How much difference will the EU’s new GSP scheme make?

The WTO encourages members to charge lower tariffs on imports from poor nations to “benefit the economic development of developing and least developed countries”. The EU has always maintained such a policy – known as a Generalised System of Preferences (GSP). Recently, it has announced a revision aimed at bringing its practices more in line with global realities.

The EU reform aims to do this through a more targeted and transparent system which increases the value of the preferences for those countries most in need. It tries to do this NOT by changing the degree of product coverage, NOR by increasing the preference margins. Instead it aims to do so by removing from the GSP scheme all “high and upper middle income” economies.

The idea is that the countries remaining in the scheme should see a “preference consolidation” effect, which would mitigate the erosion of preferences that has occurred as a result of ongoing MFN liberalisation. Additionally, there is scope for some “deepening of preferences” through the relaxation of GSP+ eligibility criteria which would see some GSP countries qualify for the more generous GSP+ regime.

The main gain from the GSP derives from its possible impact on developing countries’ exports – both in existing products/markets and from diversification into new products/markets. Indeed empirical evidence suggests that such schemes can be effective in increasing trade (Collier and Venables 2007). However, the overall impact will depend on the importance of the EU in the beneficiary country’s overall trade, on the relationship between the structure of their trade, the structure of the EU’s preferences, and the incentive for firms to utilise the preferences on offer, and on the institutional and economic environment in the countries concerned.

Building on earlier work, we have undertaken a preliminary country-by-country and tariff-line level examination of the impact of the new proposals based on examining the structural features of existing trade at a very high level of disaggregation. The analysis suggests that the EU’s proposals taken on their own are unlikely to have much of an impact on these economies. We argue that in order to deliver a system that helps those most in need, more creative policies and greater integration with aid-for-trade measures are required – otherwise the proposals may in fact create a set of incentives that could exacerbate rather than reverse preference erosion.

The changes will result in a reshuffling of the current beneficiaries into two broad groups:

(1) those who lose GSP states and are reclassified into the most-favoured-nation (MFN) category;

(2) those experiencing preference consolidation, and these fall into three categories –

            (a) those that retain GSP status,

            (b) those that retain or gain GSP+ status, and

            (c) Everything-But-Arms (EBA) countries.

The countries who stand to lose most in terms of market share are those reclassified into the MFN category and who have no FTA with the EU. This is likely to apply to 30 countries. The impact on these countries depends on the importance of the EU as a trading partner in the products utilising preferences, and on the extent of their current preference margins. Table 1 summarises the possible impacts by cross-classifying the newly excluded countries according to these criteria.1 What emerges is that most countries locate near the top left corner of the table implying that these have both a low share of trade that uses EU preferences and also receive a low preference margin. Countries such as Cuba or Uruguay (bottom right hand corner) could see the largest negative impact. In aggregate there is little evidence of a substantial negative impact on countries that graduate out of the GSP on income grounds. As a corollary, the preference consolidation effects on countries retaining preference should also be small.

Table 1. Countries moving from GSP to MFN status

Source: Own calculations based on CARIS (2010). *Loss of preference margin calculated as difference between actual weighted tariff applied and hypothetical MFN weighted tariff. Affected trade is calculated as the share of country exports that used preferences in 2008 over the total exports of that country.

For the countries experiencing preference consolidation, the extent of the impact will also depend on the importance of their trade with the EU that utilises preferences, the size of the preference margin, and the extent of competitive pressures in the EU from those countries losing their GSP status. The competitive pressure in turn will depend on the similarity of exporting structures between preference retaining and excluded countries, and the size of the trade volumes.

This is summarised in Table 2. Countries in bold are those in which the competitive pressure from the newly excluded countries is high, in all the products which have an MFN tariff above 5%.2 The first panel of this table identifies countries in group (2a); overall this panel suggests that most countries will not benefit substantially. For some the competitive pressure is larger and the impact may be greater (Uzbekistan, Kyrgyzstan, and Tajikistan). But it is unlikely to be significant, as their share of total trade affected is 4%, 3.1% and 2.4% respectively and it is only on this upper bound share that any competitive pressures would be eased. Moldova could stand to gain as it satisfies the conditions of high reliance on EU preferences, high margin of preference, and ease of competitive pressures. Ukraine on the other hand satisfies the first two conditions but excluded partners do not exert important competitive pressures hence its gains from the change should be very limited.

Table 2. Countries retaining preferences

Source: Own calculations based on CARIS (2010). *Countries which may change from GSP to GSP+ status. ** preference margin calculated as difference between actual weighted tariff applied and hypothetical MFN weighted tariff. Affected trade is calculated as the share of country exports that used preferences in 2008 over the total exports of that country

With respect to (2b) a similar picture emerges. Georgia and Honduras perhaps stand to gain most, while the potential new GSP+ countries (Pakistan, Philippines) might gain from a deepening of preferences but not from the consolidation of these, as the excluded countries trade different products with the EU and hence are not in direct significant competition with these countries. Finally, for those in group (2c) we again find little evidence of a substantial impact. Most EBA countries suffer from the structural limitations of the regime and trade mostly in MFN duty free products. The important clustering of countries (in bold) on the top left corner suggests that the spread of the preference consolidating effects will be low. The greater impact will most likely be for countries such as Eritrea, Vanuatu, Lao, Senegal, Cape Verde and Malawi.

Overall it is hard to avoid the conclusion that the change in beneficiary status is unlikely to have a substantial aggregate impact either on the newly excluded countries, or on those experiencing preference consolidation. This is not to say there will be no impact, in particular there may be important effects on particular sector/product/country combinations.

Indeed, the changes in the status induce those countries losing GSP status to engage more actively in FTA talks with the EU – and indeed in the EU is keen on this. This will tend to erode the preferences.

• First, to the extent that this effect, akin to the domino effect of regionalism, prevails and results in a rise in bilateralism, preference erosion will again occur.

This could reverse and further erode the value of the preferences for remaining countries.

• Secondly, the relaxation of the GSP+ criteria may also reinforce further preference erosion.
• Thirdly, when China graduates out of GSP as it surely will in the not-too-distant future this is likely to have a big impact on graduation and GSP+ status by increasing the GSP market share for some countries above the required threshold, and, therefore, de facto reducing the number of eligible countries.

Using trade policy as a development tool raises a wide set of issues as domestic constraints and distortions within countries affect the external environment. In this light the external trading environment, such as the GSP, can at best only be a facilitator towards meeting certain development objectives. The GSP is only likely to be successful when combined with an appropriate domestic institutional environment and appropriate domestic policies. To the extent that many vulnerable economies suffer deficiencies pertaining to supply side constraints, the current proposal should be integrated much more closely to particular aid for trade measures aimed at constraints of particular importance to these countries. The current analysis supports the conclusions arrived at in our earlier CARIS (2010) report reviewing the existing GSP scheme, which identified the limited impact that the GSP scheme can provide in helping those countries most in need because of the structural limitations of the system. We therefore underline our previous conclusions that to effectively help developing countries integrate into the world economy much more creative policies need to be developed.

CARIS (2010), “Mid-term Evaluation of the EU’s Generalised System of Preferences”, report commissioned by the EC. Brighton: Centre for the Analysis of Regional Integration at Sussex, University of Sussex.

Collier, P and A Venables (2007), "Rethinking Trade Preferences: How Africa Can Diversify its Exports", The World Economy, Wiley Blackwell, 30(8):1326-1345.

1 This is based on the Revealed Export Competitive Pressure Index (RECPI).