Identifying "Special Products" - Developing country flexibility in the Doha Round^[1]

J.R. Deep Ford, Suffyan Koroma, Yukitsugu Yanoma and Hansdeep Khaira

Developing countries recognize the potential benefits of more liberalized agricultural trade and are committed to this objective in the context of the WTO. However, many of them still seek increased flexibility in applying the agreed rules on some products in the name of advancing their longer run economic and social development. WTO members accepted the importance of different treatment for some products and agreed that based on three criteria of food security, livelihood security and rural development needs Developing Country Members will have the flexibility to designate an appropriate number of products as "Special Products". The paper seeks to advance the understanding of "Special Products" and more specifically develop and demonstrate an approach and methodology for identifying "Special Products". The paper examines indicators that can be used to measure each of the three criteria and then assesses the importance or contribution of agricultural commodities in terms of selected indicators. Factor analysis is used to evaluate the interrelationship and robustness of the indicators. The results show that several indicators are explaining the same criteria and that it is possible select a few indicators to identify special products. They show further that products need only meet one criterion to qualify for "Special Product" status.

1. INTRODUCTION

The November 2001 Doha Ministerial Declaration committed to development as an integral part of its mandate and further that this would be achieved through accommodation of sufficient flexibility for developing countries in the eventual Doha Round final agreement. These commitments are reiterated in the Doha Work Programme July 2004 document.^[2]

One central dimension to promoting development and providing flexibility in the Round is the concept, identification and treatment of "Special Products". "Special Products" is an option only for developing countries and the flexibility provided is mainly through exempting certain products or reducing the level at which its tariff is cut.

The primary purpose of this paper is to advance the understanding of "Special Products" and more specifically develop and demonstrate an approach and methodology for identifying "Special Products". It is not to identify or suggest "Special Products" for any particular country. Only the country itself in the context of the comprehensive final Doha agreement can do that.

The concept of "Special Products" is not new in multilateral negotiations. The identification of particular products, staple products and strategic products, linked to the economic growth of developing countries has been a part of multilateral trade negotiations at least since the 1950s. Similarly, the attention in the current negotiations (July Framework document) to tropical products, preferential products and sensitive products, are also all related to achieving development goals and increasing flexibility for developing countries.^[3]

Despite the commitment in the July Framework by WTO members to "Special Products" it is important at the outset to recognize that flexibility and "Special Products" remain controversial topics and members have different views about their purpose in the negotiations. On one hand, a large group of developing country members with defensive interests in mind see the purpose of "Special Products" as providing the policy space and flexibility mainly to promote development goals. On the other hand, some developing and developed countries with offensive interests see the purpose of "Special Products" as mainly a means of providing limited flexibility for some products to achieve greater ambition on all products. Thus, the challenge from a strict multilateral negotiations standpoint, is to identify products that qualify as "Special Products" for flexible treatment to advance development concerns without undermining the level of commitment considered necessary for improved market access.

Given that it is hardly likely that there will be agreement for unlimited self-selection of "Special Products", it is essential that on the basis of the criteria agreed in the July Framework a comprehensive approach to identifying "Special Products" be developed.

The rest of this paper is separated into five sections as follows. The second section of the paper provides reasons why developing countries argue for "Special Products" in the negotiations.

The third section seeks to advance the understanding of the concept of "Special Products", providing a brief background and history of the concept of "Special Products" and linking it to the related concepts of Sensitive Products and a Special Safeguard Mechanism.

Section four provides a conceptual approach for identifying "Special Products" and describes the indicator and factor analysis used to evaluate the criteria set out for identifying "Special Products".

Section five presents results from four case studies and section six draws lessons from the analysis with regard to the identification of "Special Products".

2. WHY "SPECIAL PRODUCTS"?

While developing countries recognize the importance and potential benefits of more liberalized agricultural trade and support the achievement of this in the current negotiations, the majority of them still seek accommodation through increased flexibility in applying the agreed rules in the name of advancing their development, specifically their food security, livelihood security and rural development.

