MRL Book

Understanding international harmonization
of pesticide maximum residue limits

with Codex standards A case study on rice
Food and Agriculture Organization of the United Nations

Rome, 2020
Required citation:
FAO. 2020. Understanding international harmonization of pesticide maximum residue limits with Codex standards: A case study
on rice. Rome. https://doi.org/10.4060/cb0463en
The designations employed and the presentation of material in this information product do not imply the expression of any opinion
whatsoever on the part of the Food and Agriculture Organization of the United Nations (FAO) concerning the legal or development
status of any country, territory, city or area or of its authorities, or concerning the delimitation of its frontiers or boundaries.
Dashed lines on maps represent approximate border lines for which there may not yet be full agreement. The mention of specific
companies or products of manufacturers, whether or not these have been patented, does not imply that these have been
endorsed or recommended by FAO in preference to others of a similar nature that are not mentioned.
The views expressed in this information product are those of the author(s) and do not necessarily reflect the views or policies of
FAO.
ISBN 978-92-5-133115-6
© FAO, 2020
Some rights reserved. This work is made available under the Creative Commons Attribution-NonCommercial-ShareAlike 3.0 IGO
licence (CC BY-NC-SA 3.0 IGO; https://creativecommons.org/licenses/by-nc-sa/3.0/igo/legalcode).
Under the terms of this licence, this work may be copied, redistributed and adapted for non-commercial purposes, provided that
the work is appropriately cited. In any use of this work, there should be no suggestion that FAO endorses any specific organization,
products or services. The use of the FAO logo is not permitted. If the work is adapted, then it must be licensed under the same or
equivalent Creative Commons licence. If a translation of this work is created, it must include the following disclaimer along with
the required citation: “This translation was not created by the Food and Agriculture Organization of the United Nations (FAO). FAO
is not responsible for the content or accuracy of this translation. The original [Language] edition shall be the authoritative edition.”
Disputes arising under the licence that cannot be settled amicably will be resolved by mediation and arbitration as described in
Article 8 of the licence except as otherwise provided herein. The applicable mediation rules will be the mediation rules of the
World Intellectual Property Organization http://www.wipo.int/amc/en/mediation/rules and any arbitration will be conducted in
accordance with the Arbitration Rules of the United Nations Commission on International Trade Law (UNCITRAL).
Third-party materials. Users wishing to reuse material from this work that is attributed to a third party, such as tables, figures
or images, are responsible for determining whether permission is needed for that reuse and for obtaining permission from the
copyright holder. The risk of claims resulting from infringement of any third-party-owned component in the work rests solely with
the user.
Sales, rights and licensing. FAO information products are available on the FAO website (www.fao.org/publications) and can be
purchased through [email protected]. Requests for commercial use should be submitted via: www.fao.org/contact-us/
licence-request. Queries regarding rights and licensing should be submitted to: [email protected].
Cover photographs: ©FAO/Codex Alinentarius (left); ©Ryo Yoshitake (right), ©FAO/Michael Tewelde (bottom)
Contents
Foreword v
Acknowledgements vii
Background and structure of the study ix
Abbreviations and acronyms xi
Terminology xi
Executive summary xii
PART A Level of harmonization of rice PART B Reasons behind different levels

pesticide maximum residue limits with Codex and of pesticide maximum residue limit harmonization 33
impact on trade 3
1 Global rice market 5

1 Risk assessment methodology 35
1.1. Residue definitions 35
2 esticides used on rice and harmonization

P
with Codex maximum residue limits 9
1.2. Interpretation of toxicity studies
1.3. Assessment of metabolism studies
1.4. Exposure assessments
37
39
40
1.5. Differences in Good Agricultural Practice (GAP)
2.1. Data sources and methodology 9
and selection of critical GAP 41
2.2. Codex maximum residue limits for rice 10
2.3. National maximum residue limits for rice
2.4. Codex maximum residue limits (MRLs):
10
2 Risk management considerations 43
existence of corresponding national MRLs and 2.1. Establishment of maximum residue limits
their level of harmonization 13 for specific commodities/groups 43
2.5. Most used pesticides on rice in major rice 2.2. Maximum residue limit extrapolation rules 43
producing countries 17 2.3. Application of processing factors 44
2.6. Transparency 17 2.4. Application of default maximum residue
2.7. Concluding remarks 18 limits for non-registered pesticides 44
2.5. Risk management factors applied to the
3 What does this mean for trade? 19
toxicology evaluation
2.6. Import tolerances
45
45
3.1 Food safety standards and trade 19 2.7. Commodity descriptions 46
3.2 Analysis of the effect of pesticide maximum
residue limits on trade in rice in selected
countries 22
3 Conclusions and way forward 49
3.3 Stringency of pesticide maximum residue
limits compared to Codex 23
3.4 The effect of pesticide maximum residue
limits on trade in rice 24
3.5 Concluding remarks 26 References 53
4 Conclusions and way forward 27 Glossary 59
ANNEXES PART A ANNEXES PART B

1 Sources of national pesticide maximum residue 1 ossible areas of investigation to understand
P
limits for the 19 countries/region analysed 62 better the issues that might lead to different
national and Codex maximum residue limits 75
2 Rice classification 64
2 Analysis of data on pesticides used on rice 83
3 HS codes and Codex rice classification 68
4 Methodology and data of the trade model 69
5 Detailed results of the trade model 72
iii
Figures
Figure 1: Tariffs and AVEs of NTMs, by economic sector 4
Figure 2: R
ice available for consumption (kcal/capita/day), average 2015-2017 5
Figure 3: Total number of national MRLs established for rice in the 19 countries/region analysed 11
Figure 4: S
hare of national rice MRLs for which a corresponding Codex MRL exists, in the
19 countries/region analysed 12
Figure 5: S
hare of pesticides, with national rice MRLs but not Codex MRLs, for which a
Codex MRL exists for commodities other than rice 12
Figure 6: 8
2 Codex rice MRLs: existence of corresponding MRLs at national level and level of
alignment with Codex in the 19 countries/region analysed 14
Figure 7: L
evel of alignment with Codex rice MRLs by processing level categories: Rice unprocessed (U),
Rice husked (H), and Rice polished (P)17 14
Figure 8: Alignment with the 82 rice Codex MRLs in the 19 countries/region analysed 24 16
Figure 9: M
RL indices by tariff line and country, importing countries/markets, 2018 24
Figure 10: MRL indices by tariff line and country, exporting countries, 2018 24
Figure 11: The dual effect of pesticide MRLs on imports of rice 25
Tables
Table 1: Top ten rice producers in the world (in metric tonnes – paddy) 6
Table 2: World rice production and domestic use (milled equivalent), average 2017-2019 6
Table 3: Top 20 rice importers in the world, average 2017-2019 7
Table 4: Top ten rice exporters in the world, average 2017-2019 7
Table 5: Codex MRLs for rice 11
Table 6: Existence of national MRLs and enforcement of Codex MRLs in the absence of national MRLs 15
Table 7: Definition of dependent and explanatory variables 70
Table 8: Estimated coefficients for rice imports32 72
Table 9: Marginal effects and standard errors of MRLs on rice imports 73
Boxes
Box 1. Major species of rice 10
Box 2: Effects of other country-level and bilateral determinants on trade in rice 26
Foreword
The FAO/WHO Codex Alimentarius is the most important internationally recognized standard-setting body on
food safety and quality. Its primary objectives are to protect the health of consumers and ensure fair practices in
international food trade. Codex standards are based on solid and independent scientific advice provided by FAO
and WHO and are discussed and agreed through an inclusive and transparent process that allows global trust in
the safety of food.
Since its establishment in 1963, Codex has developed hundreds of internationally recognized standards,
guidelines and codes and has defined thousands of permitted levels of additives, contaminants and chemical
residues in food. Among them, pesticide Maximum Residue Limits (MRLs) define the maximum concentration
of a pesticide residue to be legally permitted in food commodities and animal feeds, ensuring that food is safe for
consumers and public health is protected.
Globalization and growing volumes of traded food have increased the chances that the food produced in one
place affects the diets and health of people living elsewhere. Internationally recognized food standards developed
by Codex, including pesticide MRLs, ensure that trade and food safety go together and that the food reaching the
plate is safe and of expected quality. It is in this context, that the WTO Agreement on the Application of Sanitary
and Phytosanitary (SPS) Measures strongly encourages governments to harmonize their food safety regulations,
using Codex standards as the benchmark. Through that, the SPS Agreement seeks to strike a balance between
WTO Members` rights to regulate legitimate objectives, such as ensuring food safety of their populations, while
avoiding that such regulations are used as unnecessary barriers to international trade.
The harmonization of national regulations with Codex pesticide MRLs has been discussed in different fora for
several years. Within the Codex Alimentarius Commission (CAC) concerns were at first raised about the loss of
pesticide MRLs due to the periodic review policy applied by the Codex Committee on Pesticide Residues (CCPR).
Within the relevant WTO bodies, Codex pesticide MRLs and their relevance for trade have been repeatedly called
into question. At the 11th WTO Ministerial Conference, held in Buenos Aires in December 2017, a joint statement
on pesticide MRLs and trade was signed by ministers from a number of developed and developing countries,
calling for greater harmonization across national and regional MRLs.
This publication focuses on pesticide MRLs in rice. It builds on a pre-study carried out by FAO in 2017 which
showed that the level of harmonization of the pesticide MRLs of five major economies with Codex MRLs was
very low for particular commodities, including rice.
Rice is a staple food for billions of people globally and plays a significant role in food security, in particular in many
developing regions. Developing countries account for more than 96 percent of global rice production and a similar
share of world rice consumption. At the same time, eight out of the ten major rice exporters are developing
countries, accounting for almost three quarters of global rice exports.
The publication explores the harmonization of national standards with Codex pesticide MRLs from three
different angles. It assesses the level of harmonization of pesticide MRLs among the main rice producing and
trading countries, explores the possible effects on trade, and investigates the reasons behind differing levels of
harmonization. We hope that it will offer valuable insights for decision-makers and other stakeholders involved
in setting standards and designing food policy at national and international levels. Their role is important for both
shaping the international MRL standard-setting process under Codex and enhancing harmonization with Codex
MRLs at the national level.
Markus Lipp Boubaker Ben-Belhassen

Head of the Food Safety and Quality Unit, FAO Director of the Markets and Trade Division, FAO
v
Acknowledgements
This publication was prepared by the Food Safety and Quality Unit of the Food and Agriculture Organization of the
United Nations (FAO) in collaboration with the Markets and Trade Division of FAO.
For the Food Safety and Quality Unit, the analysis and write-up were carried out by Vittorio Fattori, Sarah Nájera
Espinosa and Giovanna Sartori; Ian Dewhurst (toxicologist, JMPR expert) developed Part B of the study in
collaboration with the Food Safety and Quality Unit.
For the Markets and Trade Division, the analysis and write-up were carried out by Georgios Mermigkas and
Andrea Zimmermann. Cosimo Avesani and Shirley Mustafa provided support at different stages of the preparation
of the study.
The support and guidance of Markus Lipp, Head of the Food Safety and Quality Unit, FAO, and Boubaker
Ben-Belhassen, Director of the Markets and Trade Division, FAO, are gratefully acknowledged.
The publication benefitted from comments and suggestions provided by the national authorities of the countries
considered in the study and by Gracia Brisco (Codex Secretariat), Philippe Verger (former WHO JMPR Joint
Secretary), and Yang Yong Zhen (FAO JMPR Secretary).
The authors would also like to thank Christiane Wolf (WTO) and Josef Schmidhuber (FAO) for reviewing the
publication and Thomas Heckelei (University of Bonn) for reviewing the econometric model used in chapter three.
Thanks are also extended to Jonathan Robinson for copy-editing and Adriana Brunetti for the layout and cover
design.
Finally, special thanks go to Renata Clarke (former Head of the Food Safety and Quality Unit, FAO) who guided
the analysis that underpinned the present study and initiated the development of this publication.
vii
© FAO/Hoang Dinh Nam
Background and
structure of the study
For many years, trade issues related to differences

in nationally imposed regulatory limits for pesticide
PART A.
residues have been highlighted by different countries. The level of harmonization
Within the Codex Alimentarius Commission (CAC),
with Codex rice pesticide
an internationally recognized standard-setting body,
maximum residue limits
concerns were initially raised mainly by Latin American
countries about the loss of pesticide maximum
and its impact
residue limits (MRLs), due to the periodic review
on trade
policy applied by the Codex Committee on Pesticide Part A explores the harmonization of national pesticide
Residues (CCPR). More recently, issues related to MRLs with Codex MRLs for the main players in the
the compliance with Codex pesticide MRLs and its global rice market and its possible impact on trade
relevance for trade has been repeatedly raised in other in rice. Rice was chosen as a case study because
relevant fora, including the World Trade Organization it is an important staple food in many developing
(WTO). At the 11th WTO Ministerial Conference, held in countries and was one of the commodities for which
Buenos Aires on 8-13 December 2017, a joint statement low harmonization with Codex MRLs was identified
was signed by a number of ministers from both in the pre-study. The harmonization of national
developed and developing countries, supporting actions pesticide MRLs with Codex pesticide MRLs was
“to increase the capacity and efficiency of Codex in assessed using publicly available information and
setting international standards on pesticide MRLs; to direct communications with the governments of the
improve transparency and predictability in Members’ countries included in the analyses.
setting of national MRLs; and to achieve greater
harmonization across national and regional MRLs’’. The impact of differing levels of harmonization with
Codex MRLs on trade in rice was analysed for a
In 2017, The Food and Agriculture Organization of the sample of 17 countries/markets. The results of the
United Nations (FAO) carried out a pre-study, on the analysis were peer-reviewed.
basis of publicly available information, in a systematic
effort to assess the level of harmonization among
countries and Codex MRLs on food. It was established
that countries tend to apply stricter pesticide MRLs
PART B.
than those recommended by Codex, especially in The reasons behind
the categories ‘Cereals’ and ‘Herbs & Spices’ , and in differing levels of
particular for commodities such as rice, chilli pepper harmonization
and spices. with Codex maximum
The observed low level of harmonization with Codex residue limits
MRLs triggered a series of questions and highlighted
Part B investigates the differences in risk assessment
the need to understand better the different dimensions
procedures and risk management policies that may
of this complex issue. To investigate the nature of the
lead to divergent MRLs. Data for five countries/region
problem better, the current study was designed to
were analysed in this part of the study based on
analyse two main areas:
availability and public accessibility to risk assessments
and risk management documents.
ix
Abbreviations and acronyms
ADI Acceptable daily intake JECFA Joint FAO/WHO Expert Committee on
AfCFTA African Continental Free Trade Area Food Additives
AFSA African Food Safety Agency JMPR FAO/WHO Joint Meeting on Pesticide
AoF All other Foods Residues
APEC Asia-Pacific Economic Cooperation LDC Least Developed Countries
APVMA Australian Pesticides and Veterinary LIFDC Low-Income Food-Deficit Countries
Medicines Authority LOQ Limit of quantification
ARfD Acute reference dose MRL(s) Maximum Residue Limit(s)
AVEs Ad Valorem Equivalents NAFTA The North American Free Trade
BMD(L) Benchmark Dose (Level) Agreement
CAC Codex Alimentarius Commission NOAEL No-Observed-Adverse-Effect Level
CIFOCOss FAO/WHO Chronic Individual Food NTMs Non-Tariff-Measures
Consumption database OECD Organization for Economic Co-operation
CCPR Codex Committee on Pesticide Residues and Development
CSAF Chemical-specific adjustment factor OLS Ordinary Least Squares
EFSA European Food Safety Authority PCPA Pest Control Products Act
FAO Food and Agriculture Organization of the PMRA Canada’s Pest Management Regulatory
United Nations Agency
FAOSTAT FAO Corporate Statistical Database POD Point of departure
FQPA Food Quality Protection Act PRIMO Pesticide residue intake model
GAP(s) Good Agricultural Practice(s) SDGs Sustainable Development Goals
GECDE Global estimate of chronic dietary SPS Sanitary and Phytosanitary Measures
exposure STCs Specific Trade Concerns
GEMS Global Environmental Monitoring System STDF Standards and Trade Development Facility
GDP Gross Domestic Product STMR Supervised trial median residue
HBGVs Health-based guidance values TBT Technical Barriers to Trade
HS Harmonized Commodity Description and TRR Total Radioactive Residue
Coding System TTC Threshold of Toxicological Concern
IEDI International Estimate of Daily Intake USA United States of America
IESTI International Estimated Short-term Intake US EPA United States Environmental Protection
IMR Inverse Mill’s Ratio Agency
ITC International Trade Centre WHO World Health Organization
WTO World Trade Organization
Terminology
The definitions of some terms that are used for the Corresponding national MRL: an MRL for the same
purpose of this study are reported below. combination pesticide/commodity exists at national
level, regardless of its value (does not imply alignment).
Alignment with a Codex MRL: a national MRL is
aligned with the Codex MRL when it has the same Harmonization with Codex MRLs: it is used as a
value as the Codex MRL. synonym of alignment with Codex MRLs.
Corresponding Codex MRL: an MRL for the MRL higher than Codex = less conservative/less strict
than Codex.
same combination pesticide/commodity exists at
Codex level, regardless of its value (does not imply MRL lower than Codex = more conservative/stricter
alignment). than Codex.
xi
Executive summary
Through the joint FAO/WHO Codex Alimentarius below 25 percent. This is generally due to countries
Commission, governments establish science-based not adopting Codex MRLs and not deferring to them
food standards. Codex standards aim at protecting when national MRLs do not exist. Differences in
consumer health and ensuring fair practices in commodity classification also represent an important
international food trade. Among food standards, obstacle towards achieving better harmonization.
pesticide MRLs are the maximum concentration Overall, great transparency was observed in relation to
of a pesticide residue to be legally permitted in public availability of national MRLs. Conversely, limited
food commodities and animal feeds. Codex MRLs information was publicly available for policies related to
are based on solid, independent scientific advice MRL establishment and enforcement.
jointly provided by FAO and WHO Scientific Advice
Programme and are established following an inclusive The effects of different levels of harmonization of
and transparent consultative process that ensures pesticide MRLs on trade in rice are explored for a sub-
that MRLs are set at the appropriate level to protect sample of 17 economies, where data were available
health and facilitate trade. Nonetheless, despite long- The economic analysis, conducted using a gravity
standing efforts towards international harmonization model, found that MRLs can affect trade in two ways.
of allowable thresholds for pesticide residues in MRLs stricter than Codex in the importing country are
foods, differences in national implementation of MRLs associated with relatively more rice imports, possibly
continue to exist, raising questions with regard to reflecting high consumption rates combined with strong
their impact on trade. This publication assesses the consumer food safety awareness in those countries. At
international harmonization of national MRLs with the same time, if MRLs on the importer side are stricter
Codex pesticide MRLs from different angles, using rice than on the exporter side, this may lead to additional
as a case study. costs for exporters in order to comply with the
importing country’s applied standards and dampen their
Part A of the publication examines the extent of exports to the markets with stricter MRLs. These higher
harmonization of national MRLs with Codex rice costs may impede exports by developing economies to
pesticide MRLs in 19 major rice producing and trading the countries with stricter regulation.
economies and investigates its impact on trade. To
define the level of harmonization, the analysis looks at Part B of the publication explores the reasons behind
the rate of adoption of Codex MRLs at national level, various levels of harmonization, investigating different
at the different MRL enforcement policies adopted aspects of risk assessment procedures and risk
by countries and at factors influencing harmonization management policies that may lead to divergent
levels, such as food classification. The analysis also MRLs. This part of the study focuses only on Australia,
considers the level of transparency in relation to Canada, the European Union, Japan and the United
pesticide MRL-related processes and policies. While States of America and is based on publicly available
it was found that many pesticide MRLs registered risk assessments and risk management documents.
at national level do not have corresponding Codex Considerable variation has emerged in how countries
MRLs, the analysis also revealed that the majority of are aligned with the FAO/WHO Joint Meeting on
Codex rice MRLs are not adopted at national level. The Pesticide Residues (JMPR)/Codex process for the
level of harmonization with Codex rice MRLs varies development and establishment of pesticide MRLs.
greatly across countries and regions. In general, most In general, many of the observed differences in
of the developing countries analysed are found to risk assessments do not seem to have a significant
rely strongly on Codex MRLs, showing high levels of impact on the overall outcome of the pesticide safety
harmonization with Codex. For the rest of the sample, evaluation. Some of the major differences in MRLs
alignment with Codex tends to be quite low, usually and residue definitions are due to the consideration
xii
of different data for the various countries/region and The study shows that any changes towards further
to inconsistency among the commodity descriptions international alignment of pesticide MRLs for rice to
in different countries. Automatic harmonization with facilitate trade will have to find a balance between
Codex MRLs is not the norm because such practice is the demand for strict food safety regulation on the
not embedded in national legislations. For MRLs not importer side and the additional costs incurred on the
established at national level, the default practice is, in exporter side.
general, to set a default value, usually at the limit of
quantification, or not to establish any tolerance level or Different steps could be taken by countries towards
MRL. improving harmonization with Codex standards.
For example, Codex MRLs could be taken into
While Codex MRLs continue to be an important point consideration in the absence of national MRLs. For
of reference during national policy setting processes, the optimal functioning of the Codex MRL standard-
the way in which they are considered varies among setting process, it would be important that countries
countries. Many developing countries continue to rely proactively notify whenever they have reservations
strongly on Codex MRLs when setting their national and are not in a position to adopt a newly established
MRLs, or in the absence of national MRLs, while other Codex MRL. Attention should also be given to
countries seem to use Codex MRLs as a reference developing country needs for a better and more active
when carrying out their own risk assessments but do participation in the Codex standard-setting process.
not necessarily harmonize with them. Findings of this analysis could be used to stimulate an
international dialogue to improve harmonization.
xiii
© Mgg Vitchakorn
PART A
Level of harmonization
of rice pesticide maximum
residue limits with Codex
and impact on trade
PART A Level of harmonization
of rice pesticide maximum
residue limits with Codex and
impact on trade
The value of agricultural and food exports grew almost Trade and SPS measures are closely related. While
threefold, in nominal terms, over the past decade, governments apply food standards, for example, to
reaching USD 1.8 trillion in 20181, with exports of ensure that food is safe and meets quality and labelling
emerging economies and developing countries requirements, to access international markets and to
growing much faster than those of developed be able to trade internationally, producers must be able
economies. The significant role of trade as an enabler to meet the food standards of their trading partners.
for sustainable development is acknowledged in As such, when countries apply different national
the Sustainable Development Goals (SDG targets requirements and specifications, trading food across
17.10, 17.11 and 17.12). Enabling agricultural and borders may become difficult (FAO and WTO, 2017).
food producers in developing countries to access
international markets is crucial for the economic In this context, the WTO Agreement on the Application
growth of regions and nations (UN-DESA, 2018). of Sanitary and Phytosanitary Measures (SPS
Agreement) sets out the framework for the application
While the reduction in tariffs through bilateral and of international standards by governments to ensure
regional agreements and the result of the agreement the safety of internationally traded food products.
establishing the WTO in 1995 contributed to the The SPS Agreement states that no member should
expansion of global trade, concerns have increasingly be prohibited from adopting or enforcing measures
grown about the impact of Non-Tariff-Measures (NTMs) necessary for protecting human, animal or plant
on agricultural and food exports. NTMs in agrifood life or health. However, these measures should
markets are policy measures, other than ordinary not be employed in a way that would “arbitrarily or
customs tariffs, that can affect international trade by unjustifiably discriminate between Members where
changing quantities traded or prices, or both (MAST, identical or similar conditions prevail […] or constitute
2008). NTMs are used by governments with the aim a disguised restriction on international trade.” The
to ensure food safety and to protect animal and plant Agreement encourages governments to harmonize
health (Sanitary and Phytosanitary measures – SPS) or base their national food safety measures on the
or to regulate technical characteristics of the products, international standards developed by the joint FAO/
such as marketing standards (Technical Barriers to Trade WHO Codex Alimentarius Commission (WTO,
– TBT). NTMs are particularly important for agricultural 1995)2. The SPS Agreement seeks to strike a balance
trade. Figure 1 shows that the Ad Valorem Equivalents between Members` rights to regulate legitimate
(AVEs) of NTMs are much higher for agriculture than objectives, such as food safety, while avoiding that
for other economic sectors, while for food products, in such regulations be used as unnecessary barriers to
particular, AVEs of NTMs are on average almost three international trade. It does so inter alia by strongly
times higher than normal tariffs. encouraging harmonization with Codex standards.
SPS measures are the most important NTMs The Codex Alimentarius is a collection of international
concerning agricultural products, covering almost food standards, guidelines and codes of practice that
20 percent of the world’s total merchandise imports, have the dual objective of protecting consumer health
with an average of six measures per product in each
country, more than any other NTM category (UNCTAD
and the World Bank, 2018). 2 The SPS Agreement allows WTO Members to deviate from
international standards where they wish to achieve a higher level
of health protection than that reflected in international standards,
so long as their measures are based on an appropriate assessment
1 Information retrieved from WTO data in March 2020. of risks and the level of health protection sought is consistent.
3
Figure 1: Tariffs and AVEs of NTMs, by economic sector
Tariffs AVEs of NTMs
30
25
20
AVE percent
15
10
0
Tobacco, Beverages
Food Products
Vegetable Products
Oils and Fats
Animal Products
Oil , Gas, Coal
Mining and Metal Ores
Rubber/Plastics
Precision Instruments
Transport Equipment
Basic Metals
Tanning
Non-Metallic Mineral
Petroleum Products
Chemicals
Metal Products
Textiles
Office Machinery
Paper Prod, Publishing
Comunicatin Equip.
Wood Prod. Furniture
Machinery Various
Motor Vehicles
Apparel
Electrical Machinery
Agriculture Industrial
Source: UNCTAD and the World Bank, 2018
and facilitating trade through harmonized regulations an independent expert scientific panel, administered
(FAO and WHO, 2019). jointly by FAO and the WHO4 (FAO and WHO, 2019).
Codex standards are based on sound science provided This publication addresses the issue of harmonization
by independent international risk assessments of national pesticide MRLs with Codex pesticide MRLs
carried out by the Joint FAO/WHO Scientific Advice from different angles, by taking rice as a case study.
Programme and cover the whole food safety Part A identifies the level of harmonization in the main
spectrum, from microbiological to chemical issues, rice producing and trading countries and explores the
including, among many other provisions, pesticide possible effects on trade, while Part B investigates the
MRLs in food . 3
reasons behind differing levels of harmonization.
Codex MRLs are established by the CCPR, through an The broader objective is to offer insights for decision-
open, participatory and transparent process that allows makers involved in setting standards and designing
all member states to participate in the development food policy at national and international level on the
of the standards. Once approved by the CCPR, MRLs significance of harmonization of pesticide MRLs,
are forwarded for adoption by the CAC. These two but also on areas of improvements for the standard-
bodies carry out the risk management functions of the setting international process and the ways that these
MRL-setting process. Their decisions are supported by standards are implemented at the national level.
the work of the FAO/WHO Joint Meeting on Pesticide
Residues (JMPR), which is responsible for conducting
the risk assessment and estimating MRLs. JMPR is
3 Codex Pesticide Maximum Residue Limits (MRLs) are the 4 JMPR evaluates the toxicology of pesticides and estimates health-
maximum concentration of a pesticide residue (expressed as mg/ based guidance values (HBGVs), including acceptable daily intake
kg), recommended by the CAC to be legally permitted in or on food (ADI) and acute reference doses (ARfD). Based on this, and with
commodities and animal feeds. MRLs are based on GAP data and data on registered use patterns, fate of residues, animal and plant
foods derived from commodities that comply with the respective metabolism, analytical methodology and residue data derived from
MRLs are intended to be toxicologically acceptable (definition from supervised residue trials, JMPR proposes residue definitions and
the Codex Procedural Manual, 27th ed., 2019). maximum residue limits for the pesticides in food and feed.
4
PART A
1 Global rice market

Rice is one of the most important food staples, playing Moving to international trade, although world trade in
an essential role in the food security of a large part rice has grown significantly over time, rice remains
of the global population. Rice is produced throughout a relatively thinly traded commodity, compared to
the world, although most of the leading rice producers other grains that are also used as feedstock for animal
are located in Asia (Table 1). Although rice production fodder and the food industry.6 Rice volumes traded
has become increasingly important in other regions, around the world represented 9.1 percent of global
namely in Africa, Asia remains the global rice hub, production in 2017-2019, compared with 23 percent for
accounting for close to 90 percent of world production. wheat and 14.2 percent for maize (FAO Markets and
Trade Division, CCBS database). However, it is worth
Rice is predominantly a food crop, with great
mentioning that rice plays a more significant role in the
significance in human diets, particularly in the Far East,
exports of developing countries than wheat or maize.
western Africa, Latin America and Near East (Figure 2).
According to calculations based on UN Comtrade
The global rice market is dominated by developing data, rice exports of the developing countries reached
countries , where rice provides a livelihood for many
5
USD 20.5 bn on average for the 2017-2019 period. The
poor farming households. Put together, developing respective amounts for maize and wheat were USD
countries account for more than 96 percent of global 17.3 bn and USD 13.5 bn.
rice production and a similar share of world rice intake
On the import side, the global rice market exhibits
(Table 2).
limited concentration, with the top 20 importers
Figure 2: Rice available for consumption (kcal/capita/day), average 2015-2017
Source: Based on FAOSTAT data
6 A non-negligible share of rice trade is also conducted informally,

