1. Problem Definition

1.1 Introduction

The Generalised System of Preferences ("the scheme") helps developing countries, and particularly Least Developed Countries (LDCs), to reduce poverty by offering them import preferences in order to generate or boost revenue from international trade. In addition, the scheme provides incentives in the form of additional tariff preferences, to countries that commit themselves to sustainable development and good government. The scheme currently aims to achieve the objectives set out in the Communication from the Commission to the Council, the European Parliament and the Economic and Social Committee on the function of the Community’s Generalised System of Preferences (GSP) for the ten-year period from 2006 to 2015. The scheme grants preferential access to EU markets on a generalised and non-discriminatory basis to 176 eligible countries and territories. It has three arrangements:

· the general arrangement (often described simply as "GSP");

· the special incentive arrangement for sustainable development and good governance (described as "GSP+") which offers additional preferences as incentives to support vulnerable developing countries in ratifying and implementing 27 international conventions on human and labour rights, environment and good governance;

· the Everything But Arms arrangement (EBA), which provides duty-free, quota-free access to LDCs.

The present GSP scheme is implemented through successive regulations, each applying for three years. The current GSP regulation will expire on 31 December 2011. On 26 May 2010 the Commission adopted a proposal to extend the validity of the current regulation until 31 December 2013, in order to allow time to prepare the revision of the GSP scheme in view of the longer legislative procedures introduced by the Lisbon Treaty. A recently completed mid-term review provides the background for the planned Commission proposal for a revised regulation to replace the existing scheme upon expiry in 2013. Both the EBA arrangement and the rules of origin provisions fall outside the scope of this revision: the former, because it is not subject to periodic reviews; and the latter, because new legislation on rules of origin has entered into force in 2011.

1.2 Consultation and expertise

This impact assessment has been prepared following extensive consultations with Member States and other stakeholders (including civil society, industry, beneficiary countries, the European Parliament, and WTO members). The views of stakeholders were taken into account, as is highlighted in several instances in the main report. The Commission's minimum standards for consultations were met. To assess the extent to which the EU's scheme meets the needs of developing countries, a mid-term evaluation was conducted by an external consultant, the Centre for Analysis of Regional Integration at Sussex (CARIS). The final report was published on 26 May 2010 on DG Trade's website[1]. The results of this study are reflected in the main impact assessment report where relevant.

1.3 Strengths and weaknesses of the existing GSP scheme

The CARIS evaluation of the existing GSP scheme (2010) concluded that:

· there is clear evidence that the EU's GSP preferences can be effective in increasing developing countries' exports and welfare;

· utilisation rates for the GSP scheme(s) are high, and positively correlated to the extent of the tariff and preference margin;

· the exporting countries capture about half of the rents derived from the preference margins;

· the GSP+ regime has had a positive impact on the ratification of the 27 international conventions required for eligibility, but progress on effective implementation of these is much less clear.

Nevertheless, the scheme is subject to a number of structural and other constraints (detailed both in the CARIS study and in the main report). There are also a number of specific issues that need to be addressed in the review process; these are summarised in the problem tree on the following page.

Sub-optimal targeting of beneficiaries

A large degree of competitive pressure on LDCs is exerted by other GSP beneficiaries. Several high-income countries (HICs) continue to be beneficiaries, on the grounds that they are not sufficiently diversified. Such countries have the resources to attain higher levels of diversification without the help of EU preferences. Much the same is true for so-called Upper Middle Income countries (UMIs). And countries that benefit from preferences derived from another bilateral preferential arrangement with the EU also continue to benefit from the GSP scheme. The use of GSP preferences by HICs, UMIs and countries that already benefit from other bilateral preference arrangements increases the competitive pressure on exports from poorer, more vulnerable countries, whose needs are far greater and thus deserve increased attention.

