to be realized. 3.14 Whether because of the latter factor or some other reason, the Constitution of PNG contains a strong statement on the necessity to retain forests: '4. Natural Resources and the Environment. We declare our fourth goal to be for Papua New Guinea's natural resources to be conserved and used for the collective benefit of all and to be replenished for the benefit of future generations. 3.15 The Forestry Subsector Medium Term Development Strategy 1989-1993 does not refer directly to the Constitution but describes certain 'impera- tives' (p5). These are similar to those raised in a 1986 FAO review (Campbell and Gane 1986) which sources them to the Constitution. FAO in that document was in no doubt that the sustai-ed yield objective was at the time of writing not being met, in that the rate of timber cutting 'greatly exceeds the rate at which it is being replaced so that PNG's forest capital is being depleted. This clearly contravenes the fourth aim of the Constitution... ' 3.16 The Medium Term Development Strategy makes liberal reference to the sustained yield objectives "(a) The guiding principle for all forest development shall be sustained yield management. The productivity of forest areas which are harvested must be maintained and, where possible, increased. (b) The responsibility to ensure that the country's forest resources are replaced and managed rests with the Government but with the under- standing that those involved in the utilization must contribute to the cost of replacement. (c) Resource replacement and regeneration will be pursued by adopting a careful balance between management options including intensive plan- tation forestry and natural regeneration management. Perpetuation of the country's high value indigenous species will play a key part in the management program.' 3.17 It iR clear that the Foresiu Department sees plantation forestry as a major contributor to sustainable yield: The Strategies component of the same docAment prescribes that 101 of cut-over area should be converted to planta- tion forest (to be funded from a K 2/m3 levy on exported natural forest logs.)  - 19 - 3.18 In effect, this objective calls for the ultimate establishment of some 600,000 ha of plantation in PNG, and the cumulative cost of such a program, at current plantation establishment and tending costs, would be in the order of R 900 million - a very substantial investment indeed. 3.19 On the other hand, the Department of Finance and Planning indicated to the TFAP Mission an interest in divestment of the small existing plantation estate, as opposed to a large increase in government involvement in this area. 3.20 A plantation program should be determined on the basis of economic and environmental criteria: an analysis of markets for the specific output intended; calculation of the costs and benefits of planting; an appraisal of the environmental consequences. Plantations should not be regarded as a necessary endeavor to allow attainment of some predetermined physical output level for forestry as a whole, nor as a substitute for sustainable management of natural forests. Sustained yield management of natural forests and plantation forestry are not particularly closely related: they produce very different products; they require quite different investment schedules; and they have very different environmental and social impacts. On any given area of existing forest, sustainable management of tho. natural resource is either the best alternative, or it is not. Alternatives to it on that area may or may not logically include plantation forestry. 3.21 To return now to the economic argument for sustained yield: it can be argued that the objective of national policy in the forestry sector should not be sustainability per se, but optimal resource usage. In an open economy, not exploiting standing forest involves the country in an opportunity loss - the more so in a case (such as PNG) where the volume of logs placed on the international market is unlikely to have any major price effect, within the supply limits that apply. The magnitude of this loss will depend, inter alia, on the market value of the resource, the cost of managing it (ranging from minimum cost 'mining' operations, through controlled harvesting, to plantation options), and the effect of such management on the subsequent growth of the resource. 3.22 It is not possible, given current knowledge, to determine where this optimal point of resource usage might be: that would require a study of the economics of management of forests in the various locations, and of the comparative economics of other forms of land use in each place. Such an analysis would also need to account for the high probability that the social profitability of natural forest management will change over time. 3.23 The best that can be done at this stage is to speculate on how such an analysis might turn out: (a) Most land occupied by forest in PNG will probably never be converted to production of non-forest crops, due to problems of access and the level of managerial skills and financial resources currently avail- able, as well as adverse topographic and edaphic factors in much of the area. In short, on much of the area in question forestry may be the only logical alternative, given management and investment constraints, and the inherent environmental cost of exposing forested land with no imediate conversion alternative. Much traditional  - 20 - agricultural practice in PNG is environmentally sympathetic, and investment in improvement of this land use would be probably cost effective. (b) Tropical forest output is readily disposable through a known system (albeit one which is in need of improvement, as discussed later). By contrast, non-forest crops demand greater inputs and skills in marketing, and are by nature higher risk investments due to the high initial costs of establishment. (c) The unvalued benefits of biodiversity, non-wood fcrest products and retention of the basic environmental role of forest cover in the traditional agriculture of PNG (a landuse that will need to support a large proportion of the population for some considerable time yet), will almost certainly greatly exceed any unvalued benefits generated in conversion options. 3.24 These observations suggest that a conservative approach to conversion of forest is rational, given present knowledge. Clearly, there are areas of PNG where natural forest should be, and will be, converted either to forest plantations, or non-forest crops. There is a requirement, however, that this be done only after a comprehensive analysis of all the forestry alternatives- -frcun sustainable yield management based on natural regeneration, through the range of timber stand improvement and enrichment planting options, to the conversion options. Much of the tree crop and other major agricultural deve- lopments which have occurred to date in PNG have been done on areas where either degraded forest, or no forest, existed--so the need for such comprehen- sive analysis was limited in the past. 3.25 Although as argued above, area targets may be more rational than physical output limits so far as sustained yield management is concerned, the more traditional approach is volume based, and, ultimately, volume estimates will be required for planning. 3.26 In the PNG case, the first major difficulty in arriving at a physical estimate of sustainable yield is the paucity of resources data: this matter was raised in paras. 3.6 and 3.7, and is discussed further in paras. 3.53 to 3.61 below. 3.27 Even if the physical and biological data were available, there are other factors involved in the calculation of sustainable yield. The actual distribution of log sizes by species is a critical factor, because smaller logs, whilst permissible to log in the prescription, are much less valuable than larger logs, per unit volume: the higher the proportion of smaller logs (above minimum logging diameter) the less commercially attractive the logging area. Slope, terrain, waterways and other natural variations also impact upon commercial viability of logging. The form of logging used (especially the degree of damage done to the residual stand which is the basis of sustained yield), and the market for logs, also directly influence the area of forest which is commercially loggable. Stocker (1988) has pointed out that in PNG, control over logging will be the major silvicultural mechanism in the natural forests. The treatment of a residual stand after logging i8 also a critical factor.  - 4 . - 3.28 Obviously, markets, price-size gradients, the acceptability of certain species, logging methods and efficiency, and the treatment of the residual stand all change over time, and the allowable cut from the forest will change accordingly. 3.29 The magnitude of the data problem in arriving at such a figure can be simply illustrated: on the field trip, the TFAP Mission frequently encoun- tered opinion that as little as 302 of concession areas may actually be loggable, due to topographic and associated factors. Moreover, while official estimates of commercial standing volume range in the area of 50-60 m3ha, actual turnoff of logs from operations is actually more in the order of 20-30 m3ha. 3.30 The TFAP mission originally provided, on request from GOPNG, an estimate of allowable cut from the PNG resources. This estimate was based on present harvesting and utilization technology, and some crude estimates of increment in the resource, net utilization area, and the cutting cycle recovery to produce a high valued crop on second entry. The conservative figure produced with this process was an allowable cut of 3.6 million n3 p.a. The mission pointed out that the planning figure then in use by GOPNG--6 million m3 p.a.