5 For this purpose, developing countries are classified according to through undocumented cross-border exchanges. These flows are
the United Nations Statistics Division (UNSD) M49 classification. not accounted for in the official trade data reported in this study.
5
Table 1: Top ten rice producers in the world (in metric tonnes – paddy)
Total average
Country
(2016–2018)
World 767 904 795
China 211 966 333
India 168 260 000
Indonesia 81 180 120
Bangladesh 53 672 728
Viet Nam 43 307 314
Thailand 30 064 074
Myanmar 25 571 946
Philippines 18 656 562
Brazil 11 612 049
Pakistan 10 750 577
Source: Own calculation based on FAOSTAT database
Table 2: World rice production and domestic use (milled equivalent), average 2017-2019
PRODUCTION TOTAL AVERAGE Global share

Rice, milled equivalent (thousands of tonnes) (%)
World 502 060 100.0
Developed regions 16 473 3.3
Developing regions 485 587 96.7
Low-Income Food-Deficit Countries (LIFDC) 199 869 39.8
Least Developed Countries (LDC) 76 013 15.1
TOTAL DOMESTIC USE TOTAL AVERAGE Global share

Includes use of rice as food, feed and other uses (thousands of tonnes) (%)
(namely seed, non-food industrial uses and post-harvest
losses), milled equivalent
World 499 754 100.0
Developed regions 17 948 3.6
Developing regions 481 807 96.4
Low-Income Food-Deficit Countries (LIFDC) 192 855 38.6
Least Developed Countries (LDC) 83 480 16.7
Source: FAO Markets and Trade Division, Country Cereal Balance Sheet (CCBS) database
accounting for 58.7 percent of the total value of In contrast to imports, on the export side there is
imports in 2017-2019 (Table 3). China, Islamic Republic strong concentration, with the top ten exporters
of Iran, Saudi Arabia and the European Union (intra- 7 accounting for 85 percent of total exports in
European Union trade was excluded) were the top 2017–2019. India, Thailand and Viet Nam were the top
importers during that period in value terms. exporters during that period, accounting for more than
59 percent of the global value of rice exports (Table 4).
7 The analysis is based on data until 2019. All data referring to the
European Union therefore includes the European Union and the
United Kingdom of Great Britain and Northern Ireland.
6
Global rice market 1 PART A
Table 3: Top 20 rice importers in the world, average 2017-2019
USD thousand Share

(%)
World 24 014 054 100.0

China 1 560 409 6.5
European Union 1 501 362 6.3
Iran (Islamic Republic of) 1 490 884 6.2
Saudi Arabia 1 250 266 5.2
United States of America 926 850 3.9
Benin 838 263 3.5
Iraq 711 750 3.0
Philippines 675 016 2.8
United Arab Emirates 620 267 2.6
Côte d’Ivoire 507 012 2.1
South Africa 498 415 2.1
Japan 446 925 1.9
Indonesia 420 207 1.7
Senegal 415 707 1.7
Bangladesh 415 394 1.7
Malaysia 401 454 1.7
Mexico 377 120 1.6
Canada 352 201 1.5
Yemen 329 669 1.4
Ghana 305 429 1.3
Note: Intra-European Union trade is excluded

Source: Own calculations on the basis of data from UN Comtrade database
Table 4: Top ten rice exporters in the world, average 2017-2019
USD thousand Share

(%)
World 24 292 576 100.0
India 7 183 787 29.6
Thailand 4 976 079 20.5
Viet Nam 2 233 323 9.2
United States of America 1 762 037 7.3
Pakistan 1 658 662 6.8
Myanmar 847 689 3.5
China 843 846 3.5
Uruguay 416 769 1.7
United Arab Emirates 395 073 1.6
Cambodia 360 022 1.5
Source: Own calculations on the basis of data from UN Comtrade database
7
© Sebastian Liste/NOOR for FAO
PART A
2 Pesticides used on rice and

harmonization with Codex
maximum residue limits
This chapter examines the extent of harmonization 2.1. Data sources and
of national rice pesticide MRLs with Codex MRLs in methodology
19 selected countries/region8. To define the level of
Codex pesticide MRLs
harmonization, the analysis looks first at how many
Codex MRLs for pesticide residues in specific food
Codex MRLs have a corresponding MRL at national
commodities are adopted annually by the CAC. Every
level; then examines the different MRL enforcement
year the CCPR, on the basis of scientific expert
policies adopted by the covered countries/region and
advice provided by JMPR, prepares a list of proposed
explores the factors influencing harmonization levels,
MRLs and forwards it to the CAC for adoption. For the
such as food classification. The analysis also considers
purpose of this study, values for Codex pesticide MRLs
the level of transparency in relation to pesticide MRL
adopted between 1971 and 2018 were collected from
processes and policies.
the CCPR annual reports (FAO and WHO, 1971–2018a),
while confirmation of their adoption was verified in the
Key messages:
CAC annual reports (FAO and WHO, 1971–2018b).
JJ The level of harmonization with Codex rice MRLs
varies greatly across countries and regions. Many of
National pesticide MRLs
the developing countries analysed strongly rely on
National rice MRLs were taken from national resources
Codex MRLs and the level of harmonization is very
that were publicly available online (online databases
high or even total in some cases. For the rest of the
countries analysed, alignment with Codex tends to or official documents/regulations) as of the end of
be quite low, usually below 25 percent. October 2019. When no resources could be found
online, national authorities were contacted to seek their
JJ Many Codex MRLs do not have corresponding
support in identifying the relevant documents. For all
MRLs at national level, and at the same time many
countries/region, national authorities were contacted to
MRLs registered at national level do not have
corresponding Codex MRLs. confirm the validity of the sources used in the study. A
complete list of references to these sources is reported
JJ Deferral to Codex MRLs when national MRLs do
in Annex A-1. The Bryant Christie Global pesticide MRL
not exist is not a common practice – default limits
database was also considered as a cross check (Bryant
are usually applied.
Christie Inc., 2019).
JJ Differencesin commodity classification represent
an important obstacle towards achieving better
The MRL values have been taken as they appear in
harmonization.
the national legislation (and/or national database when
JJ While great transparency was observed in relation available). Residue definitions were not verified when
to public availability of national MRLs, limited comparing national MRLs with Codex MRLs. The issue
information was available for policies related to of residue definitions is investigated in more detail in
MRL establishment and enforcement.
part B of the study (see Part B, chapter 1.1).
Compilation and data analysis

Codex pesticide MRLs for rice adopted between 1971
and 2018 were set as the reference values. National
MRLs units were verified to be consistent with Codex
8 Australia, Bangladesh, Brazil, Cambodia, Canada, China, European MRLs units (i.e. mg/kg). Subsequently values from
Union, India, Indonesia, Islamic Republic of Iran, Japan, Myanmar, national MRLs were compared with Codex MRLs,
Pakistan, Philippines, Saudi Arabia, Thailand, United Arab Emirates,
United States of America and Viet Nam. and classified as (i) aligned with; (ii) higher than; or
9
(iii) lower than, the corresponding Codex MRL value. 2.2. Codex maximum residue
A database was created for the analysis of the data. limits for rice
Food classification There were 82 rice pesticide MRLs adopted by Codex

between 1971 and 2018.10, 11, 12, 13 These refer to five
Box 1. Major species of rice
different Codex commodities: three of them denote
There are only two major species of cultivated rice: Oryza
different levels of processing of common rice (rice; rice
sativa (OS), or Asian rice, and Oryza glaberrima (OG), or
African rice. The rice varieties grown across the world are husked and rice polished), a fourth one is a different
overwhelmingly OS, while cultivation of OG is confined to rice variety (wild rice), and the last one is a commodity
Africa. Even in that region, however, OG varieties are fast group (cereal grains) that includes all cereals, including
being replaced by OS, which produces much higher yields rice and wild rice14. The majority of Codex rice MRLs
than OG, a characteristic that has prevailed over the special
are set for “rice (with husk)”, as shown in Table 5,
advantages afforded by OG, in the form of weed tolerance,
followed by rice husked and rice polished.15
pest resistance or fast growth (FAO, 2006).
Codex sets rice MRLs for Oryza sativa for three 2.3. National maximum residue
different processing levels: rice (with husk); rice limits for rice
husked (brown rice); and rice polished. Rice (with
At national level the total number of MRLs established
husk) is also regulated under “cereal grains”9 when
for rice varies widely from one country to another, as
group MRLs are established. One Codex MRL is set
presented in Figure 3, where the highest number is
for wild rice (Zizania aquatica).
observed in the European Union (486 rice MRLs), and
the lowest number is found in Cambodia (11 rice MRLs).
To match national food commodity descriptions to
the Codex food classification, reference was made to In general, very few national rice MRLs have a
Codex-relevant documents: i) the draft revision of the corresponding Codex rice MRL, for the same
Codex Classification of foods and animal feeds (FAO combinations of pesticide/commodity, as presented in
and WHO, 2006); ii) the Codex Draft and Proposed Figure 4.
Draft Revision of the Classification of Food and Feed
In about half of the countries/region analysed, less than
(Appendix VIII to XII) (FAO and WHO, 2017); and iii) the
half of national rice MRLs have a corresponding Codex
Codex Standard for Rice – CXS 198-1995 (FAO and
rice MRL (Australia, Brazil, Cambodia, China, European
WHO, 1995).
Union, India, Islamic Republic of Iran, Japan, Philippines,
and United States of America). In a handful of other
Rice classifications from the countries/region analysed
countries (Canada, Indonesia, Saudi Arabia, Thailand and
were taken from their national legislations when
Viet Nam) most national MRLs do have corresponding
available and confirmed with the countries/region to
Codex MRLs, while in four countries (Bangladesh,
match them properly with Codex rice classification.
Myanmar, Pakistan and United Arab Emirates) all
Some countries (i.e. Australia, Canada) also set MRLs
national MRLs have a corresponding Codex MRLs.
for generic groups of food (such as “All other foods”).
The high (or sometimes complete) correspondence
When available, and in the absence of specific rice
MRLs, they have been used as a match for any type
of rice for which a Codex MRL existed. A detailed 13 The Codex MRLs analysed in the study can be retrieved on the
matching of rice classification across Codex and Codex Pesticides Database, at (please refer only to MRLs adopted
between 1971 and 2018):
countries/region is reported in Annex A-2. Rice: http://www.fao.org/fao-who-codexalimentarius/codex-texts/
dbs/pestres/commodities-detail/en/?lang=en&c_id=158
Rice husked: http://www.fao.org/fao-who-codexalimentarius/codex-
texts/dbs/pestres/commodities-detail/en/?lang=en&c_id=78
9 Cereal grains comprise rice (with husk) of Oryza sativa and Oryza
glaberrima, and also wild rice. Rice polished: http://www.fao.org/fao-who-codexalimentarius/
codex-texts/dbs/pestres/commodities-detail/en/?lang=en&c_id=75
10 For a more detailed description of each Codex rice commodity,
please refer to Annex A-2. Cereal grains: http://www.fao.org/fao-who-codexalimentarius/codex-
texts/dbs/pestres/commodities-detail/en/?lang=en&c_id=164
11 Although Codex provides MRLs for both food commodities and
feed, this study only focuses on food commodities. Feed and rice Wild rice http://www.fao.org/fao-who-codexalimentarius/codex-
by-product pesticide MRLs were excluded from the analysis (these texts/dbs/pestres/commodities-detail/en/?lang=en&c_id=653
are: rice bran, rice hulls, and rice straw and fodder). 14 Codex MRLs can be set for single commodities or for a group of
12 Codex MRLs adopted in 2019 were excluded from the analysis based commodities (group MRL).
on the consideration that it might take some time for countries to 15 For the matching with HS codes and share of trade, please refer to
receive and eventually adopt new Codex MRLs in their regulations. Annex A-3.
10
2 PART A
Pesticides used on rice
and harmonization with
Codex MRLs
Table 5: Codex MRLs for rice
Codex name Codex code Total # of MRLs
Rice (defined as “rice with husks that remain attached to kernels even after GC 0649 25
threshing: kernels with husks”)
Rice husked CM 0649 18
Rice polished CM 1205 13
Wild rice GC 0655 1
Cereal grains (= rice; wild rice) GC 0080 25
Total: 82
Figure 3: Total number of national MRLs established for rice in the 19 countries/region analysed
National MRLs Codex MRLs
500 486
450
Total number of rice* MRLs
400
350 317
300 288
263
250
200
150 119
92 99
100 82 82 82 82 82
65 62 65
37 50
50 29 29
11
0
ng lia
sh
il
da
on
Pa ar
ilip n
es
Em d
Am s
Vi ica
am
s
of
az
di
in
di
si
bi
of ate
RL
pa
Ph ista
n
nm
Ba tra
de
na
Sa pin
ni
ab ila
ne
er
bo
Ch
In
ra
Br
N
lic
Ja
M
U
ir
Ar ha
s
Ca
iA
k
la
ya
do
m
ub
et
Au
an
x
T
ud
M
Ca
de
In
p
pe
Re
Co
ro
es
ic
Eu
at
m
d
St
te
la
(Is
ni
d
U
te
n
ni
Ira
* Rice refers to any type of rice regulated at national level (rice with husk, rice husked, rice polished and cereal grains).
between national and Codex MRLs is usually explained to develop new MRLs is another reason that could play
by national policies of deferring to Codex MRLs: either a role here. Even more so when pesticides have been
automatically, for all MRLs established at national assessed at Codex level but not for all commodities of
level, or only in the absence of national MRLs (national interest, such as for rice in this case. The absence of
policies are further analysed in section 2.4.3). MRLs for use in traded products is a significant concern
that has been brought to the attention of the CAC
To gain a better idea of the type of pesticides not recently. With food safety being increasingly recognized
regulated at Codex level (and to verify if this could be as a priority issue in many countries, the request for
due to toxicity concerns), pesticides that did not have scientific advice has also increased. However, despite
a Codex MRL for rice were screened to check if they the significant efforts that are being made by FAO/WHO
had Codex MRLs for other commodities. It was found to streamline procedures and to manage the process of
that on average around one third of these pesticides scientific review as efficiently as possible, the number of
did have Codex MRLs for other commodities, for all requests for evaluation by the JMPR of new compounds
countries/region analysed (see Figure 5). and new uses (as well as for periodic re-evaluations of
existing MRLs) far exceeds the current capacity.
Non-existence of MRLs for these pesticides at Codex
level may hint at toxicity concerns, but this was not There could be many other reasons behind the low
explored further. Limited capacities/resources of Codex number of Codex rice MRLs compared with national
11
Figure 4: Share of national rice MRLs for which a corresponding Codex MRL exists, in the 19
countries/region analysed
Codex MRLs
500 467
450
Total number of rice* MRLs
400
350
260
300 271
207
250
200
150
102
100 66 60
82 31 82 82 6 82 1
56 14 57 35 56 64
50 6 18 12 39
26 23 19 17 34 15 17
5 11 17
0
ng ia
sh
il
da
on
ia
ia
an
Pa ar
an
es
Th ia
Em d
a
am
of
az
di
in
te
ic
n
Ba tral
es
b
m
de
na
Sa pin
st
ni
ila
er
bo
Ch
In
ra
Br
ira
N
lic
Ja
n
on
U
ki
a
s
Ca
iA
Am
la
ya
et
m
ub
ilip
Au
ud
M
Vi
Ca
ea
In
Ph
of
Re
b
p
ra
ro
es
ic
A
Eu
at
am
St
te
l
(Is
ni
d
U
te
n
ni
Ira
U
* Rice refers to any type of rice regulated at national level (rice with husk, rice husked, rice polished and cereal grains).
Figure 5: Share of pesticides, with national rice MRLs but not Codex MRLs, for which a Codex MRL exists
for commodities other than rice
Pesticides for which Codex MRLs exist Pesticides for which

for commodities other than rice no Codex MRLs exist
100%
Percentage of total number of pesticides
90%
80%
3 4 1
70% 19
162 66 70 35 18
60% 3 14 159 429 239 53
50% 3 4 1
40%
80 10
30%
25 1 151 26 13 7 18
20% 4 48
59
10%
0%
lia
il
da
on
an
es
ia
nd
am
of
az
di
in
di
si
ic
ab
ra
na
n
ni
ila
ne
er
bo
Ch
In
Br
N
ic
Ja
pi
st
Ar
U
a
Ca
m
bl
do
m
ilip
et
Th
Au
an
pu
A
i
ud
Ca
Vi
In
Ph
of
e
Re
Sa
op
es
ic
r
Eu
at
m
St
la
(Is
d
te
n
ni
Ira
Note 1: Bangladesh, Myanmar, Pakistan and the United Arab Emirates were excluded from this analysis as all their national MRLs have
corresponding Codex MRLs.
Note 2: Labels show the total number of MRLs.
12
2 PART A
Codex MRLs
rice MRLs, nevertheless further analysis of this was 2.4.2. Rice classification and processing level of rice
outside the scope of the current study. The classification of rice was very heterogeneous
across the countries/region analysed and Codex. During
2.4. Codex maximum residue the analysis difficulties emerged regarding how to
match national MRLs to codex MRLs (see Annex A-2).
limits (MRLs): existence of
Sometimes the same name referred to different stages
corresponding national
of processing of rice in different countries/region
MRLs and their level of
(e.g. “rice” in Codex refers to rice with husk; in the
harmonization
European Union and in Japan it refers to rice husked;
2.4.1. Codex MRLs and existence of corresponding in Canada and Brazil it refers to both rice husked and
MRLs at national level unhusked). Differences in rice classifications can also
The level of alignment with Codex MRLs varies greatly be determined by national consumption patterns (i.e.
the type of rice most commonly consumed in the
across the countries/region analysed, as shown in
country and the way rice is distributed and sampled),
Figure 6. National MRLs that are aligned with Codex
which may be shaped by consumer preferences or
MRLs are marked in dark green. Higher and lower values
habits. For example, in Japan, rice is kept in husked
than Codex are reported light green and red, respectively.
form (brown) until the near end of the marketing
Figure 6 also indicates the number of missing MRLs for
chain. That is, rice is polished just before reaching the
each country as compared to total Codex MRLs (grey
consumer as a way to ensure that the product remains
section of the bars). For most countries/region (11 out
fresh and of higher quality.
of 19), more than half of Codex rice MRLs do not have a
corresponding MRL at national level.
To study if these differences could affect the number
of national MRLs and their level of harmonization with
There might be many reasons behind the low number
Codex MRLs, the 82 Codex rice MRLs were classified
of Codex MRLs for which corresponding MRLs exist
into three main categories, reflecting different
at national level. For example, some economies may processing levels of rice16, and analysed by category:
not have procedures for establishing import MRLs, or
may have policies that do not allow for consideration JJ U - Rice unprocessed: which includes rice (with husk);
of foreign Good Agricultural Practices (GAPs) when wild rice; and cereal grains (51 Codex MRLs)
there is a domestic GAP in place. There might also be
JJ H - Rice husked (18 Codex MRLs)
a tendency for countries to adopt Codex MRLs only
for export purposes and not for import. As explored JJ P - Rice polished (13 Codex MRLs)
in Part B of the study, missing national MRLs could
Figure 7 demonstrates that for some countries/region
partly be explained by the differences in the time
the number of national MRLs corresponding to Codex
of MRL adoption at Codex and at national level. For
MRLs, and their compliance with Codex MRLs, is
some markets, there may not be resources or routine
determined by these processing categories.
procedures in place to review national MRLs at the
time Codex adopts new ones. This difference in time
For the European Union and Japan, for example, the
may also impact the level of alignment with Codex
number of Codex rice MRLs for which corresponding
MRLs for markets that set MRLs prior to Codex
national MRLs have been established is very low,
MRL establishment. The time difference may entail covering 23 percent and 21 percent of all Codex
changes in the scientific data packages evaluated by rice MRLs, respectively (see Figure 6). However,
the different authorities (e.g. different GAPs reflecting this figure can be partially explained by the fact
different pests and diseases, different pesticide labels, that the establishment of MRLs for unprocessed
availability of different studies, etc.) leading to different rice (U) and polished rice (P) is almost inexistent
results, which could explain the low level of alignment. in these country and region. When looking only at
However, time difference may not be the only reason. husked rice (H), the number of Codex MRLs for
Rice classification seems also to play an important role
in terms of establishment of national MRLs and their
alignment with Codex MRLs, as explained below. 16 For more details on each Codex rice commodity, refer to
Annex A-2.
13
Figure 6: 82 Codex rice MRLs: existence of corresponding MRLs at national level and level of alignment
with Codex in the 19 countries/region analysed
Aligned with Codex Higher than Codex Lower than Codex Codex MRLs but no national MRLs
80
21 18
70 24 25 26
Total number of rice MRLs
60 48 43
51 51 1
50 13 63 62 65 1 64
67 69
29 82 77 82 82
40 8
1 11 64
30 61
11 54
14 16
20 36 16
11 8 33 5 2
10 4 2
10 11 4 6 6 4 5 16
15 2 12
4 1 9 5 6 5 8 6
0 2
ia
sh
il
da
on
ia
ar
ia
ia
an
nd
am
of
az
di
in
ne
te
ic
pa
l
b
es
nm
ra
de
na
st
ni
la
er
bo
Ch
In
ra
Br
ira
N
ic
Ja
pi
st
on
ai
U
ki
Ca
iA
Am
la
bl
ya
m
ilip
et
Em
Th
Au
Pa
ng
an
pu
Ca
ud
Vi
In
Ph
pe
of
Ba
Re
ab
Sa
ro
es
Ar
ic
Eu
at
am
St
te
l
(Is
ni
d
U
te
n
ni
Ira
U
Figure 7: Level of alignment with Codex rice MRLs by processing level categories: Rice unprocessed (U),
Rice husked (H), and Rice polished (P)17
Aligned with Codex Higher than Codex Lower than Codex Codex MRLs but no national MRLs
100%
5
3
Percentage of rice MRLs alignment
90% 13
80% 5
9 10
70% 19 11 8 5
13 10
60% 41 11
50 47 12
50% 13 3 13 51 18
4
40% 13 4 16
30% 5
5 4 1
20% 5 3 2 6
14 5 1 12
10% 1 1 4 2
1 1 2 1 3 2 1
0% 3
U H P U H P U H P U H P U H P U H P
Australia European India Iran Japan United States
Union (Islamic Republic of) of America
U: Rice unprocessed: including rice (with husk), wild rice and cereal grains
H: Rice husked
P: Rice polished
Note: Labels show the actual number of MRLs.
which corresponding national MRLs exist rises to As further described in Part B, to reduce the potential
100 percent and 83 percent, respectively. The same for confusion over multiple MRLs for different forms
is true for Australia and the United States, but in of a single pesticide/crop combination, consistency in
this case, unprocessed rice (U) is the type of rice for commodity classification across countries and Codex
which higher rates are reported of national MRLs should be improved where possible.
corresponding to Codex MRLs.
17 The countries that did not show marked changes were not
reported in the graph.
14
2 PART A
Codex MRLs
Table 6: Existence of national MRLs and enforcement of Codex MRLs in the absence of national MRLs
Procedure
Establishment
Enforcement procedure followed reported
Country/region of national
in the absence of national MRLs in nat.
MRLs
legislation
Thailand Yes Defer to Codex Yes 18
Brazil, Cambodia, Yes Defer to Codex No

Indonesia
Bangladesh, Myanmar No Defer to Codex No
Saudi Arabia Yes Defer to Codex first - then MRLs of the No

European Union or United States of America
United Arab Emirates No Defer to Codex first - then MRLs of the Yes 19
European Union, then default limit at: 0.01
ppm
Pakistan No Defer to the lowest MRL among Codex, No

European Union and United States of
America
European Union, Japan, Yes Apply a default limit at: 0.01 ppm Yes 20
India
Iran (Islamic Republic of) Yes Apply a default limit at: 0.05 ppm No
Canada Yes Apply a default limit at: Yes 21

0.1 ppm
Australia Yes Apply zero tolerance Yes 22
United States of America Yes The crop is considered adulterated and may Yes 23
be seized
Philippines Yes Apply zero tolerance No
China, Viet Nam Yes Not confirmed – it was assumed that in No

the absence of an official procedure, zero
tolerance applies
Source: Based on procedures outlined in national legislations of the countries/region analysed and/or confirmed by the national authorities
22 See: (FSANZ, 2016)

18 See Annex A-1: Thai Agricultural Standard TAS 9002-2016 point 3.4 23 See: (United States Environmental Protection Agency (EPA), 2018).
19 See Annex A-1: UAE.S MRL 1: 2017 “Maximum Residue Limits Pesticide Registration Manual, Chapter 11 - Tolerance Petitions,
(MRLs) for Pesticides in Agricultural and Food Products” – point 3 under “Tolerances and Exemptions from Tolerances”: “If residues
of a pesticide exceed the established tolerance, or no tolerance
20 Europe - See: (European Union, 2005) has been established, the crop is considered adulterated and may
Japan - See (Japan, Ministry of Health, Labour and Welfare, 2006) be seized by the U.S. Food and Drug Administration (FDA), the
India - See Annex A-1: The Gazette of India: Gazette Notification on U.S. Department of Agriculture (USDA), or a state enforcement
Food Safety and Standards (Contaminants, Toxins and Residues) - agency.” Most pesticides analysed for enforcement purposes can
Part III-Sec.4 (note) be quantified at LOQ of 0.01 mg/kg, hence in practice, the USA
21 See: (Canada, 2019) under DIVISION 15 Adulteration of Food enforcement outcomes are similar to those in countries/region that
B.15.002 apply default values at 0.01 mg/kg.
15
Figure 8: Alignment with the 82 rice Codex MRLs in the 19 countries/region analysed 24
Aligned with Codex Higher than Codex Lower than Codex
2 1 1 2
80 1 1
16 21 18
70
37
Total number of rice MRLs
11 38
60 48
53 58 54
50 64 60
71
40 82 79 8 81 82 80 80 82
30 31 61 64
55
14 7 23 16
20 36 5
4
10 15 17 17 5
14 14 11 12
6
0
ia
Ca azil
da
on
ar
Ph stan
Am s
am
of
di
in
di
si
ne
bi
te
ic
es
pa
an
l
nm
tra
na
ni
ne
er
bo
Ch
In
ra
Br
ira
N
ic
ad
Ja
pi
i l
U
ki
a
s
Ca
iA
bl
ya
do
m
ilip
et
Em
Th
Au
Pa
gl
an
pu
ud
M
Vi
n
In
of
pe
Ba
Re
b
Sa
ra
ro
es
ic
A
Eu
at
m
St
te
la
(Is
ni
d
U
te
n
ni
Ira
U
2.4.3. Codex MRLs missing at national level have neither the financial nor the technical means to
As mentioned earlier, the analysis revealed that there submit such an application.
are many Codex MRLs that are missing at the national
2.4.4. Harmonization with Codex MRLs
level (Figure 6, grey section of the bars). For most
Figure 8 reports the level of alignment with Codex
countries/region, more than half of Codex rice MRLs
MRLs after applying national policies for missing
do not have a corresponding MRL at the national level.
MRL values (e.g. deferral to Codex MRLs, LOQ,
It was found that in the absence of national MRLs,
or zero tolerance). National MRLs that are aligned
different countries/region follow different enforcement
with Codex MRLs are marked in dark green. Higher
procedures. Some countries automatically defer to
and lower values than Codex are reported in light
Codex MRLs, others use default limits, mostly at the
green and red, respectively. The level of alignment
limit of quantification (LOQ), or apply zero tolerance.
improves significantly for those countries that defer
Table 6 reports the different enforcement procedures
to Codex MRLs, while fewer changes are noted for
followed by each country.
the other countries/region. Overall, for more than half
of the countries analysed, the level of alignment with
The differences in the enforcement procedures
Codex is higher than 50 percent, with eight countries
followed in the absence of national MRLs can also
showing complete or nearly complete alignment. The
have implications for the ability of some countries, in
remaining seven countries/region present a much
particular developing countries, to export. For example,
lower level of alignment, usually below 25 percent.
as further discussed in Part B of the study, in Australia,
Canada, the European Union, Japan and the United
The countries presenting stronger alignment with
States of America when a national MRL is missing, an
Codex MRLs (dark green sections of the bars) are
application can be made to have an MRL established.
Bangladesh, Brazil, Cambodia, Indonesia, Myanmar,
However, the process can be complex and lengthy
Pakistan, Saudi Arabia, Thailand, the United Arab
(from a few months up to six years), depending on
Emirates and Viet Nam.
the country. This affects exporting partners negatively,
especially in developing countries where they might
24 All national values have been taken from national databases,

legislation or official documents. When national MRLs were
missing, the national procedures reported in chapter 1.3.3. were
applied (please refer to Table 6 for a more detailed overview of
national MRL adoption procedures).
16
2 PART A
Codex MRLs
Because most of the countries do not automatically Access to national MRLs

defer to Codex MRLs in the absence of national MRLs, National MRLs of most countries/region analysed were
differences in the MRL levels among trading partners publicly accessible online (see Annex A-1: Sources of
could create obstacles to trade. national pesticide MRLs for the 19 countries/region
analysed). Out of 19 countries/region, only three did
As further explored in chapter 3, stricter MRLs in not have publicly available information, because they
importing than in exporting markets may entail higher do not establish national MRLs: Bangladesh, Myanmar
costs for the exporters to comply with the importing and Pakistan. For most countries/region (11 out of 19)
market MRLs. the information was available in English. National MRL
databases were available only in the national language
2.5. Most used pesticides on in Brazil, China, the Islamic Republic of Iran, Thailand
rice in major rice producing and the United Arab Emirates. Some countries present
countries their legislation in online documents, others in online