Sub-optimal graduation mechanism

Emerging developing economies have generated very successful export-oriented manufacturing sectors that are highly competitive at world level. These sectors receive benefits under the scheme, although arguably they no longer require preferences to achieve a substantial presence in the EU. They exert competitive pressure on EU industry; and raise the barriers to entry for poorer countries which consequently need even greater efforts to diversify their export base. The GSP scheme has a mechanism to weed out competitive sectors from specific countries and withdraw preferences – the graduation mechanism. Yet under the current scheme, it has been barely used. Out of a total of over 2400 country-sectors, only 20 have been graduated – 13 of which are Chinese sectors. This indicates that the current graduation mechanism is insufficiently responsive to ensure the effectiveness and efficiency of the scheme. Another important weakness of the graduation mechanism is that graduation is based on sections of the EU Customs Tariff which are so large and heterogeneous, that products which are not necessarily competitive are excluded just because they fall within a category where products from a totally different, highly competitive industry predominate.

Insufficient product coverage

The GSP scheme has broad product coverage but it is not complete. Currently, 9% of all tariff lines are outside the scheme and are subject to positive tariffs. Countries most in need sometimes fail to gain access to the EU market because they would want to export some of these products. Another de facto limitation on product coverage results from the division of product lines into sensitive and non-sensitive ones: non-sensitive products enjoy duty-free access, but sensitive products obtain only a tariff reduction of 3.5 percentage points on ad valorem duties.

Insufficient support to diversification of exports

The original goal of generalised preferences schemes lent support to diversification by industrialisation. Yet the 2010 evaluation found that when all beneficiaries and products are taken together, the evidence for diversification is largely limited to products with low preference margins exported by emerging economies. The inclusion in the current scheme of GSP countries that scarcely qualify as countries most in need (HICs and UMIs) and which exert significant pressure on competing EBA and GSP+ products in the scheme – as well as the relatively weak graduation mechanism – makes diversification by poorer and vulnerable countries more difficult, because the GSP countries capture much of the preferences.

Inconsistency with overall trade objectives

The existence of GSP benefits might weaken the incentives for beneficiary countries to negotiate bilateral or multilateral trade agreements. By contrast, the objective of focusing GSP benefits on countries most in need might have the unintended consequence of providing more advanced developing countries with a greater incentive to enter into and conclude reciprocal trade negotiations with the EU.

Low utilization of preferences by some countries

The competitive pressure exerted by GSP beneficiaries can reduce GSP+ countries and LDCs to the status of residual and irregular suppliers to the EU market. Given the low value of transactions concluded under such conditions, importers have less incentive to bear the costs entailed in claiming preferences (e.g., to obtain or administer origin certificates). As a result, many preferences are simply not utilised.

Insufficient support for sustainability and good governance

The current vulnerability criteria that determine eligibility for GSP+ are excessively restrictive. This limits the extent to which GSP+ can promote sustainable development and good governance, in the sense that a less restrictive eligibility requirement can be an incentive for a greater number of countries to ratify and implement international rules and standards and engage in internal reforms. The condition for access to GSP+ (that the country has not only ratified, but also 'effectively implemented' the conventions) is unnecessarily limiting; it does not support the incentive based nature of the scheme. The existence of entry windows for GSP+ (open only once every 18 months) prevents potential beneficiaries from entering the scheme as soon as they have fulfilled all entry requirements. Under the current scheme the Commission must monitor the status of ratification and effective implementation of the 27 specified conventions by examining available information from the relevant monitoring bodies. Yet the monitoring mechanism for implementing the conventions presents a number of significant weaknesses.

Inadequate safeguard mechanism

Several weaknesses have been identified in the current GSP safeguard mechanism, notably the lack of definition of key legal concepts; the lack of definition of the rights and obligations of parties to an investigation; and the ill-defined procedural framework.

2. Analysis of Subsidiarity

The legal basis for Community action in this matter is Article 207 of the Treaty on the Functioning of the European Union (hereinafter TFEU). The principle of subsidiarity does not apply in this case. The principle of proportionality is satisfied inasmuch as the regulation is the only appropriate type of action that the European Union can take to establish unilateral, non-reciprocal, preferential market access for developing countries.