--was also within feasible bounds on the basis of what little is known about the resource. 3.31 Since the TFAP mission was completed, GOPNG has commenced the process of revising and refining the allowable cut figure, although clearly it will not be possible to greatly increase the precision of allowable cut estimates until new resource estimates are available (see Section D of this chapter, and Projects 1 and 2 in Annex 10 to this report). '.32 In the meantime, GOPNG has undertaken to continue to review its allowable cut figures, and to take a conservative approach to determination of this figure, to accommodate potential conservation area requirements and propose other options. It must be emphasized that the allowable cut figures under discussion are based on current management and utilization practices. Significant inputs to improve these are suggested in the Action Program found in this review. These aspects, and the revenue implications for PIG of doing so, whilst still adhering to the principle of sustainable management of the resource, are considerable. At present, the TFAP mission suggests that the conservative figure of 3.6 million m3 p.a. allowable cut should be applied until more definitive estimates can be prepared. 3.33 The TFAP mission visited several logged over sites of varying ages in the field in PNG, and some experimental enrichment planting and timber stand improvement areas as well. Basically, there is no formal application of the sustained management system in the commercial forestry operations of PNG--i.e. no planning based on an inventory and on selection of species for second and subsequent cycle harvests. Rather, forests which have been selectively logged for more valued species have been left to recover without silviculture or management input. Such forests do recover but relatively few of the valuable species which tend to appear later in the regeneration cycle were seen in the sites visited, and diversity appears to be lost. It may be that more of these species would return naturally after sufficient timet this is a matter for  - 4 . - systematic research, and also for consideration by forest managers who vill need to know how much time must elapse for secondary species to return in appreciable volumes. 3.34 The basic conclusion from the foregoing observations is that, despite the difficulties inherent in sustained yield management, it does seem to represent the best option for most of the PNG resource. A major effort will be needed to implement this form of management in the commercial resource in PNG. C. Plantations 3.35 In paras. 3.17 to 3.20 above, aspects of PNG planting policy were discussed, and it was argued that the role of plantations should not be to obtain some predetermined national sustainable yield, but to maximize economic and environmental objectives. 3.36 Reforestation in PNG has been a sporadic activity, varying consider- ably from year to year with availability of funds, land and supervisory personnel. In the past decade, government plantation areas have remained static, at around 25,000 ha. Private planting has moved from virtually nothing at the beginning of the decade, to about 10,600 ha now. It should be noted that the statistics on private plantings include only larger scale planting: smallholders plots are not assessed. 3.37 The plantation scene in PNG is still dominated by the 11,000 ha of government owned coniferous plantings at Wau and Bulolo in Morobe province (principally for use in the plywood mill there). There are 4,200 ha of government Pinus and Eucalvntus species planted at Laipegu and Norikori in the Eastern Highlands, and the bulk of remaining government plantings are in Central, Western Highlands, Madang and East New Britain provinces. There are now some 4,000 ha of private plantations in Madang, and a further 6,500 ha in East and West New Britain. Table 12 below summarizes plantation areas by Province since 1979.  Table 12: PLANTATION AREASs 1979, 1985, 1988 (ha) Province 1979 1985 1988 Govt Pvt Total Govt Pvt Total Govt Pvt Total Central 1,963 1,963 2,080 2,080 2,103 2,103 Northern 36 36 47 47 47 47 Morobe 11,994 11,994 10,886 10,886 11,196 11,196 S Highlands ''.9 119 133 133 216 216 W Highlands 1,740 1,740 1,793 3 1,793 2,039 3 2,042 Enga 152 152 108 108 108 108 Simbu 11 11 11 67 78 11 67 78 E Highlands 3,979 3,979 4,520 88 4,608 4,741 88 4,829 Madang 504 504 1,450 3,618 5,068 2,001 3,918 5,919 E Sepik 123 123 123 123 123 123 EN Britain 2,563 2,563 2,103 107 2,210 2,120 2,122 4,242 WN Britain 666 666 126 2,503 2,629 126 4,426 4,552 Milne Bay 399 399 N Ireland 3 3 42 42 Total 23,850 23,850 23,383 6,386 29,769 25,272 10,624 35,896 3.38 About 33Z of plantation area in 1988 was comprised of exotic species--principally Pinus and Eucaplytus--and the remainder was indigenous speciess the local Eucalypt (E. dedluDta) known as Knmarere; Araucarias (Klinkii and Hoop pines); and some rainforest species such as Terminalia and Octomeles. 3.39 It is probable that the pilantation figures given above are over- estimates: losses due to fire and other factors have not been accurately assessed in PNG. 3.40 Table 13 below details the age classes of the plantation resource in PNG: I  - 24 - Table 13: AGE CLASSES OF PLANTATIONS, 1986 1,000 ha Age Conifers Hardwoods Total Years tPinus Other Sub-Tot Eucs Fast Hwds Teak Other Sub-Tot 0-5 1.3 0.2 1.5 3.9 1.0 0.2 5.1 6.6 6-10 3.2 1.1 4.3 4.2 1.1 0.3 0.1 5.7 10.0 11-20 2.5 3.8 6.3 1.0 0.5 1.6 0.1 3.2 9.5 21-40 3.7 3.7 0.4 0.1 1.0 1.5 5.2 Total 7.0 8.8 15.8 9.5 2.7 3.1 0.2 15.5 31.3 Sourcet Resource Development Division, Department of Forests 3.41 It can seen from this table that most coniferous plantation in PNG is older than 10 years. whereas most hardwood plantations--especially the fast- growing eucalypts and others--are less than 10 vears old. 3.42 Table 14 below summarizes results from research into growth and yield as has been carried out in PNG: Table 14s MEAN ANNUAL INCREMENT DATA, MAJOR PLANTATION SPECIES Species MA, m3ha/yr Pinus spp 10-15 (P.caribaea Araucaria 10-15 (Hoop and Klinkii) Eucalyptus dealupta 25-30 (Kamarere) Eucalyptus robusta 8-10 Terminalia brassii 25-28 Octomeles sumatrana 30-35 (Erima) Tectona grandis 5-10 (Teak) Ochroma laRopus 25-35 (Balsa) Source: Forestry Department, various documents  - 25 - 3.43 To date, the principal motivation for government planting has been to maintain supply in areas where few alternatives are available: the obvious examples being the plymill at Bulolo, and in the Highlands Province. 3.44 Most private planting in PNG has been done by firms primarily concerned with logging natural forests, and it is reasonable to suggest their major motive has been the requirement to plant which was included in their logging contracts. An exception to this is the JANT company, a Japanese owned woodchipping operation in Madang Province, which are planting E. dealupta on a 10 year rotation to supply its existing plant. Whilst the density of E. deglupta is at the lower end of the acceptable scale, JANT considers that it represents a better pulp furnish than mixed tropical hardwoods, which are the current supply. The other companies involved in planting do so with varying degrees of willingness, but all seem to have an open mind on the prospects of the hardwood being grown to produce chips, and ply and timber in longer rotations (20 to 25 years to achieve a minimum diameter at breast height of 50-60 cm). 3.45 Both the government planting program, and private plantation invest- ment, have to date been severely constrained by one major problem in PNG: the availability of land. A particular case in point is the program undertaken by GOPNG at Ulabo, Kerivat, Hadang and Kaut, with the assistance of the New Zealand bilateral aid agency. These plantations have had major setbacks due to fire and other factors, and it appears that misunderstandings between the government and landholders on the nature of the lease agreement or of the terms and conditions for payment and compensation have been major problems. A review of this project by the New Zealand government was under way at the time of the TFAP Mission and will identify the nature of difficulties encountered. In general, leasing land for plantation purposes from customary landowners in PNG has been fraught with difficulties: (a) in some custom laws, planting a tree on land is regarded as a claim upon it, and in this context the idea of a lease is not clearly understood; it is difficult for landowners to comprehend that owner- ship of trees and land can be separate; (b) even when conceptual difficulties have been overcome in lease negotiations, there have been what appear to be excessive delays in processing leases (and thereby, lease payments), leading to renewed landowner dissatisfaction. 3.46 These major problems (which in fact represent a set of individual negotiating difficulties in obtaining customary land for plantation purposes in PNG) have become so severe that in some areas landowners who have been involved in the process are actively discouraging neighboring landowner groups from entering into negotiations to lease land for plantation. 