VoxEU Column International trade

Trade tensions mount

The last Global Trade Alert report back in July 2011 raised concerns that a deteriorating macroeconomic climate would lead to greater protectionism. The fear has come to pass. This column, which introduces the latest GTA report, shows that the incidence of protectionism in the third quarter of 2011 is as high as during the most troubling of 2009, when protectionist fears were at their peak. Several large trading nations have taken across-the-board measures that adversely affect many trading partners. The world trading system may face its greatest test in the year ahead.

The 10th Global Trade Alert report documents several factors that together imply that the protectionist threat to the world trading system is probably as significant as it was in the first half of 2009, when such concerns were last at their peak. In our last report, published in July 2011, concerns were raised that a deteriorating macroeconomic climate would lead to greater protectionism. This fear has come to pass. The initial reports of the quantum of protectionism in the third quarter of 2011 are as bad as comparable early reports on protectionism in the first half of 2009. Less than a third of these protectionist measures taken are tariff increases or trade defence measures; worse, some of these measures have been taken by large trading nations and affect many sectors or trading partners. Recent protectionism cannot be dismissed as a large number of small pinpricks.

Looking forward, the macroeconomic climate is expected to deteriorate further. For example, it is telling that the most recent estimate for growth by the EU economies in 2012 was only half a percentage point, and that was on the assumption that the Greek and Italian sovereign debt concerns are contained and will abate quickly. The European Commission's forecast openly acknowledged that worse outcomes, i.e. a recession, were possible and, in a telling aside, noted that they could be worsened by growing protectionist pressures.1 The growth slowdown in Europe has already caused the pace of Chinese export growth to Europe to lessen. A recession in Europe would also affect North American multinationals, many of whom still earn a disproportionate amount of sales and profits from European customers.

What is particularly troubling is that in recent months, trade disputes between leading trading nations have widened in scope. For much of 2010 and early 2011, the highest profile disputes concerned so-called currency wars and misalignments - and arguably these were only taken so far. Nowadays, many of the subsidy regimes instituted early in the crisis are becoming the subject of disputes between leading trading nations (see Box 1). The disagreements between China, India, the US, and the EU over local content requirements, technology transfers, and subsidies in the solar power industry are cases in point.2

Now that the scale of discriminatory government intervention in markets during the crisis is adding to trade tensions, one has to ask how strong are the domestic political restraints should another global economic downturn lead to pressures on governments to "save jobs," "protect local industries", etc. As remarkable as it may seem given the tumult of 2008 and 2009, the open world trading system may face its greatest test in the year ahead.

Box 1. The aggregation of crisis-era measures has begun; the period of denial is over.

A recent exchange at the World Trade Organization between China and the US highlights the growing reluctance of leading trading partners to overlook the discriminatory measures that others have taken during the crisis era. In October 2011, the US submitted a notification to the WTO, which is said to include around a thousand pages of translated Chinese legislation, requesting that China notify its WTO partners of 184 subsidy regimes. According to the WTO's website, the US position was characterised as follows at a meeting on 26-27 October 2011:

"The Committee carried out the transitional review of China’s subsidies regime. The United States said China has made many impressive steps to reform its economy, but expressed concern that it still pursued an industrial policy in which subsidies are widely used to protect domestic industry. It said China has an opaque subsidies regime, and that the US had had to file counter-notification on China’s unreported subsidy programmes. Canada expressed concerns about China’s subsidies in the iron and steel sector. Japan urged more transparency, and welcomed China’s recent subsidy notification. Mexico, the European Union and Norway shared the US concerns."3

Days later, the Chinese submitted a formal notification listing 93 subsidy programmes. Some of these programmes4 had been notified by the Chinese government before, but they could well have been scaled up during the crisis era. Other notifications were new. Most of the notified subsidies were part of schemes implemented by the Ministry of Finance and are forms of tax relief (some of which are directly trade-related), cash transfers and other subsidies.

Trade policy developments since July 2011: New protectionist measures outnumber liberalising measures by nearly three to one

Information on trade policy developments discovered after our last report was published in July 2011 to augment the GTA database, available at www.globaltradealert.org. In total, 199 announcements of state measures were found, taking the total number of reports in the GTA database to over 2,000 for the first time. Two-thirds of those new entries (132) relate to state measures that are likely to, or almost certain to, increase the discrimination against some form of foreign commercial interests.5 These measures outnumber the 47 neutral or liberalising measures by almost three to one.

