PAPER TITLE: Global inequities in pesticide legislation: Nearly half of pesticides approved in major crops in Latin America are not allowed in the European Union AUTHORS: de Groot, G. S.*, Morales, C. L., Aldea-Sanchez, P., Aizen, M. A., Antúnez, K., Arbulo, N., Basualdo, M., Branchiccela, B., Correa Benitez, A., Gutiérrez-Gamiño, E. Y., Herrera Salazar, N., Juri, P., Martinez, S.I., Molina, G. A. R., Pimentel Betancurt, D., Rodriguez, J., Salvarrey, S., Silva-Neto, C. M., Vázquez, D. E., Bogo, G. * grecia.degroot@comahue-conicet.gob.ar ABSTRACT: Pesticide use is a core strategy to control agricultural pests. Although international treaties and health recommendations call for banning hazardous pesticides, globally harmonized pesticide governance remains elusive. In Latin America (LATAM), the main net food exporter worldwide, agricultural pesticide use has increased by ~500% since 1990, compared to only ~3% in Europe. To assess the environmental and health rigor of LATAM’s pesticide legislation, we reviewed active ingredients (AIs) approved for ten major crops, and AIs banned at national level, in eight LATAM countries, assessed their hazardousness according to international standards, and legal status in the European Union (EU) and explored sources of variation in the number of AIs approved in LATAM. We identified 523 AIs approved and 236 banned in LATAM; ~50% of the approved and ~85% of the banned AIs in LATAM were either not approved or prohibited in the EU. Higher crop production and, to a lesser extent, export value were linked to more AIs approved in LATAM. These findings imply weaker regulatory frameworks for hazardous pesticides in LATAM, reflecting more permissive environmental policies compared to the EU. This reinforces the urgency of addressing asymmetries in pesticide governance and rethinking the prevailing agricultural paradigm. PUBLICATION INFORMATION: https://doi.org/10.1098/rspb.2025-0267 ________________________________________ GENERAL INFORMATION 1. Title of Datasets: Agriculture and pesticide regulation in Latin America: A comparative dataset with the European Union 2. Author Information Principal Investigator Contact Information Name: Dra. Grecia Stefanía de Groot Institution: Grupo Ecología de la Polinización (ECOPOL), Instituto de Investigaciones en Biodiversidad y Medio Ambiente (INIBIOMA), Consejo Nacional de Investigaciones Científicas y Técnicas (CONICET), Universidad Nacional del Comahue (UNCo) Email: grecia.degroot@comahue-conicet.gob.ar 3. Date of data collection: 2021-01-19 to 2021-03-14 4. Geographic location of data collection: Argentina, Brazil, Bolivia, Chile, Colombia, Costa Rica, Mexico, Uruguay, European Union 5. Information about funding sources that supported the collection of the data: This work was partially supported by “Safeguarding Pollination Services in a Changing World” (SURPASS2), funded by the Newton Fund Latin American Biodiversity Programme, awarded through the UKNERC(NE/S011870/1) and CONICET Argentina (RD 1984-19), Eva Crane Trust Fundation, (ECT_20240908C, ECTA_20230313 SOLATINA_V) through funds to the 2023-2024 SOLATINA workshops where the authors developed this project and by CYTED (Red COLMENA, grant 122RT0125). GSDG holds a postdoctoral scholarship from the National Research Council of Argentina (CONICET). 6. License information: CC-BY-NC-SA 7. HANDLE for data and code repository: http://hdl.handle.net/11336/273702 ________________________________________ DATA & FILE OVERVIEW Data - Agriculture & pesticides in LATAM & EU.xlsx: xlsx file containing all datasheets together - S1_agriculture_by_country.csv: csv file containing raw agricultural data by country for LATAM. - S2_agriculture_by_crop.csv: csv file containing raw agricultural data by crop across LATAM countries. - S3_AI_approved.csv: csv file containing raw data on AIs approved for each crop in LATAM countries, including their legal status in the EU and additional variables. - S4_AI_banned.csv: csv file containing raw data on AIs banned in each LATAM country, along with their legal status in the EU. - S5_GLMM_dataset.csv: csv file containing the processed dataset prepared for statistical modeling. - data_map.txt: txt file containing data processed to plot the LATAM map. - data_crop.txt: txt file containing data processed to plot estimated number of AIs approved in LATAM, and number and proportion of them not allowed in the EU, according to GLMMs output. Code - 01_Spearman_correlation.R: Spearman correlation between the number of harvested crops and the number of AIs approved per country. - 02_Pearson_correlation.R: Pearson correlation between the number of