Trade liberalisation involves the lowering of barriers to trade including tariffs, subsidies and other measures such as laws and regulations that constitute non-tariff barriers to trade. The process is commonly governed by trade agreements that bind countries to maximum levels of tariffs, subsidies and limit the implementation of non-tariff barriers to trade. In the context of tobacco control, trade liberalisation is said to pose at least three risks. Firstly, trade liberalisation may increase competition in tobacco markets, thereby stimulating demand and increasing consumption.^1–4 Secondly, where the implementation of non-tariff barriers to trade is restricted, domestic regulatory freedom can be limited to such an extent that effective tobacco control is not lawful.⁵ Thirdly, agreements governing the implementation of non-tariff barriers to trade have been said to create a “regulatory chill”, whereby uncertainty as to the scope of trade obligations discourages regulation.⁶

The response of the tobacco control community to these risks has predominantly been in two forms. Firstly, calls were made during negotiation of the Framework Convention on Tobacco Control (FCTC) for the inclusion of a provision to the effect that the FCTC would prevail over trade agreements in the event of conflict.^7,8 These calls were ultimately not heeded, with the FCTC remaining silent on the question of its relationship to other treaties concluded prior to it.

Secondly, calls were made for the exclusion of tobacco from trade agreements.^6,9,10 These calls have also been unheeded to a large degree. One exception to this trend is the temporary exclusion (to be reviewed in 2007) of tobacco and alcohol from the Pacific Island Countries Trade Agreement (PICTA).¹¹ Although some countries such as Vietnam have refused to lower tariffs on tobacco products in specific trade agreements, this is not a complete exclusion of tobacco products from those agreements since provisions governing non-tariff barriers to trade continue to apply to tobacco.

The purpose of this paper is to question whether the policy of exclusion is the ideal means to address the risks posed by trade processes to tobacco control. That there is a need to eliminate the risks in question ought not to be contentious and is assumed for the purposes of this paper. Similarly, that exclusion may be beneficial to tobacco control is also assumed for the purposes of this paper.

PROBLEMS WITH THE POLICY OF EXCLUSION

Despite the lack of implementation of the policy of exclusion, it remains the most commonly articulated trade-related tobacco control policy, especially in the context of US free trade agreements. The policy of exclusion is, however, problematic for at least three reasons.

Firstly, the effectiveness of a general policy of exclusion is subject to considerable practical constraint. The exclusion of tobacco from one trade agreement may be undermined by its inclusion in another similar agreement. As such, the extent to which the proposal will be effective depends largely on it being universally accepted and applied.

For example, Fiji is party to the PICTA and also to the World Trade Organisation (WTO) Agreement.¹² With respect to non-tariff commitments, to the extent that such commitments found in the PICTA and the WTO Agreement are the same in their content, the only benefit of exclusion for Fiji is that PICTA parties who are not also WTO members would not have access to a forum in which they may challenge Fiji’s tobacco control measures. Unless the terms of PICTA impose upon Fiji’s freedom to regulate tobacco to a greater degree than those of the WTO Agreement (which admittedly is the case with some non-WTO agreements such as some US free trade agreements), the effect of the exclusion from PICTA is minimal from Fiji’s perspective.

However, this is not to say that the exclusion from PICTA will not benefit other PICTA parties. For example, in the case of other PICTA parties that are not WTO members or party to other trade agreements with substantially similar obligations, the exclusion would be an effective means of protecting freedom to regulate tobacco.

In the context of binding tariff commitments, the effectiveness of exclusion is similarly limited. For example, the PICTA does not currently oblige Fiji to lower the tariffs on tobacco leaf or tobacco products with respect to goods originating in the territory of other PICTA parties. This exclusion does not, however, prevent Fiji from committing to lowering its tariffs through other agreements with respect to other PICTA parties or to other countries not party to PICTA.

Secondly, a broad call for exclusion contradicts other tobacco control policies in its failure to take account of issues surrounding domestic support. The provision of domestic support (such as agricultural subsidies) to tobacco sectors may sometimes have the effect of lowering the price of tobacco leaf and the products produced with such leaf, thereby increasing consumption.^13,14 Where such support is in place, the exclusion of tobacco from a trade agreement also removes reciprocal incentives to eliminate such subsidies. This result clearly undermines the purpose of the policy of exclusion and is contrary to calls by tobacco advocates that price-depressing subsidy schemes be eliminated.

