South African Merchandise Trade with China

Transcription

1 South African Merchandise Trade with China by Ron Sandrey tralac Working Paper No 3/2006 April 2006

2 Key Points This data analysis examines South African imports into China and Chinese imports into South Africa. Data through to December 2005 is sourced from the respective countries using the World Trade Atlas data. During 2005 imports into China from South Africa totalled some US$3,444 million, a figure that increased by 16.5 per cent from In real terms, this represented 0.52 per cent of Chinese global imports, a figure that has been relatively stable over the last ten years. By commodities, the main imports were iron ores ($967 million), precious metals and stones (($957 million), a particular Chinese classification of small lines ($462 million) and iron and steel ($427 million). The average duty on these imports was calculated to be 3.45 per cent. South African imports into China are highly concentrated, with the top ten lines at the HS 1 2 chapter level accounting for 92.5 per cent of the total trade, and in several of the main imports at the more detailed level (precious stones in particular) these imports held a major share in the Chinese market. Detailed analysis showed that both Australia and Brazil, two countries of potential FTA (Free Trade Agreement) defensive interest into China, are major competitors in several important trade lines. South Africa is doing relatively well into China, with some 40 per cent of the imports (the Stars ) gaining market share in Chinese imports sectors that are themselves growing. Chinese imports into South Africa have increased dramatically over the last ten years, and by December 2005 they were $4,926 million or 9.0 per cent of the total, with the latter figure up from a 2.1 per cent market share in These 2005 imports were up 37 per cent on the 2004 figure, much higher than the global increase into South Africa of 15.5 per cent. Chinese imports are not as highly concentrated as South African imports into China, with the top 10 HS chapters making up a lesser 72 per cent of the imports. Although 1 Where HS is the Harmonised System of merchandise trade classification that operates in a sequentially more detailed level from internationally harmonised (hence the name) HS 2 to 4 and 6 levels, and often down to even HS 10 for individual countries. 1

3 45 per cent of the trade enters duty-free, the average duty that would have been assessed on these imports in the absence of any rebates was per cent. These duties are mostly paid by the 30 per cent of the imports that enter where the duties are assessed at 20 per cent or more, a grouping that is dominated by textiles, clothing and footwear. A look at revealed comparative advantage shows that aside from the spectacular growth a remarkable feature of Chinese imports is their domination of the market where they do compete, with most of the textile, clothing, footwear and leather chapters holding a market share of at least 50 per cent. Associated with this are the equally remarkable growth rates to achieve this dominance, as some 97 per cent of the individual HS 4 import lines have increased their market share over the last ten years. Data reconciliation shows that, as expected with South African imports into China valued at two and a half times the reported exports from South Africa to China, there is little coherence in reconciling the trade flows to China. Some of this is accounted for in the transport costs (iron ore) given that imports are valued at cost, insurance and freight (cif) and exports at Free on Board (fob), while another large difference is that there are almost no reported exports of platinum and diamonds from South Africa but these are major imports into China. Much remains unexplained. Trade chilling suggests that there may be potential areas where South Africa could export to China and where an FTA could help. By value the main potential export items from South Africa may be motor cars and aircraft (duties of 15 and 5% respectively), as these are both massive imports into China. Other lines include apples, apricots, pineapples and avocados, chocolate products and processed fish and meats in agricultural goods, titanium oxide and other ores, and some lines of iron and steel and other manufactured products in general. Finally, an analysis of the so-called intra-industry trade between South Africa and China shows that, as expected between two developing countries, these index values are low at about the 6 (out of a 100) level. This index compares trade between partners in like-products, and is a feature of trade between developed countries. However, the more interesting feature is that these index values are 2

4 steadily increasing from very low levels over the last ten years, suggesting an increasing sophistication in the trading relationship. Introduction This paper will assess the trading relationship between the Republic of South Africa (RSA) and China over the most recent period. We will start by examining the importance of RSA to China by using their import data, and then examine the mirror of the importance of China to RSA. The analysis will concentrate upon using the respective partners import data, as this is generally more reliable than export data. Two points must be made here. The first is that imports are assessed using cif (the value of the goods plus the costs of freight and insurance conveying them from the export dock to the unloading at the import border) while exports are assessed as fob (free on board, or the actual value of the goods at the export dock). This means that we are inflating the import value by including freight costs, and in the case of, for example, iron ore, the largest import into China from RSA by value, this is significant. The second point is that export and import values seldom if ever agree, and we will introduce a reconciliation exercise to explore some reasons for this, with the cif versus fob values one reason. All data is expressed in US dollars. All data is sourced directly from primary sources through the commercially obtainable World Trade Atlas data (WTA, from Dr John Brasher, colloquially known as Brasher data ). The WTA South African data is directly from SARS, while the Chinese data is directly through the Chinese official sources. The Chinese official (but of course, as always, preliminary) data for the year ending 2005 was available the first week in February A salutary lesson for the Southern African Customs Union (SACU) can be drawn from this achievement! The study will also extend traditional trade analysis by looking at trade chilling as defined where tariffs (or indeed other factors) may be repressing trade from RSA to China. We do this by looking at (a) where China is importing from the world but not from RSA and (b) where RSA is exporting to the world but not to China in these product lines, and importantly then (c) where tariffs may be a factor in this lack of bilateral trade. This can provide valuable information to policy makers. 3

