Competitive Impact on LAC in World Markets

We now turn to the five-fold matrix of competitive effects of PRC on LAC economies, starting with exports to world markets and then considering the US market alone (see table 1 for definitions of the 'threat' categories). We work at the 3-digit SITC level and over the period 1990-2002 calculate changes in world market share (WMS) based on a comparison growth rates for LAC countries and PRC. For the two years 1990 and 2002 we show the proportions of trade that taken by the five 'threat categories.'

As noted earlier, these calculations can only be suggestive – they cannot prove causation – but nonetheless they are plausible and interesting. Table 7 [ PDF 67.7KB | 1 page ] summarizes the position for Latin America as a whole (LAC-18) in the world market and Table 8 [ PDF 62.4KB | 1 page ] gives the same for the US market. Figure 1 [ PDF 44.6KB | 1 page ] and Figure 2 [ PDF 44.8KB | 1 page ] show the shares of exports for each LAC country under these five categories in 1990 and 2002, for the world and US markets, respectively. Appendix Tables 2 to 7 give the detailed data on the values of exports by the five categories for each country as well as the main five products that fall under each category.

There are large variations by country in the competitive threat from PRC and the nature of the threat changes significantly for several countries. For the world market for all the LAC 18 countries together, the average weighted share of 'threatened exports' – under direct plus partial threat – is surprisingly stable at 45.1% in 1990 and 39.4% in 2002 (Table 7 [ PDF 67.7KB | 1 page ]); there is also a shift in the composition of the threat, from direct to partial. Recall that the direct threat is where a country loses WMS and PRC gains and by this measure the intensify of the Chinese threat decreases significantly over time (this is also true of EA, although there the degree of threat is much higher with on, an unweighted basis, 75% of exports under some form of threat; (see Lall and Albaladejo, 2004).¹³ By our direct threat measure in 2002 on 11% of LAC exports are in this category.

Figure 1 [ PDF 44.6KB | 1 page ] shows the share of 'threatened exports' booth direct and partial in the two years, ranked by the total threat in 2002. The least threatened is Venezuela (less than 20% of exports), shielded by its heavy dependence on oil-based exports. The countries with the largest reduction in the competitive threat in these two categories are Paraguay, Peru and Argentina: all countries that have moved over time into primary or RB products where PRC does not have a strong competitive position or into products like automobiles where PRC is not yet a significant exporter. Countries like Guatemala and Colombia appear to place PRC under threat, because they gain market share in primary products where PRC is a small exporter and is losing market share.

The most 'threatened' countries in LAC in total are Costa Rica, El Salvador and Chile (over 70% of total exports are under threat for the first two countries and around 60% in the case of Chile). While the presence of Chile as a highly threatened country may appear surprising, it reflects the large share of its exports in copper, where PRC gains WMS while Chile loses. Its large exports of fish appear partially threatened because PRC gains more WMS than it does. In Costa Rica the Chinese threat is overwhelmingly partial, with PRC gaining WMS in electronics, instruments, apparel and processed food exports. In El Salvador, it reflects a direct and partial threat in the textile and clothing industry. In terms of the more serious category 'direct threat' all countries see a decline as a share of their exports 1990-2002 and seven (Costa Rica, Ecuador, El Salvador, Guatemala, Mexico, Panama and Venezuela) have less than 10% of their exports in this category in 2002. In terms of direct threat the most threatened are now Chile, Bolivia, Brazil, Uruguay and Colombia) all with more than 20% of their exports in this category (see Appendix Table 4 for details of the main products involved).

While earlier export structure comparisons show that Mexico faces the greatest potential threat from PRC, this calculation shows that because of its very rapid gains in WMS it has not actually faced a significant threat over 1990-2002. The 'directly threatened' exports, in which Mexico loses WMS and PRC gains, only constitute 1.6% of its exports in 2002, down from nearly 10% in 1990. The ‘partially threatened’ exports are much larger, 32% in 2002, up from 19% in 1990, and comprise mainly electronic and electrical products and furniture. These may turn into direct threats if PRC continues to gain market share and actually takes markets away from Mexico. Brazil faces a larger competitive threat (28% direct and 23% partial threat in 2002) but the extent of the direct threat declines substantially, from 46% in 1990. The largest threatened exports for Brazil in the ‘partial threat’ category are telecoms and footwear. On the other hand, its largest single export, aircraft, faces no threat from PRC.