databases with search functions, and some others
Numerous pesticides are used in rice production. in online databases without search functions. Most
Codex has 82 rice MRLs for 67 different pesticides. databases proved to be user-friendly and easy to
For each country/region analysed, the total number navigate.
of national MRLs ranged from 11 to 486. The analysis
MRLs-related national regulations or policies
did not characterize the pesticides in terms of type,
use and sales volumes. However, in an attempt to MRL-related national regulations or policies on
verify whether Codex MRLs cover the pesticides most procedures followed in the absence of national MRLs
used in developing countries, national authorities of were more difficult to find and access. Out of 19
ten developing countries were approached, through countries/region, nine did not have these procedures
the FAO country offices, to obtain an indication of the reported in their national legislations (See Table 6 and
pesticides most used on rice (in terms of volumes) in Part B of the study) and national authorities were
their countries. The information was self-reported by contacted to obtain this information. In some cases,
the countries with no uniform criteria for the collection national authorities reacted promptly, providing all
of the data, so it is to be considered as indicative only. necessary information. In other cases, obtaining such
Nevertheless, it was noted that for the majority of information was a long and difficult process that took
these “most used” pesticides, very few Codex MRLs several weeks.
exist. Interestingly, the “most used” pesticides with
Documents on risk assessment methodologies,
no Codex MRLs, in most cases do not have national
risk management considerations and MRL adoption
MRLs either. This might be because developing
procedures were not always easy to locate: this matter
countries do not have enough capacity to assess
is addressed in detail in Part B of the study for five
MRLs at national level and strongly rely on Codex
major markets (Australia, Canada, Japan, the European
MRLs for the establishment of their national MRLs.
Union and the United States of America).
This indicates that Codex MRLs remain important as
many countries continue to make use of and rely on Countries’ positions towards the development and
them. Codex MRLs are therefore crucial not only for adoption of new Codex MRLs
trade, but also for ensuring food safety in the domestic A low level of transparency was found on countries’
markets of many developing countries. positions towards the development and adoption of
new Codex MRLs. By going through the annual CCPR
2.6. Transparency and CAC reports, there appear to be limited notifications
by countries on their reservations on new Codex MRLs,
The study implicitly ascertained the level of during the Codex standard development process.
transparency offered by countries in relation to Among the countries/region analysed, the European
pesticide MRLs, particularly in making relevant Union seems to be the only Codex member openly
information publicly available. This is particularly raising reservations and communicating to Codex when
important as transparency helps trading partners to not in a position to adopt a new Codex MRL, providing
become aware of the requirements they need to meet scientific reasons for the reservations and consequent
when engaging in international trade. non-alignment (see also Part B of the study).
17
2.7. Concluding remarks few exceptions, the developing countries analysed

tend to defer to Codex MRLs, either entirely, or in
Alignment with Codex MRLs
combination with regional or national MRLs of major
The study unveiled different outcomes. On the one
importing markets. This also explains the findings
hand the analysis showed that many pesticide MRLs
presented above, namely that most of the developing
(around 2/3) registered at national level do not have
countries have better alignment with Codex MRLs.
corresponding Codex MRLs (see Figure 4). This
Developed countries showed different approaches.
would call for increased resources and capacities for
While some apply default limits, usually at the limit of
Codex and the scientific advice programme for the
quantification, others apply zero tolerance.
development of new MRLs. On the other hand, the
analysis also revealed that the majority of Codex rice
Rice classification and processing level of rice
MRLs do not have a corresponding MRL at national
It was also found that differences in rice classification
level (see Figure 6), with several markets covering
(or level of processing of rice) have a significant impact
less than 25 percent of Codex MRLs. This raises
on the level of harmonization with Codex MRLs (see
the concern that increasing numbers of countries
are establishing their own national standards, not Figure 7). Because inconsistency in food classification
harmonized with Codex, thus undermining the across countries and Codex is common to many
relevance of Codex standards in international trade. commodities (see also Part B of the study) it is of
the utmost importance that harmonization of food
A similar trend was noted in the level of alignment classification be given consideration at international
with Codex MRLs, which varies greatly across the level. In the specific case of rice for example, Codex
countries/region analysed. For developing countries, MRLs could be established for all different processing
the alignment is very high or total in some cases. For levels of rice, rather than for a single one, so as to
the other countries, alignment tends to be quite low, accommodate different classifications by different
usually lower than 25 percent. countries – see also Part B of the study.
It seems that low levels of alignment are usually due Transparency

to the lack of national MRLs, non-deferral to Codex The study also gauged the degree of transparency of
MRLs in the absence of national MRLs and differences various countries regarding pesticide MRLs, especially
in food classification. More specific insights on the regarding public availability of important information.
reasons behind limited harmonization are given in Great transparency was observed in relation to the
Part B of the study. public availability of national MRLs, which for most
countries were easily accessible online. MRL-related
Absence of national MRLs and reference to Codex national regulations or policies on procedures followed
MRLs in the absence of national MRLs were more difficult
Deferral to Codex MRLs when national MRLs do not to find and access – only few countries report these
exist is not a common practice. For pesticide MRLs procedures in their national legislations (See Table 6
that are not registered under national legislations, and Part B) and obtaining them proved difficult in some
different countries apply different approaches. With cases.
18
PART A
3 What does this mean for trade?

Sanitary and phytosanitary (SPS) measures have been 3.1 Food safety standards and
used by governments for centuries in their efforts to trade
ensure that food is safe for their populations and that
Ensuring that food is safe to be consumed has been
public health is protected. However, when standards,
among the important tasks that populations have
including pesticide MRLs, are not harmonized among
entrusted on their governments. Throughout history,
trading partners, this can represent major obstacles
many countries have independently developed food
and impede the trade of food across borders. This
laws and regulations, and often found different
chapter explores the possible effects of the limited
solutions to ensure that food was safe and met the
harmonization of pesticide MRLs among countries,
quality expectations of consumers. However, the
using rice as a case study.
differences between national requirements and
regulations can make it difficult to move food across
Key messages:
country borders. At the same time, consumers have
JJ Many of the main rice importing countries have become increasingly concerned about food-related
stricter MRLs on pesticides used in the production risks, including health hazards due to pesticide
of rice than those established by Codex. Almost residues, other contaminants and unsafe food
half of the main exporting countries also have MRLs
additives (FAO and WTO, 2017).
stricter than Codex, while the rest of the countries
are aligned with Codex. 3.1.1 High consumer awareness of food safety in
JJ MRLs stricter than Codex in the importing country importing markets
are associated with relatively more rice imports, Wealthier countries with more information about
possibly reflecting high import demand combined food safety risks tend to require more stringent food
with strong consumer awareness of food safety in safety standards for both domestically produced and
these countries. imported food products and are generally willing to pay
JJ If
more for higher levels of food safety (Buzby, 2001).
MRLs on the importer side are stricter than on
the exporter side, this may lead to additional costs The growth of exports from developing countries
for exporters in order to comply with the importing in particular has been accompanied by increasing
country’s standards. These higher costs may
attention to food safety standards in many major
impede the exports of developing countries to the
importing markets (Maertens and Swinnen, 2009),
countries with stricter regulations.
with the rising consumer awareness often translated
into more stringent public food safety standards. Even
stricter private food safety standards imposed by
large trading and retailing companies have emerged
(Maertens and Swinnen, 2009; Unnevehr, 2015).
For example, 54 percent of 236 surveyed households

in Georgia, the United States of America, in the early
2000s, perceived pesticide residues to be a serious or
extremely serious food safety threat (Rimal et al., 2001).
Food safety concerns in high-income markets also apply
to imported products. Between 31 and 43 percent of
387 respondents to a survey in Japan in 2002 perceived
imported rice from the United States of America,
Australia and China as being less safe than domestically
produced rice, whereas only 7 to 16 percent perceived
the rice imported from the three countries to be safer
19
or equally safe compared to domestically produced rice products from developed countries, which are believed
(Peterson and Yoshida, 2004). to adhere to higher standards and credible certification
and control systems. Although similar products from
MRLs are often stricter in high-income countries than
developing countries are cheaper, credible institutions
in emerging economies and developing countries
allow developed countries to compete successfully in
(e.g. Xiong and Beghin, 2014 and Figure 8), while
the Thai market (Nuttavuthisit and Thøgersen, 2019). In
most food safety standards in export-oriented
another study, consumers in Beijing had a high demand
developing countries were initially imposed to meet the
for food safety assurance in beef products and were
requirements in the major import markets. This process
willing to pay more for Australian beef, which is believed
started in the 1970s and 1980s with public standards
to be safer than domestic (Chinese) beef or that from
and continued with the emergence of private sector
the United States of America (Ortega et al., 2016).
standards applied by supermarkets, fast food chains
and large processors (Reardon et al., 2019). 3.1.2 Costs to comply with food safety standards
in export markets
With rising incomes, increased education and
If food safety regulations in importing countries are
changing lifestyles, consumer demand for food
stricter than in the exporting countries, it implies higher
safety has gained momentum also in emerging
costs for the exporting country to meet the more
economies and developing countries (Ortega and
stringent standards and certification requirements
Tschirley, 2017; Unnevehr, 2015). Food safety was,
of the importing country. The costs for the stricter
for example, identified as the most important
food safety management in the exporting country are
sustainability attribute for rice consumers in Nigeria
incurred at all levels of the supply chain and comprise
(Okpiaifo et al., 2020) and became a societal issue that
investment costs as well as expenses for improved
received considerable attention in Viet Nam (Pham
monitoring and certification. The acquisition of new
and Dinh, 2020). The intention of the African Union
equipment for reducing risks and monitoring outcomes,
to establish an African Food Safety Agency (AFSA)
the establishment of management and quality control
is another relevant example. The Agency is expected
systems, and capacity development cause extra costs
to coordinate and provide leadership and support to
(Unnevehr, 2015). Due to often weak rule enforcement
meet the mounting need for enhanced food safety
and a typically large informal economy in agriculture
in the continent. That the food traded within the
and food supply sectors, in particular in developing
African continent is safe and satisfies the increasing
countries, a number of challenges arise also from daily
consumer awareness is imperative considering Africa’s
risk management (Pham and Dinh, 2020).
push towards its integration agenda, notably through
the African Continental Free Trade Area (AfCFTA).
High investment costs can imply a heavier burden on
The establishment of the AfCFTA is expected to
small firms and farms than on larger enterprises, given
significantly promote intra-regional trade in agrifood
their small capital and often limited access to credit
products, which is projected to be 20-30 percent
and other financial resources. However, the evidence
higher in 2040, compared to without the AfCFTA.25
on whether higher food safety standards have led to
Consumers often prefer domestic over imported food the exclusion of smallholders from markets is mixed.
products when both options are available (Nuttavuthisit Higher costs are incurred during the adaptation
and Thøgersen, 2019). However, in developing phase to stricter standards, but the overall impact of
countries this can sometimes co-exist with a food safety compliance on livelihoods in developing
preference for food products imported from developed countries tends to be positive due to declining
countries, due to higher trust in the standards and transaction costs over time, better employment
certification schemes in these countries (Nuttavuthisit opportunities, higher wages and longer employment
and Thøgersen, 2019). periods for low income workers in export supply chains
(Unnevehr, 2015).
For example, consumers of organic food in Thailand
are found to prefer domestic production in general. At 3.1.3 Effects of SPS measures and MRLs on trade
the same time, they also show high trust in imported
The high demand for food safety and related standards
in importing markets and the additional costs incurred
25 https://www.uneca.org/stories/african-trade-agreement-catalyst- by exporting countries to comply with these standards,
growth
20
What does this mean
for trade? 3 PART A
have repeatedly featured in the empirical literature on mixed results. This is not only due to the opposite
the effects of food standards on trade. effects that these various measures can have, but the
results also depend on many other factors, including
Reflecting the strong demand for food safety,
the selection of countries, products and measures,
accompanied by a greater willingness to pay, food
the studies considered and the methodology
standards, including MRLs, tend to be stricter in
applied in conducting the analyses (Santeramo and
high-income markets. A higher public and sometimes
Lamonaca, 2019).
private regulatory stringency, including its enforcement
in these markets, increases trust in domestically Also meta-analyses, which summarize the results
produced and imported food products and may increase of a multitude of similar studies, come to different
effective demand by relieving consumer concerns conclusions. A meta-analysis of 27 papers in 2012,
about product quality and safety (Thilmany and Barrett, for example, found that many SPS regulations tend
1997). Regulatory distance between trading partners to impede agricultural and food trade flows from
often reflects strong consumer preferences for stricter developing country exporters to high-income importing
standards in one country and does not necessarily markets (Li and Beghin, 2012). A qualitative review
translate into a barrier to trade (Drogué and DeMaria, of the literature conducted in 2017 concluded that
2012). Without the use of food safety standards in studies predominantly report negative effects of MRL
international trade, trust in imported products would stringency on trade flows (Grant and Arita, 2017).
cease, which would likely imply reduced imports, in However, the most recent and comprehensive meta-
particular from countries with a real or perceived higher analysis, using statistical methods, and considering
risk of pesticide contamination. 62 papers on the topic, suggested that stricter MRLs
tend to favour trade (Santeramo and Lamonaca,
The Spanish horticultural sector, for example, suffered
2019). Some examples of the detailed effects found in
from several crises affecting consumer perception and
empirical studies are presented below. These studies
demand in importing countries. In the “pepper crisis”
are also partly considered in the meta-analyses just
in 2006, a residue of a non-authorized pesticide was
described.
discovered in one batch of Spanish sweet peppers
exported to Germany. This led to a significant decline Several studies analysed the effects of specific MRLs
in the exports of sweet peppers from Spain between on trade. Stricter MRLs applied by importing countries
2006 and 2007. The image loss spilled also over to on the insecticide chlorpyrifos, for example, were
other Spanish horticultural products as well as to found to reduce China’s exports of vegetables (Chen,
horticultural exports from third countries, including the Yang and Findlay, 2008). Stricter MRL regulations
Netherlands (Serrano-Arcos, Sánchez-Fernández and also had a negative overall effect on exports of fresh
Pérez-Mesa, 2019). fruits from Chile and an even larger effect if the
standards were imposed by a developed country
Food safety standards and MRLs can also hinder
(Melo et al., 2014). The European Union pesticide
trade due to the higher costs implied for producers,
standards on tomatoes tend to inhibit exports from
processors and traders to comply with the standards.
African countries, while standards on oranges, limes
Producers may need to adapt production practices
and lemons were shown to have a demand-enhancing
to meet the standards and also obtain certification.
effect and stimulated new trade relations (Kareem,
Inspection and testing procedures at customs points
Martínez-Zarzoso and Brümmer, 2018). Changes in
may prolong delivery time or result in rejection of
European NTMs on green beans over the time period
particular shipments if standards are not met. The
1990–2011 appear to have led to adjustment effects
higher costs are also transmitted as higher product
of Kenyan suppliers with an initial reduction of trade
prices, which may lower demand on the importing side.
followed by a catch-up with the previous situation
The interaction of these two effects, the demand- (Henry de Frahan and Nimenya, 2013).
enhancing and the trade cost-raising, determines
Mixed effects on trade were also found for SPS
if MRLs facilitate or hinder trade overall (Xiong and
measures applied to animal products. For example,
Beghin, 2014).
MRLs imposed by importing countries on the antibiotic
Many analyses have been conducted on the effects of Oxytetracycline appeared to reduce the exports of
NTMs, SPS measures and MRLs on trade, often with fish and aquatic products from China (Chen, Yang
21
and Findlay, 2008), while MRLs on the antibiotic MRLs on trade. The gravity theory in international
Chloramphenicol were found to negatively affect the trade assumes that countries trade in proportion
intensity of imports of crustaceans by Canada, Japan, to their respective economic size and geographical
the United States of America and European Union-15 and regulatory proximity (UNCTAD and WTO, 2012
(Disdier and Marette, 2010). However, while some SPS and 2016). This means that countries more similar in
measures were shown to significantly reduce trade economic size and income trade more with each other.
in meat among the world’s ten biggest importers and Trade between two countries, however, is reduced
exporters, others had substantial positive impacts on if they are located farther away from each other
trade (Schlueter, Wieck and Heckelei, 2009). and their standards, for example MRLs, are more
heterogeneous. The relatively intuitive gravity model is
A broader study by the Organization for Economic Co-
generally found to explain a large part of the variation
operation and Development (OECD) on the effect of
in bilateral trade flows and has proved to be consistent
NTMs on merchandise trade acknowledged that NTMs
with theoretical models of trade (Feenstra, Markusen
can affect trade through both a demand-enhancing and
and Rose, 2001; UNCTAD and WTO, 2016).
a trade cost-raising effect. In particular SPS measures
including MRLs were found to be associated with The empirical specification of the gravity model applied
expanding trade, even though trade costs rose (Cadot, here distinguishes between the likelihood that two
Gourdon and van Tongeren, 2018). countries trade or do not trade rice with each other,
i.e. the extensive margin, and the magnitude of trade
Some studies explicitly refer to the level of
of rice between two countries if they already trade
harmonization of standards or specific harmonization with each other, i.e. the intensive margin. The empirical
events and their impact on trade. Harmonizing industry approach is described in more detail in Annex A-4.
standards in general tends to increase trade flows by
incentivizing investments and generating additional The trade analysis considers eight rice importing
demand (Schmidt and Steingress, 2019). For instance, markets (Australia, Canada, China, the European
increasing the similarity of MRL regulations among Union, Indonesia, Japan, Saudi Arabia and the United
countries appeared to increase the trade of apples and States of America) and 12 exporting countries
pears among these countries (Drogué and DeMaria, (Bangladesh, Brazil, China, Indonesia, India, Cambodia,
2012). A study on the regulatory diversity of NTMs on Myanmar, Pakistan, Philippines, Thailand, the United
eleven animal and plant products of the European Union States of America and Viet Nam).26 Intra-European
and nine of its major trade partners showed that stricter Union trade was excluded. The countries were
pesticide MRLs in one country relative to its trade selected based on their importance in the rice
partners reduce exports to that country (Winchester market (Table 3 and Table 4 refer) and considering
data availability. Together, these countries account for
et al., 2012). Pesticide MRLs imposed by high-income
around 90 percent of the global trade in rice.
OECD countries were found to be associated with
more trade. However, the divergence in food safety The Harmonized Commodity Description and
regulations between importers and exporters reduced Coding System (HS) classifies rice under tariff item
exports by imposing additional costs to comply with the 1006. Within that, rice is classified according to its
importer’s standards (Xiong and Beghin, 2014). production and processing stages: rice in the husk
(paddy or rough) (HS 100610); husked (brown) rice
3.2 Analysis of the effect of (HS 100620); and rice, semi-milled or wholly milled,
pesticide maximum residue whether or not polished or glazed (HS 100630).
limits on trade in rice in Moreover, broken rice is classified separately (HS
selected countries 100640). The HS classification is compared with the
classification used by Codex in Annex A-3.
So far, no study in the literature has specifically
focused on the effects of MRLs on trade in rice. These
effects are explored in this chapter.
26 The assessment of the harmonization of national MRLs with
Codex in chapter 2 covered also the Islamic Republic of Iran and
To study the effects of MRLs on trade in rice, a the United Arab Emirates. These two countries were not included
gravity model is used, which is a widely applied in the analysis here due to missing tariff data (the Islamic Republic
of Iran) and the unclear effects of a high share of re-exports in total
methodological approach to estimate the effects of trade (the United Arab Emirates).
22
What does this mean
for trade? 3 PART A
In this analysis, bilateral trade flows of rice are With regard to national MRLs, where a Codex MRL
distinguished according to the aforementioned existed, but no specific national MRL was adopted,
classification. The trade flows were represented using this was replaced by the national default tolerance
both the value and quantity of trade with equivalent level (Table 6). For all the case studies except the
results, but the results presented below are based on European Union, only the most recent national MRLs
the value of trade from 2012 to 2018. could be retrieved, as MRLs from previous years were
not available. As such, potential changes of national
The main focus of the analysis is to explore the effects
MRLs in the other countries could not be taken into
of the different levels of harmonization of pesticide
account. For the European Union, the values of four
MRLs observed among trading partners on trade in
MRLs changed during the 2012–2018 period. The MRL
rice. In order to identify better these effects, other
indices therefore do not vary much over time.
determinants of international trade are also considered
in the analysis. These include the production capacity The MRL indices by rice tariff line and country/
of the exporting countries, the tariffs imposed by the market for the year 2018 are shown in Figure 9 and
importers, as well as several factors describing the Figure 10. Out of the 17 countries/markets covered
relationship between trading partners. The variables and by the analysis, ten have stricter pesticide MRLs
underlying data sources are described in Annex A-4. than Codex. These are Australia, Canada, China, the
European Union, India, Japan, Pakistan, Philippines,
3.3 Stringency of pesticide the United States of America and Viet Nam.27 Half
maximum residue limits of the countries/markets with stricter MRLs than
compared to Codex Codex are high-income economies (Australia, Canada,
the European Union, Japan and the United States
Chapter 2 of the present report analysed national and of America), while the other half are emerging
Codex MRLs and the level of their harmonization. economies or developing countries (China, India,
To accommodate the large number of MRLs for rice Pakistan, Philippines and Viet Nam). With the exception
in the trade model, an index of MRL restrictiveness of Saudi Arabia, all of the high-income countries
was created that summarizes the level of compliance considered in the analysis have considerably stricter
with Codex in a single number per rice tariff line, year MRLs than Codex.
and country (the calculation of the index follows Li
The MRLs of the remaining seven countries
and Beghin (2014) and Xiong and Beghin (2014) and is
(Bangladesh, Brazil, Indonesia, Cambodia, Myanmar,
described in Annex A-4).
Saudi Arabia and Thailand) are on average equivalent to
The MRL index equals one if national MRLs are fully Codex MRLs.
aligned with the Codex standards. It is greater than
This implies that all importers considered in the
one if, averaging over all pesticide MRLs, national
analysis, with the exception of Indonesia and Saudi
MRLs are stricter than Codex. If national MRLs are
Arabia, have, on average across the pesticides,
less stringent than those established by Codex, the
significantly stricter MRLs than Codex (75 percent,
index takes a value less than one.
Figure 9). Seventy-five percent (six out of eight) of
It needs to be noted that the analysis focuses on the the markets considered on the importer side are
level of harmonization of national MRLs with Codex high-income economies, five of them have stricter
MRLs. As a result, the study only considers the MRLs than Codex (i.e. 62.5 percent of the importing
MRLs for which a standard was adopted by Codex, countries are high-income countries with stricter
i.e. it ignores those national MRLs for which no MRLs than Codex).
corresponding Codex MRLs exist.
Half of the exporters considered in the analysis have
Concerning the Codex MRLs, if a new MRL on a their national MRLs aligned with Codex (Bangladesh,
specific pesticide was adopted by Codex during the Brazil, Indonesia, Cambodia, Myanmar and Thailand).
period under examination (2012–2018), this pesticide
MRL was only considered starting from the year of
27 In some of the countries with stricter MRLs, for example the
its adoption. This means that trade flows in 2014, for Philippines, the MRL stringency is strongly determined by those
example, were matched with the MRL index only for Codex MRLs for which a dedicated national MRL does not exist
and the stringency of the default tolerance level which is applied in
those Codex MRLs that had been adopted up to 2014. these cases.
23
Figure 9: MRL indices by tariff line and country, importing countries/markets, 2018
Rice in the husk Husked (brown) rice Rice, semi-milled or wholly milled Rice, broken
3.0
2.5
2.0
MRL index
1.5
1.0
0.5
0.0
Australia Canada China European Indonesia Japan Saudi Arabia United
Union States of
America
Note: The broken red line indicates alignment with Codex. The higher the MRL index, the stricter the national MRLs compared to Codex.
Figure 10: MRL indices by tariff line and country, exporting countries, 2018
Rice in the husk Husked (brown) rice Rice, semi-milled or wholly milled Rice, broken
3.0
2.5
2.0
MRL index
1.5
1.0
0.5
0.0
h
zil
na
ar
an
nd
am
si
di
di
ne
ic
es
m
a
st
ila
ne
er
Ch
In
bo
Br
tN
ad
pi
n
ki
Am
ya
do
ilip
Th
Pa
gl
e
M
Ca
Vi
n
In
Ph
of
Ba
es
at
St
d
te
ni
U
Note: The broken red line indicates alignment with Codex. The higher the MRL index, the stricter the national MRLs compared to Codex.
Exporting countries with stricter than Codex standards 3.4 The effect of pesticide
are China, India, Pakistan, Philippines, the United Maximum residue limits on
States of America and Viet Nam (Figure 10). Eleven trade in rice
of the twelve considered exporting countries are
In order to reflect the dual effect that MRLs can
emerging economies or developing countries, of which
have on trade, the analysis considers both the MRL
five have stricter MRLs than Codex.
stringency of the importing country (MRL importer)
and the difference in MRL stringency between
importer and exporter (MRL divergence).
As described above, the MRLs on the side of the

importers may increase effective demand by relieving
consumer concerns about product quality, while
24
What does this mean
for trade? 3 PART A
Figure 11: The dual effect of pesticide MRLs on imports of rice
Likelihood to trade (extensive margin) Intensity of trade (intensive margin)
3.0
2.5
Magnitude of the effect on trade
2.0
1.5
1.0
0.5
0.0
-0.5
-1.0
-1.5
MRL Importer MRL Divergence
Note: The figure shows marginal effects that are evaluated at their means. The error bars show 95 percent confidence intervals. A confidence
interval is an indicator of the certainty of an estimate by giving the range in which the true value would fall with a probability of 95 percent.
the regulatory divergence may affect trade due to more about the stringency of MRLs related to both
the higher costs incurred by producers, processors their domestic production and imported products.28,29
and traders to comply with the stricter standards. If
The impacts of the regulatory divergence on trade are
domestic food safety regulation with respect to MRLs
less pronounced than those of the importers’ MRLs.
on pesticides in rice of the exporting country is stricter
However, as expected, they are clearly negative.
or equal to that of the importing country, no additional
Although the negative effect is relatively small, this
trade barrier (i.e. costs related to trade) should apply.
means that the stricter the importing country’s
However, if the MRL regulation in the exporting
MRLs relative to those of the exporting country, the
country is laxer than in the importing country, this
less likely they are to trade with each other (MRL
could imply higher obstacles to export to this country.
divergence, extensive margin). If two countries already
Evidently, if both trading partners adopt internationally
trade with each other, stricter MRLs on the importer
harmonized Codex standards, the MRLs should not
side compared to those of the exporter are also
cause any additional cost to trade. The results of the
negatively associated with the magnitude of trade
analysis for each of these opposite mechanisms are
(MRL divergence, intensive margin).
presented separately (Figure 11). As the results are
based on an index, i.e. an aggregate measure of MRL As indicated by the confidence intervals in Figure 11, the
stringency composed based on 82 rice pesticide MRLs positive association between MRL stringency on the
adopted by Codex, the actual numbers are difficult to importer side and both the likelihood to trade and the
interpret. It is for this reason that the interpretation magnitude of trade is more certain than the negative
below and in the following sections focuses on the effect of the MRL divergence on trade. As the positive
direction of the effects and their relation to each other. relationship between MRL stringency and trade on the
importer side is also much stronger than the negative
The results reflect the dual effect that MRLs can
effect of the MRL divergence, the net effect on trade is
have on trade. The effect of MRL stringency on the
positive, indicating that MRL stringency per se does not
likelihood to trade on the importer side is positive, i.e.
necessarily impede trade. Effects of other determinants
a higher MRL stringency is associated with a higher
on trade in rice are summarized in Box 2. Detailed
likelihood to import rice (MRL importer, extensive
results appear in Table 8 and Table 9, Annex A-5.
margin). There is also a significantly positive effect of
importers’ MRL stringency on the magnitude of the
imports, i.e. stricter MRLs are associated with larger 28 The potential endogeneity bias due to effects on policy design is
import volumes (MRL importer, intensive margin). largely accounted for by including fixed effects in the model setup
(Xiong and Beghin, 2014).
However, this does not imply that the stricter MRLs 29 The model identifies the relationship between MRL stringency and
imports net of income effects. Although high-income countries
lead to more imports of rice. Rather, it is more likely tend to have stricter MRLs, income effects are covered by other
that countries with a high import demand tend to care model parameters and do not interact with the results on the
effect of MRL stringency.
25
Box 2: Effects of other country-level and bilateral with each other in proportion to their economic size, two
determinants on trade in rice countries with the same development status are more likely to
In addition to the effect of the MRLs, other determinants of trade with each other and the traded volume is likely to be much
trade were also considered in the analysis. higher. Pointing to path dependency, a colonial relationship
between two countries in the past is still associated with a
A greater volume of exportable production in the exporting higher volume of trade in rice today. The likelihood to trade
countries is shown to increase the likelihood of trade and the at all, however, is no longer clearly associated with former
volume of trade in rice (Table 8, Annex A-5). By contrast, and colonial ties. Two countries, however, that traded rice in 1995
as expected, higher import tariffs in the importing countries are still more likely to trade today with tendency for a larger
are associated with both a lower likelihood to import and a traded volume.
lower magnitude of trade.
The likelihood to trade and the scale of trade between two
The farther two countries are located from each other, the less countries are also larger if the main rice varietal family (e.g.
likely they are to trade and the lower is the traded volume. aromatic, indica, japonica) imported in one country matches
Moreover, in line with the gravity theory that countries trade the one the exporting country is specialized in.
3.5 Concluding remarks Nonetheless, greater international harmonization of