3. Objectives

3.1 General objectives

The scheme has three general objectives:

1.           Contribute to poverty eradication by expanding exports from countries most in need (G-1);

2.           Promote sustainable development and good governance (G-2);

3.           Ensure a better safeguard for the EU's financial and economic interests (G-3).

3.2 Specific and operational objectives

For the period from 2006 to 2015 the Commission GSP Communication established the following objectives for the scheme:

1.           To maintain generous tariff preferences that continue to provide real incentives for developing countries to expand their exports in a sustainable manner;

2.           To target the preference on the countries that most need it, in particular by terminating preferential access for countries that no longer need it, and by ensuring that GSP preferential rates are withdrawn from competitive products;

3.           To offer a simple, predictable and easily accessible preference scheme;

4.           ³To further encourage sustainable development and good governance;

5.           To provide withdrawal mechanisms and safeguard instruments in order to ensure that the sustainable development and good governance aspects of the GSP as well as the EU's financial and economic interests are protected.

In order to ensure that policy options considered are the most appropriate for reaching the scheme’s general objectives in a changing global economic environment, these objectives have been translated into specific and operational objectives:

The specific objectives are as follows:

1.           To better focus the preferences on the countries most in need (S-1);

2.           To remove disincentives for diversification for countries most in need (S-2);

3.           To enhance consistency with overall trade objectives (bilateral and multilateral, S-3);

4.           To strengthen support for sustainable development and good governance (S-4);

5.           To improve the efficiency of safeguard mechanisms ensuring that the EU's financial and economic interests are protected (S-5);

6.           To enhance legal certainty, stability and predictability of the scheme (S-6).

The operational objectives are as follows:

1.           To revise the beneficiary country list by deferring benefits to those countries, which based on their development, financial and trade needs, no longer need the preferences;

2.           To target graduation on the prime beneficiaries ensuring that GSP preferential rates are withdrawn from competitive products;

3.           To redefine product sections to reflect more homogeneous product categories;

4.           To simplify GSP+ entry mechanism;

5.           To develop a more effective and transparent mechanism for monitoring and evaluating the GSP+ countries' commitment and progress in the implementation of GSP+ conventions;

6.           To develop credible and efficient procedures for temporary withdrawal of the preferences and procedures for renewal of the preferences;

7.           To improve the administrative procedures of safeguard mechanisms.

4.Policy options

The following summary table presents a set of core policy options that have been identified as being representative of the main avenues that can be selected.

Option || Main features

Option A: Discon-tinuation || Preferences are abandoned for GSP and GSP+ beneficiaries. EBA arrangement would remain.

Option B: No policy change BASE-LINE || The current policy continues without change. This option has two baseline scenarios: B1 (short run) – the continuation of the scheme taking into account the current status of multilateral and bilateral agreements. B2 (long run) – the continuation of the scheme based on the assumption that all on-going but unfinished multilateral and bilateral negotiations have been concluded successfully

Option C: Partial redesign || This comprises two sub-options. They have certain common elements, and certain differences—C1's changes being less extensive than C2's. Elements common to the 2 sub-options: 1. Preferences are deferred for certain eligible countries: overseas countries and territories; high and upper middle income countries; countries with a preferential trade agreement covering substantially all preferences. 2. Graduation principles are revised: product sections are redefined; graduation does not apply to GSP+ countries. 3. GSP+ entry mechanism is simplified and made more flexible: countries must ratify, not fully implement, conventions, while providing binding commitments to guarantee their implementation; countries can apply for GSP+ at any time. 4. GSP+ monitoring mechanism is redesigned to enhance implementation of the conventions. 5. More transparent and efficient procedures for temporary withdrawal of preferences are introduced. 6. The administrative procedures of safeguard mechanisms are improved. Elements that differ between the 2 sub-options: 1. Graduation threshold Option C1 Graduation threshold remains unchanged. Option C2 Graduation threshold is reduced to 7.5% and 50% safety net is eliminated. 2. GSP+ vulnerability criteria Option C1 The import-share threshold is relaxed (increased from 1% to 2%). Option C2 Vulnerability criteria are eliminated. 3. List of GSP+ conventions Option C1 The list of GSP+ conventions remains unchanged. Option C2 The list of GSP+ conventions is expanded.