Three fundamental reasons are offered here as the basis for "Special Products". First and foremost, that developing countries view this option as essential to the success of their rural area development strategies, including the stability and sustainability of their national food and livelihood security. Secondly, that based on the Uruguay Round experience they are skeptical about the gains from liberalization and their own individual capacities to reap the benefits from increased liberalization. Thirdly, that the gains that need to be made in the current negotiations for effective liberalization will most likely not be achieved given the national interests of WTO member countries.

Rural area development strategies

Domestic markets in developing countries are critical initial outlets for poor, small agricultural producers. These producers are not operating in an environment that enables them to currently compete on export markets or against imported products given that the public investment in communication, education, rural roads, and technology development that has generally been afforded the competing products has not yet been available to them. The dynamic between returns to rural area public investment and agricultural commodity markets needs careful analysis. While it is recognized that rural area public investment with liberalization has the potential to increase returns more than without liberalization, it is also believed that this can be better achieved with a phased rural development strategy that allows time to increase agricultural sector production capacity and competitiveness.

WTO Uruguay Round experience

The existing differences between countries in terms of goals, endowments, and capacities have led to winners and losers as a result of Uruguay Round liberalization. Over the past decade numerous countries have seen their export market opportunities decrease (through WTO dispute panel rulings, unilateral reform, bilateral agreements) and their food imports increase (through the removal of restrictions and their lower applied tariff levels). Their agricultural trade surpluses have been shrinking and there has been a rapid rise in imports of cereals and livestock products. The current model results based on increased liberalization again reflect aggregate gains in welfare but these are clearly skewed towards developed countries and developing countries that have the greatest domestic supply response capacity. Therefore, countries have sought accommodation through Special Products to mitigate some of the negative impacts of liberalization on their economies, especially as this might be related to import surges that could undermine some livelihood systems.

Low levels of liberalization on key products

The levels of liberalization that might lead to the gains suggested have not been achieved and are unlikely to be achieved given the goals of both developed and developing countries. Critical products for both export and domestic consumption of developing countries are products subject to the most distortions in the international trading environment, among them rice, sugar, milk, and maize. It is not anticipated that the high levels of subsidies and tariff protection provided to these and other products, especially in developed countries, will decline considerably. Very simply, almost all of the countries have national goals that require the maintenance of some level of agricultural production for food security and rural area activity for what has been long referred to as non-trade concerns. Thus, where full liberalization would undermine these national goals they will not be agreed to. The ability to identify and designate "Special Products" is an accommodation that would allow developing countries to pursue their own development objectives, especially as they might relate to key products for food security.

In conclusion, in an effort to develop their supply response capacity, identify the markets and commodities in which they will participate, and share in the benefits from trade, countries seek accommodation through "Special Products". Thus, identification of "Special Products" for lighter treatment under the market access pillar should not be seen only as introducing protection but rather as creating an enabling environment for longer run development. It is important to underscore that infant industry or import substitution considerations which were negative processes during a 1970’s political climate that turned away from liberalization can be positive processes in an era that embraces increased liberalization.

3. "SPECIAL PRODUCTS" AND THE RELATIONSHIP TO SENSITIVE PRODUCTS AND A SPECIAL SAFEGUARD MECHANISM

This section provides a brief background to the concept of a "Special Product" and makes the link to two other very related concepts in the current Doha Round, Sensitive Products and a Special Safeguard Mechanism. This paper seeks to contribute only in the area of "Special Products" identification and treatment and to serve this purpose accepts the proposal by Hoda (2005) that "it would appear to be a good strategy for developing countries to keep ‘Sensitive Products’ and "Special Products" quite apart during the negotiations".^[4]

"Special Products"

From the earliest days of the GATT there has been concern about how international trade was affecting "primary products" in view of their importance for the trade and development of a number of exporting developing countries. Some of these "primary products" began to be referred to around 1957 in GATT documents as "tropical products" (WTO, 2005). Over time as trade and development goals became more balanced in terms of the focus on imports as opposed to exports, products labelled "sensitive" and "special" have gained more prominence in the discussions.