AIs approved in LATAM and the number of them not allowed in the EU. - 03_Target_organisms.R: Calculates the number of AIs approved in LATAM classified according to target organisms. - 04_LATAM_approved-banned.R: Calculates AIs approved, banned, and overlapping across LATAM. - 05_GLMM.R: GLMM analyses for active ingredients approval patterns in Latin America (LATAM). - 05_GLMM_applied.R: GLMM coefficients applied for estimation of changes in the number of AIs approved in Latin America (LATAM) related to changes in production and exportation, considering crop ubiquity. - [Figure 1].R: Plot map of LATAM showing total number of AIs approved in LATAM plus pie charts showing the proportion of AIs in each LATAM country that were approved and not allowed in the European Union. - [Figure 2].R: Plot AIs approved in LATAM classified following WHO recommendations, grouped by country and legal status in the EU. - [Figure 3].R: Plot total number of AIs banned in each LATAM country and their legal status in the EU. - [Figure 4].R: Plot total number of AIs banned in each LATAM country and number of them that are included in international treaties and/or health recommendations. - [Figure 5].R: Plot estimated values based on pure random models of the number of AIs approved in LATAM, and number and proportion of them not allowed in the EU by country and crop. ________________________________________ METHODOLOGICAL INFORMATION 1. Description of methods used for collection of data: Data on AIs approved for agricultural use were collected from eight LATAM countries: Argentina, Bolivia, Brazil, Chile, Colombia, Costa Rica, Mexico and Uruguay. Consulted sources of pesticide regulations were governmental open sources at the country level, including agricultural ministries or their dependencies, plus non-governmental open sources from the private sector (e.g., chambers of commerce). We recorded the number of AIs approved until the 31st of December 2020 in the selected LATAM countries in the following ten crops, according to FAO denomination: corn, soybean, wheat, sunflower, rice, sugar cane, coffee, apple, avocado and grapes. These crops were selected due to their importance in terms of cultivated area, relative to the total cultivated land, and (or) their economic value. Five of these crops (i.e., soybean, sunflower, coffee, apple, and avocado) depend to some extent on entomophilous pollination to produce either the seeds or fruits we consume, while the remaining five rely on anemophilous pollination. We focused our search on a subgroup of Plant Protection Products (PPPs), specifically those applied against other organisms (e.g., insecticides, herbicides, fungicides). When information was supplied as formulations, each AI was individually recorded. Compounds were standardized by common names, International Union of Pure and Applied Chemistry (IUPAC) nomenclature, and Chemical Abstracts Service Registry Number. Information regarding target organisms (i.e., animals, plants, or fungi), WHO hazardous classification, and legal status of each AI in the EU were extracted from the EU Pesticides Database, the Pesticides Properties DataBase (PPDB) and the Bio-Pesticides DataBase (BPDB). Data regarding AIs banned in the eight LATAM countries was obtained from official regulations of Agricultural Ministries and (or) Sanitary Authorities of each country. For each AI banned in each LATAM country, we recorded whether they were included in the Montreal Protocol, Rotterdam and Stockholm Conventions and WHO, as well as their legal status in the EU as of 31 December 2020. We obtained the legal status in the EU of each AI from PPDB and the European Chemicals Agency (ECHA). PPDB classifies each AI in one of the following categories: approved, not approved, pending and not yet assessed, while ECHA lists banned AIs. Therefore, according to the respective sources, we considered AIs as ‘not allowed’ in the EU if they were either not approved or banned. Data regarding the agricultural relevance of the eight Latin American (LATAM) countries were retrieved from Food and Agriculture Organization (FAO) and World Bank. Values were calculated as the average of five years (2015-2019). Data regarding harvested area and gross domestic production of the 10 selected crops during 2019 were retrieved from FAO. 2. Software-specific information: R version 4.4.2 (2024-10-31 ucrt) Platform: x86_64-w64-mingw32/x64 Running under: Windows 11 x64 (build 26200) Matrix products: default locale: [1] LC_COLLATE=English_United States.utf8 LC_CTYPE=English_United States.utf8 [3] LC_MONETARY=English_United States.utf8 LC_NUMERIC=C [5] LC_TIME=English_United