Codex in Motion: Food Safety Standard Setting and Impacts on Developing Countries’ Agricultural Exports

DOI: 10.1111/1746-692X.12293 Codex in Motion: Food Safety Standard Setting and Impacts on Developing Countries’ Agricultural Exports Le Codex en mouvement : établissement de normes de sécurité sanitaire des aliments et effets sur les exportations agricoles des pays en développement Codex in Bewegung: Festlegung von Lebensmittelsicherheitsstandards und Auswirkungen auf die Agrarexporte von Entwicklungsländern

The Codex Alimentarius Commission (CAC) plays a central role in the global food system.Codex food safety standards are considered to be the 'international standard' used as guidance for national food safety regulation and often used as a reference to litigate World Trade Organization (WTO) trade disputes.While the work of the Codex committees 1 receives considerable attention and reference, they have also been recognised as 'battlegrounds [that] are at the center of some of the most prominent disputes over food regulation within the global trading system' (Josling et al., 2004, p. 54).
Diverse food standards across countries reflect national regulatory traditions and food culture, but are also highly contentious, because they imply trade costs and market access difficulties for countries with differing regulatory systems.The CAC attempts to bridge these differences across countries by agreeing on internationally accepted food standards; these include provisions on food hygiene, food additives, pesticide residues, veterinary drugs residues, contaminants, labelling and presentation, methods of analysis and sampling, and import and export inspection and certification.In practical terms, this means that standard by standard, food item by food item, for each potential contaminant, residue level, additive or hygiene issue, a standard has to be negotiated in an often time-consuming procedure.According to the Food and Agriculture Organization and World Health Organization (FAO/ WHO, 2018), over its 56-year history, Codex has produced: more than 4,100 maximum levels (MLs) for food additives; over 5,200 maximum residue limits (MRLs) for over 300 pesticides; over 600 MRLs covering 63 veterinary drugs; more than 200 commodity standards; and over 100 MLs covering 18 contaminants in food.
While harmonisation of food standards globally may be the ultimate goal, pragmatically it is difficult to achieve.Both developed and developing countries must balance reductions in human health risks against economic benefits in production and trade associated with setting food safety standards that are 'appropriate' and 'reasonable' among 188 Codex member nations and one member organisation, the European Union (EU).There are several instances in which the adoption of Codex standards has been protracted, put on hold, or discontinued.Factors contributing to delays in reaching agreement on Codex standards include: (1) concerns with respect to WTO implications of the standards; that is the 'status of national regulatory requirements which may differ from the international standard being developed, and hence the potential for a WTO challenge'; (2) 'non-science issues' specified as 'consumer opinions and preferences, impact on consumer confidence, prohibition in national legislation, etc'; and (3) the question of how to deal 'with situations where objections to the adoption of a standard are not based either on sound science or on other factors that are globally applicable' (CAC, 2012: para. 18, 40, 43, 51).
Examples of food safety standard setting in which the Codex process was discontinued include: (i) General standard for processed cheese, discontinued in 2017 after As Josling and Tangermann (2016) note: '… the issue of [EU] banning so-called "chlorine chicken" is now one of the battle cries of those in the EU opposed to the TTIP' (p.3).As Josling et al. (2004) conclude, the global food system has much to gain from a well-designed and rigorously enforced set of international food safety regulations that target hazards that threaten consumer health and undermine confidence in the food supply.
These situations pose challenges to exporters, particularly from developing countries, for at least two reasons.First, lower-income nations often lack sufficient capacity and scientific technical expertise to evaluate, develop and implement their own food safety standards.Second, empirical evidence suggests that developing countries benefit more than developed countries from an internationally harmonised system of food safety standards (Beghin et al., 2015;Murina and Nicita, 2017).
The objective of this article is to assess the duration of selected Codex case-study decisions and evaluate the implications for agrifood trade with a focus on developing countries.Specifically, (1) we review the decision procedures in the CAC; (2) summarise for the last ten years the number of standards that were under consideration by the CAC, and (3) evaluate the trade flow effects of five case studies of food safety standards: (i) three cases in which Codex consensus was achieved and standards adoption was accelerated, and (ii) two cases in which the adoption of Codex standards have been delayed and remain on the agenda for future discussion.