When agriculture became a formal part of multilateral trade negotiations under the Uruguay Round (UR), what is today "Special Products" under the Doha Round was clearly foreshadowed. That is, through "special and differential treatment" (SDT) under the UR and other exemption measures, the link between particular products, flexibility in applying trade policy and food security is established. For instance, in the context of market access, Annex 5 (URAoA) allows special treatment to accommodate food security and environment concerns. This clause, also known as the ‘rice clause’ allows for the postponement of tariffication and was included on the insistence of Japan and Korea for some level of protection of their staple product. The very close overlap/relationship between sensitive and special products can immediately be identified as some of the most frequently selected products as "sensitive", by both developed and developing countries (sugar, rice, dairy products, meat) are undoubtedly linked to the criteria for "Special Products".

Further, under the domestic support exemptions in Annex 2 (URAoA), there are references implying different treatment for some products linked to benefits to the rural community, products which form an integral part of a food security programme, criteria related to nutritional objectives, issues related to food aid, and income protection or income safety net programmes (WTO, 1994).

As developing countries considered the experience and results of the Uruguay Round and perhaps especially in anticipation of the continuation of the reform process as mandated by Article 20, countries made submissions that called for increasing flexibility in their agricultural policies and specifically as these measures were needed to address food security, livelihood security and rural development. One submission (WTO, 2000)) stands out in this regard and called for policy instrument flexibility that would, among other goals^[5]:

(i) protect and enhance developing countries’ domestic food production capacity particularly in key staples
(ii) increase food security and food accessibility for especially the poorest
(iv) protect farmers which are already producing an adequate supply of key agricultural products from the onslaught of cheap imports

The same submission called for the ability of countries to declare products that they felt served these goals and thereby products that would not be subject to commitments. The link between "Special Products" and a Special Safeguard Mechanism is made in a December 2002 Overview paper (WTO, 2002) submitted by the Chairman of the Committee on Agriculture. Under a section entitled "Special Safeguard Mechanisms" questions are raised about a new safeguard mechanism for all agricultural products or for a limited number of products such as strategic/food security/or livelihood products.

In the March 2003 "First Draft of Modalities for the Further Commitments" (WTO, 2003), the concept of "Special Products" is explicitly introduced in considerable detail, indicating that agricultural products would be declared at either the six or four digit level and identified as "SP" in their schedules. In the July Framework agreement of 2004 WTO members agreed on Special Products in paragraph 41:

"Developing Country Members will have the flexibility to designate an appropriate number of products as Special Products, based on criteria of food security, livelihood security and rural development needs. These products will be eligible for more flexible treatment. The criteria and treatment of these products will be further specified during the negotiation phase and will recognize the fundamental importance of Special Products to developing countries."

This paragraph in the July Framework of 2004 document is the point of departure for the identification of "Special Products" in this paper.

"Sensitive Products"

The declaration of a product as Sensitive is not required to be linked to any of the three development related criteria to which "Special Products" by agreement must be linked. However, when one considers lists of Sensitive Products declared by both developed and developing countries in the past, it is clear that criteria related to rural development, food security, and livelihood security influence the choice of Sensitive Products. Yet, Sensitive Products in the current framework are simply considered as products for which an appropriate number of tariff lines can be negotiated for more favourable treatment, meaning lower tariff reductions associated with tariff rate quota commitments. There is no stated link to any political or economic objective and as a result it is difficult to speculate what tariff lines might be chosen as sensitive. However, experience suggests that the lines likely to be chosen are those that are potentially the most negatively affected by imports, that have the highest degree of market regulation (have high tariffs, designated for Special Safeguard, have a tariff rate quota), that would likely suffer the greatest reductions in tariff revenues, and those where proportional reductions in domestic prices would be very large (Jean et al., 2005).

"Sensitive Products" are open to both developed and developing countries and the July Framework agreement (WTO, 2004) states that there will also be SDT in relation to sensitive products. Thus, developing countries not only have access to "Special Products" but they are also allowed to negotiate more favorable conditions (than developed countries) for the products they identify as Sensitive Products.

Yet, the overlapping dimensions have to be recognized between the two concepts of "Sensitive Products" and "Special Products". This is very much so given that "Sensitive Products" are generally considered to relate to non-trade concerns, and certainly the criteria for "Special Products" can be similarly classified. In the case studies analysed in this document there is a clear overlap between products formerly classified as sensitive that could under the Doha Round be classified as a "Special Product". However, one distinction between developing and developed countries from a sensitive product perspective was the increasing frequency with which tobacco and alcohol showed up as a sensitive product in developing as opposed to developed countries. In the case of Belize analysed below the only agriculture HS Chapter that has applied tariff lines greater than 45 percent is HS22 for which many of the tariff lines for alcohol and spirits is 91 percent. Under certain conditions tariff revenue could be considered one of the objectives distinguishing sensitive and "Special Products".