States.utf8 time zone: America/Buenos_Aires tzcode source: internal attached base packages: [1] stats graphics grDevices utils datasets methods base other attached packages: [1] arm_1.14-4 ggrepel_0.9.6 ggnewscale_0.5.2 scatterpie_0.2.6 [5] RColorBrewer_1.1-3 rnaturalearthdata_1.0.0 rnaturalearth_1.1.0 sf_1.0-22 [9] broom.mixed_0.2.9.6 performance_0.15.2 rsq_2.7 DHARMa_0.4.7 [13] MASS_7.3-61 sjPlot_2.9.0 lme4_1.1-37 Matrix_1.7-1 [17] ggpattern_1.2.1 lubridate_1.9.4 forcats_1.0.1 stringr_1.6.0 [21] purrr_1.0.4 readr_2.1.6 tidyr_1.3.1 tibble_3.2.1 [25] tidyverse_2.0.0 ggplot2_4.0.1 dplyr_1.1.4 loaded via a namespace (and not attached): [1] tidyselect_1.2.1 farver_2.1.2 S7_0.2.0 tweenr_2.0.3 digest_0.6.37 [6] timechange_0.3.0 lifecycle_1.0.4 Deriv_4.2.0 magrittr_2.0.3 compiler_4.4.2 [11] rlang_1.1.5 tools_4.4.2 knitr_1.50 classInt_0.4-11 abind_1.4-8 [16] KernSmooth_2.23-24 withr_3.0.2 grid_4.4.2 polyclip_1.10-7 e1071_1.7-16 [21] future_1.68.0 globals_0.18.0 scales_1.4.0 insight_1.4.3 cli_3.6.4 [26] reformulas_0.4.2 generics_0.1.4 rstudioapi_0.17.1 tzdb_0.5.0 minqa_1.2.8 [31] DBI_1.2.3 ggforce_0.5.0 proxy_0.4-27 splines_4.4.2 parallel_4.4.2 [36] vctrs_0.6.5 yulab.utils_0.2.1 boot_1.3-31 hms_1.1.4 listenv_0.10.0 [41] units_1.0-0 glue_1.8.0 parallelly_1.45.1 nloptr_2.2.1 codetools_0.2-20 [46] stringi_1.8.7 gtable_0.3.6 mcr_1.3.3.1 furrr_0.3.1 pillar_1.11.1 [51] rappdirs_0.3.3 R6_2.6.1 Rdpack_2.6.4 evaluate_1.0.5 lattice_0.22-6 [56] rbibutils_2.4 backports_1.5.0 broom_1.0.10 ggfun_0.2.0 deming_1.4-1 [61] robslopes_1.1.3 class_7.3-22 Rcpp_1.1.0 coda_0.19-4.1 nlme_3.1-166 [66] xfun_0.54 fs_1.6.6 pkgconfig_2.0.3 ________________________________________ DATA-SPECIFIC INFORMATION FOR: S1_agriculture_by_country.csv 1. Number of variables: 10 2. Number of rows: 10 3. Variable List: Country: each LATAM country evaluated Mean land area 2015-2019 (ha): mean land area between 2015 and 2019 in hectares Mean cultivated land 2015-2019 (ha): mean cultivated land between 2015 and 2019 in hectares Percentage of Cultivated land relative to Land area Mean percentage of Agriculture participation in Gross Domestic Production 2015-2019: mean gross domestic production between 2015 and 2019 LATAM cultivated land (ha): Total cultivated land in LATAM in hectares Percentage of Cultivated land relative to LATAM cultivated land LATAM primary production (tn): primary production in tons Percentage of LATAM primary production Total harvested area (2019) (ha): total harvested area by country in 2019 4. Missing data codes: NA S2_agriculture_by_crop.csv 1. Number of variables: 8 2. Number of rows: 81 3. Variable List: country: each LATAM country evaluated crop: each LATAM crop evaluated Harvested area in 2019 (ha): total harvested area by crop in 2019 in hectares Total harvested area in 2019 (ha): total harvested area in each country in 2019 in hectares Percentage of 'Harvested area' relative to 'Total harvested area' Gross production (USD): gross production in United States Dolar Mean production 2016-2020 (tn): mean harvested production in tons between 2016 and 2020. data was first log transformed Mean export value 2016-2020 (USD): mean gross production value in USD between 2016 and 2020. data was first log transformed 4. Missing data codes: NA S3_AI_approved.csv 1. Number of variables: 23 2. Number of rows: 1791 3. Variable List: country: each LATAM country evaluated Active_ingredient: list of AIs approved in each LATAM country Common_name: common name of each AI IUPAC_name: reference name according to IUPAC CAS_number: reference CAS number found_in_EU: whether each AI was found or not in the EU sources state_in_EU: legal status in the EU WHO: hazardous category according o WHO Date EC 1107/2009 inclusion expires: date of approval expiration in the EU Information on expiry dates: whether and AI is already expired source: EU source of information category: classification according to target organisms main_target: main target organism SUNFLOWER: dichotomous variable that indicates whether the AI ​​is approved (1) or not (0) for sunflower CORN: dichotomous variable that indicates whether the AI ​​is approved (1) or not (0) for corn APPLE: dichotomous variable that indicates whether the AI ​​is approved (1) or not (0) for apple SOYBEAN: dichotomous variable that indicates whether the AI ​​is approved (1) or not (0) for soybean WHEAT: dichotomous variable that indicates whether the AI ​​is approved (1) or not (0) for wheat COFFEE: dichotomous variable that indicates whether the AI ​​is approved (1) or not (0) for coffee SUGAR: dichotomous variable that indicates whether the AI ​​is approved (1) or not (0) for sugar RICE: dichotomous variable that indicates whether the AI ​​is approved (1) or not (0) for rice AVOCADO: dichotomous variable that indicates whether the AI ​​is approved (1) or not (0) for avocado GRAPE: dichotomous variable that indicates whether the AI ​​is approved (1) or not (0) for grape 4. Missing data codes: NA S4_AI_banned.csv 1. Number of variables: 17 2. Number of rows: 355 3. Variable List: Common_name: common name of each AI CAS_number: reference CAS number Montreal Protocol 1987: whether the AI is included or not in the Montreal Protocol Rotterdam Convention 1998: whether the AI is included or not in the Rotterdam Convention Stockholm Convention 2001: whether the AI is included or not in the Stockholm Convention OMS 2019: WHO recommended classification of pesticides by hazard and guidelines to classification by 2019 EU banned: whether the AI is banned in the EU according to ECHA EU not approved: whether the AI is not approved in the EU according to PPDB Argentina: whether the AI is banned in Argentina according to SENASA Bolivia: whether the AI is banned in Bolivia according to MMAyA Brazil: whether the AI is banned in Brazil according to MAPA Chile: whether the AI is banned in Chile according to SAG Colombia: whether the AI is banned in Colombia according to ICA Costa Rica: whether the AI is banned in Costa Rica according to SFE Mexico: whether the AI is banned in Mexico according to INECC Uruguay: whether the AI is banned in Uruguay according to MGAP Total LATAM bans: summarize of total bans in LATAM for each AI 4. Missing data codes: NA S5_GLMM_dataset.csv 1. Number of variables: 9 2. Number of rows: 81 3. Variable List: country: each LATAM country evaluated crop: each LATAM crop evaluated n_pesticides: number of AIs approved in LATAM Approved: number of AIs approved in LATAM and approved in the EU Not_approved: number of AIs approved in LATAM and not approved in he EU Entom_poll: whether each crop is dependent (1) or not (0) on entomophilous pollination EU_crop: whether each crop is cultivated (1) or not (0) in the EU ExpValue_mean: mean gross production value in USD between 2016 and 2020. data was first log transformed Prod_mean: mean harvested production in tons between 2016 and 2020. data was first log transformed 4. Missing data codes: NA data_map.txt 1. Number of variables: 6 2. Number of rows: 9 3. Variable List: geounit: each LATAM country evaluated código: code for each country n_pesticides: total number of AIs approved in each LATAM country EU_approved: number of AIs approved in LATAM and approved in the EU EU_not_approved: number of AIs approved in LATAM and not approved in he EU EU_pend: number of AIs approved in LATAM and pending of resolution in he EU 4. Missing data codes: NA data_country.txt 1. Number of variables: 14 2. Number of rows: 9 3. Variable List: country: each LATAM country evaluated código: code for each country n_pesticides: total number of AIs approved in each LATAM country EU_approved: number of AIs approved in LATAM and approved in the EU EU_not_approved: number of AIs approved in LATAM and not approved in he EU EU_pend: number of AIs approved in LATAM and pending of resolution in he EU cult_land: total cultivated land in each LATAM country agricultural_land: total agricultural land in each LATAM country land: total land area of each LATAM country rel_cult_toAgric: cultivated area relative to total agricultural area for each LATAM country rel_cult_toLand: cultivated area relative to total land area for each LATAM country agric_PBI: total gross domestic production for agricultural production of each LATAM country pesticide_use: tons of active pesticide ingredients used in agriculture relative_pest_use: pesticide use relative to cultivated land for each LATAM country 4. Missing data codes: NA data_crop.txt 1. Number of variables: 10 2. Number of rows: 81 3. Variable List: country: each LATAM country evaluated crop: each LATAM crop evaluated n_pesticides: total number of AIs approved in each LATAM country Approved: number of AIs approved in LATAM and approved in the EU Not_approved: number of AIs approved in LATAM and not approved in he EU animal_pol: whether each crop is dependent (1) or not (0) on entomophilous pollination EU_crop: whether each crop is cultivated (1) or not (0) in the EU VolExp_mean: mean gross production volume in tons between 2016 and 2020. data was first log transformed ValorExp_mean: mean gross production value in USD between 2016 and 2020. data was first log transformed prod_mean: mean harvested production in tons between 2016 and 2020. data was first log transformed 4. Missing data codes: NA