Has the EU’s generalised system of preferences been effective? Lessons for the future
The EU's Generalised System of Preferences (GSP) is a central component of  the EU's strategy toward developing countries. This strategy is aimed at the promotion of sustainable development where trade is seen as an essential vehicle in facilitating the achievement of both economic and social objectives. In a recent study, the Centre for the Analysis of Regional Integration (CARIS) undertook a mid-term review of the EU's GSP , to assess the regime's effectiveness in meeting developing countries' needs, and provide informed recommendations for future improvements.
The EU's GSP Regimes
The EU‘s GSP covers three separate preference regimes: the GSP, the GSP+ and the EBA (Everything But Arms). These regimes offer differentiated degrees of preferences to developing and Least Developed Countries (LDCs). The standard regime of GSP applies to all 176 developing countries and offers duty free access into the EU market in an array of non-sensitive products  and preference reductions of 3.5 percentage points on a range of more sensitive products. The special incentive arrangement for sustainable development and good governance, known as GSP+ offers the same treatment as the GSP in non-sensitive products but extends duty free access to the more sensitive products where the standard GSP offers the 3.5 percentage point preference margin. Countries eligible for such treatment (currently 15) must meet certain economic criteria and have ratified and effectively implemented international conventions on human rights, labour and environmental standards. The EBA is the most generous of the three regimes and provides duty-free, quota-free access for LDCs on 99% of all tariff lines. Table 1 provides the details of the tariff coverage of the GSP regimes in 2008.
The wide-ranging CARIS (2010) mid-term review assessed the effectiveness of the GSP schemes and discussed the limitations in attaining development objectives via unilateral preferential schemes. It showed evidence of some positive impact in five ways:
· On the basis of detailed gravity modelling, there is evidence of GSP preferences increasing developing countries’ trade and investment, although interestingly the EU’s FTAs appear to have a larger impact.
· Formal Computable General Equilibrium modeling, suggests a modest welfare gain for a number of GSP beneficiaries. For EBA eligible countries the greatest impact was for Cambodia and Bangladesh, while the impact was smaller across the Sub-Saharan Africa region.
· GSP utilisation rates are typically high. These rates appear positively related to the extent of the tariff and preference margin, i.e. the higher the preference margin, the higher the rate of utilization, as well as to the wealth of the beneficiary, and negatively related to the cost of business and the stringency of the rules of origin.
· The rents generated by preference margins are typically transmitted to developing country exporters, suggesting a degree of effectiveness in the exploitation of preferences.
· The current vulnerability criteria of GSP+ are broadly consistent with the selection of smaller, landlocked countries with limited export diversification and prone to terms of trade shocks. The GSP+ scheme also appears effective in promoting ratification of the 27 required UN conventions.
However, there are a number of important caveats and structural features which inevitably constrain the effectiveness of the GSP regimes. Even with high utilization rates, the structure of the EU’s preference regimes combined with low average MFN tariffs limits the scope for offering significant preferential access to developing countries to a few sectors. The assessment of the importance of preferences by country groupings also indicates that, on average, a high proportion of GSP country trade already entered into the EU duty free through the MFN regime. In 2008, on average just over 7 percent of GSP countries' exports used GSP preferences when exporting to the EU. For the GSP+ and the EBA countries this rate was just over 24.5 percent and 23.4 percent respectively. On average, the preference regimes do not account therefore for a significant amount of the relevant countries’ trade with the EU. If we consider their share of total trade, versus their trade with the EU only, the impact and outreach of the system of preferences shrink. Given low MFN tariffs, relatively few tariff peaks and the composition of LDC exports, the bilateral preference regimes have reduced scope to help developing countries. The preference margin in most sectors tends to be low, therefore the scope for offering preferential access via tariff reductions is limited, and by extension the study showed that full(er) utilisation of preferences would not generate significantly greater welfare gains.
