In particular, TAFTA has significant implications for the debate surrounding the discriminatory effects of rules of origins (RoOs) on the use of tariff references.
RoOs are eligibility criteria imposed to identify the true ‘originating status’ of products. In an FTA, the participant countries maintain their own external tariffs, which usually differ between member countries, while offering concessional tariffs to the member countries. It is therefore necessary to combine tariff concessions with RoOs to prevent ‘trade deflection’: imports from non-member countries into the member country with the lowest tariffs for trans-shipment to other FTA members. Stringent and complicated RoOs can diminish, or even render worthless, the preferences offered to traders.
The preference utilisation rate is the percentage share of trade accounted for by RoO certificates and the actual trade of products eligible for trade preferences. According to our estimates, on the export side (Australian imports from Thailand) the average annual utilisation rate ranged from 60 to 70 per cent during 2005–2010. The utilisation rate on the import side (Australian export to Thailand) was much lower, at around 15 per cent. An inspection of the distribution of preference margins and preference utilisation patterns across products/product groups suggests that the restrictiveness of RoOs are a major factor in determining preference utilisation.
For instance, fully-built automobiles account for over 50 per cent of Australian preferential imports from Thailand. From the late 1980s, the export-oriented motor vehicle industry has grown rapidly and a strong domestic parts and component supply network has evolved side by side with final assembly. As a result, exporters can easily meet the RoOs. By comparison, the figure for electrical and electronics goods, which account for a much larger share (about 20 per cent) of total Thai world exports compared to automobiles (14 per cent), amounts to a mere 5 per cent. Tariff preferences for most products belonging to the latter product group are subject to more complex RoO criteria. Australian preferential exports to Thailand are dominated by primary products like wheat, malt, zinc and aluminium. All these products, being entirely domestic-resource based, easily satisfy RoOs.
The emerging patterns of Australian trade with Thailand under the TAFTA closely mirror the preference utilisation patterns. There has been a notable increase in the total bilateral trade (imports plus exports), from US$5billion in 2004 to US$15.1 billion in 2010. But these figures hide a notable asymmetry in the growth of imports and exports: trade expansion has occurred predominantly on the import side. The share of imports from Thailand in total Australian imports increased from 2.8 per cent in 2004 to 5.2 per cent in 2010, whereas the share of exports to Thailand in total Australian exports varied only slightly around an annual average of 3.5 per cent.
In Australian imports from Thailand, the most striking development has been a sharp increase in automobile imports. Thailand’s share in total automobile imports to Australia increased from 5.4 per cent during 2000–04 to 17.2 per cent in 2010. Thailand is now the second-largest source country after Japan. Thailand’s gain in market share largely mirrors the market share losses of Japan (from 55.9 during 2000–04 to 38.8 per cent in 2010). Given the tariff preferences, Japanese car producers have opted to meet part of the demand in Australia by expanding their assembly operations in Thailand instead of exporting directly from Japan. Notwithstanding the same tariff preferences offered to the US under the Australia–USA free trade agreement, which came into effect in the same year as TAFTA, the US share in total Australian imports of automobiles has remained virtually unchanged during the ensuing years.
Processed food (in particular fish and fish products, chicken meat, canned fruits) and electrical goods and electronics have been Thailand’s other dynamic export products over the past ten years. Yet these products still account for a much smaller share in Australian imports from Thailand, even though most of these products are at the upper end of the distribution of FTA tariff preferences.
In Australian exports to Thailand, the share of manufacturing in total exports has declined from 42.5 per cent to 36.1 per cent between 2000–04 and 2005–10, even though most of the tariff concessions offered by Thailand are concentrated in this product category and these concessions are much larger in magnitude than those on the import side. Interestingly, the share of exports to Thailand in some manufacturing product categories has recorded a decline between the two periods. The post-TAFTA average annual growth rates have been negative in some product categories.
What these figures suggest is that RoO set at the individual commodity level in discriminatory fashion, and commodity-specific supply-side factors which enable traders to meet these criteria, determine the rate of preference utilisation and hence the actual trade flow effect of an FTA. This means that the use of officially announced preference rates in trade flow modelling, as is commonly done in computable general equilibrium modelling exercises, is likely to exaggerate the trade flow effects of FTAs. In addition, the notable expansion of automotive imports to Australia from Thailand at the expense of major traditional source countries points to the importance of taking into account the growing role of international production fragmentation and the resulting shifts in export locations when analysing the trade flow effects of FTAs.
Prema-Chandra Athukorala is Professor of Economics at the Crawford School of Economics and Governance, ANU.
Dr Archanun Kohpaiboon is Assistant Professor of Economics at Thammasat University, Thailand.