5 1 Section One: Chinese imports from RSA During 2005 Chinese data shows that imports from South Africa to the value of some US$3,444 million entered that country, a figure up by 16.5 per cent over the previous year. This was 0.52 per cent of total Chinese imports, a figure that has fluctuated around this level for the past 11 years. This information is displayed in Fig 1, with the dramatic climb in both South African imports and Chinese imports in total (as inferred by the rise in South African values but not market share) clearly evident. Thus, South Africa is doing well, but then so is almost everyone else. Again, the values in Fig 1 (and the rest of this paper unless otherwise specified) are denominated in US dollars. As the Chinese currency has remained pegged to the US dollar for most of this period, Fig 1 is effectively shown in Chinese currency also but with a different scale. Expressing these figures in rand would alter the line for the value of imports but not the market share. Fig 1: Chinese Imports from RSA $m & % Share $ m 4,000 3,500 3,000 2,500 2,000 1,500 1, % 0.50% 0.40% 0.30% 0.20% 0.10% 0.00% RSA Share RSA $m RSA % Source: WTA We have hypothesised that both Australia and Brazil will be major competitors to RSA in the Chinese market. As both are possible FTA partners with China, we considered it useful to analyse the performance of RSA against these two countries in the Chinese market, both in aggregate and throughout the report at a line-by-line level where appropriate. The aggregate import performance of these three countries is shown in Fig 2. 4

6 Fig 2: RSA, Australia & Brazil Comparison $M 18,000 16,000 14,000 12,000 10,000 8,000 6,000 4,000 2, RSA Australia Brazil Judged against both Brazil and more specifically Australia, the performance of RSA has been less impressive over the last three years in particular. The main HS 2 Chapter details for 2005 Details on the top ten HS 2 chapters for RSA imports into China are shown below in Table 1. The complete table for all HS chapters is in Annex One. The table shows, by column: 1. The average duty assessed on that chapter, assessed at a line-by-line basis; 2. The description of the goods; 3. Imports by US$ million during 2005; 4. Growth during This is assessed by taking an average of the values over the most recent six years for that chapter and calculating the percentage that 2005 is above that value (note for comparison that the average is 76% for RSA imports overall); 5. The RSA market share for 2000; 6. The RSA market share for 2005; 7. The RSA average market share over the last six years this and columns 5 and 6 enable one to see whether the RSA imports are gaining or losing market share, both over the period and during 2005; 8. The average Australian market share in this chapter over six years (we would note that this and the Brazilian market share in the chapter is built up from imports at the HS 8 line level, and therefore it represents those lines that are in direct competition to RSA and not necessarily the entire chapter imports); and 9. The average Brazilian market share over the last three years only. 5

7 Table 1: top HS 2 Chapters. $m, % share and growth Duty HS 2 Chapter RSA Details Competitor share av % Description $ 2005 Growth 2000 share 2005 & 6 yr av Aust Brazil 0.0% Ores % 7.9% 3.8% 4.9% 27.6% 16.7% 3.8% Prec. metal/gems % 16.9% 29.0% 24.2% 0.5% 0.4% 3.5% Small Lines % 21.4% 23.0% 22.5% 22.0% 0.7% 5.2% Iron & steel % 0.8% 1.8% 1.6% 0.9% 2.7% 5.6% Aluminium % 0.5% 2.1% 1.7% 10.1% 0.1% 5.5% Organic chemicals % 1.7% 1.5% 1.5% 0.0% 0.4% 11.8% Miscell. chemicals % 0.7% 1.3% 0.9% 0.3% 0.1% 6.8% Fuels % 1.1% 0.6% 0.4% 3.8% 0.1% 3.0% Nickel % 3.1% 3.2% 2.6% 15.5% 1.7% 4.7% Machinery % 0.2% 0.1% 0.1% 0.3% 0.2% Sub Total as % total 92.5% Source: tralac calculations from WTA data Several points are evident. The RSA imports are highly concentrated, with the top ten lines shown in Table 1 accounting for 92.5 per cent of the total trade. The main chapter is metal ores, with imports of $967.2 million and a growth rate as assessed of 94 per cent (above the average of 76% for RSA). During 2005, RSA held a market share of 3.8 per cent, a reduction from both the 2000 share of 7.9 per cent and the six year average of 4.9 per cent. The competitors of Australia and Brazil held market shares of 27.6 per cent over six years and 16.7 per cent over three years respectively. Of particular importance to note is that there are no duties on these ore imports, therefore there are no direct offensive or defensive interests at stake for an FTA with China. Conversely, by examining the tariff rates, RSA growth rates and the relative market shares, the table (and, more specially, the complete table in Annex One) shows those HS chapters where an FTA with either or both of Australia and Brazil may damage South African trading interests. Processed aluminium and nickel imports from Australia are two possible cases above, while there are other more obvious examples in the Annex table. RSA is performing especially well in the next chapter from the table above, that of HS 71, precious metals and stones. Both the growth rate and market share are impressive. Later in the RSA export data and the trade reconciliation analysis, we will show that there are major discrepancies between the two trading partners in these imports, and trace likely reasons for that. The third chapter is HS 98, defined as special small trade lines of low value. It is a Chinese definition with no extra explanation, but notably RSA holds nearly a quarter of these lines. We have no other explanation of these imports, including no information on the tariff duty where we have proxied the overall average for RSA in all other imports into China. 6