MRLs on pesticides in rice would ease access by
In this case study on rice, a gravity model analysis was
exporting countries to the markets of the main rice
conducted to assess the implications of different levels
importers. Although there is a growing consumer
of harmonization of national with Codex pesticide
awareness about food safety in many developing
MRLs on trade. The analysis covered eight importing
countries, including a proliferation of public and private
markets and 12 exporting countries. Seventy-five
standards and traceability systems (Pham and Dinh,
percent of the main rice importing markets and
2020), most of these countries would not be able to
50 percent of the rice exporting countries considered
move immediately to the same standards and food
in this study are found to have stricter on-average
safety outcomes found in high-income countries
pesticide MRLs than those adopted by Codex. The
(Unnevehr, 2015).
MRLs of the other exporting case study countries are,
on average, aligned with Codex MRLs. Any changes towards further international alignment
of pesticide MRLs on rice to facilitate trade will
Codex MRLs are based on solid scientific analysis and
have to balance between the demand for stricter
assessment and are established following an inclusive
food safety regulations on the importer side and
and transparent consultative process reassuring that
the additional costs incurred on the exporter side.
MRLs are set at an appropriate level for public health
Different steps could be taken by countries towards
to be protected.
improving harmonization with Codex standards,
However, in the majority of the main rice importing considering also the technical issues elaborated in
markets there appears to be a high tendency for Part B of this study. For example, it would seem that
applying stricter MRLs on pesticides than those consideration of Codex MRLs could be enhanced in
established under Codex. The analysis shows that the the absence of national MRLs (as also foreseen by
effect of the strong demand for rice imports under the SPS Agreement), as a first step towards better
strict food safety regulations outweighs the trade- harmonization.
impeding effect of the non-harmonization of national
To increase developing countries’ capacity to comply
MRLs with Codex MRLs. A relaxation of the stricter
with the required food safety standards in both their
standards of the importing countries could, therefore,
export and domestic supply chains, commitment to,
lead to unpredictable effects on rice trade or the
and consistent investment in, continuously improving
replacement of public by private standards.
food control systems and regulatory capacities are
fundamental.
26
PART A
4 Conclusions and way forward

Lack of Codex MRLs of Codex MRLs that do not have a corresponding MRL
One of the findings of the study is that many at national level. For example, some economies may
pesticides (around 2/3) registered at national level do not have procedures for establishing import MRLs, or
not have corresponding Codex MRLs (see Figure 4). may have policies that do not allow for consideration of
The absence of MRLs for use in traded products is foreign GAPs when there is a domestic GAP in place.
a significant concern that has been brought to the Sometimes, setting an MRL is part of the pesticide
attention of the CAC in the past few years. With food approval process, or is only possible after a pesticide
safety increasingly being recognized as a priority has been approved for use in a country. There might
issue in many countries, the request for scientific also be a tendency for countries to establish MRLs
advice has also increased. However, despite the only for export purposes and not for import. However,
significant efforts that are being made by FAO/WHO this was not further analysed in this case study.
to streamline procedures and to manage the process
of scientific review as efficiently as possible, the Missing national MRLs could also be partly explained
number of requests for evaluation by the JMPR of by differences in the time of MRL adoption at Codex
new compounds and new uses (as well as for periodic and at national level, as indicated in Part B of the
re-evaluations of existing MRLs) far exceeds the study. This difference in time may entail changes
current capacity. in the scientific data packages evaluated by the
different authorities (e.g. different GAPs reflecting
At the same time, the absence of Codex MRLs for different pests and diseases, different pesticide labels,
specific pesticides registered at national level, could availability of different studies, etc.) leading to different
be due to different factors, including, for example, results. While Australia, the European Union and Japan
concerns related to the toxicity of the chemicals, or review Codex MRLs upon their establishment, Canada
the inability of countries to engage in the Codex work and the United States of America report to have no
to advocate for pesticides of interest to them or to routine procedures in place to review national MRLs
submit the data needed for the evaluation of MRLs. at the time Codex adopts new MRLs. The United
This first analysis provides grounds for further research States of America and Canada review MRLs as part
on these matters. of a national active substance review programme (see
Part B for further details).
Because of the importance of Codex MRLs, both
for developing countries lacking capacity for risk It seems that consideration of Codex MRLs could be
assessment and for international trade, addressing enhanced in the absence of national MRLs, as also
the need for more Codex MRLs should be further recommended by the SPS Agreement (see Annex C
considered. (1) “chaussette”, and Article 8 of (WTO, 1995)): “the
importing Member shall consider the use of a relevant
Missing national MRLs international standard as the basis for access until a
It was established that there are many Codex MRLs final determination is made”.
that do not have a corresponding MRL at national
level. The analysis showed that many markets analysed Approaches taken by countries in the absence of
have established only a very limited number of the national MRLs
MRLs set by Codex for rice (see Figure 4). Because In the absence of national MRLs, the procedures in
most of these countries do not automatically defer place to deal with these missing MRLs differ among
to Codex MRLs in the absence of national MRLs, the countries. In general, most developing countries
result is that there could be a regulatory gap for trading analysed tend to defer to Codex MRLs, either entirely,
partners wishing to export rice to these countries. or in combination with regional or national MRLs of
There could be many reasons behind the high number major importing markets. Developed countries/region,
27
on the other hand, either apply established default of this study, similar issues also apply to other
limits or set the limit at zero (see also Table 6). As commodities such as cereal grains, nuts, pulses and
further discussed later in the study, in all the five many others. This triggers questions on how different
countries/region analysed in Part B, an application definitions among commodities may also affect
can be made to have an MRL established when a international trade and calls for enhanced international
national MRL is absent. This is termed an import harmonization on this matter.
tolerance (in Canada, the European Union, Japan
and the United States of America) or Schedule 20 As further described in Part B, to reduce the potential
amendment in Australia. The process can vary greatly for confusion over multiple MRLs for different forms
in both complexity and duration (from a few months of a single pesticide/crop combination, consistency
up to six years), depending on the country. Australia could be improved where possible, i.e. set multiple
offers a streamlined process that takes place yearly MRLs for all pesticides to cover all likely fractions, or
and accepts JMPR and/or other national evaluations reduce the number of MRLs for some pesticide/crop
in whole or in part when they exist (FSANZ, 2018). In combinations.
Canada and the United States of America, The North
American Free Trade Agreement (NAFTA) guidance Transparency
indicates that the presence of a JMPR evaluation could Transparency is considered an important asset for
result in a less extensive procedure (NAFTA, 2005). either a country’s capacity to comply with or be aware
Japan and the European Union appear to require a of regulatory changes/adoptions, to reduce its internal
full new assessment even if other evaluations are costs and so increase its trade.
available. Obviously, the re-evaluation of studies While for national MRLs good public access has
and data adds significantly to the time and cost for been ascertained for most countries/regions analysed
concluding the import tolerance. This negatively affects (see Annex A-1), less transparency was evident
exporting partners, especially in developing countries for MRL-related national regulations or policies.
where they might not have either the financial or the Some difficulties have been faced when searching
technical means for such application. for procedures followed in the absence of national
MRLs, as well as for documents on risk assessment
Codex MRLs are based on solid, independent scientific methodologies and risk management procedures to
analysis and assessment and are established following establish national MRLs.
an inclusive and transparent consultative process that
reassures that MRLs are set at the appropriate level Having policies related to MRL establishment and
to protect health and facilitate trade. Deferral to Codex enforcement available online would increase overall
MRLs when national MRLs do not exist could be transparency and would be useful for trading partners.
considered as a default practice (see also Part B). Transparency provisions in this regard are also included
in the WTO SPS Agreement to ensure that measures
Food classification taken to ensure food safety are made known to the
Food classification proved to be very heterogeneous interested public and to trading partners.
among the countries/region analysed. During the
analysis, difficulties emerged as of how to match Finally, very little transparency was evident for country
national MRLs to Codex MRLs. This holds true positions towards the development and adoption of
also for the analysis of the reasons behind low new Codex MRLs. The European Union seems to be
harmonization (see Part B). In general, due to varying the only Codex member openly raising reservations
rice classifications across countries (see Annex A-2), and communicating to Codex when not in the position
matching the different Codex rice commodity types to adopt a new Codex MRL, providing scientific
with the national ones proved challenging. Sometimes reasons for its reservations and non-alignment (see
the same name would refer to different rice types in also Part B of the study).
different countries. The study showed how differences
in rice classifications can influence the extent of For the optimal functioning of the Codex MRL
international harmonization for pesticide MRLs. While standard-setting process, it would be important that
this issue emerged for rice, as the focus commodity countries actively notify Codex whenever they have
28
Conclusions and way
forward 4 PART A
any reservation and are not in the position to adopt a engage in international fora and take advantage of the
newly established Codex MRL, providing a science- system. In this respect, it is worth noting that there has
based rationale. never been a WTO trade dispute triggered by pesticide
residues. Beside the dispute settlement mechanism,
Ongoing efforts undertaken by countries to which requires considerable legal expertise and financial
improve harmonization with Codex MRLs resources, the WTO also provides a mechanism to
Despite the observed low level of harmonization, express trade concerns during quarterly international
countries seem to be actively involved in the committee meetings (i.e. the SPS Committee meetings).
development and promotion of various regional and When looking at the total share of SPS Specific Trade
international initiatives to improve harmonization. Concerns (STCs) on pesticide residues, WTO statistics
For example, Canada, through its Pest Management report that out of 408 SPS STCs raised between 1995
Regulatory Agency (PMRA), and Australia as co-chair and 2017, 31 percent were related to food safety, of
of the APEC Food Safety Cooperation Forum (APEC, which 6 percent were on pesticide residues (WTO,
2016), have been very active in the acquisition of 2016). The small share could be explained by many
National and Codex MRLs through science-based factors, including lack of capacity by the concerned
research and workshops in their respective regions to countries to make use of the WTO forum for voicing
achieve greater regulatory convergence of MRLs and and solving their trade concerns on pesticide residues.
greater alignment with international standards. More Nevertheless, the trend seems to have been changing in
specifically, Australia is actively considering whether the past three years, when pesticide MRL discussions
Codex MRLs adopted at each CAC meeting could be gained momentum at the WTO. Five new STCs on
routinely included in their annual MRL harmonization MRLs were raised in the SPS Committee during 2017
proposal to amend the Food Standards Code (subject and two new STCs were raised both in 2018 and in
to certain tests which include an acceptable Australian 201930, which represents 17 percent of all new STCs
dietary exposure assessment). raised between 2017 and 2019. Many of these STCs
have been supported by a large number of countries.
There is ongoing work also in the areas of residue
definition for enforcement and risk assessment (by Building on this new trend and demonstrated interest,
OECD, jointly with JMPR and JECFA) and chronic consideration should be given to developing countries’
exposure assessments to improve harmonization also needs for better and more active participation in the
between JMPR and JECFA, by using individual food Codex standard-setting process and to the needs of
consumption data. These aspects are further explored those countries when developing new MRLs both at
Codex and national level.
in Part B.
Role of international organizations

Need for capacity development
In this context it is important to recognize the role
Many developing countries continue to make use of
played by FAO in enhancing capacity of developing
and strongly rely on Codex MRLs. As Codex MRLs are
countries in food control and food safety management,
critical to developing countries for the establishment of
as well as other relevant international organizations
their national MRLs, greater attention should be given
such as WHO and international partnerships such as the
to the needs of those countries when developing
STDF. It is also important to note the role of the Codex
new MRLs, as well as to means to facilitate their
Trust Fund31 to support countries to build strong, solid
engagement and participation.
and sustainable national capacity to engage in Codex.
There seems to be limited understanding of the

international processes around standard-setting and
limited capacity to engage in the process by some 30 2017: See STCs number 419, 422, 426, 427 and 428 in http://
spsims.wto.org/en/SpecificTradeConcerns/Search
countries. Throughout the analysis, it was noted how 2018: See STCs number 447, 448 in http://spsims.wto.org/en/
developing countries seem to lack access to relevant SpecificTradeConcerns/Search
2019: See STCs number 453, 457 in http://spsims.wto.org/en/
information (e.g. unable to identify problematic SpecificTradeConcerns/Search
For WTO statistics see: http://spsims.wto.org/en/PredefinedReports/
pesticides and provide data on relevant negative STCReport?Year=2020&YearFrom=2017&YearTo=2019&FilterType=0
economic impact, etc.), as well as the capacities to 31 See: http://www.fao.org/fao-who-codexalimentarius/about-codex/
faowho-codex-trust-fund/en/
29
Dual effect of MRLs on trade than the MRLs of the exporting country, are shown
Seventy-five percent of the main rice importing to have trade-impeding effects. Further international
markets and 50 percent of the rice exporting countries alignment of pesticide MRLs on rice to facilitate trade
considered in this study have stricter pesticide MRLs will have to balance these two effects.
than those adopted by Codex. The MRLs of the other
exporting case study countries are on average aligned To increase developing countries’ capacity to comply
with Codex MRLs. The on average stricter standards with food safety standards in both their export
of the importers appear to reflect high consumer and domestic supply chains, commitment to and
awareness of food safety issues and are associated consistent investment in continuously improving
with a greater demand for rice imports. At the same food control systems and regulatory capacities are
time, higher market access costs incurred by exporting fundamental.
countries, if the importers’ MRLs are more stringent
30
PART B
Reasons behind different
levels of pesticide MRL
harmonization
PART B Reasons behind
different levels of pesticide
maximum residue limit
harmonization
This part of the study aims at better understanding comments and inputs, while they were asked to
the reasons that underlie the different levels of identify any other issue that might have not been
harmonization with Codex pesticide MRLs. In considered. In addition to the information supplied
collaboration with the Secretariat of the JMPR, two by the countries/region analysed, publicly available
main areas that may lead to divergent MRLs were legislation and guidance documents were used to
identified and evaluated: provide additional insight.
Some of the data presented in Part A were also used

1. Differences in risk assessment procedures for
to inform and support the analysis in this part of the
pesticides (see section 1.1)
study; for example, the list of pesticides for which
2. Differences in risk management and policies for Codex MRLs exist for rice was used to compare
pesticide MRLs (see section 1.2) some of the different conclusions on health-based
guidance values (HBGVs), residue definitions and
In contrast to Part A of the study, this analysis
MRLs established in different countries. The JMPR/
focuses only on five countries/region (out of the 19
Codex, Australia, the European Union and the United
countries/region analysed in Part A of the study):
States of America were chosen as an example as
Australia, Canada, the European Union, Japan and
comparators for various aspects of the report, due to
the United States of America, where regulations on
the relatively easy access to relevant databases. The
risk assessment and risk management are advanced,
comparison was only intended to provide an indication
publicly available and accessible online.
of reasons for differences among countries and Codex.
The study explores the different procedures and This information was then used as part of the analysis
approaches applied in the risk assessment process of residue definitions and risk assessment practices.
and in risk management decisions that might lead to The values were obtained from a number of publicly
differing MRLs. Factors considered in this analysis available databases, having various levels of detail,
include: variances in the residue definition approaches, covering differing dates and because the ongoing
differences in interpretation of toxicological studies, evaluations are not included, should be considered as
differences in exposure assessment methodologies, a worked example rather than providing a completely
differences in extrapolation rules and other accurate description of the current situation.
methodological issues.
Prior to finalization, the draft study was circulated to
The five countries/region analysed were first the authorities of the countries/region analysed and
approached to gather their feedback. A list of key revised to incorporate comments submitted.
issues identified under the two main areas of
investigation was shared for their (countries/region)
33
© FAO/Vasily Maximov
PART B
1 Risk assessment methodology

This chapter explores various aspects of the risk 1.1. Residue definitions
assessment methodology that could lead to different
The topic of residue definitions for pesticide is
MRLs. The elements of the risk assessment that are
complex. Not only are there two types of residue
analysed are:
definition, one for monitoring (MRL compliance/
enforcement) and one for risk assessment, but the
1.1. Residue definitions
components of the residue can vary across crops and
1.2. Interpretation of toxicity studies over time. In setting residue definitions for a pesticide,
the aim is to have a simple definition that can be
1.3. Assessment of metabolism studies
applied to most, if not all, commodities.
1.4. Exposure assessments
There can also be differences induced by:
1.5. Differences in GAP and selection of critical GAP JJ Different databases e.g. additional crops with
Key messages: different metabolite profiles.
JJ There JJ Different radiolabel positions in the chemical

is considerable variation in how countries
are aligned with the JMPR/Codex process for the structure of the pesticide molecule.
development and establishment of pesticide MRLs. JJ New studies with increased analytical sensitivity.
JJ Many of the observed differences in risk JJ Differentconclusions/considerations of related
assessments do not seem to have a significant metabolites (e.g. the inclusion, or not, of
impact on the overall outcome of the pesticide thiophanate-methyl in the residue definition for
safety evaluations. carbendazim).
JJ Some of the major differences in MRLs and residue
Residue definition for monitoring/compliance/
definitions are due to the presentation of different
enforcement
data to the various countries/region.
There appears to be general agreement across
JJ Harmonization also depends on national authorities different countries/authorities that the residue
supplying updated consumption data to FAO/WHO, definition for monitoring compliance should be:
via the GEMS and CIFOCOss databases.
JJ as simple as possible;
JJ Many of the guidance/procedural documents related
JJ compatible with multi-residue analyses; and
to MRLs and human health risk assessments of
pesticides were initially drafted 10 to 20 years ago, JJ based on parent or a suitable marker compound.
with subsequent, only occasional, updates and
partial revisions. Consideration could be given to an For many pesticide/commodity combinations, the
update process that can be agreed internationally. definition for compliance is unlikely to be a problem.
However, it has been identified as an issue for the
acceptance of Codex MRLs in some cases (e.g.
saflufenacil, flunicamid).
One aspect that might prove a hindrance to accept

Codex MRLs is that some pesticides consist of a
number of isomers (e.g. synthetic pyrethroids). In
some developing countries, racemic mixtures are on
sale and the Codex MRLs relate to total pesticide,
without any resolving of individual isomers. In other
countries only resolved isomer versions are approved
35
and the definition for compliance is based on individual the burden, but other metabolites are each at low
isomers. This is an area where there is the scope for levels. In this case the need to cover 75 percent of the
further harmonization, particularly when the health- toxicological burden could involve two or three extra
based guidance values are the same for both resolved compounds in the definition. This is a more complex
isomers and racemates. approach than that of JMPR and other countries, and
it might lead to further differences in risk assessment
Residue definitions for risk assessment
outcomes.
The residue for risk assessment is frequently more
complex than that for monitoring as it needs to It should be noted that there is ongoing work
include metabolites/degradates, which will contribute by OECD, jointly with JMPR and JECFA, on the
significantly to the risk assessment. At present there development of new guidance to standardize and
is agreement that: improve harmonization on residue definition.
JJ Significant/major contributors should be included, Level of consideration of Codex MRL
although there is no precise definition for this. EFSA prepares detailed briefing documents for the
JJ Compounds representing <10% of the Total European Union risk managers prior to the meetings
Radioactive Residue (TRR) or at levels <0.01 mg/kg of the CCPR (e.g. (EFSA, 2018)). These identify where
are normally considered to contribute negligibly to there is agreement or disagreement with JMPR/Codex
the dietary risk. proposals. A number of points relating to residue
definitions, which could lead to differences in approach
Comparing Codex, the European Union and the United
included:
States of America residue definitions for pesticides
used on rice (Annex A-2) indicates that Codex and JJ Including compounds for which there are no
the European Union definitions had a good degree analytical standards (e.g. captan).
of overlap, but that the United States of America JJ Where the Codex residue definition covered
definitions tended to include more components even metabolites in addition to those listed in the
though the approaches to setting the definitions appear European Union definition (e.g. fenpyroximate,
similar. Australia had a limited number of MRLs for the carbendazim), the Codex definition would tend to
rice pesticides on the list considered, but the residue be more precautionary in terms of risk assessment,
definitions were generally consistent with JMPR. as additional components are included.
JJ Where the European Union residue definition
Where risk assessment residue definitions contain
included additional components to those in the
more, or different compounds than the JMPR
Codex definition, the European Union assessment
evaluation, it is likely that the additional components
could be more precautionary if the additional
would result in a higher intake estimate. However, components were found in all commodities.
this does not mean that the overall conclusion of However, this would not automatically result in an
the risk assessment using the more extensive adverse risk assessment conclusion.
residue definition will be unacceptable. Therefore,
under the terms of the SPS agreement, a different It is not clear whether such disagreements
residue definition, in isolation, is not a reason for not automatically lead to non-harmonization with the
harmonizing with a Codex MRL. It is also possible that Codex MRL: brief summaries of the reservations of
the additional components are only seen in a small the European Union are presented in the reports of
number of crops and therefore would not impact on the CCPR (e.g. (FAO and WHO, 2017; FAO and WHO,
the risk assessment for other crops. 2018)). While the detailed reasoning behind the final
decision is not in the public domain, the European
In the European Food Safety Authority (EFSA) guidance
Union has reported that deviations from the positions
on residue definition for risk assessment ((EFSA,
in the published document are typically limited.
2016) – not yet ratified), all metabolites detected in
crop metabolism studies for pesticides in the quartile This level of information provided by EFSA/European
with the lowest ADIs/ARfDs need to be considered; Union aids transparency, is interesting and useful but
and a minimum of 75 percent of the toxicological equivalent information did not appear to be available
burden needs to be covered by the residue definition. from other countries. To obtain information on the
The latter could lead to different definitions in cases level of consideration of Codex MRL proposals from all
where two or three metabolites cover 60 percent of countries/region, a workshop/discussion forum might
36
Risk assessment
methodology 1 PART B
be a useful way forward in identifying the main issues JJ ADIs had been derived by both JMPR and APVMA
preventing harmonization of residue definitions. for 60 of the 68 pesticides used on rice.
JJ For37 of these the values were the same (taking
1.2. Interpretation of toxicity account of rounding33) and for nine others the
studies APVMA value was higher than that of JMPR.
The basic risk assessment methodologies used by JJ Ofthe 14 ADI values where JMPR was higher, two
JMPR and other regulatory agencies around the world were higher by a factor of two or less.
contain many common elements in terms of data JJ ARfDs had been derived, or determined to be
considered, interpretation of studies, determination unnecessary, by both JMPR and APVMA for 39
of a NOAEL (or equivalent point of departure) and the of the 68 pesticides. For several of the pesticides
application of safety factors to derive the HBGVs (ADI there was variability in whether or not an ARfD was
or ARfD). necessary.
JJ For26 of these the ARfD values (or conclusion on
In most instances the HBGVs derived by different
unnecessary34) were the same (taking account of
countries or regions are either the same or differ
rounding) and for six the APVMA value was higher.
by a factor of two or less. This is confirmed by the
brief analysis of the pesticides used on rice (see JJ Ofthe seven ARfD values where JMPR was higher,
Annex B-2). When comparing JMPR (WHO, 2019a) and two were higher by a factor of two or less
the European Union values (European Union, 2019a) JJ For26 of the 39 pesticides where ADIs and ARfDs
for pesticides used on rice it was found that: were derived by both groups, the values for both
JJ ADIswere derived by both the European Union and ADI and ARfD were consistent within a factor of
JMPR for 49 of the 68 pesticides. two or the APVMA value was higher than the JMPR
equivalent.
JJ For 36/49 the ratio of the ADI values fell between
JJ The major variation for ARfDs was carbaryl, where
0.5 and 2.
both APVMA and the European Union were 20-fold
JJ ARfDs were derived, or there was agreement lower than JMPR.
that an ARfD was unnecessary, for 45 of the 68
pesticides. For most pesticide risk assessments, a difference
factor of two in HBGVs is not critical to the overall
JJ For30 of the 45 ARfDs, the ratio of the values was
conclusion of acceptability, or not, and, if necessary,
between 0.5 and 2.
could be readily mitigated by minor changes to GAP
JJ Fora number of pesticides there was divergence such as pre-harvest intervals. For many of the rice
over whether an ARfD was unnecessary or not. pesticides considered, the JMPR intake estimates
JJ Where there is a large ratio between the European were <10% of the applicable HBGVs.
Union and JMPR values these can often be linked
with fundamental differences. For example, for For the majority of pesticides there is nothing in the
mesotrione (ratio 50 for ADIs), JMPR discounted derivation of health-based guidance values that would
the rat data as inappropriate for human risk trigger an automatic concern in respect of the health
assessment, but in the European Union the rat data risks from consuming commodities containing the
were used in HBGV derivation.
pesticides at the Codex MRL. For a number of those
JJ For23 of the pesticides where both ADI and ARfD where there are significant divergences the reasons
were derived, the values for both ADI and ARfD can be identified readily, e.g. use of human data or
were consistent within a factor of two. non-default safety factors.
A similar exercise was performed for JMPR and Reasons for differences between HBGVs can be due
Australian (APVMA, 2019) HBGVs, and the conclusions to a variety of factors either singly or in combination:
are summarized below . The findings are broadly
32
JJ Different databases
in line with those described above in respect of the
European Union:
33 The HBGVs are considered ‘the same’ where the basis is the
same study and dose but due to different conversions or rounding
procedures the actual values differ.
32 The actual values are not presented in this report, but can be 34 A conclusion of ‘unnecessary’ is taken as being ‘higher’ than
checked at: APVMA, 2019. where an ARfD with a value was derived.
37
New studies on pesticides are often undertaken be mitigated by the use of BMD, but this can be
to meet new data requirements or to investigate resource intensive to perform.
or refine existing findings. As the review cycles JJ Derivation of an ARfD:
of JMPR and regulatory agencies are not
synchronized, there is the potential for different a. Deciding whether or not a pesticide requires
conclusions to be reached due to a different an ARfD can have a major impact on the overall
database available at the times of the reviews. risk assessment. If an ARfD is considered
Pesticide review programmes typically have a 10 to unnecessary, then no acute exposure/intake
20-year cycle. assessment is triggered. If no acute assessment
is performed as part of a JMPR assessment, yet
For some older pesticides with multiple
other agencies consider one is required, this can
manufacturers, the company supporting the review
be a reason for non-acceptance of a proposed
through JMPR can differ from that supporting it
Codex MRL. However, in such circumstances an
through national or regional evaluations. Studies
acute risk assessment should be performed by
performed to supplement the original core database
the agency, using appropriate criteria, to ascertain
are likely to differ between the companies, resulting
if the risk assessment is unacceptable or not.
in differences in the databases.
b. As shown by the analysis of the rice pesticides
JMPR considers that ethically performed human
(Annex B-2: Analysis of data on pesticides used
volunteer studies can provide relevant data for the
on rice) there were a number of differences
human risk assessment of pesticides. This is not the
position in the European Union due to the applicable between the European Union and JMPR
legislation (European Union, 2009). The US position conclusions on when an ARfD was considered
is currently that data from volunteer studies should unnecessary. This was also true for Australia
not be used for determining points of departure (see above), but Australia tended to derive fewer
(PODs) for human health risk assessments of ARfDs than JMPR.
pesticides. Australia has existing HBGVs for c. Although guidance documents on deriving ARfDs
pesticides based on human volunteer data. have been produced by many agencies, and
JJ Differing point of departure (POD): there is much in common, there is a high degree
of uncertainty surrounding ARfD derivation. The
a. The United States Environmental Protection main reason is that a well performed, single
Agency (EPA) normally uses NOAELs in its dose study, measuring critical end-points at the
assessments. However, it also uses benchmark key time, is not available for the vast majority
dose (BMD) methodology as a tool in refining of pesticides. Even acute neurotoxicity studies
PODs (USEPA, 2002a). Evaluations have shown
can have study designs that often preclude the
that in many instances a BMDL and a NOAEL for
determination of NOAELs (e.g. use of three high
a specific study are not markedly different. Both
doses) and contain no or few measurements
JMPR and the European Union have made use
of parameters other than those linked to
of BMD on an occasional basis for pesticide risk
neurotoxicity.
assessments.
d. Many end-points are only measured after four
b. For many end-points in toxicity studies, there
weeks of dosing, or longer. There is therefore
is no clearly defined criterion for adversity. In
considerable uncertainty and extrapolation in
determining adversity, and hence a NOAEL,
deciding if any changes seen after four weeks
experts will take account of many factors such
would have been produced by a single dose
as statistical significance, magnitude of the
or dosing over a single day. The precautionary
change, associated findings, normal background
approach is to assume all findings could be
variation etc. It is therefore possible that a
related to a single dose.
marginal change (e.g. 11 percent) in an end-point
can be considered adverse by one group, but not e. Companies have the option to present additional
adverse by another. If there is a five-fold spacing data to refine ARfD conclusions. Due to animal
between doses this could result in a five-fold welfare considerations, this should only happen
difference in an HBGV. In the future, this could after other approaches (e.g. alternative GAPs)
38
Risk assessment
have been investigated and in the reviewer’s factor is appropriate (USEPA, 2002b). Similar
experience refinement of an ARfD occurs very to the US FQPA factor, Canada also applies the
infrequently. PCPA (Pest Control Products Act) factor, which
is also a factor of ten by default, but can be
f. It might be useful to have an international
removed or reduced depending on the available
discussion on criteria for determining when an
data and underlying concern. It is important
ARfD is unnecessary, based on experience gained
to note, however, that unlike the other safety
since original guidance documents were prepared.
factors (e.g., interspecies factor), law in Canada
JJ Choice of ‘safety’ factors: and the United States of America, respectively
mandates the requirement to apply the PCPA
a. The default approach of applying a 100-fold
and FQPA factor.
overall factor to a POD in driving an HBGV is
JJ Rounding of values for HBGVs:
used for the vast majority of pesticides.
b. For non-threshold cancer end-points, the US EPA The EFSA typically has two significant figures
and the Canadian PMRA apply linear low dose (after the zero(s)) in its HGBVs. Other agencies and
extrapolation (USEPA, 2000); (USEPA, 2002a); organizations (US EPA, JMPR, APVMA) normally
(Canada, 2014). round to one significant figure, to acknowledge
the overall level of uncertainty in the value. This
c. JMPR has adopted chemical specific assessment
rounding can result in numerically different values,
factors (CSAFs; (WHO, 2005)), where this
but the margin is relatively small (<2) and will have
approach has been supported by appropriate no impact on the majority of risk assessments.
data, for a number of HBGVs, typically ARfDs.
JJ There are a number of factors that potentially can
This has resulted in HBGVs higher than those
result in numerically different HGBVs for the same
that would have been derived with the default
pesticide. However, in the majority of cases, as
100 factor. Such an approach is not possible
supported by the rice analysis, the differences
in the European Union where the applicable cover a relatively small range. Such differences will
legislation (Article 3.6.1 of Annex B-2: Analysis of only have an impact on Codex MRL acceptability
data on pesticides used on rice of EC 1107/2009) if predicted exposures/intakes represent a high
requires a minimum factor of 100 to be applied proportion of the JMPR HGBV.
to a NOAEL. Australia and the United States of
America documentation indicates that CSAFs will 1.3. Assessment of metabolism
be considered on a case-by-case basis. studies
d. JMPR and other agencies consider the Rat metabolism studies are used to determine if the
application of additional factors where there is toxicity of metabolites found in the residue chemistry
a small margin between a NOAEL and a serious database (e.g., plant and livestock metabolism studies,
effect, or there is a higher than normal level confined rotational crop studies) could have been
of uncertainty in the derivation of the HBGV. addressed in the toxicity studies using the pesticide
The decision whether to apply an extra factor active ingredient. There is general agreement that if a
is mainly down to expert judgement and can metabolite is present in the rat at 10 percent or more
vary greatly between different organizations. of the systemic dose its toxicity can be considered
However, the magnitude is typically three or five to be covered by the toxicity studies on the parent
and might not be critical to the overall conclusion compound. Studies with in vitro metabolic systems
of the risk assessment. If factors of >10 are permit multiple sampling times and can identify
considered, the level of uncertainty is so great transient metabolites not identified in the in vivo
that a HBGV might not be set. In addition to the studies. However, the in vitro systems are normally
traditional safety factors, the US EPA applies based on hepatic cells or slices and do not cover non-
an additional FQPA (Food Quality Protection hepatic metabolism. There are no agreed approaches
Act) factor of ten to account for sensitivity of to the use of such in vitro studies in the assessment of
susceptible subgroups, especially children. The metabolites found in agricultural commodities.
FQPA factor may be removed or modified based
on available data showing that an alternative
39
Where metabolites are found in crop metabolism 1.4. Exposure assessments

