THE CURRENT STRUCTURE OF LICENSES AND THE STRUCTURE AND LEVEL OF FOREST CHARGES

This section describes the current structure and level of forest charges in Zambia and the total amount of revenues collected from forest charges. It then assesses the structure of the current licensing arrangements (already described in Section 3.3.2) and the structure and level of forest charges. The current arrangements are assessed in terms of the principles set-out in Section 2.

The structure and amount of official charges levied on forest product production in Zambia are set-out in SIs and licence agreements. These are periodically revised by the Forestry Department. In addition to these, there are also a number of other charges levied in the sector.

Official charges specified in Statutory Instruments and other legal documents

The first charge that a forest concessionaire has to pay is the Commitment Fee. Although the title would suggest that this is a fee for acquiring the concession or the licence, the licence says that the Commitment Fee has to be paid "within thirty (30) days of the new calendar year of operation" (Clause 6c) and it is later referred to as a Deposit Fee. The licence also says that the fee will be based on "average annual production capacity" (Clause 6c) but it seems to be set at K 200,000 (Clause 1h) irrespectively of the size of the concession or level of production.

With the exception of the export charge (already described in Section 3.3.3), all other official charges to the central government are set out in the charging schedules that are issued as SIs.

Once a concessionaire has acquired a forest concession, they then have to pay every month a fee for each tree that they plan to cut. These fees have to be paid in advance. The fees currently charged per tree are shown in Table 3 along with the fees set in 1994 (for comparison). As the table shows, the fees have increased by only 80% since 1994. This compares with an inflation rate of just under 200% for the period as a whole. Thus, in real terms, stumpage charges have gone down by about 55% after adjusting for inflation. In US$ terms, stumpage rates have also fallen by over 50% (and by even more, if inflation is taken into account).

According to discussions with Forestry Department staff and sawmillers, many concessionaires routinely declare and pay for the minimum number of trees (i.e. 200 in a sawmilling concession or 20 in a pitsawing concession). The concessionaire has to mark, measure and record all trees and timber logs taken in the month and supply this information to the Forestry Department within 15 days of the following month. They are then billed in the following month for any trees taken above the number originally declared. The upper limit on production is 600 trees per month and, if the concessionaire fells more trees than this, the excess trees are charged at twice the normal rate.

Once the concessionaire is ready to move their product from the forest, they then have to pay the conveyance fees shown in Table 4. The assumption implicit in this list of charges is that the concessionaire will convert the round logs into sawn timber in the forest and only convey a sawn product. However, this is becoming less common and the conveyance charge for "timber (sawn)" is now widely interpreted to include conveyance of roundwood and cants.

For the production of other forest products, the fees shown in Table 5 should be paid. The production of these products is currently covered by issuing casual licences and the exact process for issuing such licences, monitoring production and collecting fees is not clearly set-out in legislation.

Currently, it appears that some of these charges are levied in an ad-hoc manner depending upon whether producers ask for licences and whether Forest Officers discover people producing these products without a licence. Some provinces appear to take a more systematic approach to monitoring production outside forest concessions and this may be reflected in greater success at awarding licences and collecting revenues.

Charges collected by local authorities

In addition to the charges collected by central Government, some local (district) councils also raise revenues through charges on exports of products from their areas. These charges are set by the local councils, but have to be approved by central Government. Currently, some districts in both Copperbelt and Central provinces charge charcoal producers for taking charcoal from their districts. These charges can vary a lot between districts and figures quoted during interviews ranged from K 300 to K 1,000 per bag. Charges are collected at roadblocks along major roads.

Informal charges

During discussions with Forestry Department staff and sawmillers, it was revealed that a number of informal charges are also currently being levied in the sector.

In some areas, local Chiefs are asking for payment from sawmillers for permission to harvest forest products. Some Chiefs ask for a monthly payment (one figure quoted was K 300,000 per month plus various other free goods, such as cement and building materials). Others ask for a fee per truckload (K 100 – K 500 per truckload, depending on the amount being conveyed). These fees are questionable in terms of their legality, but it seems likely that such practices will continue unless Joint Forest Management Agreements bring these sorts of payments out into the open. Certainly, from the point of view of good governance, it would be desirable to have such arrangements formalised gradually, but this may be difficult to achieve.