Thirdly, although this paper is neutral on the questions of whether further liberalisation of the tobacco sector ought to occur or may produce economic gains, this possibility forms a crucial backdrop to any discussion of trade-related tobacco control policy. The theory of comparative advantage provides that two or more countries gain from international trade where they have divergent opportunity costs. Such gains come in the form of surpluses to either producers or consumers or both. The theory of comparative advantage underpins the international trading system and constitutes a key justification for the role of international trade organisations such as the WTO. Despite limitations on the validity of the theory, it is widely accepted and plays an important role in global policy making.

The potential for economic gains from further liberalisation of tobacco was the subject of a study released in late 2004 by the World Trade Institute (WTI) at the University of Berne, Berne, Switzerland.¹⁵ The WTI Study identified two main forms of protectionism in place in the tobacco sector. Those forms were the imposition of tariffs on manufactured tobacco products (such as cigarettes)¹⁶ and the provision of domestic support (such as subsidies) for tobacco agriculture.¹⁷ In particular, the study noted that developed countries were engaging in “tariff escalation”—that is, some developed countries impose little or no tariffs on the importation of tobacco leaf, but impose prohibitively high tariffs on the importation of manufactured tobacco products. The effect of such tariff escalation is to protect the domestic manufacturing sectors of those developed countries, to the detriment of the manufacturing sectors of other countries, such as developing countries.

The WTI study broadly concluded that meaningful liberalisation in the tobacco sector would lead to overall economic gains.¹⁸i Although consumers are unlikely to reap economic gains from liberalisation if taxation measures and price floors are used to ensure the stability of tobacco product prices, producers are still likely to gain. Gains are likely to accrue in the manufacturing sectors of tobacco-growing developing countries, as exporters increase their overall market share or achieve higher prices in new markets than is possible in their domestic markets. This means that it is possible for overall economic gains to accrue even when product prices remain stable.

It must also be noted that there are many and varied limitations upon the potential of a country to gain from further liberalisation ¹⁹iiand that losses and gains will not be evenly distributed between different actors in the tobacco sector and different countries. For example, goods with negative externalities can generate negative growth and as such, the ability of a country to gain from trade in tobacco products depends partly on taxation policy and its enforceability.²⁰ Nonetheless, the WTI study broadly supports the conclusion that some overall economic and equitable benefit may come from further liberalisation of the tobacco sector if it is well managed.

The policy of exclusion indirectly supports protectionism and thereby stands to undermine the potential benefits of liberalisation. Exclusion would permit the maintenance of laws and regulations that constitute non-tariff barriers to trade, even when such laws and regulations do not serve a health objective—for example, exclusion would mean that a country that is party to a trade agreement would not be prohibited from banning the importation of tobacco products originating in the territory of a trading partner even when the purpose served is purely protectionist. In addition, exclusion would remove reciprocal incentives to eliminate tariffs and domestic support. Somewhat perversely, one of the effects of this would be to protect the tobacco-manufacturing industries of the US and European Union by permitting them to use tariff escalation to effectively undermine the competitiveness of products produced in developing countries.

In conclusion, the policy of exclusion is not a panacea for the risks posed by trade agreements. The policy will usually be problematic and, for this reason, it should not have a role as a standing or generic tobacco control policy. Although some of the problems associated with exclusion will vary in their severity depending on the circumstances, its incompatibility with the objectives of trade puts the policy at odds with a global policy movement supporting freer trade and fairer treatment of developing countries in trade relations. It is particularly difficult to justify exclusion from non-tariff commitments where to do so permits protectionism in the absence of a health rationale. Rather, exclusion can only be justified in this context on a temporary basis where the content of a potential non-tariff commitment may pose a threat to one’s freedom to regulate tobacco and is not properly understood. However, even in this context, once the content of the obligation in question is properly understood, it is better to develop mechanisms such as exceptions to obligations that permit sufficient regulatory freedom than it is to seek blanket exclusion.

WHAT POLICIES OUGHT TO BE ADVOCATED?

It is not possible to fully identify policies that ought to be advocated with respect to trade agreements. However, it is possible, to identify some of the characteristics that such policies ought to possess and to identify some areas in which further research is required.