Special Safeguard Mechanism

The establishment of a Special Safeguard Mechanism (SSM) has been agreed in the August 2004 Framework Agreement of the WTO. In one sense it simply extends the option of an agricultural Special Safeguard (SSG) that was limited under the Uruguay Round to 39 WTO members to a larger number of countries. From the point of view of this paper the concept is linked directly to "Special Products" because one of the main interests by countries in naming "Special Products" is to provide protection for these products from disruption and displacement by imported products, thereby undermining food security, livelihood systems and rural development. The SSM would provide this protection by preventing surges of imports into the domestic market, often reducing prices and displacing domestic production. When products are identified as "Special Products" one of the main expected treatments is that their tariff levels will not be lowered at the same rate as other products. However, any lowering of the tariff might expose the producers to import shocks and as such interest in a further line of protection that could come from an SSM is not surprising. The two shocks that an SSM would be designed to protect against is a volume shock (import surges of the product entering the country) and a price shock (import prices depressing the domestic market price). The SSM is a formula that is characterized by a volume and price trigger which signals the need for protective action for the domestically produced product in the domestic market. In the negotiations those countries advocating identification of "Special Products" have suggested that all "Special Products" should be allowed use of an SSM.^[6]

4. "SPECIAL PRODUCTS" - IDENTIFICATION

This section presents a conceptual approach and an analytical framework for identifying "Special Products".

4.1 Conceptual approach

This first sub-section outlines an applied approach to identifying "Special Products". The point of departure is to present an identification process that results in the designation of "Special Products" that will increase the chances of achieving the development goals embedded in the concept of "Special Products". Satisfactory progress on the designation of "Special Products" in that context means that countries will be better prepared to contribute to a successful outcome in the Doha Round.

The three criteria for identifying "Special Products" represent a fundamental link between trade negotiation outcomes and development goals. As a result, the importance of understanding the role of the country’s goals and strategies to designation of "Special Products" is recognized as being a critical point of departure for the analysis.

The following questions elaborate steps presented schematically in Figure 1 as a process designed to facilitate the identification of "Special Products".

1. What are the country’s goals and strategies, including relative priorities and weights, for achieving food security, livelihood security and rural development?

2. What definition/indicators of food security, livelihood security and rural development match best with the national goals and policy commitments related to the criteria for choosing "Special Products"?

3. What products are the main contributors to the achievement of these goals and strategies? How are these products ranked in terms of the criteria indicators and goals?

4. What national and international policies exist and are needed (related to the principal products) to promote achievement of the goals related to the three criteria? What is the status (do they conform/violate) of these current and needed policies in relation to WTO regulations (market access)?

5. Which of the products most need "flexibility" and why do they need the "flexibility"? At this point the list of principal products is reduced to those needing flexibility?

6. What are the policy/product combinations that do not conform to WTO regulations and what policy flexibility is needed (also related to substitutes). At this point possible treatment of "Special Products" is addressed.

7. What are the current levels of disciplines in the WTO and ambition in the Doha Negotiations and how can the needed "flexibility" for possible "Special Products" be accommodated in the modalities to be negotiated? At this point probable "Special Products" and Flexibility are identified.

8. What adjustments can be made in the probable list of "Special Products" in order to negotiate a multilateral agreement that is beneficial to all the participating countries and their needs? At this point, one establishes the probable "Special Products" and associated flexibility for negotiation.

The above approach is laid out as a series of steps but it should also be perceived as an iterative and dynamic process as countries will change goals and policies as national and international conditions change and are better understood.

The agreed framework for analysis on "Special Products" is underpinned by the criteria set as the basis for identifying the "Special Products". In a trade context the next sub-section addresses definitional and measurement issues related to the three specific criteria agreed for designating "Special Products".