3.47 Given these difficulties, it is hardly surprising that the bulk of private planting in PNG has so far taken place on government land--purchased, for the most part, from landholders prior to Independence in 1974. Presently, there remain some 10,000 ha of such government land immediately available for plantation in East New Britain, an- a similar amount exists in West New Britain. Also, in these provinceb, about 90,000 ha of government land currently earmarked for agricultural development exists.  3.48 In the short term, therefore, considerable areas exist on government owned land which could be planted to forest species or used for the develop- ment of tree crops. Much of this land offers exceptional plantation potential: the topography is flat to undulating; the soils are volcanic, deep and highly fertile; the climate is basically tropical, with rainfall distrivu- ted evenly through the year; the latitudes are above the cyclone belt; and, in the New Britain provinces, infrastructure is basically good. It is not possible at this stage to comment accurately on the degree to which forest plantations would need to compete for land with other crops, in areas such as New Britain. According to the World Bank Agricultural Assessment Review (World Bank 1988), there are currently some 19,500 ha of oil palm on West New Britain. Oil palm outputs (i.e. palm oil and palm kernel) are projected to grow by 121Z from 1988 to 1995. Even if this occurred, it would seem from observation that considerable land--especially the more undulating land which would be more suitable for forest plantations than for the export tree crops- -would still be available for forestry, if investments in it could be justified. According to the Agricultural Assessment Review, oil palm currently generates a net present value of K 3,123 per ha at 121 discount. 3.49 Whether or not forest plantations could generate comparable returns would depend largely on the purpose to which the output is used. Madang already has a ;oodchipping operation, and there would appear to be no tschnical reason why more such plants--perhaps integrated with sawing or ply operations to utilize larger sized material--could not be created. If this were done, then the returns to the whole growing/processing operation may well be quite high. The outcome of industry feasibility studies, as suggested in paras. 7.45 to 7.49 of this report, will be important elements in analyzing this option further. 3.50 There are technical difficulties in the existing plantation program: (a) The costs of plantation were estimated by some of the private companies involved to be between K 1,000 and K 2,000 per ha, exclu- sive of road construction and clearing costs. These cost levels seem excessive. The government estimates its reforestation costs at K 1,000 - K 1,500/ha, inclusive of site preparation and roading. This would seem a more reasonable figure. (b) There is a shortage of expertise for reforestation at the project management and technical levels. Many of these positions in the private sector are filled by expatriates mainly from Japan and the Philippines. (c) Even on government lands, fire from surrounding subsistence farmers is a major hazard for plantations, given its widespread use as a land clearing agent. Except for the major plantation area at Bulolo, the TFAP Mission saw little evidence of fire equipment, detection facili- ties and emergency procedures in place. (d) The level of research into plantation forestry is inadequate in PNG. In particular, many permanent growth and yield plots appear to have been lost or abandoned, and relatively little is known about growth  - 27 - rates. What is known suggests that growth in plantations in PNG- -especially in the hardwoods--is lower than might be expected from such sites. Virtually no selection or tree improvement has been done, and investment in this area would undoubtedly produce early and significant improvements in plantation performance. 3.51 Table 15 below reproduces the GOPNG Department of Forests plantation development strategy: Table 15: PLANTATION DEVELOPMENT STRATEGY, 1986-2000 Annual Planting Targets (1,000 ha) Year Conifer Hardwood Total 1986 15.8 15.5 31.3 1988 15.9 20.1 35.9 1990 18.6 29.6 48.2 1995 21.5 52.0 73.5 2000 25.0 75.0 100.0 3.52 It is evident that, to achieve this level of planting, or anything close to it, will require solution of the major technical problems noted above, and also substantial progress with resolving the basic land availabi- lity problem identified earlier. D. Resource Assessment: the Need 3.53 The first estimate of commercially accessible volumes in PNG was made in 1946. The criteria used for these estimates is not known. It is probable that they assumed large blocks of continuous forest suitable for large-scale logging operations. 3.54 Since 1946, some re-estimation has been done, but this has not included measurement of increment in logged areas, nor volumes of trees left out of harvesting. Relatively little is known in PNG about the regenerative capacity of the natural forests, in terms of log size and species composition. There is also little known about the rate of deforestation prior to and subsequent to logging operations, due to natural causes and man. 3.55 The relationships between forests and man are especially important in PNG, given the heavy reliance of traditional farming on forest cover, and the rapid rates of increase of population in the country. Information on population concentrations is available from PNGRISs the Papua New Guinea Resources Information System. This is a microcomputer based geographical information system developed in the Department of Agriculture and Livestock, by the Australian CSIRO. The rate of development and demand for land and the expansion of urban areas and the projected development of estate crops can be identified with the PNGRIS system. These data have not been used to moderate the estimates of operable forest but should be clearly utilized in any future  - 28 - resource assessment. PNGR.S holds considerable data on forest disturbance/ usage and parameters for commercial and subsistence crop species. It needs to be developed to provide advice on alternative land uses. 3.56 Air-photo cover dates from the 1940s onwards and many areas need to be covered once more, in particular, the urban centers, agricultural develop- ment sites and forests which have been or are currently being logged. Ultimately, a total photo cover of the country is needed and would benefit both the public and private sectors. 3.57 Clear definitions of the forest types will be needed and as a first level those areas containing commercial volume must be identified. Also those forests which are immature or logged over, but are potentially commercial, must be included. The size of these units needs to be specified since small blocks of commercial forest can be assigned to small production industries serving domestic markets. 3.58 A considerable amount of information on the forest resource is avail- able. This data exists in particular in the CSIRO records, Department of Forests archives ard records. Resource inventory methods have changed little since the 1940s and re-examination of all this data, combined with resource information available in PNGRIS, might establish a useful data base, together with new forest mapping, for a national forest inventory. 3.59 There is a definite need to improve resource inventory methods and to base these inventories on appropriate stratification of the resource. Such stratification requires adequate and suitable mapping procedures. A national forest inventory should include an estimate of the biomass as well as the commercial volume available. Resource inventories for management purposes need to include stand table data from at least the 20 cm diameter class and upwards, and also information on the silvicultural condition of the stand. 3.60 It is necessary to monitor the growth and development of forest stands following logging to provide data for the management of these stands. The development of growth models is essential for estimating the effects of alternative silvicultural treatments and monitoring stand development for management purposes. Whilst permanent sample plots can be established throughout the country in logged forests there are problems of relocation, maintenance and remeasurement. As a first level of continuous forest inventory a limited country wide distribution of plots, on sites which repre- sent the range of potential productivity, is needed. 3.61 To estimate volume from inventory data there is a need for tree volume models and it will be necessary to group tree species according to their forzn or stem taper in order to reduce the number of these tree volume models to w^anageable proportions.  - 29 - IV. MAXIMIZING RETURNS TO THE FORESTRY SECTOR 4.1 An important factor in sustainable forest management is the degree to which forestry rewards both the nation as a whole, and the owners of the resource. There are five basic issues involved in this: (a) Whether existing forest pricing policies correctly value the resource, ensuring that owners receive an equitable share, that users receive the appropriate signals en efficiency of utilization, and that resource managers can spend sufficient on management to maximize output under sustainable yield. (b) Whether payments to landholders are correctly distributed through time to ensure a stable and satisfactory relationship between governments and owners of the resource. (c) Whether the means by which logs are sold are efficient. and achieve the economic, social and environmental objectives of GOPNG. *d) Whether the market for output is efficient and allocates an equitable share of the final value of products made from the forest to PNG. (e) Whether further gains to national income can be made by altering the form in which the product is sold from PNG, from less processed to more processed. It will be argued below that GOPNG is justified in pursuing two objectives in the forest revenues areas si) the general per unit revenue level should rise; and (ii) the efficiency of revenue collection needs considerable improvement. A. Government Revenues. Royalty Policy and Rents 4.2 GOPNG revenues from forestry since 1985 are summarized in Table 16. Until the early 19709, expenditure on PNG forestry always substantially exceeded the revenue generated. Following the rapid increase in international prices for tropical hardwoods, the growing volume of PNG log exports (espe- cially after 1979), the introduction of log export taxes and an upward revision of royalty rates in 1981, the revenue generated by the forest sector began to grow substantially. While the Department of Forests expenditure in the last three years remained largely unchanged at around K 5 million p.a., the combined proceeds of royalties and export taxes reached K 12.3 million in 1986, K 15.9 million in 1987 and K 19.0 million in 1988. Even after deducting the landowners' share of royalties (75 percent of the total royalties) included in the above totals, forestry in PNG has clearly become 'self-finan- cing'. Furthermore, additional resources are mobilized by the Government not only in cash but in kind as timber permit holders are contractually obliged to undertake various social infrastructure developments in their areas, thus relieving the Government of the corresponding budgetary provisions. As a  - 30 - broad estimate, such Indirect budgetary support averaged between K 3 and K 6 per m3 of exported logs in 1988. Given the total 1988 log exports of about 1.3 million m3, the Government received an indirect additional budget support of about K 4 to K 8 million. The total forest tax revenue broadly defined of about K 25 million in 1988 amounted to approximately 4.5 per cent of the total tax revenue of the country. After excluding the in-kind social infrastructure levies and the share of royalties paid to landowners, the 1988 total of about K 12 million represented about 2.5 per cent of the tax revenue total or about 6 per cent of taxes on international trade. Table 16: SOURCES OF PNG GOVER9Y#NT FORESTRY REVENUE (K million) Source of revenue 1985 1986 1987 1988 Royalty 5.1 5.3 /a 5.2 6.7 Log export tax 5.8 7.0 10.5 12.0 Other lb - 0.2 0.3 Total 11.9 12.3 15.9 19.0 la The figure in DOP ComDendium of Statistics 1986 differs from that given in the DOF Annual Report Statistics 1987. Here the former is used. lb Profits of the state purchase option. 4.3 Under the proposed changes in the structure and levels of forestry charges contained in draft forestry legislation, discussed later in this report (and assuming no log export restrictions and unchanged log prices), revenue would be likely to increase to about K 50 million by 1992. Assuming further that the country reached what this report regards as a sustainable level of log output of 3.6 million m3 in 1996, the revenue would reach around K 110 million at current prices. This would amount to approximately 10 per cent of the projected total tax revenue total. 4.4 The existing pattern of taxes in PNG forestry has three main components. i.e., royalties, export taxes and indirect charges. The corporate tax has until now played a relatively unimportant role. 4.5 Royalties are payable for true log volume and structured around a basic rate with species premia and a pattern of allowances reflecting various factors affecting the cost of production. The current basic rate of K 5.41/m3 has remained unchanged since 1981. Prior to that, the rate was K 3.181m3 from 1973 to 1976 and K 3.95 m3 from 1976 to 1981. Premia on higher value species range from K 1.4013m on, e.g., Intsia, Palaauium, or Red Planchonella, to K 9.0/m3 on New Guinea rosewood (Pterocarous indicus). The royalty allowances, graduated and subtracted from the basic rate to yield the effective royalty, attempt to compensate for variations in the access to the market, climate,  - 31 - stand quality, haulage distance, total resource volume and topographylsoils. They range from zero in the most favorable conditions to a total of K 3.85 per m3 in the most adverse conditions. 4.6 The royalty payments are collected from log producers by the national government which then remits 75 per cent of the total to the resource owners and 25 percent to Provincial governments. This 7525 division of royalty superseded a 50s25:25 division to Provincial Government, landholders and National Government respectively in 1986. 4.7 The 1981 Royalty Appraisal Guidelines stipulate thas royalties apply only to logs from the natural forest suitable for sawn timber and veneer production (but not for chipping only). There is no formalized method for royalty calculation on plantation grown forests. 4.8 The Minister's Drafting Instructions for a new Forestry Act (see para. 6.18 and Annex 9) propose to replace the present system by one which: (a) makes royalties a function (10 percent) of FOB price; and (b) gives all of the royalty income to landowners instead of the current 75 percent while compensating Provincial government through an additional levy. 4.9 Export taxes are payable to the national government. The export tax has been levied in recent years at 10 percent of the FOB value of exported logs. Following the adoption of the minimum export prices (MEP) system in 1982 (see para. 4.46), the MEP often became in practice (though not in theory) the basis of the export tax calculation. With the reduced importance of MEP after 1986 (resulting from the introduction of the State Purchase Option on exportable logs), the link between export tax and MEP has weakened, and the export tax, both in theory and practice, has been levied on actual FOB price. The Minister's Drafting Instructions propose to increase the rate of tax to 15 percent of the FOB prices of logs while exempting processed wood exports from the tax. 4.10 Indirect taxes comprise various obligations placed on logging companies by the Government in the course of timber permit negotiations. As noted in para. 4.2 such obligations typically relate to items of social infrastructure (roads and bridges of certain specification, community facilities, etc.) not required by the company for the strict purpose of logging. No systematic approach to specifying the extent of such obligations exists and, instead, a rough estimation prevails, based on the authorities' past experience of dealing with companies and applied on a case-by-case basis. More recently, in new timber permit negotiations the Department of Forests has attempted to estimate each applicant company's future profitability and then to use the social infrastructure obligation as a balancing item to secure for the Government the maximum amount of resource rents. 4.11 A nominal fee of K 0.215/m3 of logs is payable by logging companies to support the Forest Industries Council (see paras. 4.79 to 4.80 for discus- sion of FIC). 4.12 Finally, although companies operating (and incorporated) in PNG are subject to a 36.5 percent corporate tax, until 1986 no logging company is said (Jackson (1986)) to have paid any tax on declared profits or declared any profits.  - 32 - 4.13 Ideally, resource charges should extract the full amount of economic rent - that is, the difference between the market value of the resource and its costs of production and sale, adjusted for normal profits, under competi- tive conditions. This concept has a vital role to play in reforestation and the larger issue of sustainable production which underlies much of the argu- ment in this report. If a resource pricing policy drives project returns below the opportunity cost of capital over the medium term, the extraction firms will make no long term commitment to forest operations. If, on the other hand, the policy appropriates too little rent, then the resource will be undervalued: utilization will then be too rapid, or incomplete, in an envir- onment generally tolerant of inefficiency. Obviously, under such conditions, returns to the owner of the resource will be lower than desirable. 4.14 Two observations on existing resource pricing in PNG are of relevance at this juncture. First, rates of royalty have remained unchanged for the past eight years, while log prices on the international market have fluctuated extensively. Second, royalty premiums on some species in PNG seem unrelated to the price differentials these species attract in the market. For example, Intsia commands a price difference over average grade PNG logs of K 60/O3 in the market, but the royalty premium for this species is only K 1.40Im3. Since the cost of production does not vary greatly among different species, the existing system encourages 'creaming' of the mote valuable species as well as resulting in a loss of revenue for the resource owners. The proposal in the Minister's Drafting Instructions to make royalties a function of FOB price partly recognizes the above weakness without, however, overcoming it. To use a simplified numerical example: if Intsia sells for K 120/m3 and other species command a price of K 58/m3 and if the logging cost is K 50/m3, a royalty of 10 percent will leave the logging firm with a profit of K 58/m3 on Intsia and K 2.20/mr on other species. 