It was also noted that concessionaires sometimes have to pay allowances to cover the costs of inspections by Forestry Department staff. These sorts of arrangements compromise the ability of Forestry Department staff to independently monitor and control forest concessionaires, but are inevitable where a lack of resources prevent staff from doing any sort of monitoring at all without such assistance. When the self-financing ZAFCOM has sufficient resources to perform its legally mandated control and monitoring activities, such arrangements should cease and staff found accepting additional payments should be disciplined.

Total collection of charges

The total amount of revenue collected by Forestry Department in 1999 is shown in Table 6. As the table shows, the charges described above (plus penalties) account for about 60% of total revenues. The remaining 40% of revenues come from sales from plantations and associated sawmills (local plantations, not the parastatal company ZAFFICO), plus a number of other miscellaneous charges and sales of produce.

Forest licences and economic efficiency

The two licences that probably have the most impact on economic efficiency are the Forest Licence (Commercial Sawmilling) and the Forest Licence (Pitsawing). The way that these licences are currently structured reduces efficiency in the industrial forestry sector in two ways.

Firstly, as has already been noted, these licences are only awarded to companies or individuals that have sawmilling or pitsawing capacity. This reduces competition in roundwood production activities and prevents the development of independent logging operations. Independent loggers can, if properly monitored and controlled, increase competition in the sector and assist with the development of a domestic market for roundwood.

If a market for roundwood was developed, sawmillers who want to specialise in the production of certain species, types or grades of forest products could obtain them from a variety of sources rather than having to rely primarily on their forest concessions for roundwood. Increased liquidity in the domestic log market would also help sawmillers to meet short-term demands by using the log market without having to alter production in their forest concessions.

Independent logging also offers opportunities for small and medium sized enterprises to enter the forestry sector without having to make the substantial investment in sawmilling equipment. To some extent this may already be happening in the pitsawing sector, but a more flexible licensing system would formalise this and make it easier for roundwood producers to develop.

The second way in which economic efficiency is reduced is that the current licensing system reduces the options open to producers in terms of how much they want to produce and the products that they want to produce. Again, this inflexibility reduces competition and the development of a flexible domestic market for roundwood and forest products.

The structure of the current licensing arrangement effectively limits production to two scales: 200 trees per month from a 10,000 ha concession or 20 trees per month from a 5,000 ha concession. A more flexible licensing arrangement that allowed producers to obtain a forest concession of a size and production level which suited them would give producers more options and increase competition and efficiency.

A second concern is that the current licensing arrangements implicitly try to force producers to convert logs into sawn products in the forest. It may be more efficient to transport logs to urban areas and process them there, where they are closer to markets and production costs will be lower. Of course, this is not necessarily the case, but allowing sawmillers to produce their final product wherever they want to would at least give them the opportunity to do so if they wanted to.

The administrative efficiency of the forest licensing arrangements

The current arrangements for monitoring and controlling production in forest concessions are quite expensive to operate. Activities that should be performed under the current licensing system are as follows:

measuring all logs produced from the forest – the licence says that the forest concessionaire should record all trees cut in a month and measure the log volumes produced from these trees;

monitoring and marking the number of trees cut – the Forestry Department should then find and mark all stumps cut in order to check on the concessionaires reported number of trees cut;

measuring the production of sawnwood removed from the forest – for the purpose of calculating the conveyance fee, the forest concessionaire should measure the volume of sawnwood that they want to take from the forest and report this to the Forestry Department;

checking the measurement of production – before awarding the conveyance licence, the Forestry Department should then check the measurement of sawnwood that will be removed from the forest and stamp it with an official mark; and

spot checks on the conveyance of forest products – in addition to monitoring in the forest, the Forestry Department mounts roadblocks to stop trucks and check that forest products are being conveyed with the proper documentation and they also monitor production at the point of sale.

Current practice is to place a Forestry Officer permanently on site in commercial sawmilling concessions to perform the first four of the functions above. Thus, rather than getting the forest concessionaire to do all of the measuring and leaving the Forestry Department to check this, the Forestry Department does all of the measuring (at least in the commercial sawmilling concessions).