4.2 "Special Products" identification: development criteria linked indicators

A major challenge facing developing countries in the establishment of an effective ‘Special Products" mechanism rests on the perceived misunderstanding of its policy basis by some WTO Members that view the ‘Special Products" initiative as motivated by simple protectionism or opposition to liberalization. It is therefore necessary to emphasize that the ‘Special Products" modality should be seen as providing developing countries with policy flexibility to address crucial non-trade concerns (food security, livelihood security and rural development) that might not be achieved through trade liberalization, in addition to coping with the unstable nature of agricultural markets, and negative impacts from trade liberalization (which can produce damaging shocks, especially to poor and vulnerable developing economies).

How then can developing countries select products based on the three criteria agreed in the WTO negotiations as the basis for designating "Special Products"? Further, can the criteria be viewed as independent or mutually exclusive? Does a product need to satisfy all three criteria to qualify for special treatment under the ‘Special Products" initiative or need it satisfy only one of the criteria? In answering these questions we first examine the indicators that can be used to measure each of the three criteria and then assess the importance or contribution of each agricultural commodity and indicator towards achieving the objectives implied by the criteria. The evaluation of the interrelationship and robustness of the indicators themselves is the focus of the factor analysis in the next section. Box 1 provides working definitions of the criteria.^[7]

4.2.1 Food security indicators^[8]

In evaluating and monitoring food security four dimensions are considered critical:

• availability (production and supply side issues related to physical access and sufficient food)
• accessibility (market demand, income, and trade issues related to economic access),
• stability (including vulnerability both in terms of vulnerable groups and situations)
• use (food safety, nutrition and food choice issues).

The indicators used below are a few indicators considered most relevant in the context of the linking products to food security dimensions:

(a) Contribution of product to nutrition

This indicator measures the share of calories per capita from the product. The ratio used can be:

- Calories per capita per day derived from the Product/Calories per capita per day derived from all products

The degree of undernourishment is based on the calculation of three key parameters for each country: the average amount of food available per person, the level of inequality in access to that food and the minimum number of calories required for an average person. In measuring the average food availability the contribution of each food commodity is tallied based on what a country produces, imports and withdraws from stocks, subtracting the amounts that were exported, wasted, fed to livestock or used for other non-food purposes, and dividing the caloric equivalent of all the food available for human consumption by the total population to come up with an average daily food intake or dietary energy supply (DES). This indicator will be examined over time to see what products are important in reducing undernourishment in selected countries.

(b) Self sufficiency or import dependency in the product^[9]

These indicators measure the share of domestic consumption in domestic production or the proportion of consumption of the product that is imported. The ratios used can be:

- Total of product (X) consumed/total of product (X) produced
- Total of product (X) imported/total of product (X) consumed

The Self-sufficiency indicator tells us the extent to which domestic production exceeds or falls short of domestic consumption. As this ratio exceeds unity, the less will the product be affected by market access provisions. Import dependency, especially if the imported products are highly subsidized can displace local production and change consumption patterns and incentives (negatively affecting rural livelihoods). Although it has been argued that lower priced imported food is essential in alleviating hunger in developing countries, subsidized food imports in some parts of the world has produced results contrary to this observation.

(c) Stability in access of the product

This indicator reflects the production and/or price variability of the main consumed products. The production variability is focused on products mainly produced within the country. The price variability measure covers all important food products, both domestically supplied and imported. The measure used can be:

- standard deviation/coefficient of variation of production and price of product
- degree of price transmission (international vs. domestic) of product
- variability in revenue (export) generated by product activity
- share of (household) total income derived from product activity(ies)

Measuring production and price variability can indicate the adverse impacts on food security in terms of the physical access and economic access dimensions. Thus, a product which is characterised by high production variability can significantly affect the food security situations of the poor who depend on its production. Further, a sudden increase in price would often compel consumers either to spend more of their income to purchase the same quantity or to reduce the intake of the commodity. High variability in the domestic price could imply the need for domestic stabilization policies, while in the case of the international price it could imply a change in border policies.

