Protection is often viewed as a powerful instrument to help domestic firms to raise their sales at the expense of foreign importers. But this view is now being challenged by recent research showing that the effects of protection really depend on the international orientation of the firms i.e. whether they are exporters or not. Protected firms that are well integrated in global value chains may actually lose sales whenever the imports of inputs are subject to protection. This observation may not come as a surprise, but it is important to realise that trade policy has not kept pace with this aspect of globalisation.
The main reason is that many of the current WTO rules governing trade protection stem from an era where trade models predicted that all domestic firms would benefit from import protection (see below). Traditional theory models assumed that all firms in the protected industry are import-competing and only sell domestically. However, in recent years an increasing number of papers have shown that even within narrowly defined industries, firms can be very different. Some firms only produce for the domestic market, others mainly export or sell both domestically and internationally. Thus, the question that can be raised is whether all domestic firms benefit from import protection given that some of the protected firms may be exporters.
Which firms benefit from import protection?
Konings and Vandenbussche (2012) look into this question by using French firm-level data. In this paper we consider whether all French firms that get import protection also benefit from that protection. Our findings suggest that protectionist measures have a heterogeneous effect on domestic firms covered by the antidumping protection. We study the effects of protection on exporters and non-exporters (Konings and Vandenbussche 2012) and use ‘programme evaluation’ to compare French protected firms to similar but unprotected ‘matched’ firms1.
Domestic French firms without any international activity benefit from import protection and see their domestic sales rise while exporters, especially those belonging to a global network, lose sales relative to unprotected firms. More in particular we find that antidumping protection raises firm-level domestic sales of non-exporters by 5%. For exporters, we find that antidumping protection lowers their exports abroad by about 8% and this fall is not compensated by an increase in exporters’ domestic sales, which also fall by around 4%. At the product level, extra-EU French exports drop by 36% while total extra-EU exports fall by 21%. We verified that these results are not driven by retaliation policy, endogeneity or the choice of control group.
The empirical finding that trade policy hurts exporters appears largely consistent with case evidence available in the literature. For example, Isakson (2007) describes in detail a European antidumping case of imported leather shoes in the EU where the imposition of a duty was heavily supported by relatively small Italian shoe producers whose market was largely local, but was strongly opposed by large and more ‘globalised’ EU shoe producers such as Ecco, a Danish firm. The reason Ecco opposed the protection was that it imported shoes from China and, after branding and labelling in Europe, exported the shoes across the world. As a result of the antidumping duty, the importing of shoes became more expensive, resulting in losses on domestic and exported sales. Other European manufacturers such as Diesel, Adidas, Hush Puppies and Puma also opposed the same shoe tariffs for the same reasons. Indeed, oftentimes multinationals often oppose the antidumping protection they receive.
An alternative explanation for the fall in exports of French firms covered by EU antidumping protection might be reduced market access abroad due to the retaliatory action of targeted trade partners outside the EU. Such retaliatory actions are difficult to empirically capture since they may or may not occur in the same sector and often take some time to materialise.
We find that the economic significance of exports going to targeted countries is low (Konings and Vandenbussche 2012). Exports to target countries represent only 1% of the total export value of products in our antidumping cases. This suggests that while retaliation adds to the fall in exports, it cannot explain the substantial decrease in exports observed during antidumping protection.
A more plausible explanation is given by the trade theory of heterogeneous firms. Since exporters are more likely to be integrated in global value chains, typically these are firms that engage more in the fragmentation of production through importing and outsourcing. You would indeed expect exporters to be adversely affected by trade policy that interferes with global value chains. Indeed, the ‘winners’ of antidumping protection appear to be the non-exporters while the exporting firms are the ‘losers’ with a relative loss in domestic sales and exports during protection (compared to exporting firms not subject to protection).
Some trade policy uses protection as an instrument to protect its domestic import-competing sector. If this policy does not take its negative externality on protected firms’ exports into account, it may have negative long-run consequences. Firms today no longer operate within the confines of a singular country or market, and their operations are increasingly international. Two decades ago, when firms mainly sold domestically, import protection laws may have been an effective way to temporarily boost a country’s trade surplus and current account. It is no longer the case today. In an increasingly globalised world, exporters’ success seems to positively depend on the free entry of imports rather than the other way round.
Cadot O, A Fernandez, J Gordon, A Mattoo (2012), Where to spend the next Million? Applying Impact Evaluation to Trade assistance, CEPR and World Bank.
Konings, J, Vandenbussche, H (2012), “Import Protection hurts Exporters”, Review of World Economics, forthcoming.
Konings, J, Vandenbussche, H (2008), “Heterogeneous Responses of Firms to Trade Protection”, Journal of International Economics, 76, 371-383.
1 This technique which has often been used in labor economics is now also finding its way to assess trade policy issues (Cadot et al. 2011).