Option D: Full redesign || This option includes and builds upon the features of option C. In particular, the product coverage of the scheme is redesigned, with 3 sub-options: Option D1 All beneficiary countries receive full product coverage and all products are deemed non-sensitive No graduation takes place. Option D2 A number of industrial and agricultural products move from the sensitive to the non-sensitive list. Option D3 The list of products covered by the scheme is expanded to include a number of industrial and agricultural products.

5. Analysis of impacts

5.1 General

Imports benefiting from preferences account for less than 5% of total EU imports. This implies that, while impacts on beneficiaries may be large, the general impacts on the EU are likely to be of a limited nature. Impacts have been assessed using the analysis performed by CARIS, additional analysis via a SMART model[2], and the examination of official EU imports, production, consumption and employment statistics. The main variable used to analyse social impacts has been employment. Environmental impacts are invariably low, and have been analysed separately.

5.2 Comments on the baselines (B1 and B2)

There is a natural reduction in the level of import duties (and, therefore, preferences) due to preference erosion under the impact of further bilateral and multilateral trade deals. Preference erosion reduces imports from GSP beneficiaries; this is the reality against which this assessment takes place. In the long run, when all multilateral and bilateral agreements are fully implemented, duties are likely to be so low that the idea of preferences becomes largely irrelevant – and so would a generalised system of preferences. Other totally different tools may have to be designed. Until then, the question is what can be done for the countries most in need of preferences.

5.3.Option A: discontinuation

Option A ends the GSP scheme while retaining the EBA scheme which benefits LDCs. Total EU imports fall, but by an insignificant amount (about €6 billion, or less than 1%).

General assessment of economic, social and environmental effects

Relative to B1, the general effects are as follows. The economic and social effects for countries most in need are expected to be negative. LDCs would benefit, but the many other developing countries and economic sectors which are also most in need would suffer as preferential access disappears. Within the EU, three elements will impact upon general economic and social effects: producer surplus, consumer surplus, and tariff income. Negative impacts for consumers are likely to be compensated by higher tariff revenues, which are of the same order of magnitude. The net impact would thus be generated by benefits to producers. As explained above, these benefits would on the whole not be significant, but would nonetheless have significant positive effects on important sectors (sugar, fruits and vegetables, textiles and apparel) – and on the EU Member States where those sectors are important. Therefore, these impacts would be positive as a whole. Environmental impacts in the EU would be (at best) marginally positive, given that the drop of imports would be marginal. As for the countries most in need, it is possible those countries losing GSP+ would deviate from environmentally sustainable practices. Thus, on the whole, a marginally negative impact may be felt.

A vs. B1. Effects on: || economic || social || environmental

Countries most in need || -- || -- || 0/-

EU || + || + || 0/+

Relative to baseline B2, the changes would be expected to go in the same direction, but they would be significantly smaller – to the point of not being noticeable.

5.4 Option C: partial redesign

Option C has many building blocks. To explore the different angles of these building blocks, two sub-options were explored. The main differences between them concern the graduation of competitive sectors and the vulnerability criteria under GSP+. With regard to graduation, the sectors which will actually be graduated are not known at this stage – they will depend on the calculations of imports on the basis of the latest available figures prior entry into force of the new regulation. Current figures have been used as a proxy. With regard to vulnerability, C1 relaxes the 'economic' criterion from 1% to 2%. The actual list of countries which will meet the relaxed criterion is not known at this stage – again, these calculations will be made on the basis of the latest available figures prior to entry into force of the new regulation. The additional countries that would currently meet the test (Pakistan, Philippines and Ukraine) have been considered as a proxy.

C2 eliminates the vulnerability criteria while it adds additional requirements regarding conventions. Again, the actual list of countries meeting the relevant conventions criterion will be determined as close as possible to the time of entry into force of the new regulation. Currently, we would expect these to be the three C1 countries plus Namibia and Nigeria (all of them having already ratified the relevant conventions), and these have been used as a proxy for the purpose of this exercise. The assessment starts with an analysis of option C1 and then describes the main differences that arise under option C2.