(d) Product consumption expenditure

This indicator reflects the share of expenditure incurred on the purchase of a product in the total expenditure on the purchase of all products. The ratios used can be:

- Expenditure on the individual food basket item/Total expenditure on food basket

An increasing percentage can point to increasing food insecurity and the need to evaluate both food import and domestic food production and marketing policies. The importance of the different products in the food basket, in terms of income expenditure on them, provides another dimension in the process of identifying products as special. Information on this indicator can usually be found at the national level through household expenditure or living standard surveys. A critical aspect of this indicator is to differentiate between high-priced products and products consumed in large quantities, when they are not the same.

4.2.2 Livelihood security indicators

Livelihood security is an even broader concept than food security and includes several of the dimensions of food security. The aspects stressed in the indicators used here are employment and household income derived from the product.

(a) Level of employment in product/sector

This indicator reflects the product’s share of employment in total employment in a specific area and/or industry, including vulnerable aspects of the labour force linked to the project. Some measures are:

- Share of employment of the product in total agricultural labour force or in total rural employment
- Share of labour force employed in product industry in total labour force
- Gender/Age distribution of labour force employed by the product

These indicators are best disaggregated by product and geographical area. While a particular industry may influence national employment data in some commodity dependent countries, it is also common for a very different and not as important a commodity, in those same countries, to be the livelihood opportunity for a specific rural area. If a certain commodity sector employs a large proportion of farmers and workers, this commodity is considered important in the livelihood context. Measuring employment among small farmers growing basic food crops is especially difficult. One method of computing product-based employment could be by dividing the national average farmer-days per hectare for individual crops by 225 (available working days per year). This would provide an indication of employment (in person years) generated per hectare for that crop.

Gender and age distribution are important for a proper assessment of the product profile. For most developing countries, crops in which more women are engaged in production and marketing are those closely linked to livelihood security. The age distribution of the agricultural sector is also a critical indicator of livelihood sustainability in rural areas.

(b) Income from product

This indicator reflects the product share of income in household income. This can be measured as:

- Income from product industry/total household income

This indicator, linked to the first livelihood security indicator, is based on the premise that the product is a significant employer in the rural area and thus a major contributor to household income. It measures product specific income relative to total household income. Ideally, this indicator shows how much income is earned from growing a particular commodity or working in the commodity industry in the total household income. The higher the number is, the more important the particular commodity is in terms of the product specific contribution to the livelihood system and or rural development objectives. When compared with the products contribution to undernourishment, together, they will provide useful insights on the interrelationship among the criteria.

(c) Agricultural land/assets product share

This indicator reflects the product share of the agricultural land/holdings/assets under cultivation in the country or rural area. This can be measured as:

- Land acreage planted with product/total land under cultivation
- Farm holdings growing the products/total number of land holdings

This indicator shows the product specific share of land to the total land under cultivation. These land assets can be measured in terms of the growth rate in the value of farm assets related to the product. Changes in this indicator over time could be used as an indicator of a product’s contribution to livelihood security and also to rural development.

(d) Incidence of surge/displacement by imports

This indicator is a more defensive and dynamic indicator, measuring the extent to which some livelihood systems may be under threat by imports coming to the country.

- Correlation between imports and domestic production of product
- Growth rate of import substitutes/growth rate of competing domestic product

The indicator focuses on the extent to which lower priced imported products are displacing domestically produced products. It is therefore important to monitor the relationship between the growth rate of these imports and the growth rates of the domestically produced products in order to design appropriate policies to safeguard the local industry. This product specific information is available in both value and volume terms. This indicator directly raises the link between Special Products and access to a possible Special Safeguard Mechanism (SSM) in the negotiations.

4.2.3 Rural development indicators

The linkages between rural area agricultural development and increased levels of overall economic development are well documented. Thus, the key phrases for selecting "Special Products" related to rural development criteria are ‘potential growth’ and ‘economic linkages and development’, which evaluate products in terms of their potential as growth and development poles.

(a) Importance of product in rural agricultural economy

This indicator measures the share of the product in total rural agricultural production.

- Product economic activity share in total rural agricultural output

Disaggregated data measured in the most food insecure areas is essential to identifying the products for Special Product consideration, including the shares of particular vulnerable groups, small farmers, women and youth in the product share. Market access tariff reduction flexibility may not be sufficient and for some production activities "Special Products" flexibility may be needed under both the Market Access and the Domestic Support pillar.