8 The next Level of HS 4 details for 2005 Table 2 follows a similar format to Table 1, but this time at the next level of product detail, that of HS 4. For practical purposes this strikes a good balance between the aggregate of HS 2 chapters and the details of HS 8 lines. The layout of the table is similar to Table 1, but not identical. The first column is the average duty on RSA import, followed by the HS codes and descriptions. Next are the imports from RSA for 2000, 2004 and 2005 in US$ million. This is followed by market shares, RSA for both 2005 and the six-year average, and then the Australian six-year average and the Brazilian three-year average share. Table 2: RSA Imports at the HS 4 Level, $m and % Share Duty % HS code Desription RSA imports into China $m RSA Import share Competitor % Av 6 yr Aust Brazil 0.0% 2601 iron ore % 5.5% 32.2% 22.6% 5.1% 7102 diamonds % 23.0% 0.0% 0.0% 3.5% 9801 special class % 22.5% 22.0% 0.7% 0.2% 7110 platinum % 42.8% 0.1% 0.0% 5.3% 7219 stainless steel % 3.1% 0.0% 1.4% 7.5% 7209 rolled iron/steel % 1.7% 0.6% 3.7% 0.0% 2602 manganese ore % 8.7% 39.2% 11.0% 0.0% 2610 chromium ore % 7.3% 5.6% 1.7% 5.4% 7601 aluminium % 3.0% 15.8% 0.2% 2.8% 7202 ferro-alloys % 13.9% 2.5% 0.0% 0.0% 2615 niobium etc ore % 15.9% 34.9% 2.2% 3.0% 7502 nickel % 3.3% 19.5% 2.1% 13.0% 3823 industrial alcohol % 20.1% 1.2% 0.0% 6.1% 2710 refined petrol % 0.2% 0.1% 0.0% 0.0% 2605 colbat % 11.1% 2.8% 0.0% As expected from Table 1, HS 2 chapter 26, that of ores dominates, with five of the 15 individual entries at the next level down some category of metal ores. Similarly from Table 1, both platinum and diamonds figure prominently at the detailed level. In five cases during 2005, RSA held a market share at around 20 per cent or better, and there were another four examples of a market share roughly between 10 and 20 per cent. Again, Australia and/or Brazil feature as major competitors in at least six product lines. The detailed line-by-line HS 8 Level A sample of the full Annex Table 2 is shown below in Table 3. These two tables look at the imports by the most detailed level possible, that of the HS 8 lines. The lines 7

9 are ranked by the average value for the import lines aggregated over the last three years, with the 2005 imports shown for comparative purposes. On the right hand side are the market shares for 2005 and the three years respectively. This configuration enables the 2005 imports to be seen in perspective with the most recent data, and the market shares to be analysed as well. The Annex table looks at the top 100 HS 8 lines, and this covers 95.8 per cent of the 2005 imports and a similar 95.1 per cent of the combined imports. There is a degree of variability in the individual lines that is easier to see in the full Annex table. Table 3: RSA Imp by HS 8, ranked by 2003 to 2005 combined, $M & market % RSA Imports $ m RSA Market share Duty % HS 8 Line Description 2005 $m Av 3 yrs 2005 Av Non-agglomerated iron ores & concentr % 5.4% Low value import items, = RMB % 24.5% Diamonds non-industrial unworked or s % 24.2% Platinum unwrought or in powder form % 50.0% Diamonds non-industrial nes excluding % 25.5% Al unwrought, not alloyed % 6.1% Duties paid at the Chinese border by on RSA imports by tariff rate Table 4 shows the calculations, based upon the most recent Chinese Tariff Schedule, of the duties that would have been paid at the Chinese border on South African imports during These rates are assessed on the most recent tariff data for aggregated imports over the last six years to even out line-by-line fluctuations in the trade. As the Chinese tariff rates were reducing over this period there may be some biases introduced here, but the alternatives are also vulnerable to bias. Note that the second-largest single individual import HS 8 line, which is the small valued items category, does not have an official duty rate associated with it. As outlined earlier, we have taken an average of the remainder of the imports as a proxy for that duty rate, and plugged this value in. The key points from Table 4 are: 1. the overall average duty was 3.452%; 2. some 40% of the lines by value entered duty-free; 3. another 32.4% paid between 1 and 4.5% (including the special lines); 8