Table 1: Number of standards for Adoption in Codex Commission, 2010-Present
Step 1 Step 5 Step Codex in motion: duration of standard-setting for selected cases Five cases are selected for review and analysis.They were chosen because: (i) the products impacted are important to at least some developing country regions in terms of export earnings; and (ii) to contrast the potential trade implications for developing countries when the setting of international standards was delayed compared to situations where the adoption of standards was expedited.
The five case studies are as follows: Codex Standards that were Accelerated and Adopted 1) Sulphur Dioxide (SO 2 ) levels impacting Sri Lankan cinnamon exports.
2) Melamine contamination impacting China's milk and powder exports.
3) Guidance document for microbial hazards concerning Honduran melon exports.
Codex Standards that are Delayed 4) Cadmium MLs in chocolate.
It should be noted that while the development of Codex standards is a deliberative process, given the nature of performing a case study analysis, some limitations regarding representativity and inference towards a larger set of countries apply.In particular, we focus on developing country exports that have been affected by the lack of a Codex standard and their difficulty meeting standards set by national authorities.However, as one reviewer noted there are likely countries that gain from the lack of an international standard if they can meet an importing countries' often stricter national standards.Thus, the case study trade flow impacts presented here may not reflect the full spectrum of winners and losers from the lack of a Codex standard.However, empirical evidence suggests that similarity of standards across countries is generally supportive for trade (Disdier et al., 2019;Grant and Arita, 2017).Moreover, the Agreement on the Application of Sanitary and Phytosanitary Measures (Article 3) encourages (but does not obligate) members to harmonise their measures with international standards.Thus, focusing on these somewhat polarising Codex cases and the CAC standard setting process allows us to shed light on the potential economic and trade implications associated with Codex delays.In addition, case-study impacts have also been econometrically analysed supporting our descriptive case study findings (see Wieck and Grant, 2020).
Differences in length of Codex decision making can be seen when comparing the selected case studies with the average standard setting duration of approximately 4.2 years (Figure 1).In particular, when food incidences were reported, Codex shows rather quick adoption times with the duration of the standard work ranging from 1-2 years.In the case of cadmium in chocolate and aflatoxins in ready-to-eat peanuts however, the duration of standard setting has taken three times as long, In what follows, we summarise each case study.It is important to note that a food scare incident in a country does not automatically start the process of standard setting in Codex.
It is up to the companies in the affected country to bring the issue to their government and trade officials so that it can be raised in the CAC where a decision by the Codex Members must be taken to initiate the process of drafting standards, if none exist.

Sulphur dioxide (SO 2 ) impacting Sri Lankan cinnamon exports. In 2004, Sri
Lanka encountered problems with consignments of Ceylon cinnamon exported to the EU.At issue was the detection of consignments containing SO 2 , for which the EU's regulation on food imports prevented a sizeable share of Sri Lanka's cinnamon exports from entering the EU market.Given the importance of cinnamon exports to Sri Lanka, the 29th Session of the CAC, held in Geneva in July 2006, 2 adopted a maximum level (ML) of 150 mg/kg for sulphites (including sulphur dioxide) in food category 12.2.1 'Herbs and spices' of the Codex General Standard for Food Additives (GSFA).