(b) Product and rural area growth

This indicator seeks to capture the importance of a particular product to the growth taking place in a particular rural area.

- Product growth rates relative to rural area growth rates

This indicator selects commodities based on their contribution to rural area income and compares actual and/or potential growth rates of production and/or exports with growth rates of economic activity in the area. It would pay particular attention to products that represent diversification, value added and new market opportunities. Growth rates may be assessed both in value and volume terms.

(c) Domestic value-added potential of product

This indicator focuses on the value linkages of the product as a catalyst and contributor to rural development.

- Degree to which the product can be transformed into other products/uses

This indicator should be evaluated in national level analysis to capture the products that are potentially high-valued and have multiple uses and linkages along the production and marketing chain. These products are often important for diversification and for attracting those leaving the agricultural sector because of the labour demands of traditional cropping systems and the low and limited income earning opportunities.

(d) Tariff revenue from product import/export

This indicator recognizes the role of some products as critical suppliers of revenue for rural development investment in areas such as infrastructure, utility services and social services.

- Tariff revenue generated by the product

The products under this criteria display certain characteristics, for instance, high import values, high current applied tariffs, and sources of its imports being mostly extra-regional. In most cases however, it may be difficult to establish a direct/clear link between revenue generated from a particular product and its use for rural development objectives. However, high tariff rates on these products do not necessarily imply a revenue objective (cultural influences can often be the motivating factor).

4.2.4 Issues related to implementation of the indicator analysis

The indicators above facilitate the identification of Special Products based on the criteria of food security, livelihood security and rural development. One of the main considerations in presenting them is to have quantifiable measures on which to base consideration of "Special Products". This facilitates comparison across commodities and countries, but most importantly, in the context of the on-going WTO negotiations, ensures objectivity. However, possible shortcomings of this process may emanate from several standpoints:

- the indicators may not capture all the products, especially small and remote area products

- not all important dimensions of the three criteria can be easily quantified

- data for all the indicators may not be easily obtainable from both national and international sources

- there is a strong level of inter-dependency amongst the indicators both within the same criteria and between different criteria..

- accurately identifying substitute products and the degree of value addition for a product may prove difficult in some circumstances. Despite the several challenges related to measuring the indicators, they provide a sound basis for identifying special products and data available in the public domain for four countries is used to present results in section four of this paper.

4.3 Statistical analysis of the indicator outcomes

In the preceding section, Special Products have been identified by linking products to indicators based on the criteria of food security, livelihood security and rural development as reflected in the July Framework Agreement. In this section, the statistical technique of Factor Analysis is used to evaluate the relationship between the products and the indicators and the indicators themselves in order to understand their interrelated nature.

As definitions of food/livelihood security and rural development vary amongst developing countries based on their specific national goals and objectives, the menu of indicators that are likely to emerge in the negotiations will be numerous and overlapping. Thus, the methodological approach of Factor Analysis is used to evaluate the pattern and structure of the indicators, the specific contribution of each product to specific indicators, and the relationship between the product scores and the indicator. Importantly, to evaluate the specific contribution of each indicator to a factor that may converge around more than one indicator. Results for such analysis can be used to draw conclusions on two very important aspects in the debate relating to SPs in the ongoing negotiations:

(i) whether the dimensionality of indicators/products can be reduced (i.e. can a country use just a few indicators for designating products as Special Products). This will be based on the relationship amongst the indicators and the factor scores.

(ii) can a product that meets only one of the criteria (food security, livelihood security and rural development) qualify for Special Product status. This will be based on the interrelationship amongst the indicators, the factor loadings and factor scores.

Factor analysis^[10] thus refers to a family of statistical techniques concerned with the reduction of a set of observable variables in terms of a small number of latent factors. The underlying assumption of factor analysis is that there exists a number of unobserved latent variables (or "factors") that account for the correlations among observed variables, such that if the latent variables are distributed out or held constant, the partial correlations among observed variables all become zero. In other words, the latent factors determine the values of the observed variables.

Each observed variable (y) can be expressed as a weighted composite of a set of latent variables (f’s) such that

y = a_i f_i1 + a_i2 f₂ +... + a_ik f_k + e_i

where y_i is the ith observed variable on the factors, and e_i is the residual of y_i on the factors. Given the assumption that the residuals are uncorrelated across the observed variables, the correlations among the observed variables are accounted for by the factors.

