Trade agreements

A review of trade articles and agreements, and their implications for local/sustainable supply chains (adapted from MacRae, 2014)

The trade agreements have reduced instrument choice, though not consistently and coherently (Guthman, 2008; Hatanaka et al., 2012). In Canada, many pre-AoA programmes directly associated with increased production intensity have been altered in part because of trade agreements (Skogstad, 2008). These include income stabilization schemes that were deemed production distorting, such as the Western Grain Stabilization Program and provincial meat stabilization programmes (Schmitz, 2008),  regional production supports and development schemes, and subsidies to specific production sectors that were deemed underdeveloped or prime export opportunities (Wiebe and Wipf, 2011). Associated with these programme changes have been overall reductions in support to producers (as defined by the Producer Support Equivalent) since the mid-80s (AAFC, 2013).  These instruments were part of the state assistance paradigm that somewhat protected agriculture from market forces and indirectly supported local food systems.  Such instruments, if they still existed, would now likely be categorized as red or amber box by the WTO Agreement on Agriculture (AoA) and restricted (see below). Instead, the favoured tools for the federal and provincial governments are tripartite-funded business risk management programmes, sectoral contribution agreements and information programmes to drive market development.  The contribution agreements are not typically framed around very specific programme parameters, but rather serve multiple purposes.  These measures typically qualify as blue, green and amber box (see below) and are therefore more acceptable to AoA signatories.  As measured by two indices, Canadian policies are only very modestly trade-distorting compared to most other industrialized countries (Anderson and Croser, 2010), again reflecting the reconstruction of state interventions to comply with trade agreements.

Given the instrument choice focus of this analysis, this section provides a review of key trade articles and agreements of the GATT, NAFTA, and WTO  viewed through a local/sustainable lens.  The focus is on currently favoured instruments, and those proposed by local/sustainable proponents, including public procurement, private and government voluntary eco-labeling schemes, government-funded programmes and direct payments to farmers. For example, a government might wish to impose local/sustainable food purchasing targets in food service contracts.  A government/NGO partnership might certify local/sustainable farmers and processors and have a label that alerts buyers to the different attributes of the products to differentiate them from conventional products.  Governments might construct direct payment schemes for farmers to encourage them to shift to sustainable production practices and sell their products to local purchasers.  Such instruments generally fall into the WTO categories of market access and domestic support / subsidies (see below), though numerous complications arise for this analysis from having many instruments affected by multiple trade provisions.  This will be highlighted through the rest of this section.

GATT Articles (and associated WTO agreements and disciplines)

Many GATT rules remain the foundation of trade agreements. They are given “discipline” by many subsequent agreements developed under the aegis of the WTO. Although many rules were first enacted when agriculture was de facto exempt, they were applied to agriculture with the GATT Uruguay Round Agreement on Agriculture, then institutionalized within the WTO when that body was created in 1995.  Two GATT articles are particularly pertinent to this discussion.

GATT 1947 Article III (and the WTO Technical Barriers to Trade (TBT) agreement)

Article III of the GATT is based on the idea that “like products” have to be treated similarly, regardless of source. This article is referenced in many WTO documents to support its provisions.

The contracting parties recognize that internal taxes and other internal charges, and laws, regulations and requirements affecting the internal sale, offering for sale, purchase, transportation, distribution or use of products, and internal quantitative regulations requiring the mixture, processing or use of products in specified amounts or proportions, should not be applied to imported or domestic products so as to afford protection to domestic production. (Article III (1)) (GATT, 1947)

Governments cannot apply a wide range of instruments, including taxes, charges, or non-tariff trade barriers (NTBs), to imported products to support their domestic industry.  Historically, country disputes regarding this rule have been triggered when such tariffs, charges and NTBs are involved.  They include those measures currently used and proposed to support local/sustainable foods.  However, it would be permitted to use border adjustments (tariffs, etc) to treated imported goods the same as domestic goods.  If an environmental charge is placed on domestic food, then an equivalent charge could be placed on the "like" imported goods.  Article II:2(a) provides the exemption (specifically to Article II:1(b)):

