Component report-Part B-10-Dec-01

Several issues facing the RGC concerning trade policies and the strategy for implementing WTO trade rules emerged in the course of the diagnostic study. These were:

An element of Cambodia’s trade policy that may not be consistent with WTO rules is the prevalence of administered minimum or fixed dutiable values from which customs calculates trade taxes. This valuation system is clearly inconsistent with WTO rules. The WTO agreement on customs valuation provides for customs valuation based on the transaction value, that is, the price actually paid or payable for the goods when sold for export to the country of importation. Cambodia will need to look at ways of phasing out administered dutiable values currently used for several product groups. In this regard the RGC is seeking assistance from the New Zealand government on customs valuation. A second area where treatment might offend WTO rules is the differential tax treatment between imports and domestically produced goods. The WTO requires that the products of the territory of any WTO Member shall not be subject to internal taxes (for example, excise and VAT) in excess of those applied to like domestic products. In addition, those taxes shall not be applied so as to afford protection to domestic production. One example mentioned earlier was the exemption from VAT on domestically produced motorcycles for the domestic market, whereas imports of motorcycles are not VAT exempt.

A second issue relates to devising a strategy for setting binding tariff offers to other WTO member countries. Binding tariffs are the maximum tariff rates the government commits to when joining the WTO. In other words, the government promises other WTO member countries that it would not raise tariff rates above these bound levels. In practice, most countries set and keep applied tariff rates below these bound rates. There are two approaches on the level to bind rates. The first approach involves setting initially high bound tariff offers; the second view advocates setting low binding tariff offers. Both approaches have advantages and disadvantages. Setting high bound tariff rates might calm concerns of domestic interest groups that oppose trade liberalization by assuring them that the government will not give away the choice of raising tariffs in the future. Setting binding rates relatively low, say equal to or very close to current applied rates also has advantages. In particular low bound rates signal to the international investment community that the country is committed to an open trade regime and wishes to become a platform for investment. Crucially important, low binding tariff rates provide stability and certainty in the country’s trade policy framework. As discussed in the section on AFTA, investors like a policy framework that is transparent and predictable. Setting bound rates close to current applied rates would achieve this predictability.

A third issue relates to building institutional capacity to implement WTO rules. As noted the RGC has set out a very ambitious legislative agenda to implement the WTO requirements. Examples include various intellectual property laws (four laws are proposed in this area); competition law, bankruptcy law, law on business enterprises, commercial contracts law, secured transaction law, law on rules of origin and various laws on antidumping and countervailing duties. Most of these laws are complex and will require the establishment of separate agencies to implement them. For example, a competition law would require a separate competition commission with its own staff of highly trained economic and legal technical experts. These laws are in addition to many other laws and institutions that the RGC has or intends to establish in both the economic and non economic areas not related to the government’s WTO accession efforts.

Given this legislative load, the RGC administrative capacity will be stretched. It is also becoming clear that at this time and for some years to come the RGC may not have the administrative capacity to implement some of these complex laws. While the technical assistance canvassed in the preceding section will help, there may need to be a strategy to reprioritize legislative efforts. In particular, those commercial laws that are not fundamental to WTO accession or economic efficiency may need to be pushed back in time. Clearly some laws are important for improving economic governance and reducing the risk of doing business in Cambodia such as a land law, contracts law and bankruptcy law. Other laws which though important, may be deferrable include laws on antidumping and countervailing duties and competition. There are strong arguments against introducing laws on antidumping duties as many countries often use them as a disguised form of protection of local industries without considering the economic costs (that is, higher prices) this instrument has on end users and consumers of the product. Similarly, it is not clear that a competition law is appropriate at this stage of Cambodia’s institutional development. The implementation of a competition law requires an independent commission with highly trained economic and legal experts as identifying anticompetitive behavior requires elaborate economic analysis and is data intensive. It might be more effective to concentrate limited government and donor resources on the commercial and economic laws that are key to reducing risk of doing business in Cambodia. In this regard technical assistance would be key to assisting the RGC to reprioritize the legislative agenda contemplated for WTO admittance.

A notable feature of Cambodia’s tariff structure is that the broad tariff bands corresponds more or less to whether goods are final goods or raw materials. This is referred as a ‘cascading’ tariff structure, since the rates cascade down, with higher rates for finished goods, to lower rates for intermediate goods and raw materials (see chart 2.13). This ‘cascading’ tariff structure has changed little since 1997. Chart 2.13 presents average tariffs for raw materials, semi processed manufactures and fully processed goods at the International Standards Industry Classification (ISIC) two digit level. It is apparent from chart 2.14 that several major sectors — food and beverages, textiles and garments, basic metal products — have very distinct ‘cascading’ tariff structures.