The first point to note is that, under the current arrangements, all of this measuring by the Forestry Department must consume a large amount of resources. The whole point of making the forest concessionaire do the measuring and then getting the Forestry Department to check this, is that such an arrangement can greatly reduce the Forestry Department’s costs without a significant reduction in the effectiveness of monitoring.

Secondly, the emphasis placed on checking the number of trees cut in the forest is a very costly and time-consuming exercise. In most other countries, production would be checked at the roadside or log landing, before the logs are taken to the sawmill (i.e. in the case of Zambia, when the Forestry Department issues the conveyance licence). This would normally be the only time that production would be assessed and this could probably be done in less than a day. Therefore, assuming that the forest concessionaire keeps a certain amount of stocks of production in the forest, this might only need to be done once every two to three weeks.

The concessionaire can be encouraged to maximise production and minimise waste through the structure of the forest charges that they pay and the enforcement of felling regulations. It is not necessary to check the exact number of trees that have been cut. Indeed, it is more important for the Forestry Department to enter the forest to check that the forest concessionaire is keeping within their approved compartment boundaries and to check on the overall quality of harvesting. If the Forest Officer does not have to worry about finding all of the tree stumps and measuring logs, they can pay greater attention to other important factors such as: the height of tree stumps; the amount of waste logs left in the forest; the width and layout of skid trails; harvesting damage to remaining trees; and soil erosion.

A similar argument would apply to the production of logs in pitsawing concessions and to the production of other forest products using the casual licence. There are ways of structuring forest charges that do not require all of this costly measurement of production but are still effective at raising revenues. When foresters are released from the time-consuming and low-value tasks of measuring production, they can concentrate on the much more important task of making sure that a high standard of harvesting and management is maintained.

The impact of the forest licensing arrangements on sustainable forest management

As has already been noted in Section 3.4.2, the existing licensing system used for forest concessionaires in Zambia does support sustainable forest management at a very basic level through the requirement to produce a plan of operations and to follow felling regulations. However, a major weakness is that the system currently in place does not seem to require forest concessionaires to maximise production or yield from the areas in which they are licensed to operate. Indeed, it encourages them to high grade. Nor does the licensing system seem to require them to harvest areas in an orderly fashion so that, when they have cut over the whole of their forest concession, there will have been sufficient time for the first area that they cut to have regenerated by the time that they go back to it. It does very little, therefore, to encourage the concessionaire to maximise sustained yield, which is one of the main principles underlying sustainable forest management.

The total area of forest in a forest concession, the area of forest that is cut during the licence period and the amount of timber that is cut during the licence period should all be linked together. The main factor that determines how they are linked together is the sustainable yield of the forest which, in turn, should be used to determine the cutting cycle and harvesting intensity that should be used in the concession.

Currently, many forest concessionaires in Zambia request and obtain the maximum concession area allowed under the two types of Forest Licence and produce (or rather, declare production of) the minimum number of trees set in the licence agreement. The yields that would implicitly be required to continuously replace trees felled by concessionaires at the rates set in the two types of Forest Licence are shown in Table 7.

In contrast, discussions with Forestry Department staff suggested that, in the majority of cases, unlogged forest in Zambia might contain between 4 and 12 commercial trees per hectare and that the appropriate cutting cycle might be between 40 years to 80 years, depending on the species. The sustainable yields implied from combinations of these commercial stocking levels and regeneration periods are shown in Table 8.

For the purpose of the remainder of this analysis, a stocking level of 6 trees per hectare is assumed along with a regeneration period or cutting cycle of 50 years and the sustainable yield implied by this is also shown in the last row of Table 8. This combination of production and cutting cycle would seem to set a safe minimum standard for forest management in the types of forest found in Zambia, which can be used to guide the design of the licensing and forest revenue system.

The first point to note about the current licensing arrangement is that, as shown in Table 7, the yield that would be required to replace trees in the two different types of forest concessions allowed in Zambia is very different. This is because, comparing pitsawing concessions to sawmilling concessions, pitsawyers are expected to produce only one-tenth of the production of a commercial sawmiller, but they are given up to half of the commercial sawmiller’s area in which to operate.