5.4.1 Option C1

General assessment of economic, social and environmental effects

When compared with baseline B1, the general effects of option C1 are as follows. Total EU imports fall by about €4 billion (a €1 billion increase in imports from countries that have never belonged to the scheme, offset by a €5 billion fall in imports from countries that cease to be members of the scheme. The economic and social effects on countries most in need are expected to be positive, as exports increase and welfare gains accrue.

As under option A, negative impacts on EU consumers are likely to be compensated by higher tariff revenues, which are of the same order of magnitude. The net impact would thus be generated by impacts on producers. As explained above, these benefits would on the whole not be significant, but would have significant negative effects on important sectors (rice, arable crops, oils and fats, sugar, fruit and vegetables, textiles and apparel, and leather) – and on the EU Member States where those sectors are important. Therefore, these impacts would be negative as a whole. Environmental impacts in the EU would be (at best) marginally positive, given that the drop of imports would be marginal. As to the countries most in need, the impact of expanded GSP+ membership would lead overall to a marginally positive impact.

C1 vs. B1. Effects on: || economic || social || environmental

Countries most in need || ++ || ++ || 0/+

EU || - || - || 0/+

When comparing the effects of options C1 with baseline B2, the changes would be expected to go in the same direction, but they would become significantly smaller – once again, to the point of not being noticeable.

5.4.2 Option C2

There is one important difference between C2 and C1. Lower graduation thresholds increase the level of graduation significantly for certain countries and sectors, particularly for India. This leads to a number of effects. The first one is a larger decrease in exports by GSP participants as a whole. The second is an increase in EBA exports, as the negative impacts on Bangladesh (foreseen under C1) diminish. The result is that, while the positive effects on EBA and GSP+ beneficiaries can not be underestimated, GSP exports by many beneficiaries will suffer. Dynamic effects are expected to more than compensate for this static loss, so the impact as a whole is considered to be positive, but certainly less so than C1. Given that the remaining impacts are largely similar to C1, the general assessment table for C2 would therefore read as follows:

C2 vs. B1. Effects on: || economic || social || environmental

Countries most in need || + || + || 0/+

EU || - || - || 0/+

5.5 Option D: full redesign

Most of the building blocks of the scheme have been re-defined under option C. However, a number of respondents to the consultation suggested a broad expansion of the two other key building blocks of the scheme: the range of products covered and the preference margins. We have therefore also examined a comprehensive re-design that includes the proposed changes under option C, and in addition changes to these two building blocks. In order to simplify the analysis, D sub-options are calculated as increments to C2 only. Three sub-options are assessed. D1 is a far reaching option. It provides full product coverage expansion and elimination of all sensitive products (eg, extending the duty-free, quota-free treatment of EBA countries) to all countries most in need (whether GSP or GSP+). This implies that the remaining beneficiaries are no longer subject to graduation. D2 and D3 are less far-reaching. They take all the parameters of C2 (graduation included) and add partial de-sensitisation (D2) and partial expansion in product coverage (D3).

5.5.1 Option D1: full product coverage, full de-sensitisation

General assessment of economic, social and environmental effects

When compared to baseline B1, the general effects of option D1 are as follows. Although the economic and social effects on countries most in need are expected to be positive as a whole, these gains accrue mainly to sectors which are already competitive, at the expense of those which are less advanced. There would be large distributional effects, with the additional share of EU imports captured by China, India and other countries that were previously subject to graduation having a negative impact on many other countries most in need. EBA beneficiaries in particular would suffer (Bangladesh being the prime example), as would GSP+ countries such as Pakistan. The overall positive assessment ("+") must therefore be nuanced. The positive impact for EU consumers is likely to be compensated by lower tariff revenues, which are of the same order of magnitude. The net impact would thus be generated by impacts on producers. As explained above, these impacts on the whole would not be significant, but would have significant negative effects for important sectors – and for the EU Member States where they are important. As a result, these impacts would be negative overall. Although larger than under option C, they would probably be of a similar order of magnitude. Environmental impacts in the EU would be marginally negative, given the overall increase in imports. The large increase in imports particularly from China or India can lead to negative impacts in those countries also. The impact on GSP+ countries would be positive overall because, although their exports would increase, the environmental protection framework under which (all) businesses operate would be improved by adherence to the relevant environmental conventions. The overall balance of these effects would be marginally negative.

