The impact of crisis-driven protectionism on EU exports: The “Russian doll” effect

Lucian Cernat, Nuno Sousa 09 January 2010



The trade collapse of 2009 was as severe as in the Great Depression – if not more so.

The two years following the Smoot-Hawley tariff of June 1930 saw the volume of U.S. imports fall by over 40%. Misguided trade policy was responsible for 25% of this trade collapse. The Smoot-Hawley tariff alone accounted for 4%–8%, while the combination of specific duties and deflation – which further raised the effective tariff – reduced imports by an additional 8%–10% (Irwin, 1998).

The modern-day crisis has also triggered protectionist responses. The Global Trade Alert, along with other international organisations such as the WTO, has exposed numerous protectionist measures over the past months (Evenett 2009). As argued in VoxEU's The Great Trade Collapse, however, the emerging consensus is that the decline in world trade was primarily a reaction to the financial and economic crisis – notably the slump in demand and disruption of trade finance – rather than driven by trade policy (Baldwin 2009).

In this column, we contribute to this debate by focusing on the EU. We examine the impact of protectionist measures taken by the EU's trading partners on the region's exports.

We make use of the latest WTO dataset and a number of systematic reviews of crisis-driven protectionist measures taken around the world, including the November 2009 European Commission report on potentially trade restrictive measures.

Dissecting protectionist measures

Understanding crisis-led protectionism is like opening up a Russian doll. Protectionist measures are both independent and interconnected – in the same way as Russian dolls are all a part of one another. Traded products can be affected by tariffs, non-tariff measures, regulatory issues, subsidies, import bans, and “buy local” clauses.

These interventions can have an affect on some or all traded products. Protectionist measures such as an import license can affect a narrowly defined product whereas customs procedures or balance of payment measures will apply to a broad range.

An appropriate analysis of crisis-led protectionism must therefore go beyond merely counting protectionist measures and then estimating the net effect on trade.

But opening "Russian doll"-style protectionist measures is not an exact science – labelling a specific measure as protectionist or discriminatory requires careful judgement. With these complexities in mind, we apply an intermediate level of analysis to assess the magnitude of EU exports that have been potentially negatively affected by protectionist measures.

How much EU trade has been affected by crisis-driven “border protectionism”?

We use the latest WTO and European Commission data and find that new border measures such as tariffs, quotas, import licenses, reference prices, and import bans have targeted around 1.7% of EU exports (based on 2008 trade figures).1 While this confirms the claim that “border protectionism” has been relatively minor, the EU does not compare favourably with the rest of the world. The latest WTO monitoring report estimates that import restrictions introduced since October 2008 cover a maximum of only around 1% of world trade in merchandise.2

When looking at the systemic implications of crisis-led protectionism, the data show that most new border restrictions have been imposed by a limited number of relatively small EU trading partners. The clear exception is Russia.

Figure 1. New border restriction on EU exports, by trading partner

Source: Authors’ calculations

Russia was the second-largest market for EU exports in 2008 and has been responsible for the introduction of measures that were worth almost 75% of the total value of EU global exports affected by protectionism. In total, the measures affected around 15.5% of total EU exports to Russia.

Tariff increases have had the most significant impact on EU exports. Tariffs alone target around 1.5% of the EU exports, out of the total 1.7% of exports affected by border measures. (See the Appendix Table 1 for details.)

Although the percentage of EU exports affected is small, in absolute terms this means that around €22 billion worth of merchandise exports are now subject to increased protection. The affected trade could actually be somewhat larger if the indirect effects of third-country protectionist measures on global value chains are included, such as the impact of other countries’ measures against non-EU exports of finished products incorporating EU intermediate exports.

Complications in the calculations

But it is important to note that these estimates are overestimates. Many measures target very detailed products or tariff lines, but with the WTO dataset we can only identify products affected by protectionist measures at a more aggregate level – namely HS6. This implies an overestimate, as not all products in a given HS6 category are affected by a given tariff-line change.

Given the intricate and complex nature of protectionist tendencies, however, this overestimate of border-related trade effects may still underestimate the overall impact of crisis-led protectionism in reality, given that behind-the-border measures such as “buy local” provisions are not included.

Looking inside the Russian doll

Understanding the full impact of protectionism requires an assessment not just of affected trade but, more importantly, of the lost trade flows. As with Russian dolls, however, the more layers one sheds, the less that remains. This lost trade core effect of protectionism is by definition smaller than the 1.7% affected trade that we estimated for the EU. But this trade loss cannot be quantified without further detailed analysis of each individual measure.

The analysis is complicated further by the fact that a number of countries have used non-tariff border measures. Some anecdotal evidence suggests that the introduction of non-automatic import licenses in Argentina has been burdensome for businesses; such licenses often take as many as 30 days to be processed due to a backlog of applications. A one-day delay in granting an import license in middle-income countries in manufactures could be equivalent to a considerable tariff increase such as 0.4% in textiles and 1.5% in mineral-based products (Minor and Tsigas 2008). This suggests that non-tariff border measures could have a significant impact on exports.

A thorough analysis of crisis-led protectionism should also include indirect effects, like the impact of export restrictions on intermediate imports (e.g. raw materials) that could be essential determinants of export performance. On a global scale, the introduction of new exports restrictions has been fairly limited. Only Argentina, Russia, China, Indonesia, and Vietnam have made use of these instruments.