4.15 The log export tax, in PNG as in most other log exporting countries, is also related to the FOB price, and suffers froA the same defect of mnderpricing ('undertaxing') the relatively more valuable species. In general, taxing a heterogeneous resource in proportion to the market prices will typically leave the higher grades underpriced (undertaxed) unless the cost of their production is well above that of the lower grades: this has generally not been the case in PNG. 4.16 The Minister's Drafting Instructions' proposal to make royalties proportional to FOB prices appears to have been motivated more by a desire to safeguard the landowners' share of resource rents during a period of rising log prices, than by efficiency considerations. Linking royalties to FOB prices is a double-edged sword and the administrative simplicity of the proposed revision carries with it a potential political risk of exposing resource owners to the consequences of market instability, from which they are at present shielded. In efficiency terms, making both royalties and export taxes a function of the same variable (FOB prices) effectively reduces two taxation instruments to one and lowers the system's flexibility. A structure which uses royalties to target price differentials among resource grades Clog species) and uses FOB price-related export tax to incorporate variability of average prices over time is clearly superior.  - 33 - 4.17 There is considerable variability in the efficiency of log producers in PNG, because of differences in the scale of operations, the quality of forests, logging conditions and the quality of firms' management. This presents the Government vith the problem of deciding which group of figms to take as a reference point for tax purposes. If the taxation policy is geared towards the least-cost group of producers, efficient rates of taxation will result in all of the higher-cost firms going out of business. Setting the tax rates with an eye on the 'average' firm, on the other hand, will give part of the pure profit to the more efficient among the concessionaires. 4.18 The existing system of levies recognizes the problem to a degree and attempts to cope with it through royalty rate allowances. However, the structure of allowances appears to be insufficiently related to underlying economic factors. Differences in production costs on account of physical and locational (rather than management-quality) factors are substantially greater in PNG than the maximum combined allowance of K 3.85/I3. Even if a good part of production cost differentials is due to the varying quality of the firms' management, existing royalty rate allowances in themselves seem insufficient to compensate for other sources of cost variability. As a result, the current tax structure favors companies operating in more favorable locations (unless the cost advantages to such companies are fully offset through the imposition of additional charges for social infrastructure). 4.19 These Indirect levies on logging companies have been increasingly resorted to by tte Government both as a means of compensating for incomplete appropriation oi pure profits as well as a mechanism for providing infrastructural .Acilities in remote parts of the country and, through it, facilitating the process of timber tights purchase from landowners. The system of preparing a de facto re-run of the company's profitability estimates at the time of timber permit negotiation and using the results to gauge the firm's ability to bear additional cost of social infrastructure is understandable but not necessarily efficient. First, the forest companies' performance as general contractors is untested. While, in a majority of cases, there may be no alternative civil work contractors (or the cost of their mobilization would be out of proportion to the value of the contract), the automatic selection of timber companies for non-forestry tasks is questionable. In the worst case, the taxation regime may unwittingly compound the problem of inefficient rent appropriation in logging by the support it gives to inefficient building of, say, schools or bridges. Also, the use of an indirect infrastructure levy obscures the financial conditions under which different firms operate since it is difficult to clearly relate different infrastructure packages to the value of the forest resource rents. 4.20 As will be apparent from discussion in para. 4.18 above, the effectiveness of the taxation system in force in PNs is further compromised by the inconsistency of conditions applied to logging ccntractors under the different types of logging agreement in force in PNG at present. 4.21 In summary, the process of regulating access to the country's forest resources combines elements of national taxation policy applicable to all operators at the same level (e.g. export tax), with those which, though applying to all operators, may apply at different rates (e.g. social infrastructure obligations) to, finally, those which are specific to the  - 34 - nature of the agreements used and which in some cases, as will be argued below, may not even be known with precision by national or provincial forest authorities. When reconsidering royalty and taxation systems in forestry, some general principles must be kept in mind: (a) it is not only the levels of royalty and tax systems which determine the level of revenue flowing to owners of the resource: the method of sale is also important. The amount of merchantable timber removed from a forest will, in large measure, be determined by whether a logging contractor has paid for all timber measured as merchantable, or only actual volumes removed; (b) a royalty system based on specific allowances for regional or site disabilities will, by definition, be difficult to administer and to keep up to date; (c) especially in the case where a wide range of species and log qualities exists, careful attention to, and monitoring of, pricing differentials is needed to ensure that the right signals (as defined in the management plan and objectives) for that forest are passed on to the logging conzractor. 4.22 If the existing system for administratively setting forest charges is inadequate, one alternative might be to resort to an auction system for sale of the resource. Under truly competitive conditions, a resources auction system would extract the maximum economic rent. However, two major difficulties would apply in the specific case of selling the PNG forest resources. 4.23 First, the well-known heterogeneity of PNG forest resources and the range of topographies on which commercial forests are found considerably increase the cost of obtaining accurate information about the resource. With the object of bidding (i.e. the forest) being a rather 'imperfect commodity', many potential bidders might be discouraged by the high cost of data gathering and evaluation. The cost could be reduced substantially only if potential bidders, resource owners and the Government could agree on a single independent surveylinventory firm whose results would be accepted by all or most parties. 4.24 In PNG the question of confidence also extends to the question of ownership of forest land. In existing circumstances, where the ownership of resources in PNG has been a frequent cause of serious dispute in forestry ao in other sectors it is doubtful that companies would be willing to pay in advance for the right to harvest the forest in the absence of government ownership of land and a guarantee of continuity of operations. 4.25 The second difficulty in auctioning the resource is the complexity of managing tropical forests to maximize the value of the residual stand. In some cases, in PNG as elsewhere, certain forested areas are going to be converted to agriculture or other uses, and in such circumstances (i.e. for the purpose of a forest clearing operation) an auction may be effective in maximizing returns to the resource owners. Where sustainable yield operations are intended, however, there will frequently be a conflict between the  - 35 - interests of the resource owner, who will wish to maximize long-term resource rente, and logging companies seeking to maximize short term rents. A demanding supervisory or environmental auditing procedure would be needed in such a case, to ensure that the residual stand after logging was sufficiently viable. To sum up: whether or not an auctioning system will maximize forest resource rents, its implementation alone will not ensure sustainability of natural forest operations. B. Methods of Sale of Wood 4.26 Annex 8 to this report outlines the major mechanisms by which logging rights are issued in PNG and provides information on allocations made under four types of agreements: (a) Timber Rights Purchase. Permits or licenses are issued by GOPNG to a forestry development company, to grant cutting rights over a project area, which can be on land held under lease by Government, or on customary land under a timber rights purchase area. They are granted subject to compliance with approved working plans and environmental conditions, and compliance with these is the responsibility of the permit holder. (b) Declared Local Forest Areas. LFAs are issued under the Forestry (Private Dealings) Act. on approval of the Minister. The basic purpose of the LFA is to allow landholders to negotiate directly with logging companies to sell their resource, without the need to extensively involve central or provincial forestry agencies. (c) Timber Authority. TAs are granted to allow landowners to harvest timber for their own use. They generally do not exceed 5,000 m3 per annum. (d) Freehold Property Operations. There is no formal Government approval required for commercial logging on freehold property. 