2. Nothing in this Article shall prevent any contracting party from imposing at any time on the importation of any product:
(a) a charge equivalent to an internal tax imposed consistently with the provisions of paragraph 2 of Article III* in respect of the like domestic product or in respect of an article from which the imported product has been manufactured or produced in whole or in part;

But are local/sustainable products “like” with imported conventional products?[1] Neither GATT articles nor WTO agreements explicitly define “like” (Vranes, 2011). Extending Swinbank’s (2006) discussion of like products and animal welfare, when products have ecological and ethical values embedded as credence characteristics in the minds of consumers, then GATT’s concept of “like” is not so clear.

Vranes (2011:11), a trade legal scholar, stated: “Article III should be understood as being primarily concerned with products that are in ... a competitive relationship” and further added that other articles and notes focus on those products that are also substitutable.  Swinbank (2006:697-8) recounted that a 1970 GATT working party proposed four criteria for case-by-case assessment of like: “the product's end-uses in a given market; consumers' tastes and habits, which change from country to country; the product's properties, nature and quality; and the tariff classification of the product”.  By these measures, the first three criteria would have different expression for local/sustainable compared to conventional product. As well, these criteria suggest that consumer perceptions of likeness are important, much more so than regulator ones (Vranes, 2011). Vranes (2011:10) also argued that the relevant provisions of the WTO Technical Barriers to Trade (TBT) agreement (which gives discipline to Article III) reflect the same purpose as the GATT articles. It can be argued that local/sustainable products are not closely competitive with imported conventional and therefore not actually ‘like’. They appeal to different market segments, often have different supply chains, and serve different economic and environmental purposes.  They are also not substitutable.

Trade experts have named at least four WTO disputes (WTO, undated) as somewhat pertinent to these questions, although none of them address local/sustainable production and distribution directly.  The absence of pertinent disputes suggests some latitude for governments.  In the dolphin-safe tuna dispute between Mexico and the US (ds381) and then later the EU, an example of how instruments can be affected by multiple provisions, the dispute panel did not rule on Mexico’s claims of violation under Article III, but this dispute ultimately left open the possibility that Article XX(b) and (g) (see below)  could be used to craft a WTO compliant measure (Swinbank, 2006)[2].  In the Canada – EU dispute over the approval and marketing of biotech products (ds292), most of the dispute centred on articles of the Sanitary and Phytosanitary (SPS) agreement.  This dispute was resolved among the parties so Article III was also not addressed by the dispute panel. A trade market dispute between the EU and Australia on geographic indications (ds290) did find that the procedures of establishing such indicators did not accord national treatment to countries outside the EU and was therefore a violation of GATT III:4.  However, the panel concluded that overall there was no substantial violation of WTO obligations. The EU agreed to make some amendments to their legislation and such designations remain permitted. In the Country of Origin Labeling (COOL) dispute (DS384,386) between the US and Mexico and Canada (with others as participants), the original panel and subsequently the appeal panel, concluded that US COOL legislation was according less favourable treatment to imported livestock (beef and pork) than to like domestic livestock and required that the US alter COOL.  This dispute continues to be active as the US has failed, in the eyes of the complainants, to comply with the WTO decision.  This case only addresses local, not local/sustainable, but would appear to support the view that bundling local with sustainable affords protections that are not available when measures are just designed to support local production and processing.

Additionally, paragraph 8 of Article III allows for exemptions for government procurement of products for government purposes without resale.  This might cover, for example, food purchased for hospitals, prisons and in some cases government employees.  However, recent dispute panel rulings on Ontario’s Green Energy Act, suggest a close reading might be required to determine how to structure the language of provisions in these cases so as not to trigger a complaint (Bell-Pasht, 2013).

