Uneven compliance: The sixth report of the Global Trade Alert

Simon Evenett 23 June 2010

a

A

In many industrial countries, nascent recovery has been called into question by a sovereign debt and liquidity crisis that arose in the Eurozone in the second quarter of 2010. The new macroeconomic context – with its growing emphasis on fiscal restraint – may influence government behaviour towards open borders as well. This latest twist in the financial crisis has certainly affected the value of the euro, with the unanticipated knock-on effect of making the Chinese government less willing to revalue the renminbi than before – much to the consternation of certain members of the US Congress, who in turn have threatened to impose extra tariffs on Chinese goods.

The latter example highlights the different ways in which crises can test policymakers' resolve to keep borders open to international commerce. Policymakers and trade diplomats need to align expectations with these new circumstances and should not assume that formulas used to keep borders open in the past will be successful in the future. This column first discusses the possible implications of fiscal restraint for protectionist dynamics before summarising the main findings of the sixth Global Trade Alert (GTA) report.

The case for open borders is even stronger in an era of fiscal retrenchment

In the second quarter of 2010, such was the pressure from financial markets that many European governments announced measures to reduce their budget deficits and borrowing needs. These measures involved both a combination of tax increases and expenditure reductions, both of which are likely to reduce the contribution of the state to total aggregate demand in national economies. What challenges or imperatives for commercial policymaking arise because of fiscal retrenchment?
Coming at a time when most European economies had barely turned the corner (from contraction to expansion), and with investment outlays constrained by frozen corporate debt markets, net exports are seen as one of the few sources of demand stimuli.1 This should strengthen the interest that every European government has in its major trading partners keeping their markets open.2

This argument has another side, of course, for it is net exports that contribute to aggregate demand, not gross total exports. Even with the euro depreciation, European governments should not seek to further squeeze import market shares with discriminatory policies. Doing so may well trigger retaliation which would limit access to foreign markets and detract considerably from the attractiveness of discriminating against imports in the first place. Overall, so as to maximise opportunities for export growth, governments engaging in fiscal retrenchment ought to strongly support open borders. Ultimately this will require leading by example.

Another consequence of financial markets applying pressure for fiscal retrenchment is that some floating currencies may depreciate substantially, with implications for trade frictions. Reference has already been made to the impact of the euro's devaluation on the US-Sino currency dispute. But the matter does not stop here. If trading partners are given grounds to suspect that a depreciation was deliberately induced or reinforced by a government, then this may trigger retaliation in different forms. Depreciations had better appear to be market-driven or the potential for difficult commercial policy disputes will increase.

The third potential relationship between fiscal retrenchment and commercial policy discrimination relates to the widespread use of subsidies and bailouts in the early phase of the crisis (that is, from November 2008 until second quarter 2010). One of the reasons governments have run up such large budget deficits is precisely because they chose to intervene during the crisis more often with subventions rather than price-based instruments, such as tariffs.

Pressure from the financial markets to reduce budget deficits may cause governments to scrap crisis-era subsidies, which on the face of it is desirable as many such subsidies distort the operation of market forces. However, if subsidy removal is replaced by other forms of government intervention, then the principal effect of fiscal retrenchment on commercial policy discrimination may be to alter its composition, rather than to reduce its level.

Much will depend on the viability of subsidy recipients and the extent to which the macroeconomic impact of fiscal retrenchment, including potential exchange rate depreciation, worsens or improves their financial health. A worst case scenario arises when retrenchment causes aggregate demand to fall considerably, subsidy withdrawal leads to fear of corporate viability and further job losses, and the state intervenes replacing a prior subsidy with a competition-restricting or profit-shifting measure.

To summarise, while there are good arguments why governments implementing fiscal retrenchment should argue for open borders at home and abroad, pressures from firms and unions may be at their greatest if that retrenchment involves unwinding the very subsidies that added to the budget deficit in the first place.
Since rule-based solutions are off the table for G20 members and evidently are evaded frequently enough in the EU, the only tool at hand to strengthen government resolve during this next commercial policy challenge is mutual support through enhanced monitoring and peer pressure. This begs the question: What have the existing mechanisms accomplished? The report's findings on this question are summarised in the next section.

What has the G20 accomplished since the Pittsburgh Summit?

Given the upcoming G20 summit meeting in Toronto, we have taken the opportunity to assess what has happened since the last G20 Leader's summit in September 2009. Worldwide, governments have imposed 357 more state measures that discriminate against foreign commercial interests, almost trebling the amount of observed discrimination (to 554 measures). Government measures that harm foreign commercial interests outnumber beneficial measures four to one, although it should be borne in mind that each measure may differ in impact.

Like the period before the Pittsburgh summit (November 2008-September 2009), afterwards G20 governments were responsible for just over 60% of all the discriminatory measures implemented worldwide. It should also be noted that in the latter period, 80% of the benign or beneficial measures implemented worldwide were put in place by G20 governments.