Firstly, policies must be tailored to individual circumstances in order to ensure that they properly address the risks at hand. The risks posed by trade liberalisation to tobacco control are not universal or uniform in nature. For example, the liberalisation of trade in tobacco will not—even in the absence of regulation—necessarily lead to increases in consumption in every instance or in every market. The factors that are said to lead to increased consumption, such as lower prices, increased advertising and brand proliferation, may not be relevant in all circumstances and may produce different results in different markets. In this respect, Taylor and Bettcher⁴ both concluded that trade openness has had a greater impact on cigarette consumption in low-income countries than in high-income countries.

Secondly, the need to tailor policies to individual circumstances requires that such policies be informed by, and based on, the specific risks in question. If such policy generation is to occur in the tobacco control community, it must assess the risks that are posed by individual trade agreements.

Thirdly, policies ought to be proportionate to their objectives. That is, policies ought to take account of the potential benefits of liberalisation and other public policies, and seek to preserve those benefits wherever it is possible without compromising health. Although it may not always be possible for tobacco control policy to be fully reconciled with other public policy objectives, the recognition of other objectives remains important for at least two reasons. First, the political acceptability of trade-related tobacco control policies is influenced by their acceptability to a trade audience. Second, it cannot be expected that advocates of trade liberalisation will take the objectives of tobacco control into account in policy development when tobacco-control advocates will not take the objectives of trade liberalisation into account.

Fourthly, to ensure that legitimate trade objectives are recognised in tobacco control policies, such policies must address the question of whether domestic regulation can negate negative consequences associated with a potential trade agreement before addressing the question of whether the terms of a proposed agreement are appropriate. For example, when a risk exists that consumption will increase as a result of liberalisation, the factors that may trigger such an increase must be identified. The question of whether regulatory action can negate those factors ought to be addressed before the question of whether the terms of the agreement are appropriate is addressed.

Finally, trade-related tobacco control policy must not focus solely on trade agreements but must also focus on other trade-related processes that may affect tobacco control. For example, unilateral liberalisation of trade, increases in foreign direct investment^4,21 and rising incomes¹⁴ can all lead to increases in consumption by decreasing actual or relative prices of tobacco products. Each of these examples can be related to trade processes without necessarily being caused by the adoption of a new trade agreement.

NEED FOR FURTHER RESEARCH

That trade-related tobacco control policies ought to possess the above characteristics poses problems for the tobacco control community. Identification of the tariff-based and non-tariff-based risks posed by a voluminous number of complex trade agreements under implementation or (usually confidential) negotiation requires a capacity beyond that traditionally associated with the field of tobacco control. Building the capacity to identify health impacts and to coordinate trade and health policy is also problematic for many countries, particularly developing countries. The World Health Assembly implicitly recognised this, when it recently passed a resolution calling for increased policy coordination between domestic and international trade and health authorities (http://www.who.int/gb/ebwha/pdf_files/WHA59/A59_R26-en.pdf).

Calls for assessments of the risks posed by trade to tobacco control are not new.²² The key issue, however, is how to institutionalise such assessments in a manner that overcomes capacity constraints and forms the basis for the generation of sound policy that is capable of negating the risks in question. In this respect, one issue that requires attention is the role of the FCTC. In particular, whether the FCTC can institutionalise risk assessments through reporting obligations, the adoption of a protocol or the creation of a subsidiary body merits consideration.

A number of complex yet fundamental issues are associated with such a question. Such issues include how the FCTC might compel the assessment of risks, whether countries themselves ought to be compelled to conduct assessments or whether another body ought to be appropriately empowered to do so, whether assessments should be publicly available, under what circumstances assessments ought to be conducted and what method such assessments ought to take. In considering these questions, there is also much to be gained from the experience of the environmental movement and the use of sustainability impact assessments in the context of trade agreements.

The existence of further obligations compelling countries to act upon and negate any identified risks must also be addressed alongside the question of risk assessment. This issue requires a broader examination of the current relationship between international trade agreements and tobacco control under international law. As a part of this examination, the relationship between standard provisions of trade agreements governing non-tariff barriers to trade and standard tobacco control measures must be addressed in more detail.

Most importantly, for the time being, the fact that the FCTC may still have a role in managing and reconciling the relationship between trade and tobacco control ought to be recognised. The trade-based threats to tobacco control are clearly cross-border in nature and demand a cooperative approach if they are to be effectively managed. The risks also demand capacity building in countries that are not sufficiently equipped to identify risks. Additionally, the FCTC also provides a forum in which health-based interests can drive international action on trade and tobacco issues.