4.27 There are major problems associated with both the major forms of agreement: TRPs and LFAs. In the case of the former, compliance with working plans and environmental conditions has often been poor, and the resources available to forestry agencies to enforce these (see discussion from para. 6.2 on) are inadequate. There is also some confusion evident as to what the actual level of resources committed under TRPs in PNG is. This is because annual allowable cuts stipulated in the TRP are rarely adhered to by companies operating TRPs. So far as landowners are concerned (see the example quoted in para. 6.32) TRPs have gained the reputation of being bureaucratically cumber- some in PNG. It is clear that under such circumstances, revenues from log sales are unlikely to be maximized, and payments to landholders may be seriously delayed. 4.28 It is widely agreed in PNG that LFAs have created major environmental damage, through the lack of supervision inherent in the process of direct landowner--contractor agreement. The situation in New Ireland (see para. 6.4) has come about largely as a result of LFAs being issued with too few controls.  - 36 - C. Loa ExDorts and the Transfer Pricing Issue 4.29 At present, log exports dominate the revenue situation for PNG forests, so the condition of thle log market is crucial to overall forestry returns to PNG: Table 17: PNG LOG EXPORT VIOLUMES, VALUES, PRICES YEARS VOLUME VALUE PERIOD PRICES K/m3 Period 1,000 m3 1,000 Kina Average Minimum Maximum 1951-55 28.6 582.6 20.22 ll.U4 29.17 1956-60 17.5 329.2 18.81 20.58 26.47 1961-65 185.6 2,977.1 16.04 13.88 20.38 1966-70 982.4 11,903.4 12.12 11.41 12.75 1971-75 2,242.7 35,765.1 15.95 10.79 19-07 1976-80 2,411.3 83,627.9 34.68 20.35 48.59 1981-85 5,267.5 253,547.8 48.13 42.09 54.73 1986-88 4,096.6 266,909.4 65.15 53.47 68.02 Source: Compendium of Statistics 1986. 4.30 Key points include: (a) Log exports were minor in volume and value during the 1951-70 period, m%Aerate in 1971-80 and have increased rapidly since 1980. (b) Peak log exports were achieved in 1987 with 1.4 million m3 by volume and K 105 million in value. (c) Major fluctuations in price between years are common. (d) Although increases in price have generally occurred since the late 19709 this is not always the trend. During the 1950s to the mid 19708 prices were depressed and generally declined. 4.31 Annex 3 contains more detailed volume and price data for log exports from PNG. Among other things, these data show a lack of any consistent relationship between log export supply and prices, over monthly, quarterly or yearly periods. This suggests that the high variability in volumes entering the market has more to do with production factors than market conditions. 4.32 As is evident from data presented in Table 6, North East Asian purchasers - in particular Japan - dominate the market for PNG log exports. Although that table shows the Japanese share has been reduced from a peak of 80? in the late 1970s to about 60? now, the influence of Japan remains very strong. In fact, information gathered from the Commission of Inquiry suggests that 80 of PNG log exports may still be contracted by Japanese interests, albeit indirectly in some cases.  - 37 - 4.33 Such market concentration raises the p-ospect of collusion and transfer pricing, particularly given the close-knit and complex inter-agency relationships that characterize the Japanese log market. 4.34 Ia fact, there is now a great deal of concern in PNG about the matter of transfer pricing and the bibliography to this report lists a selection of the voluminous literature on this subject. The issue has been a primary concern of the Commission of Inquiry. The Commission has identified the following mechanisms by which transfer pricing is practiced: (a) misdeclaration of species; (b) downgrading of log grades; (c) unders_aling of volume; (d) under-valuing log prices; (e) over-valuation of external management, machinery, lease and consultancy costs; (f) mistallying of ship loading (sinkers, lost logs, rejects, etc.); (g) overstated freight rates; (h) overstated claims for damages; (i) nil profit accounting; (j) third country invoicing (double invoicing); (k) complex distribution networks (importers' subsidiaries); and (1) collusion by importers. An extreme case of transfer pricing was revealed on New Ireland 'where a Commission of Inquiry member intercepted a log export ship loading logs which were not documented at all. As a result (primarily) of the Commission of Inquiry's efforts, several of the overseas log importing companies or marketing agents in PNG are negotiating with the Taxation Department for back- payment of taxes which have been revealed for the period 1985-87. 4.35 From the policy viewpoint, two questions are of fundamental importance in relation to transfer pricing: (a) How significant is it, in terms of losses to PNG? (b) How difficult would it be to control? 4.36 Table 18 below provides some representative log export prices from Sabah, Sarawak and PNG:  Table 18s: FOB LOG PRICES, SABAH, SARAWAX AND PNG, 1985-89 (US$/m3) Sabah Sarawak PRG Super Low High Standard High Standard Low quality quality quality quality quality quality quality quality 1985 94 90 79 76 74 54 45 43 1986 112 102 95 91 88 62 57 49 1987 184 158 113 136 130 92 81 60 1988 159 137 110 123 119 88 78 51 1989 160 140 108 126 120 88 78 51 Source: Japan Lumber Journals January 1988 - March 1989. 4.37 Without further analysis it would not be appropriate to infer a great deal from teS absolute price differences which this table exhibits: PNG timbers are different, in species and quality, from Malaysian logs, and there are also transporting cost differences. 4.38 The figures do serve to dispel one argument encountered frequently in PNG: that PNG FOB prices have improved relative to major log suppliers since the introduction in 1985-1986 of improved market monitoring measures. Relative improvements in FOB prices over the 1985-89 period in PNG have, if anything, been slightly below those for Sabah and Sarawak over the same period. If, therefore, FOB prices in PNG at the beginning of the 1985-1989 period were already relatively low - due to transfer pricing - then this would remain the case now. 4.39 The Commission of Inquiry estimates that, for the period 1985-87, one transfer pricing technique detected (double invoicing mainly -:hrough Hong Kong) accounted for about US$10 per O3. Considering this as a minimum, and assuming that transfer pricing in some way waa exercised over the whole PNG log export industry, the total loss of exports in 1987 would have been in excess of $14 million (see Table 19).  Table 19: ESTIMATED LOSSES FROM TRANSFER PRICINi, 1985-87 (BASED ON COMMISION OF INQUIRY ESTIMATES) Year Log export Transfer price Lost overseas earnings 1,00 ma3 US$/m US$1,000 1985 1,158 10 11,583 1986 1,314 10 13,138 1987 1,442 10 14,422 The loss of revenue to the PNG Government is about 1OX (export duty) of this. Waisime (1987) studied the dimensions and implications of transfer pricing on PNG's log export trade. Waisime had access to a broad data base within PNG, Japan and to trade journals for competing countries. Waisime assumed that PNG (Mix) logs compared directly with Low Quality Brand (LQB) from Sabah (NB: PNG logs of average grade were considered better than Sabah, Sarawak and Indonesia LQB). 4.40 Accounting for the differences in freight and scaling "outturno Waisime estimated transfer pricing losses as given in Table 20. Table 20: ESTIM&TED LOSSES FROM TRANSFER PRICING, 1982-85 (BASED ON WAISIME) Year Export volume Transfer price Lost overseas earning 1,000 m3 US8/m3 USS million 1982 1,063 19-34 20.2 - 36.2 1983 1,019 20-49 20.4 - 49 9 1984 1,284 18-68 23.1 - 87.3 1985 1,158 14-24 16.2 - 27.8 Despite the flaws in Waisime's assumptions that the Sabah (LQB) and PNG (Mix) fluctuate in unison there is reason to question the differential between the price expected for PNG logs and prices achieved. Part of the difference in price could be the preference of the end users for Sabah logs rather than PNG (e.g. continuity of supply, majority market share, consistent species mix, etc.).  4.41 An alternative method to estimate the level of transfer pricing is to derive notional FOB prices for PNG logs by deducting margins and costs from wholesale prices for PNG ex log pond in Japan, and comparing these to actual FOB prices achieved. The data and assumptions area in this exercise are described in Annex 4. Table 21 below gives results from these calculations. Table 21: ESTIMATED LOSSES FROM TRANSFER PRICING, 1985-88 (EX JAPAN WHOLESALE PRICE BASIS) Year Export volume Transfer price Lost overseas earning m3 US$/m3 USS million 1985 1.063 20 21.3 1986 1,019 11 11.2 1987 1,284 23 29.5 1988 1,158 22 25.5 4.42 While the estimates of losses vary consideraAly, transfer pricing has clearly been, and apparently remains, a significant drain on the resources of PNG. Thus the need for measures to correct it is quite clear. 4.43 Transfer pricing does not take place evenly across the gamut of exporters: there are many independent PNG log shippers who depend upon marketing agents for sale of their logs in the purchasing country. These shippers do not benefit from transfer pricing. Introduction of a uniform export tax or similar measure, calculated on the basis of average transfer pricing losses, would therefore impact differentially on the profitability and viability of exporters in PNG, and independents would encounter more problems than those linked to agencies in purchaser countries. 