There is also mixed evidence about the impact of preference margins and utilisation rates on sustainable development and preferences may not be well targeted to those countries that are most in need. There is no evidence that the GSP schemes have diversified beneficiaries’ exports or created new export products. Also, while there is some evidence that the GSP+ scheme may have a positive impact on convention ratification, the evidence on implementation is much weaker. More broadly, it is still too soon to conclude whether the GSP+ will be an effective tool to promote sustainable development and good governance. It is also important to note that these schemes are inherently discriminatory in nature, implying that granting more generous preferences to a particular country will invariably come at the cost of eroding those of another. Besides, the increase in bilateral agreements via the Global Europe initiative further exacerbates this preference erosion hence limiting the possible benefits that can be extended to developing countries.
Conclusions and policy recommendations
Given the preceding caveats, the question of the future shape of the GSP scheme becomes increasingly important. One of the overarching conclusions of the study suggests that whilst there is some room for improving the overall effectiveness of the system through greater preference utilisation, there is also a natural/structural ceiling to the benefits that these schemes can bring to developing countries. This, in turn, suggests that the future of such a system should lie in more creative forms of policies beyond the existing approach to preferences.
Nevertheless, one avenue worth pursuing involves accepting the structural limitations of the system while targeting its effectiveness. This would require improved product coverage and a greater focus on issues relating to the causes of non-utilisation of preferences. These could be tackled through more development friendly rules of origin (i.e. with regard to cumulation agreements and value added thresholds) or by ensuring a more stable and transparent regime. Binding unilateral preferences at the WTO could also introduce important elements of predictability and stability.
By acknowledging the system’s structural limitations and pursuing more creative policies, these regimes could become more far-reaching. Several avenues are open here. Pursuing and expanding Aid for Trade initiatives that emphasize the importance of developing capacity to trade in addition to access to export markets is one option. One could additionally recognize that market access is not just about tariffs and emphasize the ongoing need to deal with non-tariff measures such as SPS, TBT and rules of origin. Given that GSP as constituted is (a) inevitably discriminatory, and (b) constrained by the low level of the EU’s MFN tariffs, these limitations could also be overcome by the EU offering “negative” tariffs (effectively import subsidies). This would remove the zero lower bound of the preference margins offered. If such subsidies were to be applied in similar magnitudes irrespective of the product, this would also remove the distortionary effects of the current system that provides varying degrees of preferences across products. These distortions can cause countries to specialise in products where comparative advantages are solely attained via the preferential margin rather than the underlying advantages of the exporting economy. The somewhat radical nature of this proposal serves to highlight the limitations of the existing system.
In subsequent work undertaken by members of the CARIS team (Iteas Consulting ltd), another interesting policy option has been suggested: promoting “mirrored preferences”. This entails linking participation of certain categories of developing countries in the EU’s GSP, as well as possibly the rules on graduation, contingent on the adoption of a similar system of preferences towards other least developed countries. This recommendation aims at trying to lock in the benefits from domestic liberalization whilst ensuring greater access of developing country exports in dynamic third markets and reducing their dependence on preferences from a limited number of countries. This provides a system where preferences would be propagated across developing countries, thus fostering competition and allowing for export destinations other than the EU particularly those where demand may be growing faster.
Michael Gasiorek, Senior Lecturer, University of Sussex, Centre for the Analysis of Regional Integration (CARIS).
Javier Lopez Gonzalez and Maximiliano Mendez Parra, Researchers of the University of Sussex and Directors of ITEAS Consulting ltd.
 The share of MFN duty free imports from EBA, GSP and GSP+ countries is 62.9%, 64.5% and 61.3% respectively.
 Mid-term Evaluation of the EU's Generalised System of Preferences, CARIS, available at : http://trade.ec.europa.eu/doclib/docs/2010/may/tradoc_146196.pdf
 Sensitive products are goods that are considered to be of special importance to the EU that require some protection for one reason or another. Non-sensitive products do not have this stipulation.
 In 2006 the European Commission presented The Global Europe strategy, which seeks to reorient European bilateral trade through a new generation of Free Trade Agreements and increases European focus on areas such as intellectual property and access to raw materials.
 Securing tariff preferences for developing countries by Lorand Bartels and Christian Häberli, TNI, Volume 10-4