10 4. at the other extreme, 0.8% of the imports in the over 30% category paid 9.3% of the total import duties. These lines included sugar, ethyl alcohol, jewellery, truck parts and wool (although the latter is unclear); 5. nearly 40% of the duties are paid on a lesser 18.2% of the imports in the % band; and 6. only 5.3% of the imports by value paid duties of 10% or more, but this accounted for 25.3% of the duties. Table 4: RSA Duties into China, Duty band Av rate % Val Import $m % Imports % Duty Total , % % % % % % % % 3.2 1, free 0.0 1, The relative performance of RSA imports into China at the detailed level. The data presented above has been built up from the HS 8 line-by-line level to gather exact figures. From the detailed data information on the performance of individual lines can be assessed, and this can then been aggregated into a big picture profile. An assessment has been made on 368 lines, with these lines containing some 99.8 per cent of the total imports if the mysterious HS 98 small items are excluded and deleted from subsequent analysis in this section. The cut-off point for these lines was where the import value during 2005 was US1.0 million or more, and as this accounts for 99.8 per cent of the trade when the small items are ignored this effectively means all meaningful import trade. These lines are assessed on changes to market shares, and there are four possible categories for these 368 lines that make up almost all of the RSA imports. The first is the Stars, where the line has both an improved market share for 2005 above the 9

11 South African average over the last six years and where the Chinese imports for that line is also increasing at a rate above that for the average of Chinese global imports. The second category is the Battlers who are doing well where (a) RSA is gaining market share but (b) Chinese imports are not increasing at the average rate in that line. The third category is the Strugglers that could do better where imports from RSA are losing relative market share but where Chinese imports are increasing above the average. The final category are Dogs where RSA is losing market share in lines that are not dynamic growth ones into China. These categories, of course, make no judgement upon the relative profitability on these lines or the reasons as to why the results are as they are shown. It may be, for example, that some of the Strugglers or even Dogs are losing market share in China because RSA exporters have found other more profitable markets. The Stars shining in a growth market (of 368) lines with imports of $1,888.2m, or 39.8% of RSA total; 2. they paid 36.9% of the duties, indicating the average here is slightly above the RSA average; 3. the main lines were platinum, diamonds, manganese, steel and chromium. The Battlers working hard in a relatively declining market lines with imports of $646.8m, or 21.7% of RSA total; 2. they paid a disproportionate 45.7% of the duties, indicating this may be adding to the woes; 3. the main lines were other classifications of diamonds, aluminium and industrial alcohol. The Strugglers slipping in a Chinese growth market lines with imports of $ m, or 34.6% of the total; 2. they paid a lesser 12.1% of the duties 3. the main lines were iron ore (the top import line valued at $702 million), zirconium ore, ferro-chromium and stainless steel. The Dogs losing ground in a relatively declining market lines with imports of $110.7 million, only 3.7% of the imports; 2. they paid 5.4% of the duties; 10

12 3. the main lines were primary plastics, granite, mine dust cleaning equipment, butanone and wood pulp. Overall, a pleasing picture with a solid group in the Stars and Battlers improving their market share albeit in a declining Chinese market relative to all global imports. The Strugglers are more of a worry, while the Dogs are a small group licking their wounds. 11

13 2 Section Two: Chinese Imports into RSA During 2005 South African data shows imports from China to the value of some US$4,926 million, a figure up by 37 per cent over the previous year (well above the global increase of 15.5 per cent). These imports represented 9.0 per cent of total South African imports, a figure that climbed steadily for the past ten years in an almost straight-line manner. This information is displayed in Fig 3, with the dramatic (almost exponential) climb in South African imports from China clearly evident. Expressing these imports in rand would alter the line for the value of imports but not the market share. Fig 3: Chinese Imports into RSA $m & % Share 6,000 5,000 4,000 $ m 3,000 2,000 1, % 9% 8% 7% 6% 5% 4% 3% 2% 1% 0% China $m China % Source: WTA 12

14 Figure 4 shows the increase in market share relative to the EU, South Africa s major import source, and the US. The steady rise in China s share from 2.1 per cent in 1996 to the present 9.0 per cent has, over the most recent two years, been at the expense of both of these sources. Fig 4: Share of RSA Imports 50% 40% 30% 20% EU USA China 10% 0% Tariffs on the Imports from China into South Africa Table 5 shows the distribution of tariffs associated with these imports. Using the most recent SACU schedule we assessed the average duty during 2005 to be per cent. Other features of Table 5 are: some 45.3 per cent of the imports are duty-free; just over one third is paid in the 40 per cent plus tariff group, with this group mostly clothing accounting for 11.8 per cent of the imports; some 8 and 6 per cent of the imports in each of the 20 and 20.1 to 25 per cent tariff groups account for around 12 and 10 per cent each of the tariffs respectively; and in the lower third of the table (20% +) only 30 per cent of the imports account for 85.6 per cent of the total border duties. 13

