Previous Page Table of Contents Next Page

4. Negotiating modalities and simulation scenarios

During 1999-2002, the so-called "proposal harvesting phase" in the WTO negotiations on agriculture, a total of 44 negotiating proposals were submitted by individual countries or country groups. As the process entered the "modalities" phase in the second half of 2002, some countries submitted modalities.[66] In December 2002, Chairman Harbinson released an overview of the negotiating modalities,[67] and subsequently in February 2003 a draft negotiating modality. This was further revised and released in March 2003. This Harbinson text takes into account various proposals and modalities, and has been the subject of much debate. This text was expected to be modified further in Cancún, but failure to reach a consensus there left many negotiators wondering about the outcome of the next round of trade talks in Geneva.

Although the focus of this study is on the Harbinson modalities, the EU and US modalities are also analyzed in view of inter alia these being among the few modalities on the table, the weights these two proposals carry in these negotiations and also because they represent to a large extent both modest and deeper reforms, relative to the Harbinson text, and so are interesting for comparing results.

The final modalities for further commitment which were expected to be an outcome of the negotiations in Cancún did not materialize, and it was not expected that much would be on the table in the following months. However the basis of the negotiating modalities for further commitment would remain substantially the same as the revised version of March 2003. In what follows, the main elements of these modalities are introduced in brief, followed by the three scenarios constructed for simulations based on the three proposals.

4.1 The modalities[68]

Market access. The key US proposal on tariffs was the use of a harmonization formula that would reduce higher tariffs more deeply than lower tariffs. For this, the Swiss formula was proposed, with a parameter of 25. This means that all tariffs are reduced to below 25 percent. For example, an initial tariff of 100 percent would fall to 20 percent and a tariff of 10 percent to 7 percent. The other key proposal was to apply the formula to applied tariffs. The EU proposal was for the continuation of the UR approach, i.e., 36 percent average reduction with a minimum 15 percent cut for each tariff line. The reductions are to be applied to bound tariffs. The Harbinson draft proposed three different reduction rates for developed countries depending on the level of the initial tariff (and four bands for developing countries). Box 1 shows these bands and reduction rates in detail. This proposal also leads to harmonization of tariffs to some extent because higher rates are reduced proportionally more than lower rates. It also has the average-minimum feature of the EU proposal. The US proposal does not say anything about special and differential treatment for developing countries.

As regards TRQs, the Harbinson proposal called for expanding volumes up to 10 percent of current domestic consumption (6.6 percent for developing countries). Some proposals have been made for in-quota tariffs also. The US proposed elimination of the in-quota rate and a 20 percent expansion of the TRQ volumes. The EU modalities did not propose any increase in the TRQ volume.

Domestic support. The US proposal is for the reduction of total non-exempt domestic support (which includes in the US definition both the Amber Box and the production-limited Blue Box support) to at most 5 percent of the average value of agricultural production in the base period 1996-98, to be implemented over a five-year period. It further proposed maintaining the current de minimis threshold (5 percent of the value of agricultural production). The EU proposed maintaining all the three Boxes (Amber, Blue and Green). The Amber Box aggregate measure of support (AMS) to be reduced by 55 percent, while eliminating the de minimis provision for developed countries. Harbinson's proposal is for reducing the AMS by 60 percent over five years for developed countries and by 40 percent over ten years for developing countries. It also proposes reducing de minimis by 50 percent over five years for developed countries but maintaining it for developing countries.

Export subsidies. The US proposal is for complete elimination of export subsidies over a five-year period. Export taxes are to be phased out in developed countries and if export taxes are used in a developing country, it should be applied uniformly across all agricultural commodities. Furthermore, United States proposed that export credits need to be further disciplined. The EU proposed a "substantial" but unspecified cut in the volume of subsidized exports and 45 percent reduction in subsidy outlays. Harbinson's proposal is to reduce export subsidies using the following formulae:

Budgetary outlay, Bj = Bj-1 - C*Bj-1

Volumes subsidized, Qj = Qj-1 - C*Qj-1

where j is the implementation year. The value of C is 0.3 for developed countries (to be reduced over a five year period and then eliminated) and 0.25 for developing countries (reduced over 10 years and eliminated in the eleventh).