Melamine contamination impacting China's milk and powder exports.
In 2008, China's powdered milk was found to contain melamine involving intentional adulteration (Huang, 2014).Forty seven countries received melamine-contaminated products, and trade was disrupted as countries adopted their own MLs -sometimes at zero -for melamine in the absence of a science-based international standard.A WHO-FAO Expert given that CCCF had reached agreement on the standards based on the percentage content of cocoa solids, the committee sent the proposed ML to the CAC with the recommendation for adoption at Step 5/8.The Committee Chair said a compromise had been reached by the CCCF even though it may result in a higher rejection rate of exports from some Latin American and Caribbean countries.
The draft ML stalled in Codex's main decision body (CAC).This time, some African countries expressed concerns that the draft ML was not sufficiently stringent along the lines of the MLs advocated by the EU and Norway (i.e. an ML less than 0.3mg/kg), and could jeopardise their exports that have been able to meet the high cadmium in chocolate standards (i.e.cadmium MLs less than 0.3mg/kg) required in the EU market.The CAC adopted the ML at Step 5, but accelerated adoption, by leaving out steps 6 and 7, of the final standard was rejected and it remains on the CCCF agenda.
In the absence of a codex standard, importers can set their own (more stringent) standards provided they are based on a scientific risk assessment.

Aflatoxin MLs in ready-to-eat peanuts.
The

Melamine contamination affecting China's milk and powder exports.
The most direct trade outcomes around the period of the melamine incident can be observed in HS products 040120 (milk and cream not concentrated) and 040210 (milk powders in solid form including for infant use) (Figure 3).The left graphic in Figure 5 demonstrates that CCASIA and CCLAC are the

Peanut exports by Africa, Latin America and Caribbean and Asia.
There is not yet an internationally agreed upon international standard for aflatoxin in ready-to-eat peanuts.The trade impact of stricter aflatoxin MLs on ready-to-eat peanuts was measured through analysing developing country exports to the EU, which in 2010, adopted its own, lower than Codexproposed (10-15μg/kg) aflatoxin MLs, of 4μg/kg for both uses of nuts: ready-to-eat peanuts and those for further processing.To the best of our knowledge, these tolerances reflect the most recent adoption of MLs by the EU.
In Figure 6 Second, CCASIA's symmetric bilateral trade intensity index with the EU is also negative suggesting a weaker than expected trade relationship.However, CCASIA's index began trending higher in 2015 which may

Box 1: The Symmetric Bilateral Trade Intensity Index
Formally, the bilateral trade intensity index (BTTI) identifies destination countries in which an origin country's exports are concentrated.Let i (j) denote the exporting (importing) country, and w the rest-of-world (RoW) market.Letting X denote the value of exports, the BTTI is defined as: The numerator in the BTTI is the share of i's exports sent to j.The denominator is the share of world (w) exports sent to j.Thus, the BTTI normalises the share of i's exports to j (the numerator) by the relative importance of the RoW's exports to j (the denominator).Because both numerator and denominator are shares -one in terms of the partner country (numerator) and one in terms of the world market (denominator) -the value of the BTTI ranges from zero to infinity.
More intuitively appealing, is the symmetric BTTI (SBTTI), which scales the BTTI on the domain {-1,+1} as follows: Positive values of the SBTTI reflect a more intense trade relationship; a neutral trade relationship (i.e.neither intense nor weak) exists when the index approaches zero; and a relatively weak trade relationship vis-à-vis the world market exists when the SBTTI approaches a negative one.
EuroChoices 0(0) ★ 09 © 2021 The Authors.EuroChoices published by John Wiley & Sons Ltd on behalf of Agricultural Economics Society and European Association of Agricultural Economists suggest some of its producers and exporting firms are adjusting to meet the stricter EU aflatoxin standards.Finally, while CCLAC's trade has been impacted significantly by EU standards on cadmium (Figure 5) the reverse is true for its peanut exports (Figure 6).CCLAC's symmetric bilateral trade intensity index with the EU has remained positive and stable through time suggesting a stronger than expected symmetric bilateral trade intensity with the EU relative to competing suppliers and evidence that CCLAC is capable of meeting the stricter aflatoxin standards imposed by the EU.
Collectively, the contrasting trade outcomes for peanuts and chocolate exports from Africa and LAC to the EU may explain the dichotomy of interests and prolongation of standard-setting at the Codex level.However, in contrast to standards that were accelerated in the Codex process, delayed adoption of international standards can lead to export losses for at least some countries and regions.