studies, but not at significant levels in rat/in vitro
metabolism studies, further work is necessary to Food consumption data used by FAO/WHO Expert
determine whether the metabolites are toxicologically Committees (JECFA, JMPR etc.) are based on data
relevant. Depending on the amount of the metabolite from countries around the entire world, and collected
in the crop, the requirements can be limited to in in the Global Environmental Monitoring System food
silico methods, threshold of toxicological concern database (GEMS) and/or in the “FAO/WHO Chronic
(TTC), or involve repeat dose toxicity studies. Based Individual Food Consumption – summary statistics”
on the information available, it is unclear to the (CIFOCOss) database. While GEMS food database
author how consistent across different countries and contains average daily per capita consumption
regions the data requirements for crop metabolites data, based on food cluster diets, the CIFOCOss
are. Harmonization of approaches to determining the provides individual food consumption data based on
toxicological relevance of metabolites would help country surveys. National authorities carrying out
reach standardized residue definitions and potentially risk assessment use their own national consumption
minimize animal testing. data. In theory, the national consumption data will be
included in the FAO/WHO databases and therefore be
Metabolism studies in crops should be performed at
covered by the JMPR assessment. This depends on
the critical GAP, or at higher application rates if residue
national authorities supplying updated consumption
levels are low. Studies should be performed on one
data to FAO/WHO.
or more crops from the five main groupings (root,
leafy, fruit, pulses/oil seed, cereals). If data from three For acute intake assessments (IESTI model), there is
of these groups show good agreement, information currently a scientific agreement regarding the use of
on the other two is not required by JMPR, Australia, the highest residue (HR) combined with large portion
Canada or the European Union (1997 ref. annex VI B2- size data as the main intake component. A recent
1). Methodology requirements such as use of multiple probabilistic study conducted by WHO confirmed
label positions, extraction methods and desired that the IESTI is a conservative model for adults and
LOQs (0.01 mg/kg) appear to be common across all children. The study did not show any appreciable risk
countries/region analysed. for exceeding the ARfD based on the available data.
Some old (pre-1990) residue definitions have been For chronic assessments (IEDI model) there is a
compromised by the use of a single radiolabel position difference between US EPA’s and JMPR’s use of the
(e.g. tri-allate; compounds containing two or more supervised trial median residue (STMR) and mean
rings with labile bridging bonds). On cleavage of the consumption data and that of some agencies (EFSA)
molecule the unlabelled moiety cannot be traced, which use STMR and a high percentile consumption
potentially resulting in a residue definition that omits value. This is linked to the way the data for the 17
some significant component(s). If new studies using
GEMS/Food cluster diets are captured. An alternative
multiple label sites are made available during some but
global estimate of chronic dietary exposure (GECDE)
not all subsequent reviews, this has the potential to
model has been developed by JECFA (veterinary drugs)
result in differences in metabolite profiles and residue
using individual food consumption data. Among other
definitions.
things, the advantage of using such data is that it would
It is more likely that major differences in the allow for more refined dietary exposure assessments,
conclusions of crop metabolism studies will be due including for high-end consumers and for different
to differences in the studies and methodological population subgroups (e.g. children). A JECFA/JMPR
aspects such as position of radiolabel and limits of Working Group performed an exercise comparing IEDIs
detection/quantification, rather than differences in the and GECDEs. Dietary exposure estimates from the
assessment of the data. However, determination of high consumer GECDE model were of the same order
the toxicological relevance of crop metabolites appears of magnitude as the highest IEDI cluster estimate for
to be inconsistent across some countries/region the majority of pesticide residues considered in this
analysed. exercise. However, for some subpopulation groups,
the estimated dietary exposure using the GECDE was
higher than the highest IEDI cluster estimate. At the
2018 JMPR Meeting there was discussion on future
40
Risk assessment
work to enhance food consumption data and chronic expected that the Codex MRL will cover all uses in
dietary exposure assessment to be used by JMPR (see major Codex member countries.
section 2.1 of (WHO, 2019b)).
For approval of a pesticide in individual countries/
When providing briefing on Codex, EFSA does not region, residue studies are normally expected to
rely on the JMPR assessment but performs its own be performed in that country or area by covering a
intake assessment using the approaches of the number of sites and if necessary climatic conditions
European Union and the lower of the HGBVs from (e.g. northern and southern areas of the European
the European Union or JMPR. This is positive in that Union). This variation in conditions could potentially
any issues in terms of exposures are related to a have a greater impact on the MRL than variations
risk assessment rather than being based purely on a in the criteria used to determine the critical GAP.
perceived difference in approach or value for the ADI/ However, recent work has suggested that zone-
ARfD. It is unclear if all countries/authorities adopt specific conditions do not have a significant impact on
the same detailed approach. Australia and Japan MRLs (Nguyen et al., 2019).
indicate that risk assessments are performed using
national consumption data and Australia will perform Number of studies required
an assessment using both its own and JMPR HBGVs There are differences in the minimum number of
where the values differ. studies required to be performed at the critical GAP to
support the establishment of an MRL. The OECD has
1.5. Differences in Good described the minimum number of trials that would
Agricultural Practice (GAP) and be needed to support an MRL in all OECD member
selection of critical GAP countries. This information is set out in Appendix XII of
the FAO manual (FAO, 2016).
Data differences
Some countries/authorities have a relatively simple
JMPR will typically evaluate data from a wider range
approach (e.g. 16 trials for a major crop, eight trials
of geographical locations than those submitted to
for a minor crop with the possibility for fewer trials if
individual countries or authorities. This should mean
there is limited variability in results). Others have a
that JMPR receives the most extreme GAPs, which
detailed consideration of the importance of a particular
should give rise to Codex MRLs being as high or
crop to the national diet. This may lead to significant
higher than any others. However, looking at the
differences between closely related countries. For
analysis of the pesticides used on rice (Annex B-2)
example, Canada requires 16 trials for rapeseed,
there are a number of cases (e.g. bentazone,
while the United States of America require only eight;
chlorpyriphos-methyl, difenoconazole) where the
for maize the numbers are five and 20 respectively
MRLs of the European Union and the United States
of America are significantly higher than the Codex (NAFTA, 2005). Therefore, if a Codex MRL is supported
MRL, suggesting that the data of the European Union by the minimum number of trials for one country, it
and the United States of America were probably not might not meet acceptance criteria in other countries.
submitted to JMPR or that there are differences in The JMPR expects a minimum of six to eight trials for
crop groupings used to establish a group MRL. A major crops (with 15 or more recommended).
similar situation was identified for some Australian The European Union commented that because of
MRLs. This may indicate that there can be significant differing legislative provisions, it has more flexibility in
deficiencies in the data available to JMPR, due to
the number of trials required for harmonizing a Codex
incomplete submissions or the availability of new data
MRL than in the assessment of an import tolerance or
since the conclusion of the JMPR review. Companies
application for approval.
or countries submitting data should ensure JMPR
receives all critical GAPs to support the worldwide There appear to be differences in how the concept of
applicability of Codex MRLs. It is expected that there ‘limited variability of results’ is applied to accepting
will be differences in GAPs/MRLs across countries a reduced number of trials. In some instances, it is
or regions for various reasons such as climate and accepted when all values are below the LOQ, but not
pest pressure. However, as JMPR/Codex should accepted when results are tightly grouped around a
be provided with data covering the most GAPs it is value above the LOQ.
41
There are some differences in how many growing critical parameters is <25%. If there is variation in
seasons are required – JMPR will accept just one more than one critical parameter, read across might
season under certain conditions but some other not be possible.
authorities require two.
This seems to be a consistent area.
There is potential for greater harmonization in the
Statistical assessment of data in critical studies
number of trials required to support an MRL.
The application of the OECD calculator, supplemented
Reading across from studies not complying with by expert judgement was identified from the survey
critical GAP responses as a common approach in MRL derivation.
Studies used to support an MRL should be performed However, differences between some national/
according to the critical GAP conditions described on regional MRLs and Codex MRLs have been reported
the label (e.g. worst-case application rate, volume, to be related to the application of different statistical
formulation type, application stage, PHI etc.). There approaches to some elements of the dataset (e.g.
appears to be agreement that the majority of studies sampling error). International agreement on the
should be performed at or above the GAP but there is application of statistical methods used in the setting of
some flexibility for using studies that do not comply MRLs should be sought.
exactly with GAP, provided the variation in one of the
42
PART B
2 Risk management considerations

This chapter reviews various risk management aspects 2.1. Establishment of maximum
that could lead to the establishment of different MRLs. residue limits for specific
The aspects considered are the following: commodities/groups
2.1. Establishment of MRLs for specific
Rather than setting MRLs on a number of closely
commodities / groups
related commodities, so termed ‘group MRLs’ can
2.2. MRL extrapolation rules be set covering a number of similar commodities.
2.3. Application of processing factors Because different pesticides can behave differently
in even closely related commodities, there are
2.4. Application of default MRLs for non-registered
few default rules for using data from one or two
pesticides
commodities to produce a group MRL.
2.5. Risk management factors applied to the
toxicology evaluation The relevant commodities on which to perform trials
2.6. Import tolerances are those that would be expected to give the highest
residues (e.g. based on surface area to volume, water
2.7. Commodity descriptions
content, edible portion). Where data are available on a
Key messages: number of commodities within a group, it is expected
JJ Automatic harmonization with Codex MRLs is not that there should be a relatively small variation (ratio
the norm because such practice is not embedded in of <5; or appropriate statistical methods); and that
national legislations. For non-registered MRLs, the the representative commodity for the group: (i) will
accepted practice is to set a default value, usually be the one that gives the highest MRL, (ii) is a major
at the limit of quantification, or to not establish any contributor to the diet, and (iii) is morphologically
tolerance/MRL. representative of other members of the group.
JJ In the absence of a national MRL, an application
can be made to have an MRL established, which is JMPR uses the Codex classification of groupings. The
termed an import tolerance. The process can vary NAFTA has its own system that has much in common
greatly in both complexity (extent of acceptance of with Codex, but there are some varieties present in
evaluations performed by JMPR) and duration (from the NAFTA groupings, which are not in Codex and vice
several months up to six years), depending on the versa. The European Union has general groupings with
country.
no independent listing of varieties, but appears to use
JJ During the Codex step-process for the development the Codex system. Variations in crop groupings have
and adoption of new Codex pesticide MRLs, only been cited as reasons for reservations being raised on
the European Union actively notifies whenever Codex MRLs. Australia uses the Codex classifications
a Codex MRL is not going to be adopted in the and groupings.
European Union, and provides scientific reasons for
that decision.
There would appear to be some potential to get further
JJ Differences in the time of MRL adoption at Codex harmonization of commodities within groups.
and at national level may entail changes in the
scientific data packages evaluated by the different
2.2. maximum residue limit
authorities. Only Australia, the European Union and
Japan have routine procedures in place to review
extrapolation rules
national MRLs at the time Codex adopts new MRLs.
The concept of extrapolation of MRLs within a crop
JJ There is great inconsistency among the commodity grouping seems to be an accepted approach either
descriptions across different countries/region. by direct application of an MRL from a major crop to
43
a related minor one, or by use of field trials data on 2.4. Application of default
a related commodity to supplement the database of maximum residue limits for
another. However, there appear to be no definitive non-registered pesticides
rules regarding the application of the extrapolation
procedures and it is typically case-by-case, depending The default position in all responders except Australia
on the available information. Aspects that have been and the United States of America is to establish
identified as precluding such an extrapolation include: default MRLs for non-registered products. Europe and
Japan set the default MRL value at 0.01 mg/kg, which
JJ different residue definitions;
in most instances corresponds to the LOQ. Canada
JJ differentdatabases, giving differences in the has a “general MRL” of 0.1 mg/kg (Section B.15.002
appropriate crop to extrapolate from; of (Canada, 2019)). The United States of America
JJ minimum number of trials required by national/ and Australia commented that they do not establish
regional authorities; and by default any tolerances/MRLs for non-registered
JJ differencesin the crops within a group that can be products, so any residue detected for non-registered
used as the basis for extrapolation. pesticides could be considered violative.
2.3. Application of processing In order to remove these default values or establish

factors MRLs, it appears that a specific request (domestic
and/or import tolerance) needs to be made.
The use of default processing factors appears to be
limited in terms of number of examples and applicability For new Codex MRLs, specific requests (import
to major components of the diet. Where processing tolerances) may not be necessary in certain countries/
factors are involved, they are generally calculated region. In the European Union, new Codex MRLs are
case-by-case based on data. Different processing reviewed upon their establishment and the MRLs of
technologies might apply to a commodity in different the European Union are raised to the same level of
countries, resulting in different processing factors. Codex MRLs, if three conditions are fulfilled: (1) that
the European Union sets MRLs for the commodity
Australia calculates processing factors on a case- under consideration, (2) that the current MRL of the
by-case basis using the available data to determine European Union is lower than the Codex MRL, and
residues in the edible portion/derived product of a (3) that the Codex MRL is acceptable to the European
commodity. Union with respect to areas such as consumer
protection, supporting data, and extrapolations.
The European Union does not set MRLs for processed Automatic review of newly adopted Codex MRLs
commodities but uses processing factors in intake takes place also in Japan. In Australia, Codex MRLs
estimates. A number of default factors are incorporated
for pesticides are incorporated into the Australian
into the PRIMO spreadsheet (EFSA, 2019) (many taken
Code based on need, through an open and transparent
from an OECD 2008 citation) but these are mainly for
process. Any country, sponsoring company or other
relatively minor commodities/products.
third party can apply for the incorporation of the
individual Codex MRL into the Code whenever such
In the absence of a harmonized list of processing
a need is identified. This MRL harmonization process
factors within Europe, the European Union is
is free-of-charge to applicants and the needed Codex
developing a database of processing types and
MRL will be incorporated into the Code unless an
processing factors compatible with the EFSA food
estimate of dietary exposure exceeds the HBGVs
classification and description system FoodEx235. In
using Australian food consumption data. In addition,
assessing Codex MRLs, the European Union considers
Australia is considering automatic adoption of
available data on residue levels in peel or inedible
new Codex MRLs (subject to a suitable exposure
skin, but appears to have different criteria to JMPR in
assessment) in the near future.
determining the acceptability of such data.
In cases where Codex MRLs exist, unless a proven

35 See: https://www.bfr.bund.de/en/database_of_processing_ scientific justification is provided, the application of
techniques_and_processing_factors_compatible_with_the_efsa_
food_classification_and_description_system_foodex_2_-202963.html LOQ MRLs rather than Codex MRLs seems to be
44
Risk management
considerations 2 PART B
contrary to the SPS text, as the JMPR considerations In commenting on a draft of this report, the European
of toxicology and residues will have been performed Union stated that these risk management criteria are
according to internationally accepted procedures. for the approval of active substances in the European
Union and are not applied to import tolerance
There is scope for developing criteria for wider assessments that would follow a standard risk
acceptance of Codex MRLs for pesticides not assessment.
registered in specific countries/regions (e.g. large
margins can be demonstrated between predicted 2.6. Import tolerances
exposures and HBGVs).
If a Codex MRL for a pesticide/commodity
2.5. Risk management factors combination has not been harmonized by an importing
applied to the toxicology country, or there is no national MRL in the importing
evaluation country, an application can be made to have an MRL

established. This is termed an import tolerance (NAFTA
The US EPA applies an extra ten-fold factor to the (US + Canada), Japan and the European Union, or
default 100 when establishing HGBVs. This factor Schedule 20 amendment (Australia)). This process
arises from the Food Quality Protection Act (FQPA). can vary greatly in both complexity and duration
Canada has an equivalent approach. The factor is (from several months up to six years), depending
removed entirely or reduced depending on the on the country. There is also inconsistency among
information in the toxicology and exposure databases countries/region regarding the extent of acceptance
for assessing sensitive subgroups, particularly children of evaluations performed by other organizations (e.g.
(USEPA, 2002b). JMPR and other agencies will add JMPR) or national authorities versus performing a de
additional factors if databases are lacking, not as a novo evaluation:
default. It is also possible that an element of the FQPA
JJ The Australian guidance indicates a willingness to
factor could remain due to uncertainties in the non-
accept JMPR and/or other national evaluations in
dietary exposure, which would not be relevant to MRL
whole or in part (FSANZ, 2018).
considerations.
JJ The NAFTA guidance indicates that the presence of
a JMPR evaluation could result in a less extensive
Australia and Japan appear to have no further risk
procedure (NAFTA, 2005).
management requirements precluding acceptance of
Codex MRLs, if the national risk assessment of the JJ In the European Union and Japan, a full, new
pesticide (or its residues) has a satisfactory outcome assessment appears to be required even if other
and the biosecurity/quarantine requirements are met. evaluations are available. When considering setting
MRLs under an import tolerance application, the
European Union uses the GAP and supporting
In the European Union, any pesticide that is classified
information from the notifier. The supporting
as Cat 1A or 1B for reproduction or carcinogenicity
information is assessed against the same
or mutagenicity, or is considered to have endocrine
acceptance criteria (e.g. number of trials, sampling
disrupting properties, cannot be approved unless
protocols etc.) as would be used for a standard
exposure is ‘negligible’ (3.6 of Annex II of EC MRL evaluation of the European Union.
1007/2009). The negligible exposure applies to both
dietary and non-dietary exposures. It is possible that a The re-evaluation of studies and data adds significantly
negligible exposure could be demonstrated for dietary to the cost and time of concluding on the import
exposure but no approval would be given based on tolerance. If non-harmonization of Codex MRLs cannot
non-dietary exposures. be resolved, it would be a benefit to trade if agreement
could be reached on a common approach to import
The European Union also has a number of tolerances. In particular, the level of acceptance of
environmental criteria which preclude approval of a evaluations performed by other authorities using
pesticide, e.g. if it is a persistent organic pollutant, accepted risk assessment procedures.
failing the environmental assessment for groundwater
(0.1ug/L).
45
2.7. Commodity descriptions Japan37

JJ Rice (brown rice)
There is some inconsistency in the way commodities
JJ Rice milled
are described in respect of the MRLs. This is true
both within individual schemes and between them. The United States of America
The differences highlighted during the comparisons
Within the United States of America MRLs databases38
performed on rice pesticides are reported below as an
((Bryant Christie Inc., 2019); (e-CFR, 2019)) there was
example of this variation across countries/region36:
again great variation in the commodity descriptions:
Codex JJ Rice (Oryza sativa)

Within the Codex database (Codex, 2019), MRLs JJ Rice post-harvest
for rice were found with the following commodity JJ Wild rice (Zizania aquatica)
descriptions: JJ Rice, wild, grain
JJ Rice JJ Rice, wild, grain, post-harvest
JJ Rice polished JJ Rice grain
JJ Rice husked JJ Rice, grain, post-harvest
JJ Wild Rice JJ Rice polished
JJ Cereal grains JJ Rice polished post-harvest
JJ Grain, cereal, group 15
In some instances, the MRLs were the same for two
or more of the descriptions, in other cases there were As with Codex, there was great variability in how many
marked differences (e.g. dichlorvos). These values rice related MRLs were set for any individual pesticide.
and descriptions would presumably have related to
There was inconsistency between the commodity
the data considered during the evaluations of the
descriptions for individual pesticides across different
pesticide. However, there was great variability, with
countries/region. For example, for sulfuryl fluoride,
different pesticides having between one and four of
there are: Codex MRLs for rice polished and rice
the descriptions.
husked; but the United States of America has
MRLs for rice grain and rice polished, plus three
Australia
specified processed fractions plus a generic one for
Within Schedule 20 (Australia, 2018) there are four commodities not otherwise listed.
different commodity descriptions relating to rice:
In general, due to variation in rice classifications across
JJ Rice countries, it was not easy to match the different
JJ Rice, husked Codex rice commodity types with the national ones.
JJ Rice, polished Much investigation work and interactions/clarifications
JJ Cereal grains between FAO and the countries/region analysed were
necessary to understand how to match national rice
Canada
MRLs properly to Codex rice MRLs.
JJ Rice
Although the above is only based on rice, similar
JJ Wild rice
issues also apply to other commodities such as cereal
The European Union grains, nuts, grapes (e.g. table, wine, dried, raisins),
The data of the European Union were all presented as pulses (e.g. with pods, without pods, dried) etc.
‘rice’, covering primarily husked rice, but also a number
of types and forms of rice (European Union, 2019b).
37 In Japan, rice is not normally sold in a highly processed form hence
[Within the PRIMO intake calculation software (EFSA, the unexpectedly low number of descriptions for rice.
2019) a default processing factor of 0.4 can be applied 38 US MRLs were consulted both in the official US Electronic Code
of Federal Regulations (https://www.ecfr.gov/cgi-bin/retrieveEC-
for milling or polishing of rice]. FR?g-p=&SID=b4b3755e18a86aaeab34590c6eb9b576&mc=true
&n=pt40.26.180&r=PART&ty=HTML), and in the Bryant Christie
database (run by a commercial organization but linked from the
USDA website: https://www.fas.usda.gov/maximum-residue-lim-
its-mrl-database). When discrepancies existed between the two
sources, preference was given to the US Electronic Code of Feder-
36 See also Part A: Annex A-2 and Chapter 2.3. al Regulations.
46
Risk management
considerations 2 PART B
It is unclear why there is so much variation within Whether these differences in commodity descriptions
schemes. Some are due to the availability of data on can result in rejection of Codex MRLs was not
different fractions of rice but it is unclear why there investigated in this project. However, to reduce the
are such variations in the available data. If MRLs are potential for confusion over multiple MRLs for different
required (available) for multiple rice fractions for one forms of a single pesticide/crop combination, it would
pesticide, why does another pesticide only have one appear logical to look to improve consistency where
rice MRL? It seems very unlikely that the subsequent possible, i.e. set multiple MRLs for all pesticides to
treatment of harvested rice will vary greatly depending cover all likely fractions, or reduce the number of
on which pesticides have been applied. MRLs for some pesticide/crop combinations.
47
© Pan Species
PART B
3 Conclusions and way forward