What can one make of the G20's record on protectionism? Some have argued that it was naïve to expect G20 members to refrain from any form of protectionism during the crisis. Others argue that every G20 member has broken their pledge and every G20 member knows this, that each G20 member values the flexibility to do so, and therefore one should expect little from peer pressure initiatives and the like. On this view, the pledge on protectionism amounts to “window dressing”; cheating is acceptable to all precisely because fear of the unknown (which is heightened in a crisis) calls for preservation of options.

If “realpolitik” or “options preservation” were the factors underlying the calculations of policymakers, then the facts uncovered since the Pittsburgh summit highlight the growing costs of either “logic”. Such has been the resort to discrimination by governments, in contravention of the G20 and other such pledges, that there is a growing price paid by each country's commercial interests. The costs of the status quo mount quarter by quarter.

Perhaps, before the Pittsburgh summit, China was prepared to pay a price of 99 hits to its commercial interests abroad in exchange for preserving its option to discriminate at home. Now that the price paid by China has risen nearly 200% to 282 hits to its commercial interests; is the weak G20 peer pressure exercise still in China's economic interests? As Table 1 makes clear, several other large trading nations have paid a substantial price for going along with the current G20 trade initiative. Moreover, many of the hits taken to national commercial interests have been discovered since the Pittsburgh summit. Will each government continue to accept this damage to their commercial interests, especially when there is a lot of variation across countries in harm inflicted? If not, some should consider arguing for more vocal monitoring and peer pressure to discourage new discrimination and an unwinding or exit initiative to terminate measures taken since the G20 pledges were first made.

Table 1. Since Pittsburgh many G20 countries have seen their commercial interests hit very often; worse, there are still plenty of pending measures in the pipeline

Top 10 Targets
Number of discriminatory measures imposed on target
Number of pending measures, which if implemented, would harm target too
This report
(June 2010)
Increase from previous G20 meeting
(2nd GTA report)
This report
(June 2010)
Increase from previous G20 meeting
(2nd GTA report)
China
282
183
125
48
EU27
266
na
80
na
USA
213
127
46
27
Germany
204
120
56
26
France
188
110
46
22
UK
181
109
44
24
Italy
175
105
50
27
Belgium
170
92
42
21
Japan
168
90
47
24
Netherlands
163
92
42
24

Note: Unfortunately when our second report was prepared data on the total number of times the EU27 nations were harmed was not collected, hence frustrating direct comparisons between early June 2010 and early September 2009. Data on EU27 was reported from the third report on due to requests from users. Moreover, individual member state information in this table may indicate the extent how often some European trading nations have been harmed since the Pittsburgh summit.

Another factor that might create dissension among the G20 is the recognition that some countries inflict a lot more harm than others (see Table 2). Any notion that the current G20 process generates a parity of pain and opportunity ought to be dismissed. The GTA reports four indicators of harm done by a nation's policies and the top 10 worst offenders on each metric are listed in Table 2.

In this table the EU27 refers to the combined impact of all the actions taken by the European Commission and the 27 Member States. Together, the EU27 appear in the top 5 worst offenders on all four metrics, a dubious honour shared with no other trading entity or nation. It should be noted, however, that most of the harm done by the EU27 grouping result from measures taken by the EU member states and not by the European Commission. Some countries (Argentina, China, Ethiopia, India, Indonesia, Germany, Nigeria, Russia, and Venezuela) are found repeatedly in the top 10 lists of worst offenders. Most of these countries are large emerging markets. Going forward, greater attention should be given to holding to account those governments that violate the G20 pledges.

Table 1.2. Some countries inflict more harm than others

Rank
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
1.       
EU27 (146)
Venezuela (784)
EU27 (55)
EU27 (168)
2.       
Russian Federation (73)
Kazakhstan (719)
Algeria (54)
Argentina (161)
3.       
Argentina (41)
Nigeria (599)
Nigeria (45)
China (161)
4.       
India (31)
EU27 (437)
Venezuela (38)
Indonesia (152)
5.       
Germany (29)
Russian Federation (421)
Kazakhstan (36)
Russian Federation (142)
6.       
UK (24)
India (347)
 
Indonesia (347)
Russian Federation (34)
Finland (132)
Germany (132)
South Africa (132)
7.       
Indonesia (22)
Ethiopia (32)
8.       
China (19)
 
Italy (19)
Ethiopia (345)
Indonesia (32)
 
9.       
Argentina (336)
India (31)
Belgium (131)
Brazil (131)
10. 
Austria (17)
China (335)
Germany (27)

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. The EU27 refers to the sum of all of the measures taken by the 27 national governments of the European Union and the measures taken by the European Commission. To be included in this total a state measure must have involved discrimination against the commercial interests of another state, including potentially another member of the EU.