Another arena under discussion that is potentially pertinent to the question of “like” is process and production measures that are non-product related (npr-PPMs).   Much of the architecture of “like” revolves around physically similar products and product-related process and production measures (PPMs). Some analysts argue that non-product related PPMs, for example sustainable production practices that do not affect a product in obviously material ways, are grounds for discriminating between products, i.e, using npr-PPMs  as a way to distinguish products in the market place would not violate Article III (Vranes, 2011; see also Daughbjerg, 2012).  However, this issue remains largely unresolved in both trade and academic circles[3] (see the next section for additional dimensions of this debate).

If not deemed “like”, then Article III would not be broadly applicable to instruments supporting local/sustainable foods, however the argument would not apply in cases where governments might be favouring domestic sustainably produced foods over international sustainably produced ones.  Currently, that scenario might only arise around certified organic foods.

GATT 1947 Article XX (and the Sanitary and Phytosanitary (SPS) and Technical Barriers to Trade (TBT) agreements)

This article sets out exemptions from the GATT, including measures “relating to the conservation of exhaustible natural resources if such measures are made effective in conjunction with restrictions on domestic production or consumption.”   Such measures must not be applied in a way that ‘would constitute a means of arbitrary or unjustifiable discrimination between countries where the same conditions prevail, or a disguised restriction on international trade’ (Article XX, GATT, 1947).  This exemption is a basis for imposing environmental requirements that must be met by both domestic and international producers.

Article XX is enforced through the SPS agreement, but it only deals with process and product measures (PPM), not npr-PPM (Daughjberg, 2012).   Consequently, the SPS agreement is unlikely to address issues of sustainability, including organic trade, since these are generally viewed to be npr-PPMs.  Similarly, the TBT agreement is intended to cover technical standards, and it is clear that it does cover PPMs, but less clear that it was designed to address npr-PPMs. Vranes (2011) argues that it does, but that the TBT agreement does not per se find them discriminatory.  Canada has argued similarly, but many countries do not agree and there is also debate in the academic literature.  An additional interpretative complication is that the TBT does not specifically name the exception provided by Article XX, though Vranes (2011) believes it is in effect. This situation suggest that using Article XX to justify supports for local/sustainable food is permissible.

The SPS and TBT agreements use international standards, many set by Codex Alimentarius, to bring rigour to their articles.  For example, Article 2.4 of the TBT Agreement states:

Where technical regulations are required and relevant international standards exist or their completion is imminent, Members shall use them, or the relevant parts of them, as a basis for their technical regulations except when such international standards or relevant parts would be an ineffective or inappropriate means for the fulfilment of the legitimate objectives pursued, for instance because of fundamental climatic or geographical factors or fundamental technological problems.

According to Daughberg (2012), however, it remains unclear how WTO rules relate to private standards, especially when they are not under consideration by Codex. Codex is not apparently developing standards for sustainable production and processing systems other than organic. In Canada, other than certified organic standards and regulations, sustainability standards are largely private, with limited state facilitation and enforcement.   The implication is that private standards for local/sustainable, given the lack of attention from Codex, are in a grey zone (Swinbank, 2006). Programs with standards often also have eco-labels and according to Swinbank (2006), labeling for quality attributes, including npr-PPMs, is highly contentious within the WTO and currently unresolved.

WTO Agreements

Under the WTO, several agreements have been adopted that affect local/sustainable food development.  This section focuses on two, the Agreement on Agriculture and the Agreement on Government Procurement.

Agreement on Agriculture

With the Agreement on Agriculture (AoA), governments are supposed to remove barriers to trade.  The agreement emerged out of GATT rules and agreements, including the Subsidies and Countervailing Measures (SCM) Agreement.  The SCM was a component of GATT 1947 rules, and defined a subsidy as a transfer of funds to private firms and beneficial to that firm.  In cases, however, where the benefits are society-wide, as with subsidies for environmental improvements beyond the farm, and where the subsidy only covers part of the farms costs associated with implementing environmental improvements, there is a legitimate question about whether it would be considered a farm subsidy. These questions appear to have coloured the structure of the domestic support provisions in the AoA (see below).  The SCM agreement also prohibited export and import substitution subsidies, but interestingly, the WTO AoA, which further developed rules against export subsidies (see below), does not explicitly mention import-substitution subsidies.