The feedback obtained from the five markets analysed, to review national MRLs at the time Codex adopts
together with the additional information sources new MRLs. The United States of America and Canada
used in the analysis, were not sufficient to provide review MRLs as part of a national active substance
a complete picture for the reasons behind lack of review programme.
harmonization with Codex MRLs. Despite these
In the absence of a national MRL in the importing
limitations, some conclusions could be drawn and
country, an application can be made to have an MRL
some of the findings are reported below.
established, which is termed an import tolerance. The
process can vary greatly in both complexity (extent
Risk management considerations
of acceptance of evaluations performed by JMPR)
The analysis showed that automatic harmonization
and duration (from several months up to six years),
with Codex MRLs is not the norm because such
depending on the country. The re-evaluation of studies
practice is not embedded in national legislations. For
and data adds significantly to the time and cost of
non-registered pesticides and non-registered uses in
concluding on the import tolerance.
specific countries/region, the default practice is to set
a default value, usually at the limit of quantification, There is great inconsistency between the commodity
or to not establish any tolerance/MRL, which results descriptions for individual pesticides across different
in zero tolerance. In these cases, specific applications countries/region. While this study only considered rice,
(with varying levels of detail) are required to establish the same is true also for other commodities. It was not
a specific MRL or support harmonization with Codex investigated whether these differences in commodity
MRLs. descriptions can result in rejection of Codex MRLs.
During the Codex step-process for the development Risk assessment methodology
and adoption of new Codex pesticide MRLs, only the A considerable variation was perceived in how
European Union actively notifies whenever a Codex countries/region are aligned with the JMPR/Codex
MRL is not going to be adopted in the European Union, process for the development and establishment
and provides the (scientific) reasons for not adopting of pesticide MRLs. Australia is closely aligned, and
the Codex MRL. Such information was found very although not automatically harmonizing with Codex
informative in terms of preparing this report, but was MRLs, indicates it is willing to accept at least a
not identified for any of the other countries analysed. proportion of the conclusions of JMPR assessments
Some authorities do not harmonize with Codex MRLs as part of an application. Canada, the European Union,
pending the conclusion of ongoing reviews, some of Japan and the United States of America, although
which can take several years to finalize. This is a one- having a number of aspects of their procedures that
sided position because existing ‘national’ MRLs are are closely aligned, perform an extensive independent
not suspended during the review. evaluation of the study reports according to their own
national criteria.
Differences in the time of MRL adoption at Codex and
at national level may entail changes in the scientific Many of the differences in risk assessments do not
data packages evaluated by the different authorities seem to have a significant impact on the overall
(e.g. different GAPs reflecting different pests and outcomes. The review of pesticides used on rice, as
diseases, different pesticide labels, availability of well as the experience, showed that the health-based
different studies, etc.). While Australia, the European guidance values typically vary within a factor of +/-2. In
Union and Japan review Codex MRLs upon their addition, for a number of pesticides used on rice, the
establishment, Canada and the United States of JMPR intake estimates were well below the applicable
America report to have no routine procedures in place HBGVs, leaving a significant margin for differences in
49
HBGVs and consumption data to be accommodated drafted 10 to 20 years ago, with occasional updates
within an acceptable risk assessment. and partial revisions subsequently. Consideration could
be given to an update process that can be agreed
What is unclear is whether having a slightly different
internationally. With differing legislative constraints
ADI/ARfD or consumption value within the risk
there will not be complete agreement. However,
assessment is sufficient to trigger a reservation
there are a number of areas where there are relatively
regarding a Codex MRL – even if the risk assessment
minor variations in approaches and an internationally
is/would be satisfactory based on national/regional
acceptable approach might be determined, which
criteria.
could facilitate the acceptance of Codex MRLs.
In only a very few cases there were large (ten fold or
It is also worth recognizing the ongoing work to
more) differences between JMPR ADIs/ARfDs and
address some of the issues that might lead to
those of other countries/region analysed.
deferring MRLs. For example, this includes the work
From the limited review of pesticide MRLs for underway at OECD, jointly with JMPR and JECFA,
rice, there are indications that some of the major on the development of a guidance document to
differences in MRLs and residue definitions are due standardize residue definition approaches, or the
to the presentation of different data to the different efforts to improve harmonization on chronic exposure
countries/regions and authorities. Some of the assessments between JMPR and JECFA, by using
differences in residue definitions are due to different individual food consumption data.
forms of a pesticide being available (e.g. racemic
The following points were drawn from the general
versus resolved isomer preparations), which have no
findings of the analysis on the reasons behind different
or minimal impact on the risk assessment. The reasons
levels of harmonization, as a way forward to facilitate
behind the differences in data availability are varied,
further discussion on how to improve international
including different data requirements, availability
harmonization:
of data to different companies and differences in
timings of reviews. Increased harmonization of review 1. When not subject to national legal requirements
timetables could overcome some, but not all of these. that govern the use of a default MRL value, in line
with the WTO SPS Agreement, harmonization with
Furthermore, national authorities conduct exposure
Codex MRLs should be sought as a default position
assessments using their own consumption data, while
(because the JMPR assessments are performed
JMPR assessments are based on data from countries
to internationally accepted standards), while non-
around the world, as submitted via the GEMS food
harmonization should be supported by scientific
system or CIFOCOss database. Harmonization thus
arguments39. Where legal requirements restrict
also depends on national authorities supplying updated
harmonization with Codex MRLs, the reason for the
consumption data to FAO/WHO.
legal decision should be reviewed.
The use of the OECD MRL calculator tool, with
2. For increased transparency, it would help if
the addition of expert judgement, is common to all
Codex member countries actively notify CCPR
countries/region analysed. However, inputs to the
of reservations, or non-harmonization on Codex
OECD calculator are not fully harmonized (e.g. crop
MRLs, providing background information for non-
groupings), which could lead to different MRLs.
harmonization with newly developed Codex MRLs
National GAPs are used when setting national MRLs, (along the lines of that provided by the European
as enforcement tools for compliance with nationally Union when they do not harmonize with a Codex
registered pesticide uses (label information), with limited MRL). This information could then be evaluated with
if any consideration of other GAPs, other than those a view to identifying and resolving common areas
which are part of import tolerance applications. On the of concern.
other hand, the objective of Codex MRLs is to support
3. Bodies such as OECD and/or JMPR and/or Codex
international trade of products and, as such, they are
could consider improving harmonization/consistency
based on the highest residues of international GAPs.
in the following areas:
Many of the guidance/procedural documents, related
to MRLs and human health risk assessments of
39 WTO SPS Agreement, Art. 3: https://www.wto.org/english/
pesticides identified during this review were initially tratop_e/sps_e/spsagr_e.htm
50
Conclusions and way
forward 3 PART B
a. crop groupings; 4. As also indicated in some of the feedback received
b. extrapolation of data between crops/ by interested parties, countries could notify CCPR/
commodities; JMPR Secretariat if they receive new data that casts
c. number of field trials required to support an MRL significant doubt on the continuing validity of an
for a particular crop; existing Codex MRL. For example, results of a new
d. statistical methods applied to field trial data; crop metabolism study using additional radiolabel
e. consideration of isomers (and other minor positions, or significantly adverse toxicity data.
differences in chemical structure e.g. esters) in
5. If non-harmonization of Codex MRLs cannot be
residue definitions;
resolved, it would be a benefit to trade if agreement
f. the necessity for setting MRLs on different parts
could be reached on a common approach to import
of a crop (e.g. straw, grain, husked grain, bran);
tolerances. In particular, the level of acceptance of
g. use of existing reviews in the evaluation of
evaluations performed by other authorities using
applications for import tolerances;
accepted risk assessment procedures.
h. data requirements for crop metabolites not
present in mammalian metabolism; 6. Findings of these analyses could be used to
i. criteria for deciding if an ARfD is required or not; stimulate an international discussion to identify
and reasons behind non-harmonization with Codex
j. impact of risk management practices, especially MRLs and steps to improve the situation.
those not directly related to dietary risk
assessment.
51
© Eduardo Prim
References
Anderson, J.E. & van Wincoop, E. 2003. Gravity with Cadot, O., Gourdon, J. & van Tongeren, F. 2018.
Gravitas: A Solution to the Border Puzzle. American Estimating Ad Valorem Equivalents of Non-Tariff
Economic Review, 93(1): 170–192. (also available at: Measures: Combining Price-Based and Quantity-
https://doi.org/10.1257/000282803321455214) Based Approaches. In: OECD Trade Policy Papers.
[online]. OECD Publishing, Paris. [Cited 29 July
Asia-Pacific Economic Cooperation (APEC). 2016.
2020]. http://www.oecd.org/officialdocuments/
Import MRL guideline for pesticides. A guideline
publicdisplaydocumentpdf/?cote=TAD/TC/
on possible approaches to achieve alignment of
WP(2017)12/FINAL&docLanguage=En
international MRLs. In: Asia Pacific Economic
Cooperation [online]. Singapore, APEC Food Safety Canada. 2014. Minister’s Response to Petition 364
Cooperation Forum Sub-Committee on Standards from Mr Leroy Lowe, 1 October 2014. In: Office
and Conformance. [Cited 29 July 2020]. https:// of the Auditor General of Canada. [online]. [Cited
www.apec.org/Publications/2016/08/Import-MRL- 29 July 2020]. http://www.oag-bvg.gc.ca/internet/
Guideline-for-Pesticides English/pet_364_e_39690.html
Australia. 2018. Australia New Zealand Food Canada. 2019. Food and Drug Regulation (C.R.C,
Standards Code – Schedule 20 – Maximum residue c870). Division 15, Adulteration of Food. In: Justice
limits. F2018C00914. In Federal Register Legislation Law Website. [online]. Canada. [Cited 29 July
[online]. Australia. [Cited 29 July 2020]. https:// 2020]. https://laws.justice.gc.ca/eng/regulations/
www.legislation.gov.au/Details/F2018C00914 c.r.c.,_c._870/page-74.html#docCont
Australian Pesticides and Veterinary Medicines Chen, C., Yang, J. & Findlay, C. 2008. Measuring
Authority (APVMA). 2019. Health based the Effect of Food Safety Standards on China’s
guidance values. In Australian Pesticides and Agricultural Exports. Review of World Economics,
Veterinary Medicines Authority [online]. Australian 144(1): 83–106.(also available at: https://doi.
Government. [Cited 29 July 2020]. https://apvma. org/10.1007/s10290-008-0138-z)
gov.au/node/26596
Disdier, A.-C. & Marette, S. 2010. The Combination
Bryant Christie Inc. 2019. BCGlobal - Global pesticide of Gravity and Welfare Approaches for Evaluating
MRL database. In Bryant Christine Inc. [online]. Nontariff Measures. American Journal of
[Cited 29 July 2020]. https://www.globalmrl.com Agricultural Economics, 92(3): 713–726.(also
available at: https://doi.org/10.1093/ajae/aaq026)
Buzby, J.C. 2001. Effects of Food-Safety Perceptions
on Food Demand and Global Trade. In: A. Regmi, ed. Drogué, S. & DeMaria, F. 2012. Pesticide residues
Changing Structure of Global Food Consumption and trade, the apple of discord? Food Policy, 37(6):
and Trade. Market and Trade Economics Division, 641–649. (also available at: https://doi.org/10.1016/j.
Economic Research Service, U.S. Department of foodpol.2012.06.007)
Agriculture, Agriculture and Trade Report. WRS-01-1,
e-CFR. 2019. Electronic Code of Federal Regulations.
pp. 55-66. (also available at: https://www.ers.usda.
U.S. Government. In: Electronic Code of Federal
gov/webdocs/outlooks/40303/14978_wrs011i_1_.
Regulations. [online]. [Cited 29 July 2020]. https://
pdf?v=723.8)
www.ecfr.gov/cgi-bin/retrieveECFR?gp=&SID=b-
4b3755e18a86aaeab34590c6eb9b576&mc=t-
rue&n=pt40.26.180&r=PART&ty=HTML
53
European Food Safety Agency (EFSA). 2018. FAO & WHO. 1971–2018a. CCPR6 to CCPR50 -
Scientific support for preparing an EU position in the Reports from the Codex Committee on Pesticide
50th session of the Codex Committee on Pesticide Residues (CCPR). In: Codex Alimentarius -
Residues (CCPR). EFSA Journal, 16(7), 5306. (also International food standards. [online]. [Cited
available at: https://doi.org/10.2903/j.efsa.2018.5306) 29 July 2020]. http://www.fao.org/fao-who-
codexalimentarius/committees/committee/related-
European Food Safety Agency (EFSA). 2019. PRIMo
meetings/en/?committee=CCPR
– Pesticide Residue Intake Model. In: European
Food Safety Authority. [online]. [Cited 29 July FAO & WHO. 1971–2018b. CAC08 to CAC41 - Reports
2020]. https://www.efsa.europa.eu/en/applications/ of Sessions of the joint FAO/WHO food standards
pesticides/tools programme Codex Alimentarius Commission. In:
Codex Alimentarius - International food standards.
European Food Safety Agency (EFSA). 2016.
[online]. [Cited 29 July 2020]. http://www.fao.
Guidance on the establishment of the residue
org/fao-who-codexalimentarius/committees/cac/
definition for dietary risk assessment. EFSA
meetings/en/
Journal, 14(12), 4549. (also available at: https://doi.
org/10.2903/j.efsa.2016.4549) FAO & WHO. 1995. CODEX Standard for Rice 198-
1995 (Adopted in 1995. Amended in 2019.). Rome,
European Union. (EU). 2005. REGULATION
Italy: FAO. In: Codex Alimentarius - International
(EC) No 396/2005 of the European parliament
food standards. [online]. [Cited 29 July 2020]. http://
and of the council of 23 February 2005 on
www.fao.org/fao-who-codexalimentarius/sh-proxy/
maximum residue levels of pesticides in or on
en/?lnk=1&url=https%253A%252F%252Fwork-
food and feed of plant and animal origin and
space.fao.org%252Fsites%252Fcodex%252F-
amending Council Directive 91/414/EEC. In:
Standards%252FCXS%2B198-1995%252FCX-
EUR-Lex. [online]. [Cited 29 July 2020]. https://
S_198e.pdf
eur-lex.europa.eu/LexUriServ/LexUriServ.
do?uri=CONSLEG:2005R0396:20121026:EN:PDF FAO & WHO. 2006. Codex Classification of Foods and
Animal Feeds - Draft Revision. Rome, Italy: FAO. In:
European Union. (EU). 2009. Regulation (EC)
Codex Alimentarius - International food standards.
No 1107/2009 of the European Parliament and
[online]. [Cited 29 July 2020]. http://www.fao.org/
of the Council of 21 October 2009 concerning
tempref/codex/Meetings/CCPR/ccpr38/pr38CxCl.
the placing of plant protection products on the
pdf
market and repealing Council Directives 79/117/
EEC and 91/414/EEC. OJ L 309, 24.11.2009, FAO & WHO. 2017. Report of the 49th session of the
p. 1–50. In: EUR-Lex. [online]. [Cited 29 July codex committee on pesticide residues. Beijing, P.R.
2020]. https://eur-lex.europa.eu/legal-content/EN/ China, 24 - 29. In: Codex Alimentarius. [online]. [Cit-
TXT/?uri=celex%3A32009R1107 ed 29 July 2020]. http://www.fao.org/fao-who-co-
dexalimentarius/sh-proxy/en/?lnk=1&url=https%
European Union. (EU). 2019a. Search active
253A%252F%252Fworkspace.fao.org%252F-
substances. In: PLANTS - EU Pesticides database.
sites%252Fcodex%252FMeetings%252F-
[online]. European Comission. [Cited 29 July 2020].
CX-718-49%252FREPORT%252FREP17_PRe.pdf
https://ec.europa.eu/food/plant/pesticides/eu-
pesticides-database/public/?event=activesubstance. FAO & WHO. 2018. Report of the 50th session of the
selection&language=EN codex committee on pesticide residues. Haikou,
P.R. China, 9 - 14 April 2018. In: Codex Alimentari-
European Union. (EU). 2019b. Crop groupings
us. [online]. [Cited 29 July 2020]. http://www.fao.
– Rice. In: PLANTS - EU Pesticides database.
org/fao-who-codexalimentarius/sh-proxy/en/?lnk
[online]. European Comission. [Cited 29 July
=1&url=https%253A%252F%252Fworkspace.
2020]. https://ec.europa.eu/food/plant/pesticides/
fao.org%252Fsites%252Fcodex%252FMeet-
eu-pesticides-database/public/?event=product.
ings%252FCX-718-50%252FREPORT%252FFI-
displayAll&language=EN&selectedID=237
NAL%252520REPORT%252FREP18_PRe.pdf
54
References
FAO & WHO. 2019. Codex Alimentarius Commission Grant, J. & Arita, S. 2017. Sanitary and Phyto-
Procedural Manual, 27th edition. Rome, Italy: FAO. Sanitary Measures: Assessment, Measurement,
In: Codex Alimentarius. [online]. [Cited 29 July 2020]. and Impact. In: IATRC Commissioned Paper 21.
http://www.fao.org/3/ca2329en/CA2329EN.pdf [online]. [Cited 29 July 2020]. https://iatrc.umn.edu/
sanitary-and-phyto-sanitary-measures-assessment-
FAO & WTO. 2017. Trade and food standards. Trade and
measurement-and-impact/
food standards. (also available at: https://www.wto.
org/english/res_e/booksp_e/tradefoodfao17_e.pdf) Greene, W.H. 2002. Econometric Analysis. 5th edition.
Upper Saddle River, N.J, Prentice Hall. 1026 pp.
FAO. 2006. Rice International Commodity Profile.
Rome. In: Food and Agriculture Organization of the Head, K., Mayer, T. & Ries, J. 2010. The erosion
United Nations. [online]. [Cited 29 July 2020]. http:// of colonial trade linkages after independence.
www.fao.org/fileadmin/templates/est/COMM_ Journal of International Economics, 81(1):
MARKETS_MONITORING/Rice/Documents/Rice_ 1–14. (also available at: https://doi.org/10.1016/j.
Profile_Dec-06.pdf jinteco.2010.01.002)
FAO. 2016. FAO Manual on the Submission and Heckman, J. 1979. Sample selection bias as a
Evaluation of Pesticide Residues Data for the specification error. Econometrica, 47: 153–
Estimation of maximum residue levels in food 161. (also available at: https://www.jstor.org/
and feed, 3rd Edition. In: FAO Plant production stable/1912352?seq=1#metadata_info_tab_contents)
and protection, NSP - JMPR Guidance and related
Henry de Frahan, B. & Nimenya, N. 2013. 9 Trade
documents. [online]. Rome. [Cited 29 July 2020].
Effects of Private and Public European Food Safety
http://www.fao.org/agriculture/crops/thematic-
Standards on Horticultural Imports from Kenya. In
sitemap/theme/pests/jmpr/jmpr-docs/en/
J. C. Beghin, ed. Nontariff Measures with Market
Feenstra, R.C. 2015. Advanced International Trade: Imperfections: Trade and Welfare Implications, pp.
Theory and Evidence - Second Edition. Princeton, 215–243. Frontiers of Economics and Globalization.
Princeton University Press. 496 pp. Emerald Group Publishing Limited. (also available at:
https://doi.org/10.1108/S1574-8715(2013)0000012014)
Feenstra, R.C., Markusen, J.R. & Rose, A.K. 2001.
Using the gravity equation to differentiate among Japan, Ministry of Health, Labour and Welfare.
alternative theories of trade. Canadian Journal 2006. Introduction of the Positive List System
of Economics/Revue canadienne d’économique, for Agricultural Chemical Residues in Foods. In:
34(2): 430–447. (also available at: https://doi. Ministry of Health, Labour and Welfare. [online].
org/10.1111/0008-4085.00082) Japan. [Cited 29 July 2020]. http://www.mhlw.
go.jp/english/topics/foodsafety/positivelist060228/
Food Standards Australia New Zealand (FSANZ).
introduction.html
2016. Approval report – Proposal P1027, Managing
Low-level Ag & Vet Chemicals without MRLs. Kareem, F.O., Martínez-Zarzoso, I. & Brümmer,
Australia. In: Food Standards. [online]. Australia - B. 2018. Protecting health or protecting imports?
New Zealand. [Cited 29 July 2020]. https://www. Evidence from EU non-tariff measures. International
foodstandards.gov.au/code/proposals/Documents/ Review of Economics & Finance, 53: 185–202. (also
P1027%20Low%20level%20Ag%20and%20 available at: https://doi.org/10.1016/j.iref.2017.08.012)
Vet%20Chems%20AppR.pdf
Li, Y. & Beghin, J.C. 2012. A meta-analysis of
Food Standards Australia New Zealand (FSANZ). estimates of the impact of technical barriers to
2018. Guide to submitting requests for maximum trade. Journal of Policy Modeling, 34(3): 497–
residue limit (MRL) harmonisation proposals. In: 511. (also available at: https://doi.org/10.1016/j.
Food Standards. [online]. Australia - New Zealand. jpolmod.2011.11.001)
[Cited 29 July 2020]. https://www.foodstandards.
Li, Y. & Beghin, J.C. 2014. Protectionism indices
gov.au/publications/Documents/Guide%20to%20
for non-tariff measures: An application to
submitting%20requests%20for%20maximum%20
maximum residue levels. Food Policy, 45:
residue%20limit.pdf
57–68. (also available at: https://doi.org/10.1016/j.
foodpol.2013.12.005)
55
Maertens, M. & Swinnen, J.F.M. 2009. Trade, Ortega, D.L., Hong, S.J., Wang, H.H., & Wu, L. 2016.
Standards, and Poverty: Evidence from Senegal. Emerging markets for imported beef in China:
World Development, 37(1): 161–178. (also available Results from a consumer choice experiment in
at: https://doi.org/10.1016/j.worlddev.2008.04.006) Beijing. Meat Science, 121, 317-323. (also available
at: https://doi.org/10.1016/j.meatsci.2016.06.032)
MAST (Multi-agency support team). 2008. First
Progress Report to the Group of Eminent Persons Ortega, D.L. & Tschirley, D.L. 2017. Demand for food
on Non-tariff Barriers. Geneva: Mimeo, UNCTAD. safety in emerging and developing countries: A
research agenda for Asia and Sub-Saharan Africa.
Melo, O., Engler, A., Nahuehual, L., Cofre, G. &
Journal of Agribusiness in Developing and Emerging
Barrena, J. 2014. Do Sanitary, Phytosanitary, and
Economies, 7(1): 21–34. (also available at: https://
Quality-related Standards Affect International
doi.org/10.1108/JADEE-12-2014-0045)
Trade? Evidence from Chilean Fruit Exports. World
Development, 54: 350–359. (also available at: Peterson, H.H. & Yoshida, K. 2004. Quality
https://doi.org/10.1016/j.worlddev.2013.10.005) Perceptions and Willingness-to-Pay for Imported
Rice in Japan. Journal of Agricultural and Applied
Mittelhammer, R.C., Judge, G.G. & Miller, D.J. 2000.
Economics, 36(1): 123–141. (also available at: https://
Econometric Foundations. New York, Cambridge
doi.org/10.1017/S1074070800021908)
University Press. 784 pp.
Pham, H.V. & Dinh, T.L. 2020. The Vietnam’s food
Nguyen, J., Tiu, C., Stewart, J., & Miller, D. 2019.
control system: Achievements and remaining
Global Zoning and Exchangeability of Field Trial
issues. Food Control, 108: 106862. (also available at:
Residues Between Zones: Are There Systematic
https://doi.org/10.1016/j.foodcont.2019.106862)
Differences in Pesticide Residues Across
Geographies? Statistics and Public Policy, 6(1), 14- Reardon, T., Echeverria, R., Berdegué, J., Minten,
23. Please put journal name into italics (Statistics B., Liverpool-Tasie, S., Tschirley, D. & Zilberman,
and Public Policy) https://doi.org/10.1080/233044 D. 2019. Rapid transformation of food systems
3X.2018.1555068 in developing regions: Highlighting the role of
agricultural research & innovations. Agricultural
North American Free Trade Agreement (NAFTA).
Systems, 172: 47–59. (also available at: https://doi.
2005. NAFTA Guidance Document on Data
org/10.1016/j.agsy.2018.01.022)
Requirements for Tolerances on Imported
Commodities in the United States and Canada. Rimal, A., Fletcher, S.M., McWatters, K.H., Misra,
US Environmental Protection Agency, Office of S.K. & Deodhar, S. 2001. Perception of food
Pesticide Programs (OPP) and Health Canada, Pest safety and changes in food consumption habits:
Management Regulatory Agency (PMRA). In: US a consumer analysis. International Journal of
Environmental Protection Agency Office of Pesticide Consumer Studies, 25(1): 43–52. (also available at:
Programs and Health Canada Pest Management https://doi.org/10.1111/j.1470-6431.2001.00162.x)
Regulatory Agency. [online]. [Cited 29 July 2020].
Santeramo, F.G. & Lamonaca, E. 2019. The Effects of
https://www.epa.gov/sites/production/files/2015-10/
Non-tariff Measures on Agri-food Trade: A Review
documents/nafta-guidance.pdf
and Meta-analysis of Empirical Evidence. Journal
Nuttavuthisit, K. & Thøgersen, J. 2019. Developing- of Agricultural Economics, 70(3): 595–617. (also
Economy preferences for imported organic food available at: https://doi.org/10.1111/1477-9552.12316)
products. Journal of International Consumer
Schlueter, S.W., Wieck, C. & Heckelei, T. 2009.
Marketing, 31(3), 225-249. (also available at: https://
Regulatory Policies in Meat Trade: Is There Evidence
doi.org/10.1080/08961530.2018.1544529)
for Least Trade-distorting Sanitary Regulations?
Okpiaifo, G., Durand-Morat, A., West, G.H., Nalley, American Journal of Agricultural Economics, 91(5):
L.L., Nayga, R.M. & Wailes, E.J. 2020. Consumers’ 1484–1490. (also available at: https://doi.org/10.1111/
preferences for sustainable rice practices in Nigeria. j.1467-8276.2009.01369.x)
Global Food Security, 24: 100345. (also available at:
https://doi.org/10.1016/j.gfs.2019.100345)
56
References
Schmidt, J. & Steingress, W. 2019. No double U.S. Environmental Protection Agency (USEPA).
standards: Quantifying the impact of standard 2000. Science Policy Council Handbook – Risk
harmonization on trade. Bank of Canada Staff Characterisation. In: Environmental Protection
Working Paper, No. 2019-36, Bank of Canada, Agency. [online]. Washington DC: U.S. [Cited 29
Ottawa July 2020]. https://www.epa.gov/sites/production/
files/2015-10/documents/osp_risk_characterization_
Serrano -Arcos, M.M., Sánchez-Fernández, R. &
handbook_2000.pdf
Pérez-Mesa, J.C. 2019. Is There an Image Crisis
in the Spanish Vegetables? Journal of International U.S. Environmental Protection Agency (USEPA).
Food & Agribusiness Marketing, 1-19. (also available 2002a. A review of the reference dose and
at: https://doi.org/10.1080/08974438.2019.1599759) reference concentration process. Prepared for
the Risk assessment forum. In: Environmental
The Department of Economic and Social Affairs
Protection Agency. [online]. Washington DC: U.S.
of the United Nations (UN-DESA). 2018.
[Cited 29 July 2020]. https://www.epa.gov/sites/
Sustainable Development Goal 17: Strengthen the
production/files/2014-12/documents/rfd-final.pdf
means of implementation and revitalize the global
partnership for sustainable development. Targets U.S. Environmental Protection Agency (USEPA).
& indicators. In: Department of Economic and 2002b. Determination of the appropriate FQPA
Social Affairs - Sustainable Development. [online]. safety factor(s) in tolerance assessment. Office
United Nations. [Cited 29 July 2020]. https:// of Pesticide Programs, USEPA. In: Environmental
sustainabledevelopment.un.org/sdg17 Protection Agency. [online]. Washington DC: U.S.
[Cited 29 July 2020]. https://www.epa.gov/sites/
Thilmany, D.D. & Barrett, C.B. 1997. Regulatory
production/files/2015-07/documents/determ.pdf
Barriers in an Integrating World Food Market.
Applied Economic Perspectives and Policy, Winchester, N., Rau, M.-L., Goetz, C., Larue, B.,
19(1): 91–107. (also available at: https://doi. Otsuki, T., Shutes, K., Wieck, C., Burnquist, H.L.,
org/10.2307/1349680) Souza, M.J.P. de & Faria, R.N. de. 2012. The Impact
of Regulatory Heterogeneity on Agri-food Trade. The
UNCTAD & the World Bank. (2018). The Unseen
World Economy, 35(8): 973–993. (also available at:
Impact of Non-Tariff Measures: Insights from a new
https://doi.org/10.1111/j.1467-9701.2012.01457.x)
database. In: United Nations Conference on Trade
and Development and the World Bank [online]. World Health Organization (WHO). 2005. IPCS
[Cited 29 July 2020]. (also available at: https:// - Chemical-specific adjustment factors for
unctad.org/en/PublicationsLibrary/ditctab2018d2_ interspecies differences and human variability:
en.pdf) guidance document for use of data in dose/
concentration–response assessment. In: World
UNCTAD & WTO. 2012. A Practical Guide to Trade
Health Organization, International Programme
Policy Analysis. United Nations and World Trade
on Chemical Safety. [online]. Geneva. [Cited
Organization. In: United Nations Conference on
29 July 2020]. http://whqlibdoc.who.int/
Trade and Development and The World Bank.
publications/2005/9241546786_eng.pdf
[online]. [Cited 29 July 2020]. https://vi.unctad.org/
tpa/web/docs/vol1/book.pdf World Health Organization (WHO). 2019a. Inventory
of evaluations performed by the Joint Meeting
UNCTAD & WTO. 2016. An Advanced Guide to Trade
on Pesticide Residues (JMPR). In: World Health
Policy Analysis: The Structural Gravity Model. United
Organization. [online]. [Cited 29 July 2020]. http://
Nations and World Trade Organization. In: United
apps.who.int/pesticide-residues-jmpr-database
Nations Conference on Trade and Development
and The World Bank. [online]. [Cited 29 July 2020]. World Health Organization (WHO). 2019b. Pesticide
https://vi.unctad.org/tpa/web/docs/vol2/book.pdf residues in food 2018 - Report 2018. Joint FAO/
WHO Meeting on Pesticide Residues. In: FAO
Unnevehr, L. 2015. Food safety in developing
Plant Production and Protection. [online]. Rome,
countries: Moving beyond exports. Global Food
Italy. [Cited 29 July 2020]. http://www.fao.org/
Security, 4: 24–29. (also available at: https://doi.
documents/card/en/c/CA2708EN/
org/10.1016/j.gfs.2014.12.001)
57
World Trade Organization (WTO). 1995. The WTO Xiong, B. & Beghin, J. 2012. Does European aflatoxin
Agreement on the Application of Sanitary and regulation hurt groundnut exporters from Africa?
Phytosanitary Measures (SPS Agreement). In: European Review of Agricultural Economics, 39(4):
World Trade Organzation. [online]. [Cited 29 July 589–609. (also available at: https://doi.org/10.1093/
2020]. https://www.wto.org/english/tratop_e/sps_e/ erae/jbr062)
spsagr_e.htm
Xiong, B. & Beghin, J. 2014. Disentangling demand‐
World Trade Organization (WTO). 2016. The SPS enhancing and trade‐cost effects of maximum
Agreement and Pesticide MRLs presentation. In: residue regulations. Economic Inquiry, 52(3):
World Trade Organzation. [online]. [Cited 29 July 1190–1203. (also available at: https://doi.org/10.1111/
2020]. https://www.wto.org/english/tratop_e/sps_e/ ecin.12082)
wkshop_oct16_e/s1_anneke_hamilton.pdf
58
Glossary
Taken from EHC 240 (WHO, 2009) and CODEX definitions where
possible. It is acknowledged that other organizations might have slightly
different definitions.
Acceptable Daily Intake (ADI) – The estimate of the GEMs food – The World Health Organization’s
amount of a chemical in food or drinking-water, Global Environment Monitoring System – Food
expressed on a body weight basis, that can be Contamination Monitoring and Assessment
ingested daily over a lifetime without appreciable Programme, which maintains databases on
health risk to the consumer. It is derived on the contaminant levels in foods and estimates of dietary
basis of all the known facts at the time of the exposure to food chemicals. Collects information
evaluation. The ADI is expressed in milligrams of the and maintains databases on consumption levels of
chemical per kilogram of body weight (a standard agricultural commodities.
adult person weighs 60 kg). It is applied to food
Good Agricultural Practice (GAP) – For pesticide
additives, residues of pesticides and residues of
use, includes the nationally authorized safe uses
veterinary drugs in food.
of pesticides under actual conditions necessary for
Acute Reference Dose (ARfD) – The estimate of effective and reliable pest control. It encompasses
the amount of a substance in food or drinking- a range of levels of pesticide applications up to
water, expressed on a body weight basis, that the highest authorized use, applied in a manner
can be ingested in a period of 24h or less without that leaves a residue that is the smallest amount
appreciable health risk to the consumer. It is derived practicable. Authorized safe uses are determined at
on the basis of all the known facts at the time of the national level and include nationally registered
evaluation. The ARfD is expressed in milligrams of or recommended uses, which take into account
the chemical per kilogram of body weight. public and occupational health and environmental
safety considerations. Actual conditions include
Benchmark Dose (BMD) – A dose of a substance
any stage in the production, storage, transport,
associated with a specified low incidence of
distribution and processing of food commodities
risk, generally in the range of 1–10 percent, of a
and animal feed.
health effect; the dose associated with a specified
measure or change of a biological effect. Normally Health Based Guidance Values (HBGV) – A numerical
considered along with the upper (BMDU) and lower value derived by dividing a point of departure (a
(BMDL) confidence intervals from the analysis. no- observed-adverse-effect level, benchmark dose
or benchmark dose lower confidence limit) by a
Chemical-specific adjustment factor (CSAF) – A
composite uncertainty factor to determine a level
modified default ten-fold uncertainty factor that
that can be ingested over a defined time period
incorporates appropriate data on species differences
(e.g. lifetime or 24h) without appreciable health risk.
or human variability in either toxicokinetics (fate of
Related terms: ADI, Provisional maximum tolerable
the chemical in the body) or toxicodynamics (actions
daily intake, Provisional tolerable monthly intake,
of the chemical on the body).
Provisional tolerable weekly intake and Tolerable
Codex – The Codex Alimentarius (Food Code) is a daily intake.
collection of standards, guidelines and practices
adopted by the Codex Alimentarius Commission.
The aim is to contribute to the safety, quality and
fairness of international trade in food commodities.
59
Joint Meeting on Pesticide Residues (JMPR) – No-Observed-Adverse-Effect Level (NOAEL) –