Other proposals. As said earlier, the modalities contain proposals in several other areas that are not easy to quantify and so are not discussed in this paper. These include, for example, rules on TRQ administration, state trading enterprises, non-direct forms of export subsidies, Green Box measures, "non-trade" concerns, and so on. These are obviously important provisions and could have significant impact on outcomes, but are not as easy to study within a quantitative framework.

4.2 Scenarios simulated

Box 1 summarizes the parameters selected for simulations. For obvious reason, these are those proposals in the modalities that are amenable to quantitative analysis with the ATPSM. Thus, some elements of the modalities are not included in the model simulations, and it is not clear to what extent this will bias the results. However, the scenarios cover all the most important parameters and so the bias should be small. Some of the elements not currently included could be modelled with further technical improvement of the model, but that is left for the future. It is important to note that the EU scenario includes its preferential market access package for the LDCs, called "Everything But Arms" (duty-free and quota-free imports).

Box 1 Modalities parameters used in simulations[69]



Developed countries 3 band reduction formula

tariff > 90

reduction of 60 percent with a minimum 45 percent

15 < tariff £ 90

reduction of 50 percent with a minimum 35 percent

tariff < 15

reduction of 40 percent with a minimum 25 percent

Developing countries 4 band reduction formula

tariff > 120

reduction of 40 percent with a minimum 30 percent

60 < tariff £ 120

reduction of 35 percent with a minimum 25 percent

20 < tariff £ 60

reduction of 30 percent with a minimum 20 percent

tariff < 20

reduction of 25 percent with a minimum 15 percent


No change in in-quota rates, Expand TRQ up to 10 percent of the current domestic consumption for developed countries and 6.6 percent for developing country. LDCs exempted.

Domestic support:

60 percent cut in Amber Box for developed country, 40 percent for developing country. No reduction for LDCs

Export subsidies:




Swiss formula with parameter 25 on applied tariffs


No cuts in in-quota rates, 20 percent in expansion TRQ volume

Domestic support:

Reduce to 5 percent of the value of agricultural production

Export subsidies:




15 percent minimum cut, 36 percent on average


No cuts in in-quota rates, No expansion in TRQ volume


Duty and quota free access for LDCs

Domestic support:

55 percent cut in Amber box, (two thirds for developing countries, no cut for LDCs)

Export subsidies:

45 percent reduction (two third for developing and LDCs)

The ATPSM is designed to simulate the effects of the Swiss formula, and so this is not an issue. There are, however, some areas where incorporating the scenario parameters is not as simple. For example, simulating tariff reductions in the Harbinson proposal requires first the grouping of commodities into the three and four tariff bands as per tariff rates. Even then, it is not possible to implement the "minimum-average" reduction rule, because it cannot be known in advance how individual countries will select tariff lines for the minimum reduction. This limitation also applies to the EU proposal. In case of export subsidies, the EU proposal merely says a "substantial" reduction in export volumes without specifying the exact number. In the simulation, the same 45 percent as for budgetary outlay is also assumed for quantities. Finally, additional work was needed to incorporate the duty-free and quota-free market access terms to the LDCs in the EU scenario.

[66] Unlike the proposals, which express positions and reform measures in a broad manner, the modalities are much more specific, for example in matters such as reduction rates and formulae.
[67] WTO Document TN/AG/6, 18 Dec 2002.
[68] The presentation of modalities in this section is not comprehensive, interested readers should refer to the original WTO modalities documents.
[69] The full text of the Harbinson proposal can be found at, document number TN/AG/W/1/Rev., dated 18th March 2003, for the EU proposal, and the US proposal.

Previous Page Top of Page Next Page