The likely result of this is that sawmilling concessions will be entirely logged long before the forest has had time to regenerate, while pitsawing concessions can not possibly operate at a high enough level of production to completely harvest all of the areas allocated to them before the first area harvested is ready for logging a second time. An example of what this means is shown in Table 9.

If a concessionaire holding a Forest Licence (Commercial Sawmilling) harvests, on average, six trees per hectare, at a minimum production level of 200 trees per month, they will harvest 400 ha per year. Thus, they will work through all of their forest concession in 25 years, which is probably much shorter than the time required for the forest to regenerate. If they are even more selective and only harvest four trees per hectare, they will work their way through the entire concession in just under 17 years. The charging system encourages concessionaires to be selective and discussions with sawmillers suggested that they only take some of the commercial trees in their concessions rather than all of them. Thus, the latter scenario seems quite probable.

In contrast, the pitsawyer only has to cut 20 trees per month, which would mean that, at six trees per hectare, they would only cut 40 ha per year. Assuming regeneration period or cutting cycle of 50 years, they would only work through 2,000 ha before the first area harvested was ready for harvesting a second time. Even assuming more selective harvesting of, say, four trees per hectare, they would still only work through 3,000 ha, which is far less than the 5,000 ha allotted to them.

An alternative way of looking at this is to examine the harvesting intensity that a forest concessionaire would have to achieve given the total area of concession awarded to them, the minimum level of production they are expected to achieve and assuming a 50 year cutting cycle or regeneration period. This is shown in Table 10. If the holder of a Forest Licence (Commercial Sawmilling) was to harvest the forest on a 50 year cutting cycle they would have to limit harvesting to 200 ha per year. Thus, to meet the minimum monthly production target of 200 trees per month, they would have to cut at an intensity of 12 trees per hectare. In contrast, the pitsawyer only has to cut 2.4 trees per hectare. The first figure seems higher than the forests of Zambia can probably sustain (in most circumstances), while the latter seems far below it.

Even though the above analysis is based on a lot of assumptions (particularly with respect to what is the appropriate harvesting intensity and cutting cycle), it shows that the expectations for production from the two types of forest concession are very different and seem to be based on the type of producer rather than the growth of the forest. There is no reason to expect that the forest in pitsawing concessions is likely to be any more or less productive than the forest in sawmilling concessions, so the current licensing arrangement appears to give pitsawyers too much forest, but give the commercial sawmiller too little.

A second point worth considering is the duration of the licences awarded to forest concessionaires. In many countries, forest concessions are awarded for a relatively long period of time, particularly if the forest concessionaire is expected to make major investments in infrastructure or silviculture. This is done under the assumption that a long concession period is required so that the concessionaire will not be discouraged from making such investments.

Currently in Zambia, sawmilling concessions are awarded for five years and pitsawing concessions are awarded for three years. Forest concessionaires are not required to make significant investments in infrastructure in the forest and silviculture, so the short length of forest concessions probably does not matter very much. However, when a forest concession of such a short duration is issued, the following should be made very clear:

If the concession is to be sustainably managed as one unit, the concessionaire should be limited to only using a part of the total area of the concession during the licence period. For example, if a concession is awarded for five years and the cutting cycle is 50 years, then the concessionaire should clearly and explicitly be limited to only operating in one tenth of the total area of the concession. Once this area has been harvested, it should be closed to allow the forest to regenerate and the concessionaire should move on to the next cutting area.

If the concession is to be harvested over a period that is shorter than the regeneration period, it should be completely closed and left to regenerate once the last coupe has been felled. It appears that this is currently the case in Zambia, although it is unclear whether there is sufficient forest area or organisation of forest concessions to maintain production through a system of rotating the relatively small concession areas that are currently awarded.

Another point worth noting is that there is little reason why the sawmiller should be awarded a concession for five years while the pitsawyer is only offered one for three years. As with the differences in cutting intensities, this means that the two types of forest concessionaire are treated differently for no justifiable reason. As will be discussed in Section 5.5, it will be proposed here that the area of forest concessions that a concessionaire can apply for should be made more flexible while, at the same time, the duration of a forest concession licence should be standardised. By doing this, it should be possible to achieve more equitable treatment of the different types of concessionaire currently operating in the sector.