Consistent with previous GTA reports, only a fraction of recently documented protectionism are trade defence measures or tariff increases. Since our last report, new protectionist non-tariff barriers, discriminatory investment measures, export subsidies, and discriminatory bailouts together outnumber new trade defence measures and tariff increases by a ratio of five to two. Once again, governments appear to prefer measures that are subject to fewer, looser, or no multilateral trade rules. One possible interpretation of these findings is, to the extent that legally binding WTO rules have had any effect at all, that it is probably through the choice of protectionist trade instruments rather than the quantum of protectionism.

Once again, the G20 nations are responsible for the lion's share of the recently documented protectionist measures. Since July 2011, a further 104 protectionist measures implemented by the G20 countries have come to light. In the interests of balance, it should be noted that the G20 also implemented 37 measures that limited or reduced discrimination against foreign commercial interests.

The number of product categories (tariff lines) affected by G20 protectionism continues to rise. With the recent protectionist measures, the total number of product categories affected by some type of G20 protectionism has risen 31 to 1080, out of a maximum of 1214. Moreover, since November 2008 - the starting point of GTA monitoring of G20 policies - 215 countries' commercial interests have been harmed by G20 protectionism. As well as the individual reports on each G20 member at the end of the 10th report, these figures give some sense of the scale of the harm done by discriminatory policies of the G20.

Another important recent development has been the fact that leading trading nations have not just undertaken the selective interventions that may harm only a small amount of trade or a small number of trading partners (such as the investigation and then imposition of antidumping tariffs), but some have now put in place measures that potentially affect all or most of their trading partners, or affect a wide range of domestic industries at home. Some of these measures have already received a lot of press attention, others have not. The concern, of course, is that domestic political restraints on discrimination against foreign commercial interests are weakening. Some of the wide-ranging measures implemented in the third quarter of 2011 alone are summarised in Table 1.

Table 1. Selected significant protectionist acts in Q3 2011

Implementing jurisdiction, date
Title in GTA database6

Government measures to be taken

Commerce affected
Australia, 20 October 2011

Changes in the antidumping and countervailing policy

Antidumping and countervailing duty policies

In principle affects all trading partners
Azerbaijan, 10 October 2011

A new policy of state protectionism

Public procurement preferences
Local content requirements

Measure affects 752 industrial product categories
Brazil, 2 August 2011

The "Brasil Maior" plan to advance competitiveness

Government procurement preferences
Faster antidumping investigations
Reduced payroll taxes for selected firms

4 major sectors
France, 11 August 2011

Reduction of shortage occupations list for non-EU/EFTA citizens

Restrictive migration policies

Affects in principle all non-EU, non-EFTA migrants
Japan, 21 October 2011

Comprehensive Package Responding to the Yen Appreciation

Export and R and D subsidies

Multiple sectors affected plus additional targeting of small and medium enterprises.
UK, 4 July 2011

Employment-related restrictions for holders of student visas

Migration restrictions

Affects in principle all non-EU, non-EFTA migrants

The latest update of the GTA database has also led to a revision of the ranking of countries according to the scale of the harm done by their policies. The most significant changes relate to China, now that due account has been taken of various measures China took during the crisis to affect exports. In terms of the number of almost certainly discriminatory measures implemented, China moves up from 9th position to 7th position. In terms of tariff lines (products affected), China now enters the top 10 offenders for the first time, ranked 4th and affecting 698 (out of 1214) product categories. China now moves to the 3rd spot in terms of sectors affected by protectionism, up from 7th. In terms of trading partners harmed, China edges out the combined effect of the 27 EU member states for the top rank. Now China's measures are estimated to have harmed 195 trading partners, as opposed to the EU27's 181 affected trading partners and Argentina's 175 harmed trading partners. The dominance of these rankings by G20 countries and EU member states is apparent.

Table 2. Which countries have inflicted the most harm? Certain emerging markets and European nations


Metric, Country in specified rank, Number

Ranked by number of (almost certainly) discriminatory measures imposed

Ranked by the number of tariff lines (product categories) affected by (almost certainly) discriminatory measures

Ranked by the number of sectors affected by (almost certainly) discriminatory measures

Ranked by the number of trading partners affected by (almost certainly) discriminatory measures


EU27 (242)

Viet Nam (927)

Algeria (62)

China (195)


Russian Federation (112)

Venezuela (786)

EU27 (58)

EU27 (181)


Argentina (111)

Kazakhstan (729)

China (47)

Argentina (175)


UK (59)

China (698)

Nigeria (45)

Germany (161)


Germany (58)

Nigeria (599)

Kazakhstan (43)

India (154)
UK (154)


India (56)

EU27 (550)

Germany (42)
US (42)

Belgium (153)
Finland (153)


China (55)



France (51)

Russian Federation (439)

Ghana (41)


Brazil (49)

Argentina (429)

Indonesia (40)
Russian Federation (40)

Indonesia (151)


Italy (47)

Indonesia (388)

France (150)

Note: There is no single metric to evaluate harm. Different policy measures affect different numbers of products, economic sectors, and trading partners. GTA reports four measures of harm.