The abbreviated title for the Joint Meeting of Greatest concentration or amount of a substance,
the FAO Panel of Experts and the WHO Core found by experiment or observation, that causes
Assessment Group on Pesticide Residues in Food no adverse alteration of morphology, functional
and Environment, which has met since 1963. The capacity, growth, development or lifespan of
meetings are normally convened annually. The FAO the target organism distinguishable from those
Panel of Experts is responsible for reviewing residue observed in normal (control) organisms of the
and analytical aspects of the pesticides considered, same species and strain under the same defined
including data on their metabolism, fate in the conditions of exposure.
environment and use patterns, and for estimating
Point of Departure (POD) – A value for the dose
the maximum residue levels and supervised trials
(obtained from relevant dose-response data) that
median residue levels that might occur as a result
serves as the starting point for estimating the
of the use of the pesticide according to (GAP).
equivalent (acceptable) dose in a target human
The WHO Core Assessment Group on Pesticide
population. Can be based on a NOAEL or a BMD
Residues is responsible for reviewing toxicological
analysis.
and related data on the pesticides and, when
possible, for estimating ADIs and long-term dietary Residue Definition – A chemical or mixture of
intakes of residues. As necessary, acute reference chemicals present in food commodities resulting
doses for pesticides are estimated along with from the use of a pesticide; can include its
appropriate estimates of short-term dietary intake. subsequent metabolites or degradation/reaction
JMPR is a technical committee of specialists acting products that are considered to be of toxicological
in their individual capacities. Each is a separately significance. Residue definitions for monitoring or
constituted committee. When the term “JMPR” enforcement might differ from those for use in risk
or “the Meeting” is used without reference to a assessment.
specific meeting, it is meant to imply the common
policy or combined output of the separate meetings
over the years. Makes recommendations for MRLs
to the CCPR.
Maximum Residue Limit (MRL) – The maximum

concentration of a pesticide residue (expressed
as milligrams per kilogram) recommended by the
Codex Alimentarius Commission to be legally
permitted in or on food commodities and animal
feed. MRLs are based on (GAP) data, and food
derived from commodities that comply with the
respective MRLs are intended to be toxicologically
acceptable. Consideration of the various dietary
residue intake estimates and determinations,
at both the national and international level, in
comparison with the ADI intake should indicate
that foods complying with Codex MRLs are safe for
human consumption.
60
ANNEXES
1 Annex PART A
Sources of national pesticide
maximum residue limits for the 19
countries/region analysed
Country/ Publicly Available
National MRLs source Link
region available in English
Australia Australia New Zealand Food Yes Yes https://www.legislation.gov.au/Series/

Standards Code-Scheduled F2015L00468
20-Maximum residue limits
(Compilation 43 – Oct 2019)
Australia New Zealand Food Yes Yes https://www.legislation.gov.au/Details/

Standards Code – Schedule F2017C00330
21 – Extraneous residue
limits
Bangladesh - No - -
Brazil Regularização de Produtos Yes No http://portal.anvisa.gov.br/registros-e-

– Agrotóxicos (Monografias autorizacoes/agrotoxicos/produtos/monografia-de-
Autorizadas) agrotoxicos/autorizadas
Cambodia Proclamation No. 002 MAFF, Yes Yes http://extwprlegs1.fao.org/docs/pdf/cam81986.pdf

03/01/2017 on the list of
Maximum Residue Limit
of Pesticides in Agricultural
Product of Plant Origin
Canada Health Canada: online Yes Yes http://pr-rp.hc-sc.gc.ca/mrl-lrm/index-eng.php

database on Maximum
Residue Limits for
Pesticides
China GB 2763-2019 Yes No https://www.sdtdata.com/fx/fmoa/tsLibList
European Pesticides Database of the Yes Yes http://ec.europa.eu/food/plant/pesticides/eu-

Union European Union pesticides-database/public/?event=pesticide.
residue.selection&language=EN
India Gazette Notification on Yes Yes https://archive.fssai.gov.in/home/fss-legislation/

Food Safety and Standards notifications/gazette-notification.html
(Contaminants, Toxins and
Residues) Amendment
Regulation related to MRL
of pesticide (December,
2018)
Indonesia Main document: Permentan Yes Yes http://ditjenpp.kemenkumham.go.id/arsip/

(ministry regulation) no. 55 bn/2016/bn1757-2016.pdf
/ PERMENTAN / KR.040 /
11/2016 concerning Food (55/PERMENTAN/KR.040/11/2016)
Safety Monitoring of Fresh
Food Material Imports
If a commodity is not Yes Yes https://kupdf.net/download/sni-batas-maksimum-

specified in the Permentan, pestisida_58f8da3bdc0d60f361da97e3_pdf
then refer to: SNI 7313:
2008, Maximum Limits
of Pesticide Residues in
Agricultural Products
Iran Pesticides - The maximum Yes No http://www.standard.ac.ir/en/research-centers

(Islamic residue limit of pesticides- Select Risk Analysis Office
Republic of) cereals
62
annex PART A
Country/ Publicly Available

National MRLs source Link
region available in English
Japan Positive List System for Yes Yes http://db.ffcr.or.jp/front/pesticide_detail?id=3900

Agricultural Chemical
Residues in Foods:
Maximum Residue Limits
(MRLs) List of Agricultural
Chemicals in Foods
Myanmar - No - -
Pakistan - No - -
Philippines Philippine National Standard Yes Yes http://www.bafs.da.gov.ph/images/

PNS/BAFS 162: 2015: Approved_Philippine_Standards/PNS-BAFS162-
Pesticide residues in rice - 2015PesticideResiduesinRiceMRLs.pdf
(MRLs)
Saudi Arabia Saudi Food & Drug Authority Yes Yes http://apeda.gov.in/apedawebsite/HACCP/SFDA_
FD_382_eng.pdf
SFDA.FD 382/2018
Maximum Limits of
Pesticide Residues in
Agricultural and Food
Products
Thailand For MRLs: Thai Agricultural Yes No http://www.acfs.go.th/standard/download/

Standard: TAS 9002-2016 MAXIMUM-RESIDUE-LIMITS.pdf
For Extraneous MRLs: Thai Yes No http://www.acfs.go.th/standard/download/eng/

Agricultural Standard: TAS EMRL.pdf
9003-2004
United Arab UAE.S MRL 1: 2017 Yes No http://www.aecegypt.com/Uploaded/

Emirates Pdf/2907182.pdf
(MRLs) for Pesticides
in agricultural and food
products
United Electronic Code of Federal Yes Yes https://www.ecfr.gov/cgi-bin/text-idx?SID=6cb75

States of Regulations: Part 180 - e3393c27ab4458f4b56494cf72c&mc=true&tpl=/
America Tolerances and exemptions ecfrbrowse/Title40/40cfr180_main_02.tpl
for pesticide chemical
residues in food
CPG Sec. 575.100 Pesticide Yes Yes https://www.fda.gov/media/75151/download

Residues in Food and Feed -
Enforcement Criteria
Viet Nam Circular No. 50/2016/TT-BYT Yes Yes http://content.bcimonitor.com/DataServices/

from Ministry of Health Circular%20No.%2050-2016-TT-BYT%20
effective%2001JULY2017%20(English%20
translation).doc
Notes:
For the purpose of this study, all MRL sources were consulted during October 2019.
Australia and Thailand establish both MRLs and extraneous residue limits (ERLs): both were considered in the analysis.
63
2 Annex PART A
Rice classification
Matching rice classification of the 19 countries/region analysed to Codex
Rice Wild rice

Codex
GC 0649 GC 0655
(Oryza sativa L.; several ssp. and cultivars) Wild rice (Zizania palustris L.)
Wild Rice, Eastern (Zizania aquatica L.)
“Rice with husks that remain attached to kernels

even after threshing: kernels with husks (Note: For
rice, only about 10% of traded grains is with husk).
Portion of the commodity to which the MRL applies
Codex Food Commodity (and which is analysed): Whole commodity in trade”
Description
CCPR 2017 report - REP17/PR - Appendix XI
Rice Cereal grains

Cereal grains
Australia
All other foods All other foods
All other foods except animal food commodities All other foods except animal food
commodities
Brazil Rice / Arroz Rice / Arroz
Rice Rice
Cambodia
Cereal grains Cereal grains
Rice Wild rice

Raw cereals Raw cereals
Canada
All food crops; All food crops;
All food crops (other than those listed in this item) All food crops (other than those listed in
this item)
China Rice / 稻谷 NA
European Union NA Rice
Food grains NA
India
64
annex PART A
Rice, husked Rice, polished Cereal grains

CM 0649 CM 1205 GC 0080 / GC 2088
Group 020 - Cereal grains - Class A (up to

July 2017)
GC 0649 Rice (Oryza sativa L.; several ssp.&
cultivars)
GC 0655 Wild rice (Zizania aquatica L.)
“Brown rice (or cargo rice) is paddy “Milled rice (white rice) is husked rice
Codex Classification of Foods and Animal
rice from which the husk only has from which all or part of the bran and
Feeds-2006
been removed. The process of husking germ have been removed by milling”
and handling may result in some loss of Subgroup 020C - Rice Cereals (since July
bran” CXS 198-1995 2017)
CXS 198-1995 GC 0649 Rice (Oryza sativa L.; several ssp.&

cultivars)
GC 3088 Rice, African (Oryza glaberrima
Steud.)
GC 0655 Wild rice (Zizania palustris L.)
Wild Rice, Eastern, (Zizania aquatica L.)
CCPR 2017 report - REP17/PR - Appendix XI

Rice
All other foods All other foods All other foods

All other foods except animal food All other foods except animal food All other foods except animal food
commodities commodities commodities
Rice / Arroz Rice / Arroz Rice / Arroz

Rice (milled or polished) Rice
Rice Rice Rice

Raw cereals Raw cereals Wild rice
Raw cereals
All food crops; All food crops; All food crops;

All food crops (other than those listed in All food crops (other than those listed in All food crops (other than those listed in this
this item) this item) item)
Rice, husked / 糙米 Rice, polished / 大米 Grains / 谷物 (Rice / 稻谷)
Rice NA NA
Milled food grains Milled food grains Food grains

Rice Rice
65
Rice Wild rice

Codex
GC 0649 GC 0655
Rice / Beras NA
Indonesia Cereal grains / Cereal
Cereal grains / Biji-bijian padi
Iran (Islamic Republic

Rice / Rice /
of)
Japan NA NA
Philippines Rice NA
Rice NA
Saudi Arabia
Cereal grains
Rice paddy / ข้าวเปลือก Cereal / ธัญพืช

(Rice paddy: means non-glutinous rice or glutinous rice
Thailand
with husk)
Cereal / ธัญพืช
Rice Rice wild

Rice, grain Rice wild, grain
Grain, cereal, group 15 Rice
United States of Rice, grain

America Grain, cereal, group 15
All food commodities (including feed commodities) not All food commodities (including feed
otherwise listed in this subsection commodities) not otherwise listed in this
subsection
Viet Nam Rice NA
Note: Bangladesh, Myanmar, Pakistan and the United Arab Emirates are not reported as they do not establish national MRLs
(hence do not have national lists of MRLs) but automatically defer to Codex.
66
annex PART A

CM 0649 CM 1205 GC 0080 / GC 2088
Rice husked / Beras pecah kulit Rice polished / Beras, dipoles (disosoh) Rice / Beras
Cereal grains / Cereal
Cereal grains / Biji-bijian padi
Rice / Rice / Rice /
Rice (brown rice) Milled rice NA
Rice Rice Rice

Rice
Rice / ข้าวสาร Rice / ข้าวสาร Cereal / ธัญพืช

(Rice: means paddy in which husk has (Rice: means paddy in which husk has Rice paddy / ข้าวเปลือก
been removed or polished) been removed or polished)
(Rice paddy: means non-glutinous rice or
Cereal / ธัญพืช Cereal / ธัญพืช glutinous rice with husk)
Rice husked Rice polished Grain, cereal, group 15;

Rice;
Rice, grain
All food commodities (including feed All food commodities (including feed All food commodities (including feed
commodities) not otherwise listed in commodities) not otherwise listed in commodities) not otherwise listed in this
this subsection this subsection subsection
67
3 Annex PART A
HS codes and Codex
Rice classification
Matching of Codex rice classification with HS classification
Codex classification HS classification
GC 0649 Rice
100610 Rice in husk (paddy or rough)
GC 0080 Cereal grains
CM 0649 Rice husked 100620 Husked rice (brown)
100630 Semi-milled or wholly milled rice, whether or not polished or

CM 1205 Rice polished glazed (rice, parboiled; Basmati rice)
100640 Broken rice
68
4 Annex PART A
Methodology and data
of the trade model
Analysis of the effects of non-harmonization of estimation. In order to avoid identification problems,

pesticide MRLs on trade in rice one variable from the first step is dropped in the
In order to assess the impact of pesticide MRLs on second step (Greene, 2002; Mittelhammer, Judge and
trade, most empirical studies revert to the gravity Miller, 2000).
model, which states that countries trade in proportion
Definitions and dimensions of dependent and
to their respective Gross Domestic Products (GDPs)
explanatory variables are given in Table 7.
and geographical proximity (UNCTAD and WTO, 2012;
UNCTAD and WTO 2016). Data and variables used in the analysis
Trade data
In a general representation of the gravity model,Yij,
Bilateral trade data for the time period 2012–2018 are
the trade flow from the exporting country i to the
retrieved from UN-COMTRADE. Trade flows of rice are
importing country j is determined by a function f of
distinguished based on four tariff lines at HS-6 digit level
exporter-specific factors Si (e.g. exporter’s GDP),
(see Annex A3): (1) Rice in the husk (paddy or rough)
importer-specific factors Mj (e.g. importer’s GDP) and
(HS 100610); (2) husked (brown) rice (HS 100620);
the ease of market access ɸij to market j of exporter i:
(3) rice, semi-milled or wholly milled, whether or not
polished or glazed (HS 100630); and (4) rice, broken (HS
Yij = f ( Si , Mj , ɸij ).
100640). The eight importing markets considered in the
analysis are: Australia, Canada, China, the European
Here, a generalized gravity equation is applied
Union, Indonesia, Japan, Saudi Arabia and the United
to disentangle the effects of different levels of
States of America. The twelve exporting countries
harmonization of national MRLs with Codex MRLs
considered are: Bangladesh, Brazil, China, Indonesia,
on the trade in rice. The empirical specification of the
India, Cambodia, Myanmar, Pakistan, Philippines,
model closely follows Disdier and Marette (2010),
Thailand, the United States of America and Viet Nam.
Xiong and Beghin (2012) and, especially, Xiong and
Beghin (2014). A two-step Heckman estimation
Pesticide MRLs on rice
procedure is applied to correct for sample selection
Information on rice pesticide MRLs is retrieved from
bias (Heckman, 1979). Sample selection bias can, for
the database created for the analysis in chapter 2 of
example, occur if some countries do not trade specific
this report. MRL restrictiveness for application in the
tariff lines for unknown reasons or countries simply do
trade model is measured based on a stringency index
not report their trade.
first defined by Li and Beghin (2014) and also applied
In its first step, a probit estimation, the Heckman by Xiong and Beghin (2014) and Kareem, Martínez-
model explores the impacts of MRLs and other Zarzoso and Brümmer (2018):
explanatory variables on the propensity to trade
at all, i.e. the extensive margin. The second step
determines the impact of the explanatory variables
on the magnitude of trade, i.e. the intensive margin.
In the second step, therefore, only actually observed where MRLindexjpt is the MRL index of product tariff
(positive) trade flows are considered. This step is line p imposed by country j at time t. MRLcodex,kpt is the
estimated via Ordinary Least Squares (OLS). To correct MRL recommended by Codex for pesticide k and
for potential sample selection bias, the second step product p at time t. The index defined by rice tariff
includes an additional explanatory variable, the Inverse line, year and country is lower and upper bounded
Mill’s Ratio (IMR), which is derived from the first step between zero and e≈2.718. The index equals one if
69
Table 7: Definition of dependent and explanatory variables
Variable name Definition
Yijpt Trade flow of rice tariff line p from country i to country j in year t
MRLimporterjpt MRL index of importing country j applied to tariff line p in year t
MRLdivergenceijpt Difference between the MRL index of importing country j and the MRL index of exporting country
i in tariff line p and year t
Productionipt Exportable production in tariff line p of country i in year t
Tariffijp Tariff rate applied by country j on imports in tariff line p from country i
Distanceij Distance between exporting country i and importing country j
DevelopmentStatusij Dummy variable indicating if country i and country j have the same development status
Colonyij Dummy variable indicating if country i and country j have ever been in a colonial relationship with
each other
Trade1995ijp,1995 Dummy variable indicating if there was a trade flow in rice tariff line p from country i to country
j in year 1995
RiceVarietyij Dummy variable indicating if country i and country j engaged mainly in trade of the same rice
variety
WTOaccessionij Dummy variable indicating if country i and country j both acceded the WTO after 1995
FEit Fixed effect for exporting country i and year t
FEjt Fixed effect for importing country j and year t
FEp Fixed effect for tariff line p
national MRLs are completely aligned with Codex, Other determinants of trade considered in the
is greater than one if, averaging over the pesticides, analysis
national MRLs are stricter than Codex and less than In addition to the MRLs, standard variables considered
one if national regulation is less stringent than Codex. in gravity analyses and variables covering bilateral
The analysis considers only MRLs for which a standard relationships and trade costs were used to explain
was adopted by Codex, ignoring national MRLs for the trade flows. Following Xiong and Beghin (2014),
which no Codex MRL existed. Missing national MRLs the exportable production, i.e. the sum of a country’s
were replaced by the national default tolerance level exports to all destinations globally, was considered
(see Table 6). to proxy production capacity. The data were retrieved
from UN-COMTRADE. Apart from NTMs such as
In line with Xiong and Beghin (2014), the analysis
MRLs, trade can also be hampered through tariffs.
considers both the MRL stringency of the importing
Tariff rates expressed as AVEs were used to cover
country (MRL importer) and the difference in MRL
other policy induced trade costs than MRLs. Tariff
stringency between importer and exporter (MRL
data were taken from the Macmap database of the
divergence). If domestic food safety regulation
International Trade Centre (ITC). For the estimation,
with respect to MRLs on pesticides in rice of the
only tariff data from 2012 were used to avoid possible
exporting country is stricter or equal to the one in the
endogeneity with the dependent variable through the
importing country (MRLindeximporter,p ≤ MRLindexexporter,p),
computation of the AVEs (Xiong and Beghin, 2014).
no additional trade barrier (i.e. costs related to trade)
should apply. However, if the MRL regulation in As standard variables used in gravity equations, the
the exporting country is laxer than in the importing distance between trading partners, as a proxy for
country (MRLindeximporter,p > MRLindexexporter,p), this could trade costs (UNCTAD and WTO, 2012), and a bilateral
imply higher obstacles to export to this country. If dummy capturing colonial ties were used. These
both trading partners adopt internationally harmonized variables were derived from the CEPII gravity database
standards, the MRLs should not cause an additional (Head, Mayer and Ries, 2010). Different variants of
cost to trade. a variable indicating common language did not exert
any significant effects on trade flows and were not
70
annex PART A
considered in the final model specification. Other Multilateral trade resistance terms in both the
variables reflecting a historical trade relationship importing and exporting countries (Anderson and
between two countries are a bilateral dummy van Wincoop, 2003), were controlled for by including
indicating whether two countries traded rice in year-specific importer and year-specific exporter fixed
1995 based on FAOSTAT data and a bilateral dummy effects (Feenstra, 2015; Xiong and Beghin, 2014). In
indicating whether a pair of countries entered the addition, fixed effects are included for the four tariff
WTO only after its establishment in 1995. The WTO lines considered. Income effects in the countries are
accession variable was used as exclusion variable in entirely absorbed by the year-specific importer and
the Heckman model, i.e. it was only considered in the exporter fixed effects (Xiong and Beghin, 2014). Also,
first step, but not in the second to avoid identification specific policies referring to rice as a staple in many of
problems. To account for other factors except the the exporting countries, e.g. temporary bans on rice
MRLs that might affect trade flows due to a different exports, are captured by the year-specific exporter
economic status of exporting and importing country, fixed effects. Overall, the analysis covers seven years,
a binary variable indicating whether trade partners four tariff lines, eight importers and 12 exporters.
have the same development status was added to
the model. A binary variable indicating whether
two countries engaged mainly in trade of the same
rice variety was used to capture the importance of
countries’ specialization in specific varieties. The
variable was created based on the varietal family that
accounted for the largest share of overall volumes
traded between two countries in 2012–2016.
71
5 Annex PART A
Detailed results of
the trade model
Table 8: Estimated coefficients for rice imports40
Dependent variable:
Probit OLS
Extensive margin Intensive margin
(1) (2)
MRLimporter 1.149*** 1.551***

(0.202) (0.434)
MRLdivergence -0.431*** -0.634**

(0.161) (0.299)
ln(Production) 0.065*** 0.139***

(0.011) (0.027)
ln(Tariff) -0.255*** -0.930***

(0.091) (0.143)
ln(Distance) -0.796*** -0.805***

(0.113) (0.142)
DevelopmentStatus 0.772*** 4.807***

(0.154) (0.401)
Colony -0.130 1.352***

(0.281) (0.381)
Trade1995 0.581*** 1.453***

(0.106) (0.268)
RiceVariety 0.318*** 0.726***

(0.086) (0.213)
WTOaccession413 -0.432**
(0.172)
IMR42 1.915***
(0.335)
Observations 2,604 1,366
R2 0.629
Adjusted R2 0.587
Log Likelihood -965.557
Akaike Inf. Crit. 2,211.114
Residual Std. Error 2.405 (df = 1226)
F Statistic 14.982*** (df = 139; 1226)
Note: Significance levels *p<0.1; **p<0.05; ***p<0.01
40 For brevity, the table does not report the year-specific importer and exporter fixed effects and tariff line fixed effects. However, these were
included in the regression.
41 The WTO accession variable did not significantly influence the intensive margin of trade in model test runs. It was therefore used as the
exclusion variable in the second step estimation (the intensive margin).
42 IMR indicates the inverse Mill’s ratio, which is only used in the second step, i.e. the estimation of the intensive margin. As it is highly
significant, it points to selection bias and justifies the application of the Heckman model.
72
annex PART A
Table 9: Marginal effects and standard errors of MRLs on rice imports
Extensive margin Intensive margin
MRL Importer 0.448 *** 1.551 ***
(0.079) (0.434)
MRL Divergence -0.168 *** -0.634 **
(0.063) (0.299)
Note: Significance levels *p<0.1; **p<0.05; ***p<0.01.

The marginal effects are evaluated at their means. Due to characteristics of the Heckman model, the marginal effects differ from the estimated
coefficients in the first step (Table 8), but are equivalent with the coefficients in the second step of the estimation.
73
© Veejay Villafranca/NOOR for FAO
1 Annex PART B
Possible areas of investigation to understand
better the issues that might lead to different
national and Codex maximum residue limits
Summarized consolidated feedback received from the Feedback from Japan indicates similarities with JMPR/
five countries/region analysed. Additional conclusions Codex, but further analysis is hindered by the main
of the report author are presented in italics. guidance/requirement documents being available only
in Japanese.
Overview
Australia closely follows the JMPR approaches in 1. Differences in risk assessment procedures
most areas. Adoption of Codex MRLs currently
requires an application, but the process appears to be JMPR and different countries’/region’s authorities may
relatively simple with facilitated acceptance of JMPR utilize different procedures and methods in the risk
evaluations/Codex MRLs indicated in the guidance assessment process. A better understanding of this
document (FSANZ, 2018). Australia is considering would require an analysis that looks into the different
automatic adoption of new Codex MRLs in the near procedures and approaches applied by JMPR and the
future, subject to a suitable exposure assessment. different national authorities. This is an area where
In Australia there are two separate lists of MRLs for there may be different elements to be considered
agricultural and veterinary chemical residues in food including possible differences in the following issues:
commodities: the APVMA MRL standard based on
domestic uses; and MRLs contained in Schedule 20
of the Australia New Zealand Food Standards Code
(the Code), which contains domestic MRLs and
imported MRLs – whichever is higher. The APVMA
MRL standard is used for the enforcement of good
agricultural practice (GAP) within Australia, usually near
the farm gate. In contrast, the MRLs in the Code apply
to food at the point of sale and for imported foods at
entry into Australia.
The European Union feedback noted that there are

several areas of variation between the European
Union and JMPR/Codex. New guidance from EFSA on
residue definitions for risk assessment (EFSA, 2016) –
not yet ratified, proposes a different approach to that
used by JMPR.
The Canadian and United States of America feedback

and available supporting documents (USEPA, 2000)
(NAFTA, 2005) identified some common approaches
but a number of methodological differences continue
to exist. However, critical differences are actively being
explored through various international fora, including
the OECD and JMPR (e.g. the work underway at the
OECD/JMPR/JECFA level on residue definition).
75
Proposed areas of
Australia Canada
investigation
Residue definition JMPR approach OECD Residue Definition Guidance Document

approaches
Takes into account approaches, such as those
used by JMPR
Support development of standardised
approaches at international level.
Methodology on residue data JMPR approach • Consideration of independence of trials based

extrapolation, pooling, and on following criteria: Location;
trial
Timing; Variety. When criteria are met, average
residues from replicate trials are used for MRL
calculation.
(While in JMPR: variety is not typically
considered as a criterion for independence. If
remaining criteria are met, highest residue from
replicate trials is used for MRL calculation)
• Application of proportionality on a case-by-
case basis (while JMPR: Only when there is
insufficient data reflecting cGAP)
Interpretation of toxicological JMPR approach Broadly similar to JMPR in terms of study

studies (e.g. different end- evaluation.
points used to derive ADI
and/or ARfD) Apply default factors (ten fold or three fold)
for reproductive or severe effects for ADI/
ARfD setting, unless reassuring data available
(i.e. PCPA factor, which can defer from other
schemes).
Approach on the assessment JMPR approach Same data requirements