In terms of the casual licences awarded for the production of other forest products, it has already been noted that these mostly act as a receipt for payment rather than as a mechanism for controlling what goes on in the forest. Improved management and organisation of the production of charcoal is being developed however, and this very positive development should be continued.

A comment on the Timber Export Policy

The Timber Export Policy was introduced to stop the large-scale export of logs which, at the time it was introduced, was perceived to threaten the sustainability of the forest resource. However, in addition to banning the export of logs, the policy also placed restrictions (quotas) and imposed charges on the export of all wood products. It also introduced a rather complicated licensing procedure, requiring a lot of clearances and approvals before a shipment of wood products can be exported.

Overall, the policy seems to have been successful at stopping the export of logs. However, it has not stimulated the creation of a significant export industry in manufactured wood products. Rather, it seems to have reduced all wood product exports to a very low level. This is not an unusual result of such a policy. Many other countries have tried to use export restrictions to force exporters of unprocessed wood products to process them domestically and export the finished product. In nearly all cases, such policies have resulted in a contraction in exports overall and/or manufacturers try to minimise the impact of the restrictions by producing the product with the least amount of processing that can legally be exported. In the case of Zambia, this seems to be cants and railway sleepers.

Furthermore, the way that the restrictions and charges on exports are currently structured in Zambia does little to support the development of an export industry. Normally, when the aim of an export policy is to protect resources and/or develop a domestic processing and export industry, there are fewer restrictions on the export of high value-added products (and lower export charges) than there are on low value-added products. However, in Zambia this is not the case. Export of unprocessed products (including logs and charcoal, but not poles) is forbidden, but the percentage charge on the export of all other wood products does not vary. The exporter therefore, has little incentive to produce a more highly manufactured and more valuable product because they will just end-up paying more tax. Indeed, with the requirement in the licence to track the source of the wood, the administrative burden placed on the exporter of highly processed wood products is far higher than that placed on the exporter of simple wood products. Thus, the current licensing system may actually be penalising the exporters of more highly processed products.

In view of the above analysis, it would seem to be desirable to consider liberalising the licensing of wood exports in Zambia. At the very least, the export regulations should be restricted to only a limited number of basic wood products (e.g. roundwood, charcoal, cants and sawnwood) and should not include manufactured products (e.g. wood based panels, mouldings, joinery products, furniture and furniture components). Consideration should also be given to varying the export charge (as a percentage) by product or, more simply, to introduce fixed charges by volume or weight of exported product. This could capture some of the value of the roundwood used to make the exported product, without acting as a tax on manufacturing. The requirement to track wood back to its source should be dropped, as this does not seem to be necessary or effective.

Institutional and governance aspects of the current licensing arrangements

From the institutional point of view, the current licensing process is not working very well at all. Staff of the Forestry Department at all levels do seem to understand what is required under the licence agreements (although, as already noted, there seems to be some differences in interpretation). However, due to a lack of resources, they do not have anywhere near the capacity required to perform all of the control and monitoring activities set-out in the licences.

There are two ways in which such a problem can be addressed: increasing the resources available to the Forestry Department and/or making the monitoring and control procedures simpler to implement. Increased revenue collection should help in the first case, but simplifying procedures could also be very effective and should not be overlooked.

In terms of whether the current licensing arrangements meet the principles of good governance, there are probably two questions that should be answered:

do all stakeholders know and understand the current licensing arrangements; and

is the process for awarding and implementing licences transparent, clear and fair?

Discussions with a very small number of licensees suggested that they do understand the licensing arrangements well and, as noted above, Forestry Department staff also seem to have no problem with them. These are currently the two main stakeholders involved in the licensing process. However, there is a third group that should also be considered: companies and individuals that might like to acquire a forest concession but do not currently understand how to go about doing so. Current procedures for awarding licences are not clearly documented, so this group is at a disadvantage.

In terms of the process for awarding forest concessions, Zambia is in a transitional phase. Until recently, the system for awarding forest concessions was the wrong way around. Potential concessionaires would identify an area of forest that they were interested in, examine it and assess stocking and other operational factors, then apply to the Forestry Department for a concession in that area. The Forestry Department did not have an overall plan or strategy for identifying areas that should be harvested. Rather, it responded to requests for concessions as they came in on an ad-hoc basis.