Initial totals for Q3 2011 protectionism are particularly high

One of the surprising findings from the latest update of the GTA database is the rather large number of discriminatory measures implemented in the third quarter of 2011. A total of 72 such measures were found by early November 2011. To facilitate interpretation, it may be useful to know that, almost six months after the first quarter of 2009 had closed, the GTA team found that 77 protectionist measures had been implemented in Q1 2009. This is significant for two reasons. First, during Q1 2009 concerns about protectionism early in the crisis were at their peak or, if not, close to it. Second, as the GTA team has come to learn, reporting lags have led us to revise upwards the number of protectionist measures implemented in Q1 2009 to 150 (see Figure 1). That the initial reports for the quantum of protectionism in Q3 2011 are almost as large as those for Q1 2009 is surely a cause for concern. That concern must surely be heightened by the fact that, as shown in Table 1, several protectionist measures were implemented by large trading nations whose effects are wide-ranging and, therefore, likely to be economically significant.

Figure 1. Deteriorating prospects for the world economy since Q4 2010 coincided with an increased resort to discrimination

Note 1: The total quarterly number of harmful measures for Q1-Q3 2011 are converging quickly to the 100-120 range seen in 2009. Q3 2010 seems more anomalous as time goes by. Note 2: A harmful measure is taken to be one which has been implemented since November 2008 and is almost certainly discriminatory (coded red) or likely to be discriminatory (coded amber).

The high number of protectionist measures implemented in Q3 2011 are important for other reasons too. First, the fear that the summer 2011 deterioration in economic prospects might lead to a greater resort to protectionism has come to pass. Policymakers are not dealing with hypotheticals – an increase in protectionism has already happened. Second, the upward revisions of the total amounts of protectionism in Q1 2010 through to Q2 2011, imply that 2010 and the early part of 2011 are rapidly converging to the 100-120 range of “total number of protectionist measures implemented per quarter” that was witnessed in 2009. The last GTA report cast doubt on the wisdom of downplaying protectionism in 2010, as certain policymakers and analysts did, and our latest findings reinforce that initial skepticism. There are real dangers in reading too much into low initial quarterly estimates of the number of protectionist measures implemented.

What policy implications follow from these findings?

For sure, there is diversity across countries, protectionist instruments used, and harm done. Moreover, the evolution of protectionism (away from more transparent policy instruments such as tariffs towards measures less well disciplined by international trade rules) adds to the difficulties in making clean-cut comparisons. Still, the findings reported here suggest that deteriorating macroeconomic prospects have already induced more protectionism, and more protectionism of the most damaging (that is, across-the-board) type. If the recent numbers are anything to go by, those policymakers that were concerned about protectionism in 2009 ought to be as concerned now, possibly more so if one takes a dim view of future global economic growth prospects.

Editors' note: Trade Tensions Mount: The 10th GTA Report is available to download at www.globaltradealert.org.

1 See page 5 of the forecast's "Overview," available at http://ec.europa.eu/economy_finance/eu/forecasts/2011_autumn/overview_en.pdf. This forecast was made public on 10 November 2011.
2 In addition to media reports, other examples of criticism of crisis-era policy responses can be found in the Minutes of the more recent meetings of the WTO's Council for Trade in Goods (obtainable from the WTO's website.)
3 Text taken from http://www.wto.org/english/news_e/news11_e/scm_26oct11_e.htm
4 WTO document G/SCM/N/155/CHN and G/SCM/N/186/CHN, dated 21 October 2011.
5 For these purposes a measure in the GTA database that is both implemented and categorised amber or red is treated as protectionist. The discrimination-based scheme used by the GTA to classify state measures is summarised below Table 2 of this report, which can be found in the next chapter.
6 With the title of the measure and the implementing jurisdiction interested readers should be able to easily access on the GTA's website the report on each of these measures


630 Reads