4.44 The basic problem is that quantifying transfer pricing is, by nature, an uncertain business: essential data are hardly likely to be willingly made available. Detection and enforcement will therefore be time and resource consuming. 4.45 Other options have been tried in PNG: a minimum export price scheme, set to capture at least some proportion of rental losses; and btate purchasing. 4.46 The Minimum Export Pricing System was introduced in 1982, as a means of setting minimum prices for logs, grouped into various species and grading categories. The MEP is reviewed monthly by a committee of representatives from the Department of Forests, the Forest Industries Council, and the Depart- ment of Trade and Industry.  4.47 In practice, the MEP has not achieved any significant effect on under-pricing, which was its primary objective. Instead of operating as a minimum or floor price, it has tended to be regarded as the maximum price. It merely reflects recent offer prices, because the system hcs no access to quality market intelligence data. Moreover, the MEP in itself offers no remedy to the problem of misdeclaration of species, and other practices listed in para. 4.34 above. Finally, GOPNG has now introduced a State Purchasing Option and it is obvious that, with this measure in force, the MEP is now superfluous. 4.48 The State Purchasing Option was introduced in 1986, with the objec- tive of allowing GOPNG to participate in the marketing of PNG timber, and thus to gain an appreciation of the market realities. The option allows GOPNG to acquire up to 252 of a timber concessionaire's export volume on a first refusal basis, and then to on-sell that acquisition in the market. 4.49 Originally the SPO was administered by the Forest Industries Council, a body set up under an Act passed in 1979 with the objectives of bringing private sector and government representatives together to promote and develop forest processing (see paras. 4.79 to 4.80 for further discussion of the FIC). The FIC has suffered from basic inflicts of interest, in that strong support by the log export sector did not engender diligent attention to policing the sector's marketing practices. In 1987, the Forests Department took over the State Purchasing Option. 4.50 In theory, the concept of the SPO is appealing: it provides the government with direct market intelligence, which can be used to monitor proceeds from sale in the private sector; and it avoids the dangers of total acquisition and sale of logs by the state: a model which has not generally worked well in Asia, and for which the precedent in PNG--in the fishing industry--proved to be a costly error. 4.51 In practice, the SPO has suffered from a lack of resources and commitment in its operation to date. The SPO may have succeeded, since 1987, in maintaining the relativity between PNG and Malaysian log prices, but in effect this has really only amounted to allowing the pre-existing levels of transfer pricing practices to continue in PNG. As currently authorized, the SPO cannot operate on exports from Timber Authority Area or Local Forest Area licences (see para. 4.26), only from Timber Rights Purchase areas. The current SPO operation works out of a small unit within the Forests Department: there is no capacity to man the office on a 24 hour basis; no computer recording and monitoring facility; no ability to offer significant services to shippers; and very little capacity to research and develop interesting new market possibilities for PNG logs, such as India and Hong Kong. There has been some indication that some log shipping interests inimical to the SPO have manipulated species, grades and even shipping schedules to bring the SPO into disrepute in purchasing countries. Another complaint made by smaller log exporters is that the SPO was not operating in different market areas, and that the impact on their credibility (in having to withdraw shipments offered for sale, to supply to SPO) was not worth the (marginal) improvements in price obtained.  4.52 Despite these difficulties, the SP0 could be made effective in PNG, given increased political and administrative support. It offers a relatively simple means of improving the Government's knowledge of the market, an: of rendering transfer pricing more transparent. Several measures are needed to bring this potential to fruition: (a) the SP0 needs to be re-structured along commercial, self-financing lines, rather than as an undifferentiated adjunct to an administrative organization; (b) the SPO should be given a clear incentive to diversify and improve markets; (c) the percentage acquisition from concessionaires under the scheme should be expandable beyond 25Z, so that the private sector's incen- tive to outperform the SPO becomes, and remains, strong; and (d) as an initial accelerator to the SPO's performance, an association with an international marketing firm (preferably to be contracted on a profit-sharing, rather than a fee-for-service basis), would be advantageous. D. Forest Industries Development 4.53 At the time of the TFAP mission's visit to PNG, considerable interest was being exhibited in the development of the forest processing sector. Yet the recent history of forest industries in PNG is a rapid decline. This report will argue that the first requirement in encouraging appropriate industry development is the implementation of sound policies. 4.54 Table 22 below summarizes the historical pattern of forest product export developments in PNG. Annex 3 contains details of forest product outputs, exports and export destinations.  - 43 - Table 22: PROCESSED FOREST PRODUCE EWPORTS, 1951-88 Volume (000 S), Value (000 K), Price K/m8 limbeo Pli/Veneer Wood Chip Other Yeare Vol Vol Prc Vol Val Prc Vol Vol Prc Vol Val Pre 1951-S6 28.8 1,187 50 20.5 8,014 147 - - 1956-60 48.1 2,878 55 60.0 10,480 174 - - 1961-65 61.6 8,298 64 48.9 9,498 194 - - 1906-70 76.1 5,256 70 65.6 12,842 189 - - 1971-75 168.1 4,426 86 77.0 14,267 188 264.8 4,584 17 1.8 98 76 1976-80 246.9 28,676 117 46.8 14,551 814 872.0 25,098 29 26.5 5,298 116 1981-86 97.7 16,1 170 26.8 11,882 431 826.1 28,574 86 10.0 2,296 280 198e-88 18.8 8,128 285 0.2 189 695 298.0 17,974 60 0.9 248 276 Sourcee: Compendium of Statistics 1956, Timber D1gate 1986-68, Departmnt of Foreate 4.55 Table 22 clearly demonstrates the downturn in output from the PNG saumilling sector after 1979, when legislation preventing export of raw material was liberalized in PNG. 4.56 The current domestic market consumes about 150,000 m3 p.a. of locally produced timber, and 18,000 m3 p.a. of plywood. Total sawmilling operations in PNG have a combined log import capacity of 500,000 m3 p.a.t i.e., about a third of the current annual log harvest. The number of sawmills in operation is currently about 50. Of these, 20 have a log input capacity of more than 20,000 m3 per annum, and these large mills account for 60Z of total sawmill log input in PNG. Currently, 'opan, Australia and New Zealand take a third each of the exports of sawn tiuaber from PNG. There are also a number of joinery factories, mainly owned by building companies, in PNG. There are seven furniture manufacturers (of commercial scale). 4.57 A joint Government-private sector plywood mill (Papua New Guinea Forest Products) produces about 15,000 m3 of plywood per annum, based on the Araucaria spp. resource in the Lae-Bulolo area. PNGFP aimed to sell most of its ply and veneer production in the export market, but this declined in the face of increased Indonesian exports, and the mill has ceased production of veneer, and sells all of its plywood on the domestic market, which is protected.  - 44 - 4.58 Woodchips are manufactured from rainforest material by JANT, a Japanese company based in Madang. In 1977 the m11 was producing 193,900 m3 p.a.: this has declined to about 100,000 m3 at present. A second woodchip plant, constructed by Nam Yang Timbers at Kapuluk in West New Britian about 10 years ago, has never operated. Japan presently takes about 762 of woodchips exports from PNG; Taiwan takes the remainder. 4.59 Rattan is the only other forest product which is exported from PNG and that is only in experimental quantities to the Philippines at present. The export of unprocessed sandalwood was banned in the early 19809, and a plant manufacturing chopsticks was burnt down in 1983, and has not been rebuilt. 4.60 At present the forest industries of PNG employ about 5.000 people, with an annual wage payout in the order of K 15 million per annum. 4.61 In Hay 1989, in an attempt to reverse the decline of the processing industry, the Minister for Forests announced the following proposals: (a) A total ban be placed on the imports of sawn timber and mouldings. (b) The present five year quota system be maintained to control the imports of plywood. (c) Import duties on furniture be increased from 30Z to 50? with a view to placing a total ban pending a survey to determine the capacity of the furniture industry to meet local demand. (d) Companies involved in the import of components for pre-fabricated houses must buy at least 502 of their requirements locally. (e) That it be made mandatory for all new buildings to be built of at least 40? of local materials. (f) Tax exemptions be allowed on equipment and other materials for downstream processing. (g) Amend the merchant shipping act to enable coastal transportation of products by barge. (h) Tax holiday for 10 years for those companies processing a minimum of 502 of their annual log harvest. (i) Allow variable export duty on exports so that companies that process more timber pay less duty and vice versa. (j) Appoint marketing agents to assist the industry to sell sawn timber and other processed products. (k) Upgrade practical training to develop the industry's manpower. (1) Establish an export oriented timber processing complex with the assistance of reputable overseas companies as an incentive.  - 45 - (m) Resolve the current status of the Forest Industries Council (see paras. 4.79 to 4.80) and the Forest Industries Association In order to have an effective autonomous organization to promote and develop the PNG timber industry. 4.62 It is not clear at this point to what extent these proposals are being implemented. Their intention, however, is clear enough: protection of significant sections of the domestic forest product markets for local suppliers, and stimulation of processed exports. 4.63 The most comprehensive measure suggested is that of log export bans, and this is addressed first. In the Minister's Drafting Instructions for the new Forestry Act (see Annex 9), considerable powers are foreshadowed to allow the Minister to make a determination of the level of log output that can be exported in unprocessed form from PNG. At the time of the TFAP mission, serious consideration was also being given to imposing a log export ban, and a ban on export of some premium species is already in force. In assessing the merits of such a policy, consideration needs to be given to two matters. First, the conditions applying in the existing domestic processing sector are important; specifically, the question arises of whether a guaranteed log supply will overcome whatever constraints on its competitiveness presently exist. Second, the general economic arguments pertaining to such policies are relevant. 4.64 There is little doubt that, as presently structured, the PUG processing sector is uncompetitive internationally. Some reasons for this are strictly cost related: (a) High labor costs, and low labor productivity. General labor costs in Papua New Guinea are higher than for most forestry exporting nations of Southeast Asia, at the current value of the currency. Most processors contacted believed the productivity of that labor to be below the equivalent in Indonesia and Malaysia. (b) High energy costs. No systematic data on energy costs in forest processing were collected. However, World Bank data suggest that diesel costs are higher in PNG than in Australia or New Zealand, and are unsubsidized. While high, therefore, they may be at the appropriate economic level. The same applies to electrical power in PNG. To become competitive, the industry will need to be able to absorb these cost levels. (c) High internal and external freight costs. Annex 1 details some recent freight rates for PNG. A calculation included in the Annex shows that the total cost of shipping and handling sawn timber from a mill in Western Province to a major port in Australia would be US$143.55 per m3. This is a very high transport cost indeed. 4.65 Some reasons for lack of competitiveness in the PNG industry are more related to policy and market factors (and are therefore more debatable):  - 46 - (a) A lack of com=unication between Government and industry on policy matters, created in part by a lack of continuity in Government policies and senior staff in forestry agencies. (b) A lack of knowledge of overseas markets and requirements. An example of this is that producers in PNG seem unaware of quarantine regulations in Australiat while some PNG timbers are known to be resistant to a certain borer, they are still excluded from entry to Australia because there has been no attempt to have them exempted. (c) A lack of trained personnel. 4.66 There are a range of other factors that affect the competitiveness of the PNG processing sector, as presently structured. The relevant question, however, is: Is it likely that the inefficiencies and impediments to competitiveness noted above would disappear, under a log export ban? What would occur if the industry suddenly became the recipient of large volumes of logs (presumably at highly subsidized prices, since only a small proportion of log supply is taken up by domestic producers at current export fob prices)? 4.67 If the primary objective of a log export ban is the stimulation of local processing, and thereby value added and national income, then the experiences of Indonesia and the Philippines are relevant. In Indonesia, it has been estimated that the cost to the Indonesian economy of the export ban imposed there has been somewhere between US$2 billion and $4 billion, between 1980-85. A similar calculation fur the Philippines suggests that the loss of revenue from banning log exports there has been in the order of US$140 million per annum in recent years. 4.68 The general case against imposing log export restrictions to promote wood-based industrialization is, briefly: (a) A log export ban isolates the local processing industry from the world prices of the raw material (logs) and typically imposes a revenue loss on resource owners (whether traditional owners, as in PNG, or the State as in most of South East Asia). If local industry can compete in the market for the final product only because it pays less than the full (world market) price for the raw material, the government is merely redistributing income from resource owners to the local resource-processing industry. If, in the interest of macroeconomic efficiency the raw material, though banned from exports, is to be priced at the full world market cost, the rationale itself for the ban disappears as log producers would gain no advantage by exporting logs rather than selling them locally. If, finally, the local processing industry is willing to buy less than the entire 'without-processing' quantity of logs (while paying world market prices for them) the resource owners are once again involved in a real or opportunity loss resulting from reduced access to the more profitable foreign market. (b) If the objective of a log export ban is to slow down the rate of forest extraction, the same objective is better achieved (in the sense of imposing a smaller cost on the resource owners or society)  - 47 - not through an extort ban but a production ban (e.g., reducing the rate at which forest is made available for logging). Under certain conditions (e.g., a likelihood of increases in future log prices, or a significant rate of resource depletion through unpriced practices such as slash-and-burn agriculture, etc.) a logging ban (but not a log export ban) may be the economically optimal way of managing the resource. (c) Variants of the same policy (most notably, in PNG, a ban on export of selected - usually high-value - log species) suffer from the same fun; atal drawbacks. Indeed, the real or opportunity loss to the res; .rce owner will tend to be higher the higher is the world market val%ie of the banned species. (d) The argument that a log export ban will increase export revenue, value added or employment (over the levels prevailing under no log export ban) is not a particularly strong one. Increasing the cost of the final product over its 'raw material equivalent" is not necessarily synonymous with increasing the value added. Moreover, value added (by use of more labor and equipment) can easily become value lost (if the raw material is used less effectively by local producers than by competing foreign users). In such circumstances also, the private (let alone social) cost of additional employment generation will tend to be high while remaining hidden from the policy makers and the public. 4.69 The policy of log export restrictions in its various forms is inferior to that of removing the (de facto) price incentive in favor of logging brought about by an inappropriate taxation regime. A policy which effectively gives, or promises to give, too large a share of the resource rent to log producers, contributes to the reluctance of such producers to engage in log processing. Adopting a log export ban as a means of ensuring that no log producer makes excessive profit would represent not only an admission of fiscal impotence but a case of mismatch between a policy objective and a tool. The recent practice of allowing log producers to export a 'supplementary quota' of logs related to the volume of logs locally processed is an indica- tion of the far greater profitability of log exports. One inefficient policy (excessively profitable log exports) combined with another inefficient practice (unprofitable processing) usually add up to two inefficiencies rather than one cancelling out the resource allocation distortions caused by the other. 4.70 In addition to carrying a substantial private and social cost, the policy of log export restrictions, conceived as an instrument of industrialization, has the drawback of obscuring its true cost and, in PNG, placing the entire risk of possible loss of income on resource owners. 4.71 An example of the ctsts which could be involved can be calculated from selling prices and processing costs collected by the mission in PNG. The example chosen was Pencil Cedar (genus Palaguim), which is a high grade timber from PNG:  - 48 - Value of log if exported K 125/m3 Conversion rate (2) 33 . Value (fob) of sawn output K 375Imx i Cost of processing K 60 + Cost of treatment 20 + Cost of drying 50