The Agreement has a somewhat tortuous negotiating history. The Doha Round that commenced in 2001 to update the original rules has never actually been completed.  Negotiations were stalled until very recently[4], and the AoA is still operating largely under rules agreed upon in 1995.  The three guiding pillars of the AoA are:

  • Market access: elimination of import quotas and non-tariff trade barriers (NTBs), replaced with tariffs that are progressively reduced. Of particular significance, Canada reduced tariffs on supply managed commodities by 36% in the 1995-2000 period.
  • Export subsidy (red box) reductions: each country had different reduction targets depending on volume or value. The original intent was to eliminate the red box, but this has yet to happen, even though the US and EU have proposed it at the negotiating table since 2004.
  • Domestic support reductions (sorted into three categories: Amber, Blue and Green).

Regarding market access, much of the focus of the agreement is on import quota and tariff reduction, which could affect supply managed commodities, but many currently proposed supports for local/sustainable food systems would likely be considered NTBs. However, the structure of the AoA appears to offer numerous opportunities to exempt such supports or position them as non-distorting. The agreement does not specifically apply to the Municipal Agencies, Schools and Hospitals (MASH) sector.  Nor does it say anything about non-governmental organizations.  Multi-actor programmes, a common feature of current development efforts (as described above), might not then be covered by the agreement, even if the state was part of the partnership.  However, as Vranes (2011) cautions, under GATT rules, the activities of non-state actors can be attributed to the state, when influenced by state incentives and disincentives.

Export subsidies are not directly applicable to this discussion, but domestic supports are very pertinent.  Under the agreement, green box (or environmental measure) support is permitted, so it is feasible to design local/sustainable programming for this category, since if properly constructed, such measures can also generate significant environmental improvements.

Even if local/sustainable provisions fall into the amber box, governments are allowed to provide a considerable amount of subsidy, but not exceed the limits imposed by the AoA. The federal government’s spending on amber box programmes has been substantially below the agreed upon limit for much of the 2000s (Holden, 2005). Canada’s most recent domestic support notification to the WTO (deposited in 2012 and covering 2009) reveals that spending was only about 32% of its commitment level for the period in question (WTO Committee on Agriculture, 2012).  In part, this is because when a specific programme costs less than 5% of total farm income in the targeted commodity area, it is not counted, known as the de minimus provision (Brink, 2009).  Many small support programmes are, thus, not counted as part of the commitment.

WTO Agreement on Government Procurement (AGP)

Canada is one of a limited number of signatories to this agreement that applies to central agencies and provincial governments and their affiliated agencies.  It is in effect for central government contracts valued at over CAD$204,750 (Sept. 23 2013 SDR/CAD$ exchange rates) and sometimes higher for sub-central agencies.  This agreement is primarily about tendering procedures, supplier qualification and selection, and awarding of contracts, and may limit the choice of supplier for a contract, but the AGP does permit proposals with technical specifications regarding performance (as opposed to design or descriptive characteristics), particularly when based on international standards, national technical regulations, or recognized national standards (Bell-Pasht, 2013).

The AGP, as with many agreements, has a general exception for public interest measures to protect human, animal or plant life or health and Canada has exempted procurements with set-asides for small and minority businesses or contracts for agricultural products that further agricultural support programs or human feeding programs. Some sub-central agencies are also exempt if the procurement relates to regional development and environmental quality objectives. The provisions cover many provincial departments, but equally many MASH entities and NGOs are exempt, depending on the province.


A deal between Canada, the USA and Mexico, NAFTA was actually the first free trade agreement to apply GATT-like rules to the agriculture sector (Doern and MacDonald, 1999).  Its provisions for food and agriculture are generally aligned with GATT and WTO requirements (tariff reduction, elimination of export subsidies), except that it is largely silent on the kinds of domestic support measure restrictions set out in the AoA (Konforti, 2010). It did, however, largely protect the supply managed industries. The agreement also regulates government procurement (Chapter 10), largely also aligned with GATT non-discriminatory rules. The procurement rules between Canada and the US, however, are still only applicable at the federal level for goods valued at more than USD$25,000. Provisions were originally non-binding at the municipal and provincial levels (Carter-Whitney, 2008) but provinces are now subject to the rules of the WTO AGP.  Municipal procurement remains excluded under NAFTA.