The important role of Codex
Society and European Association of Agricultural Economists over 10 years of discussion with no consensus; (ii) MRL for ivermectin in beef, discontinued in 2017 because of new scientific information from the Joint Expert Committee on Food Additives (JECFA); and (iii) MLs for cadmium in dry mixtures of cocoa and sugars sold for final consumption, discontinued because of limited international trade.Other examples include Non-centrifuged dehydrated sugar cane juice, discontinued in 2019 after several years of discussion; MRLs for Bovine Somatotropin (BST) and Zilpateral which have been on the agenda since 1998 and 2012, respectively; and a ML for methylmercury in amberjack and swordfish discontinued in 2018 because of low levels of concentration and lack of consensus.Finally, there are other prominent examples of food safety issues that continue to plague governments and trade negotiators in high-income markets such as the US and EU, despite the existence of Codex standards and guidance (Wieck and Rudloff, 2020).
Authors.EuroChoices published by John Wiley & Sons Ltd on behalf of Agricultural Economics Society and European Association of Agricultural Economists as consensus on the final level of MLs has been held up.

Figure 5 :
Figure 5: region chocolate exports to the world and EU-1998-2019

Guidance document for microbial hazards concerning melon exports by Honduras. In
Meeting was convened in December 2008 and provided guidance to the Codex Committee on Contaminants in Foods (CCCF) when it met in March 2009.That same year, the Committee approved a proposal to establish and the CAC finalised MLs for nonintentional and unavoidable presence in July 2010.

Figure 1: Comparison of Codex work for selected cases
Source: Authors' own illustration."Wenn der Codex " EuroChoices 0(0) ★ 05 © 2021 The Authors.EuroChoices published by John Wiley & Sons Ltd on behalf of Agricultural Economics Society and European Association of Agricultural Economists the new work in 2011.A CCFH inter-session work group drafted the proposed guidance, which was taken up by the CCFH in 2011 and sent to the 2012 Commission session at Step 5/8, where it was adopted, less than two years after it was proposed.Cadmium MLs in chocolate.and cocoa-derived products by the percentage content of cocoa solids.At the 2019 CCCF meeting, there was general support for an ML of 0.3mg/kg for chocolate containing less than 30 per cent cocoa solids.The EU and Norway advocated for a lower ML of 0.1mg/kg to 'ensure sufficient protection, in particular for children' consistent with the EU's risk assessment.See https://ec.europa.eu/food/sites/food/ files/ safet y/docs/codex_cac_42_ cl_2019-46-cccf.pdf,p. 1.However,

Laboratory work on maximum limits is important for food safety compliance.
" 06 ★ EuroChoices 0(0) © 2021 The Authors.EuroChoices published by John Wiley & Sons Ltd on behalf of Agricultural Economics Society and European Association of Agricultural Economists

Right Axis) Figure 3: Exports by China of milk and milk powder including for infant use
Source: Authors' calculations from Trade Data Monitor (TDM).Note: Milk/Cream not conc.falls under HS code 040120 and Milk Powder in solid form including for infant use (Milk Pwd/Solid Incl.for Infant Use) falls under HS code 040210.EuroChoices 0(0) ★ 07 © 2021 The Authors.EuroChoices published by John Wiley & Sons Ltd on behalf of Agricultural Economics Society and European Association of Agricultural Economists

There is not yet an agreed international standard for aflatoxin in ready-to-eat peanuts.
Evaluations of the Joint FAO/WHO Expert Committee on Food Additives (JECFA), World Health Organization, Available online at: https://apps.who.int/food-additives-conta minan ts-jecfa-datab ase/chemi cal.aspx?chemID=1376 Wieck, C. and Rudloff, B. (2020).Bilateral positive integration: Different strategies for regulatory cooperation in the TTIP, in Krämer-Hoppe (ed.):Positive Integration-EU and WTO Approaches Towards the Trade and Debate, European Yearbook of International Economic Law 2020 (Springer), pp.17-34.