15 Table 5: Tariff Distribution, Chinese Imports 2005 Tariff Range % Number of hs6 % of Imports % of Duty Paid Zero Total 12.52% av 3705 R31,469m R3,941m The Main HS 2 Chapter Details for 2005 Details on the top five HS 2 chapters for RSA imports from China are shown below in Table 6. The table shows the import values (in rand) and relative market shares for 1996 and Table 6: Chinese Imports to RSA, 2005 $m & % Share HS Descript 2005 $m 2005 % 1996 $m 1996 % All 4, General Machinery 1, Electrical equip Footwear Clothing Clothing Sub Total as % of Total 56.9% 40.5% Source: tralac calculations from WTA data The Chinese imports into South Africa are not that highly concentrated, with the top ten lines accounting for 71.7 per cent of these imports. The comparable figure for South African imports into China is 92.5 per cent of the total trade. The main chapter 14

16 is general machinery, with imports of $1,007 million and a growth rate as assessed of 47 per cent over the period (but below the overall average growth of 59% for China). During 2005 China held a market share of 11.6 per cent, an increase from the 1996 figure of only 1.1 per cent. This is followed by electrical equipment, and the textile/clothing/footwear (TCF) sectors that China has come to dominate globally (the market share that China holds in these three sectors in South Africa is consistent and only slightly lower than the comparable share that it holds in the developed country non-quota markets of Japan, Australia and New Zealand). A comparison: RSA imports into China and Chinese imports into RSA at the General Level The previous section briefly looked at the main HS 2 tariff chapters for Chinese imports into South Africa, while the first main part of the paper looked at the similar picture for South African imports into China. Tariff schedules have an international category called sections, whereby the 99 chapter headings are divided into 20 sections of similar groupings. A sample of a section, that of the Textile and Clothing Section, is shown in Table 7 below, while the full listing of these sections is given in Annex 3. The data shows, by column: 1. The section in bold and the associated HS 2 Chapters shown in the centre to break the table; and 2. Two sets of three columns of data firstly relating to Chinese imports into RSA and secondly RSA imports into China. These three columns show the average duty rate in %, the value of the imports in US dollar million, and the percentage market share of total imports for the respective data sets. 15

17 Table 7: Trade between RSA and China, 2005, HS 2 and Sections. $ and % share Section/HS Chinese imports into RSA RSA imports into China Text/Cloth Duty % Imp $m % Share hs Duty % Imp $m %Share Silk Wool Cotton Other fibre Filaments Fibres Wadding carpets Woven fab Fabrics Knit fabric Clothing Small 0.0 Clothing Worn cloth Section As expected, the table duplicates part of the data shown in table 6 above with respect to the importance of clothing imports from China. It also shows that China has a dominant market share in some, but not all, of the fabrics and textiles, and that the duties are in general high in most of the HS chapters at the RSA border. On the right hand side there is, as expected, very little action excepting for the minor role played by imports of South African wool into China (note that the duty here is most likely an overestimate, as it applies to the out-of-quota rate and not the one and three per cent rates that apply in-quota for wool). 16

18 3 The remaining sections This discussion refers to the complete table given in the Annex, with reference to the main sections only. Agricultural products 1. Animal products China has a market share of 7.0% with imports of $33.8 million, and sausage casings the main product in a low-duty environment. RSA s imports into China are modest, and face medium duties. 2. Vegetable products follow a similar pattern to animal products above, but are of less relative importance. 3. Animal and vegetable fats do not feature. 4. Trade in prepared foods is also modest in both directions, although China does have a high market share in HS 20 into RSA. Non-agricultural Products 1. Ores (iron ores in this case) are the main import from RSA, where the dutyfree trade has an overall 3.7 per cent market share into China. China exports fuels to RSA. 2. Chemicals are important to both partners, with substantive flows both ways. The duties are much higher into China. 3. Plastics and rubber are substantive into the protected (13.4% duty) RSA market, where China has an 8.0 per cent market share. 4. Chinese leather goods hold a compelling market share in the highly protected RSA market with imports of $85 million. 5. Raw logs and paper products are of modest interest. 6. The key textile and clothing sector is discussed above. 7. Footwear follows the exact pattern of textiles and clothing, with China dominating the protected South African market. 8. Both ceramics and glass are important export trade items for China, where a market share of around 20 per cent is held in the South African moderately to highly protected market. 9. Precious stones are the second crucial section for South African imports into China, where the dominant shares are held in diamonds, gold and platinum. The overall duty for these products into China is 3.8 per cent. 17