The Canada – EU Comprehensive Economic and Trade Agreement (CETA)

More erosion on IPR that negatively affects seed saving
Subsidies left primarily to WTO rules and negotiations
Further erosion of supply management by allowing cheese products in tariff – free (4%), phased-in
Procurement rules apply more fully to provincial and local governments than previous agreements, but not full coverage and still also financial thresholds
There are rural development exemptions for some provinces and territories
Has provisions (at this point vague) related to environmental protection, sustainable development and trade


[needs updating]

The recently announced signing of CETA adds potentially new dimensions to this discussion, but many deal details have yet to be worked out or released.  The agreement in principle[5] builds on existing agreements but has some features that might affect the line of argument presented here.  The text:

  • does have a sustainable development and trade section, a trade and environment section, and a trade and labour section, and CSOs are to participate in monitoring the provisions when worked out;
  • calls for a ban on export subsidies pending tariff reductions;
  • proposes collaborating on regulatory measures, including animal welfare, and technical; standard equivalency through the TBT process;
  • sets out immediate EU tariff elimination on some foods, mostly processed;
  • sets out quota duty free access for certain goods to each other’s market;
  • has no changes to supply management, except increased cheese and milk proteins access for European producers.

The procurement provisions are perhaps the most significant. Gerry Ritz, the minister of agriculture, had indicated earlier that he saw nothing in the negotiations that would jeopardize buy local programmes (CBC News, 2011) and news reports prior to the announcement implied that some protections would continue to exist for local purchasing (Clark et al., 2013). Shrybman (2010) had argued, from earlier released drafts, that the CETA would go beyond the WTO AoA and AIT in changing municipal procurement provisions which could then also shift rules around food.  Access to sub-national purchasing provisions was thought to be a prime interest of EU negotiators (Sinclair, 2011). The procurement provisions of the agreement in principle:

  • are broadly based on the WTO- GPA;
  • maintain a nation’s ability to set technical standards and social and environmental criteria and favour eco-labels;
  • provide broad exceptions for key concerns regarding human, animal, plant life and health;
  • allow Canada to maintain exclusions for regional economic development, health care and other public services, and agricultural goods as part of food programs;
  • cover federal, provincial and municipal governments, an expansion of application beyond many aspects of the WTO-AGP;
  • set varying thresholds for contracts to which rules are applied: high value contracts at $560,000; federal, crown, and arms-length contracts at $205,000; the provinces and MASH sector at $315,000 (these thresholds are higher than the AIT);
  • put in place some restrictions on geographic indications (GI) to protect EU GIs.

How agricultural goods are defined and whether food resold to the public will be covered under procurement provisions are open questions.   Although provinces have indicated preliminary agreement, the adoption process will likely take several years and could be fractious (Clark et al., 2013).  Numerous Canadian municipalities have already passed council resolutions seeking exemptions from CETA provisions (Council of Canadians, 2011). Because the EU has used exemptions provided by the AGP on which to base Fair Trade and regional development initiatives (Rourke, 2011; Bell-Pasht, 2013), and has a 20-year history of supporting in more substantial ways than Canada adoption of sustainable food production and distribution measures,   it may not want to settle on final language that compromises current initiatives, leaving Canada also freer to support local/sustainable food systems development.


[1] Note that the arguments that follow do not apply when comparing local vs. imported sustainable products.

[2] Swinbank (2006) also reported in the shrimp and sea turtles case that Article XX(g) might also be invoked if the environmental legislation was suitably constructed.

[3] Note that Vranes (2011) has concluded from a legal interpretation of the pertinent texts that PPMs, whether npr or not, do not per se represent discriminatory treatment.

[4] The Bali mini-deal of December 2013 has some significant features, but does not affect  the arguments presented here (