This ‘cascading’ tariff structure has important implications for the government’s export led economic development and poverty alleviation strategy. By setting high tariffs on semi processed and consumer goods, Cambodians are required to pay above international prices for basic needs, unless smuggling of imported goods circumvents these high prices. Take the example of locally produced wheat flour. The tariff on wheat flour is 35 per cent. This duty has the effect of protecting the local wheat flour miller from cheaper imports. This means that downstream users of wheat flour (bakers, producers of processed foods, restaurants and home industries) must pay a relatively high price for wheat flour as an important ingredient into production of their goods. The higher costs of production are then passed onto consumers through higher retail prices. Since food makes up a large proportion of poor household’s expenditure, lowering import tariffs on these products would benefit poor households.

The ‘cascading’ tariff structure is not consistent with the Governments broad based industrial growth and development strategy. A central platform of the Royal Cambodian Government’s First Socioeconomic Development Plan, 1996–2000 (SEDP) and the second SEDP2 (2001 to 2006) is reducing poverty by the promotion of rural and labor intensive industries via an export led development strategy. Given Cambodia’s small market size and low per capita income, this strategy seems most appropriate in achieving the RGC goals of high economic growth, employment and poverty reduction. In this context industrial and trade policies are linked. Trade policies, particularly tariffs, influence domestic prices and these in turn impact on investment and production decisions. Industries which do not face import competition because of tariffs, subsidies, and quantitative restrictions have an incentive to produce for the domestic market and not to export. It is sometimes argued that a country can protect designated import competing industries, while at the same time allow exporters to source imported materials at competitive world prices through various duty exemption schemes, thus, achieving the dual objective of promoting exports and import competing industries. Experience of other countries such as Sri Lanka, New Zealand, Indonesia and the Philippines that had adopted ‘import substitution’ strategies in the late 1960s, 1970s and early 1980s clearly show that protected import competing sectors became inefficient and unable to compete in the international market and consequently the country’s economic growth performance lagged behind those countries with a more outward orientation. Moreover, governments in many countries found it politically difficult to remove protection once it has been granted to industry. If a broad based export oriented approach is to be pursued, as in the case of Cambodia, a cascading rate structure inevitably sends the wrong signals (see chart 2.14 and box 2.15 reports on experiences with import substitution strategies in other countries).

Recently estimated effective rates of protection for a selection of Cambodian industries indicate that this is likely what is happening now. Effective rates of protection measure the degree to which trade policy has artificially increased the profitability of producing for the domestic market relative to export activities and, thus, signals a disincentive to produce for the export market. Table 2.16 presents estimates of effective rates of protection for a sample of Cambodian industries. These are rates prevailing in 1996–97. Given that the tariff structure has changed little over the last four years, the estimated ERPs and ranking of these industries will still be relevant. The weighted average ERP for this sample is 111 per cent. This level of effective protection is quite high if it is indicative of the entire industrial sector. These estimates were based on a small sample size of 50 firms surveyed in 1996. Thus, further analysis would be needed before the wider applicability of these estimates can be assumed. Nevertheless, the general picture that emerges is of relatively high but variable incentives resulting from tariff policies. Despite the shortcomings in the sample, this is likely to be a reasonable indicator of the results that would be obtained from a larger, more detailed analysis. The implication of these ERP estimates is that Cambodia’s tariff policy is structured in such away that it acts as a disincentive to export a wider range of products than currently being exported.

In the 1960s and 1970s many countries adopted an industrialization strategy based on import substitution. Tariffs were used to provide incentives (effective protection) for selected industries. In the early stages of this strategy, activities engaged in the manufacture of consumer goods were viewed as the easiest targets for industrial development and the tariffs for these goods were raised relative to those of other goods (that is, cascading tariff structure). The original intention was that when these infant industries matured more types of final good manufacturers would be targeted and their tariff protection raised. These were to be followed by increased protection for the domestic production of intermediate goods, processed raw materials and capital goods in subsequent stages, thus, fostering industrial backward linkages. As experience in the 1970s and 1980s aptly demonstrated, this approach had serious flaws and the results were not as hoped. The first stage of the strategy was implemented and high levels of protection were established for many types of consumer goods. But in doing this the seeds of failure were sown. Many of these firms remained uncompetitive, which made it politically difficult to proceed to the second stage where the tariffs for intermediate goods used by these firms would be increased. As a result a powerful constituency was created that was able to block further implementation of the import strategy. Another significant outcome was that for many of these countries, the industrial base remained small in terms of contribution to value added and employment and relatively high cost when measured against international competition. As world markets became more integrated those countries that adopted import substitution strategies tended to grow slower than those countries that were more export oriented.