In some cases, however, bilateral effects can reach more significant levels. For instance, Indonesia's new regulation stipulating that exports worth $1 million must be supported by letters of credit will affect products that represent around 23% of the EU's 2008 imports.

How well has the trading system coped with protectionist tendencies?

The fact that only a few countries resorted to border measures can be interpreted as evidence that the WTO passed an important stress test – perhaps its most important since its inception as the GATT in 1947. Clearly a 1930s-style protectionist spiral has been avoided. Russia, which has not yet acceded to the WTO, is among the countries that have used border measures most widely, offering a glimpse of what might have happened without the current trade rules.

Countries such as Bolivia, South Africa, Egypt, and Indonesia have made extensive use of the "flexibility" that WTO rules offer by pushing up applied tariffs to bound levels.3 But, clearly the WTO system of rules has worked to restrict some policy initiatives by providing a legal basis for their withdrawal, for example in the case of Ukraine, which has inadequately resorted to trade measures to tackle a balance of payment crisis.

Murkier forms of protection

Digging beyond this first layer of the "Russian-doll"-style protectionism may reveal several other layers of protectionist initiatives that are increasingly more difficult to analyse from a trade policy perspective but may affect even larger shares of exports. There seems to be some correlation between the propensity to impose more complex forms of protectionist measures and the lack of boundaries coming from the WTO legal framework.

The G20 partially managed to address this threat at an early stage of the crisis through a self-imposed standstill on new measures that could restrict or distort investment or trade. Despite this, it is clear that the trading system has proved – as expected – to be the most vulnerable. The unprecedented public support to the economy may have helped to prevent an escalation of inward-looking policies, but it poses challenges to the competition environment.

Uncovering and quantifying the implications for trade will require more detailed analysis on an individual basis grounded on extensive information about the specific design of interventionist measures.


Compared to the Great Depression, the systemic impact of today’s protectionist tendencies has been contained. Less than 1% of global trade and less than 2% of EU exports have been affected by such measures.

But this crisis might put an additional strain on those areas where the absence of clear WTO rules offers unbounded "protectionist policy space”. This in turn could prevent trade from unleashing its full potential as an engine for economic recovery. One can only hope that the more dangerous layers of "Russian-doll"-style protectionism remain unopened.

Disclaimer: The views expressed herein are those of the authors and do not necessarily reflect the views of the European Commission or EU Member States. The authors would like to thank Denis Redonnet, Marco Dueerkop, and Iglika Yakova for useful comments.


1 Similarly to the WTO computations, only trade measures that are actually in force are covered. However, we aim to focus exclusively on crisis-related measures and therefore, contrary to the WTO, we excluded import bans related to the H1N1 virus outbreak as well as the use of trade defence instruments. Data from the World Bank show that the use of TDI has not been, at least so far, greatly affected by the crisis, see World Bank (2009).

2 We also note that some countries introduced measures aimed to liberalise imports. Such measures have targeted products that represent around 0.4% of EU exports in 2008. However, in some cases the lifting of import restrictions is specifically targeting raw materials and aim to increase the effective protection in certain downstream industries.

3 Other countries such as India, Indonesia, Mexico, Turkey, Brazil and Vietnam have also taken advantage of this policy space but they have done so in a more "selective" fashion.


Baldwin, Richard (ed) (2009), The great trade collapse: Causes, Consequences and Prospects,

European Commission (2009), “Fifth report on potentially trade restrictive measures in the context of the global economic crisis”, November.

Evenett, Simon (ed) (2009), The Unrelenting Pressure of Protectionism: The 3rd GTA Report, Centre for Economic Policy Research.

Irwin, Douglas (1998), "The Smoot-Hawley Tariff: A Quantitative Assessment", Review of Economic and Statistics, vol. 80(2), pages 326-334.

Minor, Peter and Marinos Tsigas (2008), "Impacts of Better Trade Facilitation in Developing Countries: Analysis with a New GTAP Database for the Value of Time in Trade".
World Bank (2009), "The Pattern of Antidumping and Other Types of Contingent Protection" World Bank PREM Notes No. 144, 21 October.


Appendix Table 1. Overview of measures affecting EU trade flows by type and by partners (Measures imposed since October and that are actually in force in November 2009)

  Trade restricting/distorting measures Trade promoting measures
  Tariffs Quotas







Tariffs License




    X X X   X X
Australia           X    
Bolivia X              
Brazil X         X    
Canada   X       X    
China       X X X   X
Ecuador X         X    
Egypt X              
India X     X   X   X
Indonesia X     X X      
Korea X              
Malaysia           X X  
Mexico X         X    
Mongolia X              
Russia X X     X X   X
Vietnam X       X X    
Turkey X              
S. Africa X              
Switzerland         X      
US         X      

Source: WTO.

a. Includes non-automatic import licenses, shipping inspections, etc.
b. Includes export restrictions (namely tariffs and quotas) as well as export support measures.



Topics:  Global crisis International trade

Tags:  trade policy, EU, global crisis, protectionism

Head of Global Regulatory Cooperation and International Procurement Negotiations, DG TRADE

Directorate-General for Economic and Financial Affairs of the European Commission


CEPR Policy Research