19 10. Base metals and light manufacturing products are very important to both partners, with South African exports of steel and Chinese exports of manufactured articles featuring in another moderate to high environment. 11. The importance of Chinese imports into South Africa in both general and electrical machinery is another feature of the trading relationship, with duties in the light to moderate range. General machinery exports are also important to RSA. 12. Vehicle and vehicle parts feature in the Transport section for Chinese imports into the moderately protected South African market. 13. China holds a substantive market share in the almost duty-free South African instrument market. 14. Arms show no recorded trade either way. 15. Furniture, toys and other miscellaneous manufactured items from China dominate the unevenly protected RSA market. 16. Works of art does not feature, while, as discussed at length in the early part of the report, South African imports in the Special Products category are large (and the average duty into China has been assigned to these products in the absence of any other information). The relative performance of Chinese imports into RSA at the detailed level. As was the case for the South African imports into China in Section 1, data for this exercise has been built up from the HS 8 line-by-line level to the HS 4 level in order to gather exact figures for the detailed lines where China is actually competing in the South African market. From the detailed data information on the performance of individual lines can be assessed, and this can then been aggregated into a big picture profile. An assessment has been made on 785 aggregated HS 4 lines, with these lines containing some 99.9 per cent of the total imports from China. The cut-off point for these lines was where the import value from China during 2005 was $US50,000 or more. These lines are assessed on changes to market shares, and there are four possible categories for these 368 lines that make up almost all of the RSA imports. 18

20 The first is the Stars, where the line has both an improved market share for 2005 above the South African global import average over the last six years and where the global South African imports for that line is also increasing their overall market share (i.e. China is doing very well in an increasing South African market in real terms). The second category is the Battlers who are doing well where (a) China is gaining market share but (b) the South African imports are not increasing relative to the average in that line. The third category is the Strugglers that could do better where imports from China are losing relative market share but where South African imports are increasing above the average. The final category is the Dogs where China is losing relative market share in lines that are not dynamic growth ones into South Africa. These categories, of course, make no judgement upon the relative profitability on these imports or the reasons as to why the results are as they are shown. It may be, for example, that some of the Strugglers or even Dogs are losing market share in South Africa because Chinese exporters have found other more profitable markets The Stars shining in a growth market * 247 (of 785) lines with imports of $1,822, or 37% of Chinese total imports The Battlers working hard in a relatively declining market * 208 lines with imports of $1,065, or 21.6% of the Chinese total The Strugglers slipping in a Chinese growth market * 63 lines with imports of $956, or 19.4% of the total The Dogs losing ground in a relatively declining market * 193 lines with imports of $1,081, or 21.9% of the imports Note that the definitions used in this analysis are very harsh on China, as we are judging the Chinese performance in each line against the impressive overall Chinese 19

21 performance itself, as we are looking to see if China is gaining market share not in total but relative to its spectacular growth. If we modify this and instead ask the question: Is China gaining market share in the South African imports this line? instead of asking the same question against the share of Chinese overall trade, the answer is much different. Such has been the impressive performance of China over the last few years that for some 97 per cent of the individual imports into South Africa China actually gained market share over its competitors! Data reconciliation To double-check the trade flows it is useful to seek to reconcile the trade data for the respective partners, that is, compare what one is exporting against the analysis that we have used based upon imports. Such an analysis is fraught with difficulties, as it is unlikely, for a variety of reasons, that there will be a concordance of the data. These reasons include currency fluctuations, different valuation methods (fob versus cif), timing differences, the role of an third port such as Hong Kong and intermediaries such as the European diamond market. We have undertaken a reconciliation of Chinese and South African data using the 2005 data for South African exports and Chinese imports. This data shows that imports into China were $3,443.6 million compared to a lesser export value from South Africa of $1,375.7, with the latter representing 40 per cent of the reported imports into China. We would expect exports to be lower than reported imports, as exports from South Africa are recorded as fob while imports into China are assessed at cif, with the differences possibly being in the order of at least 5 to 10 per cent 2. These results represent a major difference notwithstanding the cif versus fob valuations. In order to examine the situation more clinically we collated the data by aggregated HS 2 codes and then divided these codes into four different groups. These four groups are analysed below. 2 This is based upon New Zealand global import data, one of only two countries in the world to publish both cif and fob values for imports. The overall difference has consistently averaged around 7 per cent. 20