Several Asian countries had followed import substitution strategies in the 1970s and 1980s. The results were not as hoped and in response many of these countries implemented substantial trade policy reforms in the late 1980s and early 1990s. The experience of Indonesia aptly demonstrates that a more outward oriented strategy achieves greater results in terms of employment growth and poverty reduction. Indonesia followed an import substitution strategy in the 1970s and 1980s whereby tariffs followed a ‘cascading’ structure. However, by the early 1980s the industrial base remained small and inefficient, accounting for 16 per cent of gross domestic output. After a period of low economic growth in the early 1980s and a slump in oil prices in 1986, the government began to reform its trade policies. In 1986 the government introduced a duty drawback facility on raw materials and intermediate inputs for exporters and in a series of reform packages from 1986 to 1995 the government eliminated most non tariff barriers and reduced tariff rates on most goods. Unweighted average tariffs fell from over 25 per cent in 1986 to 7 per cent by 1999. As a result, Indonesia’s manufacturing sector grew rapidly, with annual growth in excess of 8 per cent for most years until 1997 when the currency crisis erupted. By 1995 the manufacturing sector accounted for 23 per cent of GDP. The engine of growth was exports of manufactured goods. At first, exports were heavily concentrated in garments, textiles and footwear, which accounted for about 65 per cent of total exports of manufactured goods at the end of the 1980s. Exports began to diversify in the early 1990s to include toys, electronics, furniture, and jewelry. By 2000, exports of manufactured goods accounted for over 65 per cent of total exports, up from 8 per cent in 1984.

Activity	ERP)
	%
Mfgr of basic chemicals, except fertilizer	2279
Mfgr of wearing apparel	2268
Mfgr of rubber tyres etc	1007
Mfgr of wearing apparel	458
Mfgr of Dairy products	415
Mfgr of Tobacco products	411
Mfgr of metal containers etc	365
Mfgr of wearing apparel	329
Mfgr of other rubber products	273
Mfgr of malt liquors (beer)	144
Distilled beverages	142
Mfgr of soft drinks	136
Mfgr of structural non refactory clay and ceramic products	59
Mfgr of structural non refactory clay and ceramic products	41
Mfgr of structural non refactory clay and ceramic products	35
Mfgr of other rubber products	33
Mfgr of other rubber products	25

Source: Cambodia: Strengthening the Foundation for Trade and Industrial Development, Report presented to the ADB and RGC (ADB TA No. 2570 CAM; 1997)

The RGC’s economic goals include higher rates of industrial growth and development based on an expansion and diversification of exports and improved competitiveness of import competing industries. If these goals are to be achieved to any significant degree, resources must be employed in their most productive uses. The dominant role of the private sector in allocating resources means that it is essential that policy based incentives do not distort their assessment of the returns to alternative investments. In other words, the large differences in effective rates of protection noted above need to be substantially eliminated. This in turn implies moving away from a cascading tariff structure. This indicates that the primary policy objective for the tariff system should be the adoption of a low, uniform tariff rate structure. In other words, a single tariff rate that applies to all or virtually all imports. The sole purpose of the tariff system would be as a source of revenue. The determination of the appropriate level of tariff rate would be set to achieve the required revenue. A low uniform rate on (or virtually) all imported capital goods, raw materials and finished goods would ensure that local producers are not disadvantaged vis à vis competing imports. A low uniform tariff rate typically discourages smuggling of imported goods and a low uniform of say 5 per cent with no exemptions would produce almost as much revenue as the prevailing system with probably much less bureaucratic costs..

To a large extent this is already the government’s objective under the AFTA agreement; that is, to move to a 0–5 per cent tariff rate for virtually all goods imported from ASEAN member countries. Thus, the logical extension of this strategy is to reduce MFN rates in tandem with the preferential CEPT tariff reduction schedule. By reducing CEPT and MFN tariff rates in parallel with each other, any trade diversion arising from the ASEAN preferential trade agreement is eliminated. Although as indicated earlier there are good reasons to accelerate implementation of both CEPT and MFN tariff reductions.

Three important implementation issues need to be considered. First, what would be the status of the current duty exemption scheme under a uniform tariff rate policy; second, what is the revenue implication of such a policy; and third, what would be the short term adjustment costs to this strategy.