We test in what type of product Latin America countries are losing market share most rapidly by a simple correlation analysis. At the 3 digit SITC level we correlate relative change in market share 1990-2002 (the growth of PRC exports minus the growth of Latin American exports) with firstly the growth of world exports for the product concerned and secondly with the degree of specialization of Latin American exporters (as measured by the revealed comparative advantage ratio, RCA). We carry out this correlation analysis for LAC as a group and for individual countries. For all countries we find the loss of market share to PRC is greatest in the fastest growing categories. For LAC as a group the correlation coefficient although relatively low (0.16) is significant at the 1% level. For Mexico the correlation is higher (0.32) and again strongly significant. As far the degree of specialization is concerned there is some evidence that LAC has held its position better in its more specialized product lines. The correlation coefficient between RCA in 2002 and relative export growth is negative and significant at the 1% for LAC as a group (-0.19) and for Mexico (-0.24) but not for many other individual countries. It also does not hold if we take specialization at the beginning of the period, that is the RCA for 1990.

These results suggest that while potential for a competitive threat exists, LAC faces a significantly smaller threat overall than EA for two reasons. First, export structures as compared with PRC differ far more, and second, structural similarities that do exist have yet to translate into a genuine market share challenge. This evidenced by the fact that if for 2002 one ranks LAC countries by the correlation coefficient of their total export structure with that of PRC and compares this ranking with that by the degree of direct threat (the direct threat category as a share of total exports) there is a significant negative correlation. The Spearman rank correlation coefficient is –0.504 (significant at the 1% level). In other words, the countries with the more similar export structure show lower degrees of export threat. The clearest example is Mexico, the LAC country with the greatest similarity, which has been growing sufficiently rapidly over the 1990s to avoid a loss of WMS to PRC. However, it remains to be seen whether this will continue to be the case.

We now conduct a similar competitive impact exercise for the US market (Table 7 [ PDF 67.7KB | 1 page ]).¹⁴ PRC had a share of US imports of 12% in 2002 compared with just 3% in 1990. Its gain in US market share over 1990-2002, 8 percentage points, was nearly double of that of LAC 18 (note that LAC-M lost market share in the US in this period, almost entirely in RB products). Latin America as a whole (LAC-18) had a share of 17% in 2002, but of this 11% is due to Mexico alone. PRC accounts for about double US imports of LT products as compared to LAC 18 and for almost as much of HT imports. By 2002, it overtook Mexico in HT products (it lagged in 2000) and almost matched it in RB products. For the US market Appendix Tables 5 to 7 give the value of the threat by each category for each LAC country and the five main products in each.

In comparison with the analysis of the world market there are similarities as well as differences (Figure 1 [ PDF 44.6KB | 1 page ] and Figure 2 [ PDF 44.8KB | 1 page ]). In terms of total threat (direct plus partial) Venezuela continues to be the least threatened country in LAC in both exercises. However, in 2002 Paraguay appears as the most threatened country in the US as compared to Costa Rica in the world as a whole, which now appears about half-way in the threat ranks. Mexico appears even less threatened than in world markets, while Brazil appears somewhat more threatened. Argentina also moves up the threat ranks.

By our preferred indicator of direct threat all LAC appear to have a smaller share of trade directly threatened in 2002 as compared with 1990. The most threatened countries are now Chile (around 40% of trade in this category), followed by Argentina and Uruguay with around 35% and Brazil with 30%. The main products involved are Copper (Chile), and Fruits (Chile, Brazil), Petroleum products (Argentina, Brazil), Sugar and Fish (Uruguay) and Internal Combustion Engines (Brazil), none of which appear goods in which PRC might be expected to have an obvious comparative advantage over LAC. The least threatened countries are Costa Rica, Mexico, Panama and Venezuela all with less than 10% of their trade with the US in this category. In 2002 a serious threat to Mexico in particular does not show up in these figures are only about 3% of its exports to the US are 'directly threatened'.

When we carry out a similar correlation analysis to that for the world market we find there is a tendency for the growth of PRC's exports relative to those of individual LAC to be higher in the faster growing categories of US imports, however this result is not significant for LAC as a group nor for Mexico. For LAC-Mex there is a weak correlation of 0.15 (at the 5% level).

The views expressed in this paper are the views of the authors and do not necessarily reflect the views or policies of the Asian Development Bank Institute (ADBI), the Asian Development Bank (ADB), its Board of Directors, or the governments they represent. ADBI does not guarantee the accuracy of the data included in this paper and accepts no responsibility for any consequences of their use. Terminology used may not necessarily be consistent with ADB official terms..

Comment(s)

There are [0] comment(s) for this entry. Post a comment.