The following is an example of a simple path diagram for a factor analysis model. This diagram is a schematic representation of the above formula.

F1 and F2 are two common factors. Y1, Y2, Y3, Y4, and Y5 are indicators or measures of food security, livelihood security and rural development. e1, e2, e3, e4, and e5 represent residuals or unique factors, which are assumed to be uncorrelated with each other. Any correlation between a pair of the observed variables can be explained in terms of their relationships with the latent variables. Further explanation and interpretation of this method accompanies the presentation of the results of the application to two countries in section four.

4.4 Special Products - Trade policy context

The current policy treatment of the product, including the trade policy measures, and their relationship to the current and potential trade regime commitments are clearly factors that will determine the final selection of "Special Products". This applies to both the country’s own treatment of the products and the treatment of the product by its trading partners and others in the multilateral trading system.

The July Framework document is silent on treatment of "Special Products" stating that this "will be further specified during the negotiation phase". Some WTO Members have called for self-designation and that the numbers of "Special Products" be limited. If this is the option, then one might assume that the treatment will be very liberal and "Special Products" will not face reduction commitments and countries will not have to compensate in other areas.

If the number of "Special Products" is more flexible and allowed to correspond to the needs according to the three specified criteria as presented by countries, one would assume that "Special Products" will have a cost and countries will have to compensate with greater tariff reductions in some areas, possibly the opening of tariff rate quotas on "Special Products". In these cases, the tariff reduction and the tariff rate quota would probably be less than on normal and "Sensitive Products" while similarly the level of in quota tariffs would be relatively higher.

There is also the issue of the level at which tariff lines should be designated - broader or narrower. This too is related to what is allowed in terms of numbers and categories of products to be classified as "Special Products". Protection of the domestic market from alternative related crops that displace local crops may require identification of tariff lines in different HS Chapters to protect the same product. Thus, it may be more than the issue of designation at an HS 8 digit level as opposed to at the HS 6 (or HS4 level).

Treatment could also be linked to perceived vulnerability of the different "Special Products" as reflected through the current tariff profile of the country. In some cases, countries may have to explore renegotiation of tariff levels if in the light of the flexibility needed for "Special Products" the current tariff profile does not already provide sufficient protection.

The nature of the Special Safeguard Mechanism, its availability and what it offers in terms of flexibility is also important to "Special Products" and will undoubtedly influence both the choice and treatment of "Special Products". The same applies to the determination on Sensitive Products. Given that Sensitive Products are also open to developed countries and the principle of "substantial improvement" will apply to each product chosen, it may be possible for developing countries to negotiate that in this Doha Round "Special Products" should be exempted from any commitments. This would essentially be treating "Special Products" for all developing countries the same as all products for least developed countries. Unfortunately, the controversial problem of the developing country category being too wide could again be a stumbling block.

4.5 Summary

In summary, using the criteria identified in the Doha negotiations for selecting "Special Products" raises several challenges.

- First, the concepts themselves are very broad and extremely complex to define and measure. They are to be applied to a range of very different countries and conditions within those countries. The conditions differ both in terms of levels of development but also capacities and needs.

- Secondly, given the numerous variables affecting outcomes in the criteria to be used for selection of "Special Products", it could be difficult to make definitive trade and trade policy linkages between specific products and the criteria outcomes.

- Thirdly, there is the issue of data availability, at the national level and moreso at the rural level where the importance of the criteria are relatively more concentrated.

- Fourthly, while criteria for choice of "Special Products" have been specified, it is unclear what the treatment of "Special Products" will be. This treatment would clearly influence the choice of products to be classified as special.

- Fifthly, managing the large number of possible indicators and linking them to the criteria.

The next section of this paper provides results for some countries after applying the approach detailed above using secondary information. It is intended to complement further application of the above approach at the country level by the country itself. This process is underway in several countries, including some of the countries used here only for examples. As indicated above, the main objective of this paper is to lay out the methodological approach and demonstrate it through examples so that countries can advance their own processes of selecting "Special Products" and improve their participation in the ongoing negotiations.