Such an arrangement puts forest concessionaires in a very strong position. They initiate the process of obtaining a concession, which then effectively rules-out any competition for that piece of forest.

Recently however, the process of awarding concessions has been revised and the Forestry Department has started to take more control over the overall process. Taking into account the quality of forests available, they have started to identify when and where harvesting should take place and to invite potential concessionaires to submit proposals for these areas.

Awarding concessions in this way encourages competition and enables the Forestry Department to choose the best concessionaire for each area on the strength of their proposals. It is unclear exactly how proposals are judged (and this should be made more transparent), but this new arrangement is a very positive development and should be continued.

Once a forest concession is awarded, the licence is comprehensive and clear about the procedures for suspension and cancellation of the licences and the procedures for arbitration.

The impact of the forest revenue system on economic efficiency

In terms of economic efficiency, the main concern with any forest revenue system is that the level of charges is roughly equal to the prices that would be obtained if the products taken from the forest were sold by competitive means. For the production of roundwood, this is called the stumpage value and this can be calculated from forest product prices and production costs (see below). It seems likely that the current levels of charges in Zambia are, as in many other countries, too low. It also seems likely that the range of charges for logs of different species should be much greater, to take into account the variation in the value of finished products made from each of the different species.

An example of how the stumpage value for the production of roundwood might be calculated is given in Box 1. The calculations presented here are based on the author’s knowledge of forest products markets in countries in a similar situation to Zambia, rather than detailed information from the Zambian forestry sector. They are however, probably reasonably representative of product prices and production costs in Zambia.

The analysis suggests that the current stumpage value of roundwood produced in forest concessions in Zambia might be from US$ 10 to US$ 45 per cubic metre, depending upon tree species and market conditions. At current exchange rates, this would be equal to a stumpage value of K 40,000 to K 180,000.

The other point worth noting about the example is that it shows how wide the range of stumpage values might be, depending on the end use of the wood. For example, although the higher product price used in the calculation is only 130% higher than the lower product price, the stumpage value is 350% higher. This is because production costs do not account for all of the difference in the product prices. To put this another way, part of the higher price for the more valuable product reflects the higher demand for the properties of the roundwood itself (e.g. its colour, grain, strength properties etc.). This can be reflected in stumpage values that are relatively much higher than the differences in product prices would at first suggest.

Currently, the difference between the forest charges for low value species and high value species in Zambia is relatively small. Low value species are charged at 40 Fee Units per tree, while high value species are charged at 100 Fee Units per tree. This gives a difference of 150% between the low and high values. Based on the above calculations therefore, the range of charges for different species should be extended. Alternatively, a smaller range could be maintained along with an export charge that would capture some of the additional value of the most highly valued species.

Given that the Fee Unit is currently valued at K 180, the above analysis suggests that the lowest value species in Zambia should be charged at a rate of 220 Fee Units per tree and the highest value species could be charged at a rate of up to 1,000 Fee Units per tree.

Similar calculations to those above can be performed for the production of other forest products. As with the production of industrial roundwood, it seems likely that the current level of charges for poles, fuelwood and charcoal in Zambia are also probably all lower than would be achieved in a competitive market. (A more complete analysis of the economics of charcoal production and implications for the forest charges is given in Annex 4).

Some of the effects of artificially low charges on economic efficiency were already listed in Section 2.1.1. Very briefly, low charges tend to result in the following inefficiencies:

artificially low forest product prices in the domestic market;

consumption of wood products at an artificially high level;

harvesting of forest areas at an artificially high level;

distortion of investment and manpower into forest industries when they would be more productive in other sectors of the economy;

a lack of any incentive to maximise product recovery in the forest processing industry;

a lack of any incentive to invest in more intensive forest management, especially the development of forest plantations;

high levels of waste in forest harvesting;

a lack of any incentive to work on developing forest products markets;

capital flight (if there is a significant forest products export sector); and

artificially low forest land values, encouraging the conversion of forests to other land-uses.