22 Group 1: where the differences (measured by exports as a percentage of imports) are between 60 and 150 per cent, with this figure arbitrarily taken to indicate a 'close enough' match. The key points from this group are: Some 23 per cent of the imports and 48 per cent of the exports are in this group. This includes exports of iron and steel products, one of the two major exports, with 64 per cent reconciliation. There are 21 HS 2 chapters in this group. Group 2: where exports are recorded at being represented by less than 60 per cent of the reported import values into China. The key points from this group are: Some 76 per cent of the imports and a lesser 44 per cent of the exports are in this group. There are 39 HS 2 chapters in total represented. These include (1) imports of 'special other', or 13.4 per cent of the imports that have no corresponding export classification. This classification is one of the factors complicating reconciliation, as it represents returned goods for repairs and so forth, according to the Chinese data. This is most unlikely, and we are uncertain what else is included here, but it seems to include significant 'other' and small consignments. (2) precious stones and metals that are 27.6 per cent of the imports, and in this case diamonds and platinum, where exports are only 0.8 per cent of the exports. The possible explanation is that diamonds are exported from South Africa to Europe for an auction before going to China, and gold exports from South Africa are not declared by destination in the trade data. (3) iron ore at 25 per cent of the imports and 29 per cent of the exports, but where the export value is 55 per cent of the import value. As the cif/fob difference here may be significant, we examined quantities traded. This confirms our hypothesis: aggregating quantities over the three years to 2005 show that exports are 91 per cent of the Chinese imports during this period. This gives a satisfactory reconciliation. These three HS lines significantly influence the reconciliation, as they are 69 per cent of the total imports by value. Add this 69 per cent to the 23 per cent of the imports into China in the 'close enough' group and we have a much more acceptable match. 21

23 Group 3: where exports are recorded at being represented by more than 150 per cent of the reported import values into China. The key points from this group are: The group is much smaller (although it contains 26 HS chapters), with only 0.7 per cent of the imports and a low 6.7 per cent of the exports. Exports of inorganic chemicals and vehicles and their parts make up most of this group. Group 4: with no trade reported in one by either or both of the partners is also very minor. There is virtually no trade in this category, and nine HS chapters. Analysing the trade by HS 4 lines directly adds very little light to the reconciliation puzzle. There were 141 HS lines with either exports or imports of at least half a million US dollars during 2005, and these lines accounted for 97 per cent of the exports and 99 per cent of the imports. The results of this analysis showed: 1. between 60 and 150 per cent reconciliation the acceptable above consisted of 45 lines, with 47 per cent of the exports but only 22 per cent of the imports; and 2. where exports are less than 50 per cent of the imports (54 lines) accounted for 36 per cent of the exports and 76 per cent of the imports, again with the iron ores, precious stones and special category featuring. It is likely that a more detailed analysis of quantities traded here may show similar examples to the iron ore where transport costs to China were a considerable component of the overall import price, although we are allowing for that by going down to 60 per cent reconciliation as acceptable. The overall conclusion from the reconciliation exercise is that during 2005, reported South African exports were only 39 per cent of the reported Chinese import value. This was largely influenced by a 'special category' of returned and repaired goods that were not reported as exports, by a massive underreporting of diamonds and platinum that had probably been transited through a third country, and a difference in iron ore that seems to be accounted for by the transport costs associated with a bulky low-value commodity. 22

24 The trade chilling effects of Chinese tariffs and priority sectors Both quantitative and qualitative analyses and projections of the welfare effects of tariff liberalisation traditionally focus on current flows of trade, and as found in this study this analysis is a fruitful source of inquiry in tracing future growth opportunities. Such approaches are unable, however, to estimate where new opportunities might lie. In particular, it is not possible to derive from the standard quantitative models or qualitative analyses a sense of where new areas of trade might be opened up as a consequence of tariff liberalisation in markets. It is quite possible, for instance, for South Africa to have relatively concentrated flows of trade in specific product categories, with one reason for this level of concentration being that the tariff structure outside of those specific product lines is relatively high. In short, as a consequence of these tariffs, trade may have been chilled. The trade chilling effect has already been identified in New Zealand s trade with Australia prior to Closer Economic Relationship (CER) 3, where analysis found that New Zealand s trade widened rather than deepened as a consequence of CER, in other words, as a consequence of trade liberalisation, opportunities for increased New Zealand trade were expanded beyond those products that were not heavily traded prior to the agreement. This is a significant finding. It confirms that the benefits of a free trade agreement are likely to go beyond existing trade and offer possibilities for expansion beyond the traditional (to that market) export sectors. In order to establish if South African trade into the Chinese market may be chilled by a relatively high tariff rate, it is necessary to identify trade to other markets where the tariff has been liberalised and where there is evidence of South African export activity. Global trade is analysed and compared with the trade into China. If it is found that the global trade in a particular product line is extant, but that the same trade with China is limited or non-existent and, simultaneously it is found that (1) a relatively high tariff exists and (2) that China does import from others, then it may be possible to argue that there is (a) evidence of a chilling effect; and (b) potential for new trade with that country once the tariff is liberalised. A more elegant way to undertake this analysis is to benchmark exports to China against a similar country, for example the 3 Sandrey, R Has the New Zealand/Australian Closer Economic Relationship (CER) been Trade Widening or Deepening? MFAT Working Papers Series, MFA. Wellington. 23