There is a strong economic case for retaining duty exemptions on materials required by exporters. Exporters need to be able to source materials at international prices to remain competitive. Producers of garments account for about 80 per cent of total value of duty exemptions. One approach for the garment industry would be to reduce tariff rates on imported materials and accessories to zero. (The tariff rate on the great majority of these items is currently 7 per cent.) This treatment would have little effect on customs revenue for the reason that most imported fabrics and accessories are already exempted from duties. Garment exporters would not have to maintain a Master List for imports of materials at CDC, as they are currently required to do under the duty exemption scheme and this would help reduce administration costs for both the garment industry and the government. A zero tariff policy for garment inputs would also help develop a competitive domestic oriented garment industry, as these firms would also be able to source materials at international prices. However, to avoid a distorted incentive regime, tariffs on finished garments (which are currently 35 per cent) would also be set to zero (see box 2.17). This option would only be viable for exporters operating in the garment (and also footwear) sector where it is possible to identify required imported inputs for which their demand dominates the demand of general importers. Rather than a zero tariff a low rate of 2 or 3 per cent combined with streamlining of procedures cold be an improvement over the existing state of affairs for exporters (lower costs overall) and the government (more) revenue but perhaps not for people relying on bureaucratic costs. In the case of new export industries some kind of duty exemption or rebate system needs to be examined.

Customs data is used to estimate the lost revenue if the scheduled tariff rate on textile and garments was set to zero. Assume that most raw materials imported by exporters fall under the HS Chapter Headings of 50–55, 56–58 and semi processed and finished garments fall under Chapter Headings 60–62 of the Cambodian tariff schedule. From the Customs data, we are able to identify the total value of imports exempted from customs duty (that is, imports of materials by garment exporters) and the value of imports that were subject to custom duties (that is, general importers). The table below summarizes imports and revenue by textiles and garments.

Notes: textiles refer to HS Chapter Headings 50–55, 56–58; Semi processed and finished garments refer to HS Chapter Headings 60–62.

As the table demonstrates, for those items imported by exporters of garments, the imports by such exporters dominate imports by general importers. In fact they account for 98 per cent of imports of such items and consequently only some CR3–4 billion of revenue would be lost if the tariff rate on all such items under HS Chapters 50–59 and 60–62 were reduced to zero.

Note: First semester 2001 Customs data was not available and 2000 data has been used for this exercise. There may be some underestimation of revenue loss, but since most imports under these Headings are exempted from duties this underestimation should be small. Also, tariff rates on almost all items under these Chapter Headings remain the same between 2000 and 2001 so no bias arises from the tariff structure itself.

There are good reasons for limiting exemptions to trade taxes for non exporters and on capital imports. First, it is always difficult to administer exemptions. This includes exemptions to government, donors, international organizations and NGOs. There are inevitably pressures applied to officials to expand the scope of imports receiving exemptions or to be flexible in applying regulations. Second, duty exemptions can distort purchasing decisions. It may be more efficient to purchase a good locally but it appears less expensive to import the good duty free. While it is often difficult to reduce exemptions for imports by aid and donor agencies, it is straightforward to end exemptions for government imports. This reduces distortions in allocating government resources, and also sets a proper example for the private sector. Moreover, as the tariff structure moves towards a low uniform rate over time, the need for duty exemptions on imports by non exporters will diminish. Finally, in moving towards a low uniform tariff rate, the administered minimum dutiable valuation system will need to be abolished to remove implicit surcharges inherent in this system.

A final issue relates to the adjustment costs of tariff policy reform. Countries that have liberalized their trade policy regimes have experienced to varying degrees short term adjustment costs. In particular, countries that had in the past made extensive use of tariff policy and non tariff barriers to protect their inefficient industrial bases typically experience the greatest adjustment costs. Often these costs included temporary, but high unemployment as inefficient sectors restructure and resources move to more productive uses. However, Cambodia is highly unlikely to face the same short term adjustment costs that some other countries experienced when they liberalized their trade structures. The reason is that, despite the inherent distortions in Cambodia’s tariff structure, Cambodia does not have an inefficient industrial sector built around high tariffs. For many activities, remaining high tariff rates are redundant, either because little domestic activity is undertaken in these sectors or because smuggling of goods keeps domestic prices close to world prices.

As indicated earlier, most of the rapid growth in the industrial sector has been driven by export oriented sectors, notably garments and recently footwear. Instead, further tariff reform will produce greater benefits in the medium term. A low, uniform rate would signal to investors that Cambodia wishes to be a platform for investment in activities where Cambodia has a comparative advantage. This should attract new FDI in these sectors. A low, uniform tariff rate would allow investors to source raw materials at world prices (or close to them) and this ensures that producers supplying the domestic market (including SMEs) will not be disadvantaged vis à vis imports; this in turn encourages development of competitive import competing industries.