To some extent, several of these results also work against the stated objectives of forestry policy. Therefore, the low level of charges has a negative impact in that dimension as well. It seems likely that the currently low level of forest charges in Zambia is resulting in several of these undesirable effects.

One final point worth noting is that it is only really necessary to resort to trying to calculate and set forest charges when it is not possible to use competitive means to sell forest products and establish true market prices. Due to the long-term nature of forest concessions and the complexity of selective forest harvesting systems used in natural tropical forests, it is usually considered difficult to try to introduce competition into the revenue system in these cases. However, in the case of harvesting roundwood from forest plantations and the production of other forest products (e.g. fuelwood, charcoal and NWFPs) there is less of a reason why competitive selling mechanisms, such as auctions and tenders, should not be used. At least in the case of the Forestry Department’s forest plantations, it is suggested that competitive selling might be considered as an option.

The impact of the forest revenue system on sustainable forest management

It has already been mentioned that the current charging structure does not support sustainable forest management. In particular, it encourages high-grading and discourages the maximisation of sustained yield. The reasons for this can best be explained with reference to Figure 2.

Figure 2 shows (as the solid line) the average charge that a forest concessionaire would have to pay per tree for harvesting in one month. Given the minimum production level set in the licence, the concessionaire would probably not want to produce less than the minimum, because this would increase the average charge paid per tree. For example, only cutting 100 trees per month would double the average charge per tree. However, once the minimum has been reached, there is no incentive to produce more than the minimum. The charge for each additional tree is the same. Indeed, in the current charging structure, the forest concessionaire would actually be penalised if they produced (and declared) a very high number of trees, because the excess trees would be charged at twice the normal rate.

Now, given the current level of technology in use in the forestry sector in Zambia, it seems likely that a typical forest concessionaire could harvest about 50 ha per month if they maximised production per hectare. Assuming six commercial trees per hectare, this would be equal to about 300 trees per month. However, because of the charging structure, it is more profitable for the forest concessionaire to try to find the 200 most valuable trees that they can in the month, rather than to take all of the commercial trees from each hectare and to try to maximise total production. If monitoring and control of the areas actually being harvested is weak, this search for the most valuable trees could be far reaching and result in a lot of repeated damage to the forest.

One alternative to the above situation, would be simply to charge the forest concessionaire 20,000 Fee Units for the month (i.e. what they currently pay anyway) and restrict them to the 50 ha which it is estimated they could fully harvest in one month. This would be equal to setting an area-based charge of 400 Fee Units per ha (as long as control over the area harvested is more strictly enforced).

The impact of such an arrangement on the average charge per tree is shown by the line of squares in the figure. Under such an arrangement, production of less than 200 trees per month would still result in a very high charge per tree (i.e. the dotted line and the solid line overlap in the figure). However, production of more than 200 trees per month would start to look more attractive to the concessionaire, as this would bring the average charge per tree down. Therefore, for example, if production were at the estimated maximum of 300 trees in the month, the average charge per tree would be 25% lower than under the current system. The forestry administration would still receive the same amount of revenue as before, but the forest concessionaire would be encouraged more to try to maximise the yield from the area being harvested rather than to try to find and harvest only the best trees.

Of course, in order to use only an area-based charge, it would be desirable to have very good information about the amount, type and value of trees in the forest being harvested. A system of area charges that was fixed across all areas would also require a very homogenous forest. Because of these limitations therefore, most forest revenue systems try to combine both an area-based charge and a volume-based charge. The area-based charge is used to encourage the maximisation of production and is usually set at a quite low level. This is so that the charge is not too high in areas with a low level of stocking. The volume-based charge is used to try to capture variations in value by species and to capture some of the additional total stumpage value in more highly stocked areas.

An example of what one combination of area and volume-based charges might do to the average charge per tree is shown by the line of circles in the figure. The charging structure represented by this line is an area-based charge of 10,000 Fee Units in total (i.e. 200 Fee Units per ha over a 50 ha cutting area) plus a volume-based charge of 50 Fee Units per tree. At a production level of 200 trees in the month, revenue to government would be the same, but half of it would come from the area-based charge and half from the volume-based charge.