25 duty-free destination of Hong Kong. We have eschewed this option, as Hong Kong and China have such a close trading relationship that this comparison would become meaningless. There really is no similar country to China in world trade! Note that we will only look at trade flows in this north-easterly direction, and not Chinese imports into South Africa. In a salute to Chinese traders we consider that given their remarkable performance in both global and the South African markets over recent years there is little to be gained by attempting to analyse their future opportunities in the RSA or any other market. Their remorseless conquests globally attest to their manufacturing and trading abilities. Recognising that this analysis is rather open-ended, we have concentrated upon the most recent 2005 trade data, but keeping an historical perspective at times. Analysis was undertaken on a combination of (a) the 1,094 HS 4 tariff lines where global imports into China were at least $1 million during 2005 to represent the demand side, and these were then compared with the respective HS 4 tariff lines exported to global markets from South Africa during 2005 to represent the supply side. From there, 6 categories were examined. These categories are: 1) Where South African imports held at least a 1 per cent market share in China; 2) Where at least 1 per cent of South African exports went to China; 3) Where South Africa records positive exports to China in 2005 but China does not record positive imports from South Africa; 4) Where positive imports are recorded into China from South Africa but no exports from South Africa are reported; 5) Where there is at least $5 million exported from South Africa globally but no reported imports of this trade reported into China from South Africa; and 6) As in (5) above, but where South Africa reports less than $5 million in global exports (but above $500,000). A side note on Chinese tariff quotas At present, China administers Tariff Rate Quotas (TRQs) on the following agricultural products: wheat, corn, rice, edible vegetable oil (bean oil, palm oil, and rape seed oil), sugar, wool, wool tops and cotton. Analysis of the trade data shows that there 24

26 are some TRQ products inter-dispersed among the six categories above. As these sectors represent sensitive sectors into China, it is important to consider current and potential access for South Africa in these products. The first point to make is that there are products of interest to South Africa in this list. In particular, these are wool and sugar, but potentially wheat, maize, rice, soya bean oil and cotton have all been exported from South Africa globally in the last three calendar years. These TRQ products therefore represent potential opportunities for South African trade negotiators to concentrate upon, with sugar probably the priority. Here the in-quota tariff is 15 per cent, while the out-of-quota rate is an almost prohibitive 50 per cent. It is likely that, except for sugar, the tariffs may not necessarily be the limiting constraint for agricultural products into the Chinese market from South Africa. Other priorities and analysis of the categories above The other key points from this analysis (by category as above) are: 1. South Africa has at least a 1 per cent share the main imports are again iron ore, diamonds and platinum, the ubiquitous special products, and iron and steel products. There are 66HS 4 lines here, and these imports accounted for 72.6 per cent of the total imports from South Africa into China during The overall average market share into China was 4.99 per cent. 2. At least 1 per cent of South Africa s exports go to China the main products here are iron and cobalt ores, iron and steel, and aluminum products (recall that platinum and diamonds are not recorded as South African exports). This category accounted for almost all (97.8%) of the exports from South Africa to China, reinforcing the value of the Chinese market. 3. Positive exports from South Africa that may or may not be reported as imports this is a small and rather eclectic mix where South Africa could possibly be doing better. Lines of interest include carbides and titanium oxides, but overall this group is only 2.8 per cent of South African exports to China and represents a similarly low 3.8 per cent of Chinese global imports. 4. Chinese imports but not reported as South African exports these are all very low values of the Chinese imports, although they do represent 6.7 per cent of the total. There are some where South African global exports are significant, but for products such as fuels, some iron and steel lines, fuel wood, some 25

27 precious metal ores and maize there are obvious reasons as to why these may not be traded. This category accounts for only 3.8 per cent of South African global exports. 5. Where China imports these goods and South Africa exported at least R5 million globally during 2005 (but the twain did not meet) this is potentially the important category for examination, as it clearly shows there are both supply and demand factors that are not, for whatever reason, meeting here. By value the main export items from South Africa are motor cars and aircraft (duties of 15% and 5% respectively), and these are both massive imports into China. Other lines include apples, apricots, pineapples and avocados, chocolate products and processed fish and meats in agricultural goods, titanium oxide and other ores, and some lines of iron and steel and other manufactured products. 6. As in 5 above, but this time South African exports globally were less than R5 million, making these products less of a priority if we assume that exports are facing supply constraints in South Africa. Again, we find here a very mixed list of 143 different lines that may have some limited potential, with nothing really standing out in the absence of a very detailed analysis of South African supply potential. Intra-industry trade The traditional concept of trade is the standard comparative advantage one whereby nations trade products that each can specialise in: South African natural resources for Chinese manufactured goods. However, a particular feature of modern trade, and especially trade between developed countries, is that there is an increasing trade in similar products. Examples of this may be that countries producing similar motor vehicles may see one country making gear boxes and another clutches, or French wine for Spanish wine within the EU; still specialising, but in a very narrow range. This phenomenon sent trade practitioners off in search of a theory to fit the facts a reverse of the way it is supposed to work! Consequently, the term intra-industry trade was coined in the 1970s. Here one divides the trade in a particular item into net or total trade by simply adding imports and exports in each line and intra-industry trade, with the latter a ratio of the difference between exports and imports in that line divided by the total net trade. Thus, if trade is mostly either exports or imports, the 26