期刊名称：Document de Travail / Centre d'Etudes Prospectives et d'Informations Internationales
出版年度：2007
卷号：1
出版社：Paris
摘要：According to WTO rules, countries are allowed to adopt regulations under the Sanitary and Phyto-Sanitary (SPS) and Technical Barriers to Trade (TBT) agreements in order to protect human, animal and plant health as well as environment, wildlife and human safety. These non-tariff barriers (NTBs) may play an important role in the conduct of international negotiations. Developing countries (DCs) protest regularly against the increasing use of NTBs by developed countries.
In this paper focusing on agricultural trade where such measures play a prominent role, we investigate two central questions: first, do these measures influence significantly trade flows? Second, is the impact similar for all exporting countries or are there differences (i) between OECD countries and developing (DCs) & least developed (LDCs) ones?
Our source data are WTO members’ notifications of SPS and TBTs. These notifications are collected and analyzed by the UNCTAD. For each notification, the database provides the notifying country, the affected product (at the six-digit level of the Harmonized System of classification), and the classification code of the barrier. The UNCTAD distinguishes seven broad categories of measures. These categories includes 43 different measures such as the ban of some products (SPS) or technical measures (pre-shipment inspection or quarantine requirements). Considering the number of affected products, “technical barriers” is the most frequent measure. With rare exceptions, SPS and TBT measures are applicable to all exporting countries. They do not have a bilateral dimension. However, exporters will be differently affected by these measures depending on the structure of their exports in terms of products and markets.
We firstly provide some descriptive statistics on SPS and TBTs by merging information on notifications with trade data. Data on trade flows are for the year 2004 and come from the BACI database developed by the CEPII. Our sample includes 690 agricultural and agro-food products. Among the 154 importing countries, only 92 notify measures under the SPS and TBT agreements. Using these data, we calculate a coverage ratio, which corresponds to the ratio of imports in notifying countries over world imports in affected products.
Over the 690 products, only 4 do not face any barrier in any importing country. For the remaining products, measures are notified by at least one importer; for these products, the average coverage ratio is 45%. 502 products have a coverage ratio above 25%. The average coverage ratio for them is 55%. For 20 products, one can suspect a protectionist use of barriers, identified as cases where only five or less countries enforce a measure on a product. 366 products of our sample are “sensitive”: at least 25% of importing countries notify a measure. The average coverage ratio for these products is 50%. The most affected exporters in terms of coverage ratio are developing countries. On the other hand, the most affected exporting countries regarding the number of affected products are developed countries, which export more products and face more NTBs.
This inventory approach misses two important issues: : first, it does not necessarily inform on the stringency of SPS and TBT measures; second, in case of incomplete information on traded products, SPS and TBTs can facilitate trade by signaling that products are safe to the consumer. In their absence, there might be no trade at all. Using the same data, we accordingly estimate econometrically the impact of SPS and TBT measures on bilateral trade in agricultural products.
BNTs are usually introduced as explanatory variables in models of trade flows (mainly gravity equations). Evaluations of these NTBs are usually based on frequency or coverage indexes. In the paper, we adopt another method, which consists of introducing ad valorem equivalents (AVEs) of NTBs. These AVEs are directly comparable with tariffs. Different methods have been used in the literature for estimating AVEs. First AVE could be computed directly as the difference between the domestic price of the imported product in the presence of NTBs and the reference price of the same product. However, due to the absence of detailed price data for a large number of countries and products, we prefer to use the indirect derivation developed by Kee et al. (2006). This approach consists of first estimating the quantity impact of NTBs on trade flows (by introducing a dummy variable in a trade model). This quantity impact is then converted into an AVE using import demand elasticities. In our estimations, we use AVEs estimated by Kee et al. (2006). Besides, bilateral tariff barriers extracted from the MAcMap database (jointly developed by the ITC (UNCTAD-WTO) and the CEPII) are also introduced in our estimations in order to distinguish the impact of NTBs on trade from that of tariffs. For lack of data, we restrict our sample of importers to OECD countries.
Our results first show that, on the whole, SPS and TBT measures influence negatively OECD imports. Our estimations also suggest that SPS and TBTs do not affect significantly bilateral trade between OECD members but reduce significantly DCs and LDCs exports to OECD countries. Within DCs, Cairns and non-Cairns members’ exports are similarly affected by SPS and TBTs. Furthermore, EU imports seem to be more negatively influenced by SPS and TBTs than imports of other OECD countries. Lastly, our sectoral analysis shows that SPS and TBT measures could foster trade in some sectors.
关键词：Agriculture; sanitary and phyto-sanitary norms; technical barriers to trade; ad valorem equivalents; protectionism