D1 vs. B1. Effects on: || economic || social || environmental

Countries most in need* || + || + || 0/-

EU || - || - || 0/-

*Positive economic and social impacts for countries most in need as a whole hide significant negative impacts on EBA and GSP+ beneficiaries.

When comparing the effects of option D1 to baseline B2, the changes would be expected to go in the same direction, only that they would become smaller – but still noticeable.

5.5.2 Options D2 and D3

These options build on option C. In order to simplify the analysis, only one of the options, in this case C2, has been used as a basis for D2 and D3. There is no reason to believe that significant differences would exist if C1 were taken as a basis instead. Given that D2 and D3 change only one building block at a time with respect to C, only the salient novelties are mentioned here.

D2 and D3 generate preference erosion to the disadvantage of LDCs

D2 explores de-sensitisation. As expected, the immediate impact is preference erosion for EBAs, particularly vis-à-vis GSP competitors, which are the net winners. Given that the CARIS evaluation made clear that significant competitive pressure is exerted by GSP countries on their EBA counterparts, this was to be expected. India, Indonesia, Vietnam and Thailand absorb almost all the gains, while EBA countries capture almost none. D3 explores product expansion. A similar impact as that described for D2 is foreseen: benefits to GSP beneficiaries at the cost of preference erosion and export losses for EBA countries. Both D2 and D3 thus confirm that product expansion and desensitisation have a price, a price which is paid by the poorest and which compounds the preference erosion they suffer.

D2 and D3 may place obstacles to negotiation of bilateral and multilateral agreements

Compared to option C, these options would arguably send a false signal to our trade partners, by creating the expectation that concessions regarding products which have been introduced into the GSP scheme or which have been desensitised can be obtained from the EU automatically in bilateral or multilateral negotiations. The size of the changes introduced by D2 and D3 are not sufficiently large as to vary the order of magnitude of the rest of the results of option C. However, EU producers of new products that are introduced into the GSP scheme, and of those goods which receive higher preference margins via desensitisation, would suffer additional pressure.

6. Comparing the options

6.1 Review of different options by objectives and impacts

The following table compares how well the different options examined above meet the objectives sought by a review of the scheme. This comparison is based on three criteria: effectiveness (number of objectives met, to what degree); efficiency (use of resources necessary to meet the objectives, unintended spill-over effects); and consistency with overarching EU objectives.

Options || A || C1 || C2 || D1 || D2 || D3

Effectiveness || - || ++++ || +++ || -- || ++ || ++

Efficiency || -- || +++ || ++ || -- || + || +

Consistency || ++++ || ++ || ++ || --- || + || +

A detailed analysis follows, based on the relative efficiency and effectiveness of each of the options in achieving the general policy objectives.

6.2 Effectiveness of the policy options in achieving the general and specific objectives

Option A

Option A meets general objective G-1 (contributing to poverty eradication by expanding exports of countries most in need) only partly. By focusing preferences on LDCs, it leaves many other countries with similar trade, development and financial needs without preferences (specific objective S-1) – with adverse economic and social impacts as a consequence. Moreover, removing preferences from some countries that are most in need will expose export sectors in those countries to competition from developed countries. Option A runs directly contrary to specific objective S-4 and general objective G-2 (promoting sustainable development and good governance). And it does nothing to ensure better safeguards for the EU's financial and economic interests (general objective G-3 and specific objective S-5). By contrast, option A can strengthen the EU's hand in bilateral and multilateral negotiations (specific objective S-3). It would have positive economic and social impacts for certain sectors in certain Member States at a time when so much emphasis is on boosting competitiveness, growth and job creation. And last, at a time of extreme pressure on public finances, it boosts tariff revenues.