of metabolism studies
Broadly similar approach to JMPR
Exposure assessment Same approach as JMPR, except utilizing Acute and Chronic Dietary Risk Assessment
methodologies both for the detailed national food consumption
chronic and acute exposure data that are available for Australian Consumption: US NHANES / WWEA data and:
scenarios consumers. DEEM-FCID™, Version 4.02, 05-10-c
Differences in Good Australian MRLs are based either on: Use patterns: Proposed Canadian use pattern or
Agricultural Practices (GAPs) exporting country’s registered use pattern.
and related selection of the (i) Australian GAP; or for import MRLs;
critical GAP on which the Crop field trials: Canada/US only and those from
(ii) An overseas national MRL which is exporting country [for import MRL request].
estimation of the MRL is based on their national GAP; or
based
(iii) A Codex MRL.
Where MRLs have not been established
by (i), (ii) or (iii) above, an AoF MRL may
be considered for all other crops for
registered pesticides. Since these MRLs
are not established on the basis of GAP,
an alternative risk assessment approach
is used.
AoF MRL is not a default MRL; it
is intended to allow for inadvertent
presence of residues (e.g. by spray drift,
crop rotation) and is established using
risk assessment methodology.
76
annex PART B
European Union Japan United States of America
Agreed, there are different approaches Same approach as JMPR. Currently - Simplest possible;
on deriving residue definitions. It might
be appropriate to analyse the situation In the past - Toxic residue
separately for residue definitions for E-fate given low consideration
dietary intake and for MRL compliance.
(JMPR: e-fate included in the evaluation)
New EFSA guidance could lead to
different definitions. However, the United States of America
tends to have more complex definitions
than JMPR/European Union
This is a point that should be discussed in Same approach as JMPR. Appears that current practices by
the section below on policy issues. the United States of America and
JMPR may be converging with regard
We believe that the overall impact on to extrapolation practices and data
numbers of non-aligned MRLs is small, pooling. In the past there may have
compared to the points on residue been differences related to use of the
definition approaches and interpretation proportionality concept and global zoning
of toxicological studies. and residue trial exchangeability.
Also see below on policies related to
crop groups.
Important point as well. Broadly similar to JMPR, some unique In general, diverging opinions on
studies required. toxicological points of departure may
The discussion should not only focus be related to how weight of evidence
on parent compounds but also on is approached, or different policies for
metabolites. addressing uncertainty.
Slightly different ADI/ARfD or Use of BMD, combined exposures
consumption value within the risk and FQPA 10x factor differs from other
assessment would not trigger a schemes.
reservation regarding a Codex MRL, as
long as the risk assessment remains
favourable.
Agreed. Different approach to decide on Broadly similar to JMPR, some unique See above response on residue definition
the relevance of metabolites observed in studies required. approaches.
metabolism studies.
Appears broadly similar to JMPR
EFSA Primo model Japan has conducted chronic and acute Acute: diet based, probabilistic
exposure assessment using average daily
food consumption per person (g/day) and (JMPR – commodity based,
97.5th percentile daily consumption (g/ deterministic)
day), on the basis of the results obtained Chronic: diet based, average
from “Food Intake and Frequency consumption, tiered residues
Questionnaire in Japan” in FY 2005 to
2007 (n = 40,394). (only difference JMPR: median residues).
Provided all other points are not Japan has regarded the results of The United States of America bases its
problematic, this would only apply to residue trials data with a maximum MRLs on the field trials/GAPs that are
the European Union for substance/ change of +/-25% in each parameter as submitted by the applicant and selects
commodity combinations where the confirming GAP, and adopted them for an the critical GAP to establish the MRL.
GAP of the European Union is more establishment of MRL.
critical than the one assessed by JMPR. Differences in MRLs related to GAP
However, any analysis of non-alignment selection seem most likely to arise when
with Codex MRLs should focus on different GAPs/trials are submitted to the
national/regional MRLs that are lower United States of America and JMPR for
than Codex MRLs. review. This could be the case with new
chemistries, where the United States
of America may register the substance
several years before it is evaluated by
JMPR, and new data are generated in
the interim and provided to JMPR but not
the United States of America
The reverse can also apply if the first
registration is outside the United States
of America
77
Proposed areas of
Australia Canada
investigation
MRLs calculators/tools OECD MRL Calculator (for estimation of OECD MRL Calculator
MRLs)
OECD Feed Calculator (for estimation
of livestock dietary burden, used to set
animal commodity MRLs)
Other?
2. Different risk management considerations and policies on the setting of MRLs
Proposed areas of
Australia Canada
investigation
Policies related to the For domestic use, the MRL is based on It is important to note that some policies are
establishment of MRLs for approved national GAP. rooted in legislation. For example, in Canada,
specific commodities and/or while the PMRA specifies science-based MRLs
crop groups Adoption of MRLs based on an overseas under the Pest Control Products Act, in the
national MRL (based on their national absence of such an MRL, pesticide-commodities
GAP), or the adoption of Codex MRLs, can be subject to Food and Drug Regulations
is based on need. Australia has an open GMRL of 0.1 ppm.
and transparent process to incorporate
these MRLs for pesticides into the Food Crop Groupings: International Crop Grouping
Standards Code. Any country, sponsor Consultation Committee (ICGCC)
company or other third party can apply
for the incorporation of these MRLs [While JMPR: Codex Classification of Food and
whenever such a need is identified. This Feed]
MRL harmonization process is free-of-
charge to applicants and the needed
MRL will be incorporated into the Code
unless an estimate of dietary exposure
exceeds the health based guidance
value(s) (HBGVs) using Australian food
consumption data.
MRLs extrapolation rules Australia has crop group guidelines Extrapolation between commodities is covered
which indicate representative crops for under policies on crop groups.
extrapolation purposes
https://apvma.gov.au/crop-groups
Policies related to the JMPR approach Reliance on default processing factors, where
application of processing available, when chemical specific processing
factor/s, studies are not submitted. MRLs are specified
for processed commodities only when chemical-
specific data demonstrate a concentration in
residues that exceeds the RAC MRL.
Application of default values The only exception to the JMPR approach 0.1 mg/kg (ppm)
for non-registered pesticides is in relation to default values and specific
MRLs for All other foods except animal
food commodities (AoF) because these
specific MRLs are not based on GAP.
AoF MRLs are considered for registered
pesticides, instead of default MRLs,
using an alternative risk assessment
approach.
78
annex PART B
This point is more or less harmonized. Japan has set the MRLs with OECD Both the United States of America
The OECD calculator is at least in the calculator in principle. When the number and JMPR currently utilize the OECD
European Union the standard tool that is of residue trials data is insufficient, Japan calculator. (In the past, the United States
used to derive MRL proposals. has calculated the MRL with allowance of America historically relied on best
by Japanese method taking into account judgment. The NAFTA calculator was
variations in results of residue trials data. used prior to development of the OECD
calculator).
We would like to add the processing

data: the need to submit standard
hydrolysis studies is part of the European
Union data requirements, but as far as
we understand this is not mandatory for
JMPR.
Agreed. The European Union follows with Japan has classified foods based Differences of crop groupings between
great interest (and is actively engaged in) on Codex Classification of Foods, the United States of America and Codex
the review of the crop grouping in CCPR. considering the current food situation in may result in different MRLs. Crop
Japan (daily intake, size of agricultural groups of the United States of America
products). can be found at in the Code of Federal
Regulations, 40 CFR 180.41.
(*Japan is reviewing food classification
now, not all of Japan’s food classification
is consistent with Codex Classification of
Foods.)
Agreed. Japan has set MRLs based on Extrapolation between commodities is

the extrapolation of Codex MRLs, covered under policies on crop groups.
considering whether part and form of
samples, characteristics of growing and
usage of pesticide are the same.
(*Japan is reviewing food classification
now, not all of Japan’s food classification
is consistent with Codex Classification of
Foods).
The European Union sets MRLs for raw Japan has adopted the Codex MRLs for Default for some commodities. US EPA
agricultural commodities. In contrast processed food in principle. In the case OCSPP Guideline 860.1520 describes
to CCPR, the European Union does where the values converted into the requirements for processed food/feed.
currently not set MRLs for processed concentration in the raw material using
products. See point below on products to the processing factor does not exceed The guideline is available at:
which MRLs apply. the MRL of the raw material, Japan has https://www.epa.gov/test-guidelines-
not set the MRLs for processed food. pesticides-and-toxic-substances/series-
Processing included as part of the
PRIMO intake assessment. 860-residue-chemistry-test-guidelines
In the absence of information to set 0.01 ppm The United States of America do not
MRLs based on authorized uses, apply a default MRL or defer to other
including Codex MRLs, the European MRLs in the absence of a national MRL.
Union applies the LOQ for a given An applicant may submit a request
substance/matrix combination or the to establish import tolerances for
legal default value of 0.01 mg/kg. substances that are not registered in the
United States of America.
79
Proposed areas of
Australia Canada
investigation
Other national policy decision For trading purposes and at the point of Looking to consider global GAP where possible.
related to trade or other food sale, MRLs apply to both imported
relevant considerations. For and domestically produced food.
example:
To facilitate trade, while protecting the
• More stringent policy for consumer, Australia’s approach is to have
imports than for exports an established open and transparent
process to incorporate Codex MRLs,
• Specific Level of Protection or MRLs based on third country GAPs,
for pesticides for pesticides into the Australian Food
Standards Code based on need. Further
details are given above.
Other? Submission of same data package to Codex

shortly after approval in Canada.
3. Timing of national and Codex MRL establishment

For markets that set MRLs prior to Codex MRL establishment, there may not be resources or routine procedures
in place for those markets to review national MRLs at the time Codex adopts new MRLs
Proposed areas of
Australia Canada
investigation
Procedures to review national Has a regular consideration of Codex No procedures in place

MRLs at the time Codex MRLs
adopts new MRLs
MRL harmonization process PMRA establishes MRLs in some instances

many years ahead of Codex based on scientific
analysis – it is primarily established for domestic
production – when Codex MRLs are ultimately
established there is no policy nor resources to go
back to see why there is a divergence from the
Canadian results. Codex MRLs may be considered
when the product comes up for re-evaluation or
when establishing import tolerances.
For example, the Canadian MRL of 5 ppm for
fludioxonil on pomegranates was established in
2016 and was based on residue data of the United
States of America and the use of the OECD
MRL calculator. The Codex MRL of 2 ppm was
established in 2011 based on the same residue
data of the United States of America but using
the NAFTA calculator. Based on the timing of
when each jurisdiction reviewed the data and
the policies in place at the time the data were
reviewed, different MRLs were established.
Engagement of JMPR in parallel review with
national authorities may lead to the timelier
setting of Codex MRLs and alignment between
national and Codex values.
80
annex PART B
Once an European Union MRL is None United States of America law does not
established, it applies to food on the distinguish between MRLs for domestic
European Union market regardless of the production and imported food. MRLs
country of origin, i.e. equally to products apply equally to domestic and imported
produced in the European Union and products. United States of America policy
imported from outside the European allows for the setting of a higher MRL
Union. European Union MRLs do not to facilitate trade, so long as the MRL
apply to products exported from the meets the safety standard.
European Union to a non-European Union
country, if it has been established by The United States Environmental
appropriate evidence that the country of Protection Agency must be able to make
destination requires or agrees with the a safety finding when setting MRLs,
particular treatment. i.e., a finding that the pesticide can be
used with a reasonable certainty of no
harm, taking into account the special
susceptibility of children by applying
an additional tenfold safety factor, and
also considering aggregate risk from
exposure to pesticides from multiple
sources (food, water, residential, and
other non-occupational exposure), as well
as cumulative exposure to pesticides that
have a common mechanism of toxicity.
This safety standard applies equally to
domestic and imported food.
Currently, the European Union does not

set MRLs for feed-only products (e.g.
straw) or for processed products (e.g.
ketchup). It is thus not possible to align to
Codex MRLs in these areas.
Has a regular consideration of Codex Has a regular consideration of Codex No procedures in place
MRLs MRLs
Not applicable to the European Only when Codex adopts an MRL that is Participation by JMPR in a joint review
Union. The European Union routinely higher than the Japanese MRL. when a new active substance is first
implements new Codex MRLs every brought to market would potentially help
year, unless (1) the European Union MRL The Food Safety Commission of Japan to address this issue. Further exploration
is the same or higher, (2) the European conducts risk assessments for each of the benefits and challenges to JMPR
Union does not set MRLs for that component, and Ministry of Health, participation in global joint reviews is
commodity, or (3) the European Union Labour and Welfare of Japan sets and underway for discussion at CCPR51.
introduced a reservation, for which many reviews MRLs based on their results.
of the possible reasons are captured Therefore, Codex MRLs might not be The United States of America review
above. The European Union also takes old adopted immediately. Codex MRLs are each registered pesticide at least every
Codex MRLs (i.e. those established prior not adopted in cases where estimated 15 years to ensure that it continues to
to the current European Union policy chronic and acute exposure exceeds meet the standard for registration. During
and MRL legislation) into account when ADI or ARfD in exposure assessment this review the United States of America
reviewing MRLs for substances. using “Food Intake and Frequency seek to harmonize its tolerances with
Questionnaire in Japan”. Codex MRLs.
When the residue definition of Japan An applicant may also submit a request
differs from that of Codex, Japan may for modification of an existing tolerance
establish MRLs that differ from Codex in the United States of America, including
MRLs. a request to harmonize with the Codex
MRL.
81
© Veejay Villafranca/NOOR for FAO
2 Annex PART B
Analysis of data on pesticides
used on rice
The information in this section provides some real- Notes:
world examples of the differences between residue
The JMPR ADI value is the upper bound of the range.
definitions, MRLs for rice and HBGVs across some of
X in the ARfD column indicates not considered due to the age
the countries/region analysed in this study. A similar
of the evaluation.
evaluation was performed for Australian data but has
CXL = Codex MRL
not been presented here, in an effort to keep the table
NA = No information available.
legible. Some of the conclusions from this evaluation
UN = Unnecessary – compound has no toxicity associated
of Australian data are summarized in the main text –
with a single dietary exposure
the tabulated values can be made available on request.
ADI and ARfD values were taken from the WHO
database (https://apps.who.int/pesticide-residues-
The information was obtained from publicly available
jmpr-database) and European Union database (https://
databases or reports in November–December 2018 and
ec.europa.eu/food/plant/pesticides/eu-pesticides-database/
checked for any updates in August–September 2019.
public/?event=activesubstance.selection&language=EN).
Intake as % ADI or % ARfD are taken from the most
Where there are MRLs for multiple descriptions of
recent JMPR report item found to contain the appropriate
rice, or the term cereals is used, in the databases of
information.
the United States of America and CODEX, the table
ADI and ARfD ratios are JMPR / European Union: values >1
includes information on the description of the rice
indicate JMPR Health Based Guidance Values are higher than
associated with the cited MRL. The European Union
those of the European Union.
only uses the term ‘rice’. United States of America
PTDIs are set for contaminants, not pesticides in current use.
MRLs for ‘aspirated fractions’ are not used because
The list consists of the all pesticides for which a Codex MRL
they appear to have no equivalent in the Codex system
for rice exists.
(they are typically very high relative to MRLs for grain).
The information in this table is not intended

to be a definitive description of the values, but
to provide information for internal use in the
production of this project report. IT SHOULD NOT
BE USED FOR OTHER PURPOSES WITHOUT BEING
INDEPENDENTLY CHECKED.
83
Codex Common name ADI ADI ARfD ARfD

code (mg/kg bw) Ratio (mg/kg bw) Ratio
JMPR European JMPR European

Union Union
1 20 2,4-D 0.01 0.02 0.5 UN 0.3 ∞
2 177 Abamectin 0.001 0.0025 0.4 0.003 0.005 0.6
3 95 Acephate 0.03 NA ∞ 0.1 NA ∞
4 280 Acetochlor 0.01 0.0036 2.8 1.0 1.5 0.7
5 1 Aldrin and Dieldrin 0.0001 0.0001 1.0 X 0.003 ∞
6 229 Azoxystrobin 0.2 0.2 1.0 UN UN 1.0
7 172 Bentazone 0.09 0.09 1.0 0.5 1.0 0.5
8 221 Boscalid 0.04 0.04 1.0 UN UN 1.0
9 47 Bromide ion 1.0 NA - X No entry 1.0

(see MeBr)
10 8 Carbaryl 0.008 0.0075 1.1 0.2 0.01 20.0
11 72 Carbendazim 0.03 0.02 1.5 0.1 0.02 5.0
12 96 Carbofuran 0.001 0.00015 6.7 0.001 0.00015 6.7
13 230 Chlorantraniliprole 2.0 1.56 1.3 UN UN 1.0
14 12 Chlordane 0.0005 0.0005 ∞ X NA 1.0
15 17 Chlorpyrifos 0.01 0.001 10.0 0.1 0.005 20.0
16 90 Chlorpyrifos-Methyl 0.01 0.01 1.0 0.1 0.1 1.0
17 238 Clothianidin 0.1 0.097 1.0 0.6 0.1 6.0
18 179 Cycloxydim 0.07 0.07 1.0 2.0 2.0 1.0
19 146 Cyhalothrin (includes 0.02 0.0025 8.0 0.02 0.005 4.0

lambda-cyhalothrin) (lambda- (lambda-
cyhalothrin) cyhalothrin)
20 118 Cypermethrins (including 0.02 0.05 0.4 0.04 0.2 0.2

alpha- and zeta-
cypermethrin)
84
annex PART B
Intake Intake Components in MRLs for rice (mg/kg) Comment
JMPR % JMPR % residue definition
ADI ARfD
JMPR USA European Codex European USA

Union Union
20 NA 1 1 4 0.1 0.1 0.5 USA includes conjugates

husked Grain determined as acid
5 60 1 - 3 0.002 0.01* NA USA 0.4 on aspirated fractions

husked
0 4 2 - 1 1 0.01* NA Not approved in European Union

husked
4 1 2 - 1 0.04 0.01* 0.05 Not approved in European Union

wild rice
<100 NA 2 2 2 0.02 NA 0.02 PTDI. Not approved in European

cereal cereal Union
20 NA 1 2 1 5 5 5 USA mentions E & Z isomers

husked
0 <3 1 3 4 0.01* 0.1 0.05

cereal husked
40 NA 2 - 3 0.1 0.15 0.20 USA 3 on aspirated fractions

cereal husked cereal
- - 1 - 1 50 50 50 Br ion not mentioned in USA lists.

cereal husked See also methyl bromide
560 - 1 1 1 1 0.01 15 Not approved in European Union;

polished husked grain but other cereal MRLs 0.5.
- 0 3 - 2 2 0.01* NA JMPR ARfD for general population

husked = 0.5. Not approved in European
Union
30 4 2 4 2 0.1 0.01* 0.2 Not approved in European Union.

husked grain US is import tolerance
1 NA 1 1 1 0.4 rice 0.4 0.15 CXL 0.04 for rice polished

husked
- NA 2 - 2 0.02 NA NA PTDI. Not approved in European

Union but cite JMPR PTDI. No
cereal MRLs listed in European
Union, other crops have MRLs
1 <4 1 - 1 0.5 0.5 NA

husked
110 20 1 2 1 NA 3 6 USA 30 for polished

husked grain
3 - 1 1 1 0.5 0.5 0.01

husked
50 <1 3 - 4 0.09 0.09 NA JMPR ARfD UN for general

rice husked population
9 9 1 2 1 1 0.01 1 Only lambda-cyhalothrin approved

husked in European Union. USA lists
isomers separately
30 20 1 - 1 2 2 1.50
husked rice
grain
85


Union Union
21 21 DDT 0.01 0.01 1.0 UN NA 1.0
22 135 Deltamethrin 0.01 0.01 1.0 0.05 0.01 5.0
23 274 Dichlobenil 0.01 0.01 1.0 0.5 0.45 1.1
24 25 Dichlorvos 0.004 0.0008 5.0 0.1 0.002 50.0
25 224 Difenoconazole 0.01 0.01 1.0 0.3 0.16 1.9
26 130 Diflubenzuron 0.02 0.1 0.2 UN UN 1.0
27 255 Dinotefuran 0.2 NA ∞ 1.0 NA ∞
28 184 Etofenprox 0.03 0.03 1.0 1.0 1.0 1.0
29 37 Fenitrothion 0.006 0.005 1.2 0.04 0.013 3.1
30 39 Fenthion 0.007 NA ∞ 0.01 NA ∞
31 202 Fipronil 0.0002 0.0002 1.0 0.003 0.009 0.3
32 211 Fludioxonil 0.4 0.37 1.1 UN UN 1.0
33 243 Fluopyram 0.01 0.012 0.8 0.5 0.5 1.0
34 205 Flutolanil 0.09 0.09 1.0 UN UN 1.0
35 256 Fluxapyroxad 0.02 0.02 1.0 0.3 0.25 1.2
36 175 Glufosinate-Ammonium 0.01 0.021 0.5 0.01 0.021 0.5
37 114 Guazatine 0 0.0048 ∞ X 0.04 ∞
38 43 Heptachlor 0.0001 0.0001 1.0 X NA 1.0
39 46 Hydrogen Phosphide UN 0.022 ∞ UN 0.038 ∞

(phosphine; phosphane)
86
annex PART B
ADI ARfD

Union Union
NA NA 4 3 4 0.1 0.05 0.50 PTDI. Not approved in European

cereal husked cereal Union but cite JMPR PTDI
50 - 1 3 1 2 1 1 USA mentions 3 isomers

cereal husked cereal separately
1 <1 1 - 1 0.01* 0.01 NA JMPR ARfD UN for general

cereal husked population. Not approved in
European Union. Residue as
2,6-dichlorobenzamide
30 70 1 - 1 7 0.01* NA Not approved in European Union.

rice husked 3 CXLs (0.15 to 7 mg/kg). USA 0.5
for bagged processed food
70 NA 2 1 1 8 3 7
rice husked
20 NA 1 3 1 0.01* 0.05 0.02 Pome fruit MRLs at 5 mg/kg

husked
2 2 3 3 NA 8 8 9 Not approved in European Union

rice husked CXL 0.3 rice polished
3 <1 1 1 1 0.01* 0.5 0.01

husked
780 190 1 - 1 6 0.05 NA Not approved in European Union

cereal husked
- - 4 - 4 0.05 0.01* NA Not approved in European Union

husked
- - 2 4 2 0.01 0.005 0.04 Not approved in European Union.

husked
90 NA 1 1 1 0.05 0.01 0.02 Fruit MRLs 5–10 mg/kg

80 NA 1 - 1 4 0.01 NA Other European Union cereal

rice husked MRLs up to 1.8 mg/kg. No CXLs
for cereals
1 NA 1 1 1 2 2 7 4 CXLs for rice. US & JMPR res.

husked husked grain def. is for all TFMBA components;
European Union has flutolanil other
than for fruits.
20 2 3 1 1 5 5 5 4 CXLs for rice

rice husked grain
10 5 3 3 4 0.9 0.9 1 Not approved in European Union-

rice husked Cat 1B repro
NA NA 1 - 1 0.05* 0.05 NA ADI could not be established by

husked JMPR. Not approved in European
Union
- - 2 2 2 0.02 0.01 0.03 PTDI. Not approved in European

cereal husked cereal Union but cite JMPR PTDI
- - 1 1 0.1 NA 0.1 JMPR - Based on absence of

cereal residues in food
87


Union Union
40 276 Imazamox 3.0 3.0 1.0 3.0 3.0 1.0
41 266 Imazapic 0.7 0.46 1.5 UN UN 1.0
42 289 Imazethapyr 0.6 NA ∞ UN NA ∞
43 206 Imidacloprid 0.06 0.06 1.0 0.4 0.08 5.0
44 111 Iprodione 0.06 0.02 3.0 X 0.06 ∞
45 299 Isoprothiolane 0.1 NA ∞ UN NA ∞
46 277 Mesotrione 0.5 0.01 50.0 UN 0.02 ∞
47 138 Metalaxyl 0.08 0.08 1.0 UN 0.5 ∞
48 100 Methamidophos 0.004 0.001 4.0 0.01 0.003 3.3
49 147 Methoprene 0.09 NA ∞ UN NA ∞
S-methoprene 0.05 NA ∞ UN NA ∞
50 52 Methyl Bromide Bromide ion 0.001 ∞ X 0.003 ∞

(bromomethane) only
51 57 Paraquat 0.005 0.004 1.3 0.006 0.005 1.2
52 120 Permethrin 0.05 NA ∞ 1.5 NA ∞
53 62 Piperonyl Butoxide 0.2 NA ∞ UN NA ∞
54 86 Pirimiphos-Methyl 0.03 0.004 7.5 0.2 0.15 1.3
55 142 Prochloraz 0.01 0.01 1.0 0.1 0.025 4.0
56 63 Pyrethrins 0.04 0.04 1.0 0.2 0.2 1.0
57 287 Quinclorac 0.4 NA ∞ 2.0 NA ∞
88
annex PART B
ADI ARfD

Union Union
0 0 2 - 1 0.01* 0.05 NA
husked
0 NA 1 - 1 0.05* 0.05 NA Not approved in European Union

husked
0 NA 1 2 NA 0.1* NA 0.3 Not approved in European Union

grain
5 - 2 - 1 0.05 1.5 0.05

cereal husked grain
- - 1 3 1 10 10 10 Not approved in European Union

husked husked grain
- NA NA - 1 6 6 NA Not approved in European Union.

husked husked CXL 1.5 polished
0 NA 1 - 1 0.01* 0.01* NA
husked husked
- NA 1 2 1 0.05* 0.01 0.1 Includes metalaxyl M. Some

cereal husked cereal European Union MRLs on fruit up
to 2 mg/kg
- - 1 - 1 0.6 0.01* NA Not approved in European Union

husked husked
40 NA 1 - 1 10 cereal 5 NA Not approved in European

husked Union. Includes S-methoprene in
European Union. CXL 40 for rice
hulls
60 NA NA - 1 10 cereal 5 NA Not approved in European Union.

from husked Codex does not list separately -
methoprene from methoprene
from
methoprene
- - 1 1 1 5 NA 50 Res def. as bromide ion (CXL,

cereal USA + European Union). European
Union ref doses based on MeBr
not Br ion. Not approved in
European Union. CXL 0.01 as
MeBr in cereal products
- 0 1 1 1 0.05 0.05 0.05 Not approved in European Union

husked
- - 1 - 1 2 NA NA Not approved in European Union.

USA 0.5 aspirated fraction
- NA 1 1 NA 30 NA 20 Not considered a PPP in European

cereal Union. Used in biocides
- 30 1 - 1 7 0.5 NA European Union MRLs for cereals

cereal husked up to 5 mg/kg. Codex value is for
cereal grains
- - 2 - 2 2 1 NA European Union citrus MRLs at

cereal husked 10 mg/kg
1 - 1 3 1 0.3 3 3 No CXL for rice - 0.3 is cereal grain

husked
0 1 3 1 NA 10 5 5 Not approved in European Union

rice husked
89


Union Union
58 251 Saflufenacil 0.05 NA ∞ UN NA ∞
59 259 Sedaxane 0.1 0.11 0.9 0.3 0.3 1.0
60 233 Spinetoram 0.05 0.025 2.0 UN 0.1 ∞
61 203 Spinosad 0.02 0.024 0.8 UN UN 1.0
62 218 Sulfuryl fluoride 0.01 0.014 0.7 0.3 0.7 0.4
63 189 Tebuconazole 0.03 0.03 1.0 0.3 0.03 10
64 196 Tebufenozide 0.02 0.02 1.0 0.9 NA ∞
65 223 Thiacloprid 0.01 0.01 1.0 0.03 0.03 1.0
66 143 Triazophos 0.001 0.001 1.0 0.001 0.001 1.0
67 213 Trifloxystrobin 0.04 0.1 0.4 UN 0.5 ∞
68 303 Triflumezopyrim 0.2 NA ∞ 1.0 NA ∞
* = LOQ
90
annex PART B
ADI ARfD

Union Union
20 NA 1 3 3 0.01 0.03 0.03 Not approved in European Union

0 0 1 1 1 0.01* 0.01* 0.01

cereal cereal
2 NA 3 - 1 0.02* 0.05* NA
husked husked
40 NA 2 2 2 1 2 1.5
1 0 1 1 1 0.1 0.05 0.04 Separate consideration for F in

polished husked grain some schemes. USA does not
mention fluoride but has 6 different
& husked MRLs related to rice
0.01
polished
9 5 1 1 1 1.5 1.5 NA USA 16 aspirated fractions

rice husked
- - 1 - 1 0.1 3 NA
husked husked
10 0 1 - 1 0.02 0.02 NA Some fruit MRLs at 1 in European

husked Union
NA 100 1 - NA 0.6 0.02 NA Not approved in European Union

polished husked but cite JMPR ref doses
4 7 2 2 2 5 5 3.5
rice husked
1 NA NA 1 NA 0.2 0.01 0.4 USA MRL is for import tolerance. 4

rice husked grain CXLs for rice
91
Understanding international harmonization of pesticide Maximum Residue Limits (MRLs) with Codex
standards: a case study on rice.
The FAO/WHO Codex Alimentarius is the international point of reference on food safety and quality. Internationally
recognized food standards developed by Codex, including pesticides Maximum Residues Limits (MRLs), aim
to protect consumers’ health and ensure fair practices in international food trade. Despite long-standing efforts
towards international harmonization of allowable thresholds for pesticide residues in foods, differences in
the national implementation of MRLs continue to cause trade concerns. This publication explores international
harmonization with Codex pesticides MRLs from different angles, using a case study on rice. Part A assesses the
level of harmonization of pesticide MRLs in main rice producing and trading countries and explores its effects on
trade. Part B investigates the reasons behind varying levels of harmonization, looking at differences in national risk
assessment procedures and risk management policies that may lead to divergent MRLs. Ultimately this publication
aims to offer insights for decision-makers involved in standards setting and designing of food policies to facilitate
better international harmonization.

MRL Book

Uploaded by

Copyright:

Available Formats

MRL Book

Uploaded by

Document Information

Original Description:

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

MRL Book

Uploaded by

Copyright:

Available Formats

Understanding international harmonization

of pesticide maximum residue limits

Food and Agriculture Organization of the United Nations

PART A Level of harmonization of rice PART B Reasons behind different levels

1 Global rice market 5

2  esticides used on rice and harmonization

4 Conclusions and way forward 27 Glossary 59

ANNEXES PART A ANNEXES PART B

4 Methodology and data of the trade model 69

5 Detailed results of the trade model 72

Markus Lipp Boubaker Ben-Belhassen

For many years, trade issues related to differences

Figure 1: Tariffs and AVEs of NTMs, by economic sector

Tariffs AVEs of NTMs

Oils and Fats

Oil , Gas, Coal

Mining and Metal Ores

Paper Prod, Publishing

Wood Prod. Furniture

Source: UNCTAD and the World Bank, 2018

1 Global rice market

Figure 2: Rice available for consumption (kcal/capita/day), average 2015-2017

Source: Based on FAOSTAT data

6 A non-negligible share of rice trade is also conducted informally,

World 767 904 795

China 211 966 333

India 168 260 000

Indonesia 81 180 120

Bangladesh 53 672 728

Viet Nam 43 307 314

Thailand 30 064 074

Myanmar 25 571 946

Philippines 18 656 562

Brazil 11 612 049

Pakistan 10 750 577

Source: Own calculation based on FAOSTAT database

PRODUCTION TOTAL AVERAGE Global share

World 502 060 100.0

Developed regions 16 473 3.3

Developing regions 485 587 96.7

Low-Income Food-Deficit Countries (LIFDC) 199 869 39.8

Least Developed Countries (LDC) 76 013 15.1

TOTAL DOMESTIC USE TOTAL AVERAGE Global share

World 499 754 100.0

Developed regions 17 948 3.6

Developing regions 481 807 96.4

Low-Income Food-Deficit Countries (LIFDC) 192 855 38.6

Least Developed Countries (LDC) 83 480 16.7

USD thousand Share

World 24 014 054 100.0

Note: Intra-European Union trade is excluded

Table 4: Top ten rice exporters in the world, average 2017-2019

USD thousand Share

World 24 292 576 100.0

India 7 183 787 29.6

Thailand 4 976 079 20.5

Viet Nam 2 233 323 9.2

United States of America 1 762 037 7.3

2 esticides used on rice and harmonization

2 Pesticides used on rice and