As the figure shows, it would still be very expensive for the concessionaire to produce less than 200 trees per month (i.e. the average charge per tree would still be very high), so they would be encouraged to produce at this level or more. However, if they produced more than 200 trees, both they and the Government would benefit. The government would receive slightly more revenue than if 200 trees were cut (an additional 50 Fee Units for every tree cut over 200), while the forest concessionaire would end-up paying a lower average charge per tree as production increased.

The above argument could also be applied to the structure of charges for other forest products, which are all similarly focused on the volume of production without paying much attention to the area of forest being used for production. In particular, the production of charcoal should be much more strongly controlled in terms of the areas that are used. Charcoal production seems to be having a major impact on forest resources in Zambia, so a diversion of efforts away from trying to measure and charge for volumes of production towards trying to control where such activities are actually allowed to take place would seem to be a good idea.

Revenue collection and revenue sharing under the current forest revenue system

A complete breakdown of all of the revenues collected by the Forestry Department in 1999 was given in Table 6 . About sixty percent of current revenues come from charges and penalties associated with the licensing of forest production and conveyance. The remainder comes from commercial activities and sales of other forest produce. One of the first actions that ZAFCOM should take when it is created is to separate more clearly the operating account for commercial activities from the revenues collected from charges on all forest production. This will make it easier to assess how well the forest revenue system is working and to evaluate the performance of these other activities.

The total amount of revenue collected in 1999 (from both of these sources) amounted to K 1.6 billion. This compares with a budget of K 2.3 billion, of which the Forestry Department actually managed to spend K 1.0 billion. Some of this expenditure was, presumably, spent on commercial activities (e.g. the "revolving fund" used to support plantation establishment in Local Forests), so it is not possible to tell whether the Forestry Department raised enough revenue from charges to cover the cost of non-commercial activities. It seems likely though, that revenues were sufficient to cover the existing level of non-commercial activities.

Although the Forestry Department is currently generating adequate finances to cover its costs, the current level of activities and achievements of the Department are severely limited by a lack of funds. They are also well below what would be required to support the widespread implementation of sustainable forest management across the whole of Zambia’s forest estate. There is, however, significant potential to increase revenues from the collection of forest charges. Table 11 gives an approximate estimate of potential revenue collection from forest charges, based on the existing charge rates and stated and possible levels of current forest products production.

The table shows that current revenue collection, at K 1.0 billion, is probably only about 35% of what could have been collected (based on the production figures stated in the Forestry Department’s Annual Report for 1999).

If the true level of production were known, the proportion captured would probably be much lower than this. For example, a very crude estimate of the actual level of production possibly achieved in 1999 is also shown in Table 11. Based on the size of the population and forest processing industry, it is estimated that the overall production and consumption of forest products might be several times higher than that reported and that, consequently, revenue collection might be less than 5% of the potential. Out of all of the forest products, it would seem that collection is least effective in the charcoal-producing sector.

Furthermore, these estimates have been calculated using existing charge levels. Given that these are also very low in themselves, it may be possible to multiply revenues by a factor of 20 or more.

Of course, it would be very difficult to achieve such an increase in performance quickly but, even assuming that some leakage in the system persists, this analysis suggests that it should be possible to substantially increase revenue collection by the Forestry Department.

There is, of course, currently no official sharing of forest revenues. However, given that the Forestry Department is currently unable to collect most of the forest revenues due to them, they are implicitly sharing the majority of the benefits from forests with other stakeholders in the sector.

Institutional and governance aspects of the current forest revenue system

The revenue system currently in place in Zambia is easy to understand. Almost all charges relate to some measure of production and these measures do not require any complicated calculation. As noted, there are some uncertainties (e.g. the definition of personal use) but these are few and could easily be overcome with the production of some simple guidelines.

In terms of whether the charges are applied fairly and equally, the current system is less satisfactory. The current lack of resources has led to widespread evasion of charges and, to a large extent, the amount that any producer is likely to have to pay is determined by how likely they are to get caught. This probably gives the smaller producers an advantage over the larger producers.

There also appears to be some differences between regions in terms of how effective they are at collecting revenues. Some of this could be due to differences in production, but it seems likely that greater attention should be given to the allocation of resources between regions for the purposes of monitoring activities.