The trade coverage of “jumbo discriminatory measures”

Counts of state measures reveal the frequency of beggar-thy-neighbour interventions by governments, but not necessarily the amount of trade potentially covered by such discrimination. In 2010 a number of reports by international organisations have sought to estimate the “trade coverage” of certain measures taken by governments. The latest report of the WTO secretariat (issued on 14 June 2010) computes the trade coverage of “import restricting measures” implemented since 1 November 2009. The WTO secretariat estimates that trade amounting to approximately 0.4% of total world imports is affected by these measures. The WTO secretariat did not comment on the policy implications of the magnitude of this figure.

For the first time the GTA decided to check whether the approximate level of trade coverage could, in fact, be so small. Rather than restrict our analysis to “import restricting” measures, objective criteria were used to identify so-called “jumbo discriminatory measures”, that is, those measures likely to affect a large number of trading partners and trade. If the so-called jumbo measures in fact covered relatively little trade, then this would be surprising. Indeed it would certainly be more surprising than confirming that targeted trade defence actions and tariff increases – the apparent basis of the WTO's calculation – affect small amounts of trade.3 Surely, the sensible place to start when calculating the minimum amount of trade covered by crisis-era protectionism is to sum up the trade covered by those wide-ranging discriminatory measures for which confidence in the underlying calculations is high?

Despite adopting a remarkably conservative methodology, 22 such measures were found to each cover more than $10 billion of trade and harm 15 or more G20 trading partner. (Each jumbo measure is described in section 3 of this report.) All but three jumbo measures were implemented by G20 governments. Few of those measures are tariff increases, none are trade defence measures, and quite possibly some jumbo measures are not even trade-reducing. It is quite possible that other attempts to estimate the trade coverage of crisis-era protectionism have focused on the wrong state measures. If we have learned anything from this crisis, it is that governments have chosen to discriminate against foreign commercial interests in many different ways.

The fact that the 22 jumbo measures relate to policy instruments that are either not covered by WTO accords or are covered by the weaker WTO accords (such as the Agreement on Government Procurement) is also telling. One interpretation of this finding is that, during the recent crisis, when governments wanted to discriminate on a wide scale they chose measures that fell outside the body of WTO rules or in legal grey areas. Rather than holding the line against protectionism, the role of the WTO accords in this crisis has been to channel pressure for discrimination elsewhere, into areas of government policy not covered effectively by WTO rules.4

Our most conservative estimate of the total trade covered by the 22 jumbo measures alone is $1.618 trillion, equivalent to approximately 10.45% of the total value of world imports5 in 2008. (This trade coverage figure falls short of the arithmetic sum of the trade coverage of the 22 jumbo measures; that sum exceeds $2.3 trillion.) While these trade coverage figures are nowhere near as high as those seen in the 1930s, they are very hard to square with any claims that contemporary protectionism has affected only a tiny amount of international trade.

No doubt a less conservative method – that added in the trade coverage of the other 1000 or so state measures included in the GTA database – would have obtained a far higher percentage of world imports covered. But that was not our purpose. Our goal was to cast doubt on any claims that the total amount of trade potentially affected by crisis-era protectionism is de minimus.

In sum, the sheer volume of information that has become available about the frequency and potential trade coverage of contemporary protectionism reinforces the case for vigilance, effective monitoring, peer pressure, and public warnings about the threat to open borders. Going forward, however, even more vigilance is needed as governments enter an era of fiscal retrenchment and potential deflation.

Unwinding crisis-era discrimination should be added to the G20 trade policy agenda

The task before policymakers in the year ahead is not just to resist the temptations of future protectionism but also to find the means to unwind the substantial amount of discrimination introduced into the world economy during the crisis. That task is likely made harder by the fact that fears of another Great Depression no longer keep decision-makers awake at night. Still, G20 Leaders could easily instruct their trade ministers and the WTO to draw up a list of systemic (that is, far-reaching) measures whose removal would greatly reduce the trade coverage of crisis-era discriminatory measures. The GTA's list of jumbo discriminatory measures would be a good place to start.


1 No position is taken here as to whether fiscal retrenchment is desirable. What matters for the argument that follows is that net exports will be one of the few remaining sources of aggregate demand expansion, which in turn could tempt governments to bolster exports and limit imports in ways that trigger retaliation from abroad.
2 Although this last phase of the crisis began in the Eurozone experience teaches us that it could spread to other regions of the world economy.
3 A finding that is already well established in the literature on anti-dumping, for example.
4 This finding is a consequence of the incomplete nature of the WTO accords, incomplete in the sense that the discriminatory policy instruments covered by multilateral trade rules is a subset of all of the possible discriminatory policy instruments. So long as the set of WTO accords remains incomplete, the potential for diversion to non-covered policy instruments exist.
5 The same benchmark was used by the WTO.

 

a

A

Topics:  International trade

Tags:  protectionism, Global Trade Alert

Professor of International Trade, University of St. Gallen; Research Fellow, CEPR

Events

CEPR Policy Research