Option C1

Option C1 contributes well to general objective G-1 (contributing to poverty eradication by expanding exports of countries most in need). In particular, it ensures that preferences are properly targeted on countries most in need (S-1) and it reduces the disincentive to diversification (S-2) that arises from the competitive pressure exerted by more developed beneficiaries of the current system. Option C1's combination of a more flexible entry mechanism for GSP+, more flexible trade criteria for eligibility, and no graduation, would boost the scheme's contribution to promoting sustainable development and good governance (G-2 and S-4). It improves the efficiency of the safeguard instrument (S-5) and the withdrawal mechanism, which both would contribute to G-3. It also boosts tariff revenue. It would have the unintended effects of strengthening the EU's hand in bilateral and multilateral trade negotiations (but in respect of fewer countries than under option A). There would, however, be negative economic and social impacts for certain sectors in certain Member States.

Option C2

The main difference between C2 and C1 is that the lower graduation thresholds of C2 reduce total exports by countries most in need. In addition, the more GSP+ beneficiary countries enter the scheme, the greater the competitive pressure on LDCs – the neediest of all developing countries. These effects make C2 a less effective way to achieve general objective G-1 (contribute to poverty eradication). However, it ranks superior to C1 in terms of its likely effectiveness for achieving G-2 (promoting sustainable development) because it provides for a revision of the required conventions.

Option D (only D1 is discussed in the executive summary)

Option D1 targets those most in need by deferring preferences for sufficiently rich beneficiaries and for those that enjoy preferential access via a bilateral agreement. Yet it eliminates graduation entirely, and extends EBA-equivalent treatment to all beneficiaries (which would accelerate preference erosion for the poorest). Overall therefore, it cannot be said to be meeting objective G-1. Equally the G-2 objective (promoting sustainable development via the incentives of GSP+) is totally undermined by granting EBA- equivalent treatment to all beneficiaries. Option D1 improves the efficiency of the safeguard mechanism (S-5) and of the withdrawal mechanism (S-6), and thus contributes positively to safeguarding the EU's financial and economic interests (general objective G-3). However, D1 is expected to decrease tariff revenues at a time of extreme pressure on public finances. Also, it gives rise to negative economic and social impacts in certain industrial sectors and Member States. And it would significantly weaken the EU's negotiating position in the bilateral and multilateral context (specific objective S-3).

6.3 Preferred option

The option which meets the objectives of the scheme in the most effective, efficient and coherent manner is C, and in particular C1. This does not negate that C2 contain positive aspects (review of the list of conventions) which can also be considered.

7. Monitoring and Evaluation

The table below includes suggestions for indicators that can be used to assess the progress and effectiveness of the preferred option in achieving general policy objectives.

General Objectives || Indicators || Sources of information

To contribute to poverty eradication by expanding exports from countries most in need || - expansion of developing countries exports to EU - increased share of imports from countries most in need - increased utilization of the preferences - effective graduation of the competitive sectors -increasing diversification || - Eurostat data

To promote sustainable development and good governance || - increased number of countries committing to sustainable development and good governance principles within GSP+ arrangement - overall improvement of implementation of GSP+ conventions by GSP+ beneficiaries -number of withdrawals || - reports of relevant international monitoring bodies -DG TRADE

To ensure a better safeguard for the EU's financial and economic interests. || -number of safeguard requests -number of safeguard measures -revenue foregone due to the scheme -number of preferential trade agreements signed with beneficiaries - number of preferential trade agreements signed with non-beneficiaries || -Safeguard requests -Eurostat data -DG TRADE

The effectiveness of the GSP scheme should be subject to formal and independent evaluation prior to any subsequent revision. To be effective, such evaluation is likely to require a minimum of 3 years' post-implementation data, which implies that the evaluation cannot take place before the end of 2017 at the earliest.

[1]               http://trade.ec.europa.eu/doclib/docs/2010/may/tradoc_146196.pdf

[2]               A model developed by the World Bank in collaboration with various international organisations.