Trade scenarios compensating for halted wheat and maize exports from Russia and Ukraine increase carbon emissions without easing food insecurity

The Russian invasion of Ukraine has destabilized global agricultural markets, triggering food price increases. We present scenarios of reduced exports and production affecting both countries that increase maize and wheat prices by up to 4.6% and 7.2%, respectively. Production expansion in other regions can partially compensate for export declines but may increase carbon emissions and will exacerbate ongoing global food security challenges. Global agricultural markets can partially compensate for halted crop exports from Ukraine and Russia by increasing wheat and maize production in other areas, but carbon emissions and global food insecurity will also increase.

policy option to compensate for the lack of Ukrainian exports. The last scenario can be justified by Ukrainian farmers being unable to produce due, for example, to war-induced destruction of infrastructure and equipment.
The trade model presents results after production adjustments have taken place over a year, and thus does not account for short-term spikes. The results focus on countries other than Russia and Ukraine due to the difficulty of assessing agricultural production in those places given the current situation. In scenario A1, 'no exports from Ukraine' , global maize and wheat prices are higher by 3.9% and 3.6%, respectively, compared with the baseline ( Table 1). Assuming that Russia reduces its grain exports by 50% (in addition to no Ukrainian exports), price increases of 4.6% and 7.2% for maize and wheat, respectively, are observed. Russia is more important than Ukraine in exporting wheat, which explains the sharper price increase in this scenario. Those increases are lower that the 8-22% price surges estimated by the Food and Agricultural Organization of the United Nations (FAO) 9 . This can be explained in part by the inclusion of higher inputs costs in the FAO analysis as a result of the conflict, and by the reduced capacity of alternative producers to increase output. Our model assumes full adjustment capacity. The increases are on top of the already elevated and rising preconflict levels that were reflecting the difficulty of some countries, such as Pakistan and Egypt, to acquire enough food for a healthy diet 10 . The biofuel reduction policies in the presence of no Ukrainian exports have a price impact that is substantial for the feedstock crops (mostly maize for ethanol and rapeseed for biodiesel) but modest for wheat because the supply gap remains due to no Ukrainian wheat exports. The supply gap caused by the decline in exports is partly closed by an increase in crop production in other countries (Table 1). In the 'no exports' scenario, some of the major wheat-producing countries increase their exports by double-digit percentages; these include India (72.4%), the European Union (36.1%), the United States (24.2%), and Argentina (11.4%). There is also a substantial increase in Australian and Canadian wheat exports of 9.4% and 8.6%, respectively. For maize, which is the other commodity affected by the war in Ukraine, Brazil and the United States increase their exports by 13.8% and 15.6%, respectively.
In scenario A1, 'no exports from Ukraine' , aggregate (on a caloric basis) per-capita consumption of barley, maize, rice, sorghum and wheat changes to between −1.2% and 0.1% across countries due to price increases (Fig. 1). The range increases to −2.0% to 0.4% if Russian Trade scenarios compensating for halted wheat and maize exports from Russia and Ukraine increase carbon emissions without easing food insecurity Miguel Carriquiry 1 , Jerome Dumortier 2 ✉ and Amani Elobeid 3 exports are reduced by 50% in addition to no Ukrainian exports. At the lower end, the decrease almost doubles because of the importance of Russia as a wheat exporter. In scenario A3, 'no exports from Ukraine and 50% biofuel reduction in the European Union and the United States' , the range of per-capita food consumption changes to −1.0% to 0.3%. In the model region 'rest of the world' , which contains many low-income countries, per-capita food consumption is least changed (−0.2%) in the scenario in which the European Union and the United States reduce their biofuel use. The reduction in food consumption is more pronounced for the individual crops. Maize and wheat consumption is reduced by up to 1.5% (Argentina) and 1.1% (Brazil), respectively, without Ukrainian exports. This is in stark contrast to the scenario where less biofuel is used in European Union and the United States. Per-capita maize consumption increases between 0.6% and 1.3% in all countries, whereas wheat consumption is still reduced by up to 1.0%. The decrease in the food consumption of wheat and maize in scenario A1, 'no exports from Ukraine' , is partially compensated by an increase in the consumption of rice. However, this increase for all countries is below 0.4%. The region 'rest of the world' accounts for approximately 2.5 bn people including the world's poorest. None of the scenarios analysed increases their per-capita food consumption. Many of those countries, especially in Africa, have low baseline per-capita caloric intake, a high number of people in food-insecure situations and a high reliance (as a share of total caloric intake) on grains, roots and tubers. Thus, even a seemingly small decrease in caloric intake can have adverse effects of undernourishment and food security.
To increase production and close the supply gap in scenario A1, 'no exports from Ukraine' , major agricultural wheat producers (other than Ukraine and Russia) increase crop area. Australia, China, the European Union and India increase their wheat area by 1.0%, 1.5%, 1.9% and 1.2%, respectively. Across the four scenarios analysed, the total global cropland area (excluding Ukraine and Russia) increases by at least 6.6 Mha (in scenario A3, 'no exports from Ukraine and 50% biofuel reduction in the European Union and the United States') and up to 18.2 Mha (in scenario A2, 'no exports from Ukraine and 50% export reductions from Russia'). The increase in Brazilian area is 1.3 Mha (with maize being responsible for more than half of the increase) in the no-Ukrainian-exports scenario, which is problematic in view of greenhouse gas (GHG) emissions due to the country's biomass and soil carbon stock as well as biodiversity 11 .
Total increase in global crop area (not including Russia and Ukraine) is 11.1 Mha, an increase of 1.4% relative to the baseline in scenario A1, 'no exports from Ukraine' . This crop area expansion due to the invasion of Ukraine by Russia can lead to significant carbon emissions from land-use change. Using mean carbon coefficients, land-use change emissions are 1,011.2 MtCO 2 e (Extended Data Fig. 1). Land-use change and emissions from Russia and Ukraine are excluded in this number because it is difficult to quantify area changes at this point. Thus, the emission numbers mentioned represent upper bounds. Compared to other estimates in GHG emissions from different macroeconomic developments or policy scenarios, the land-use change emissions from the Russian invasion into Ukraine are significant. For example, analysing the emissions from a 30% reduction in biofuels in the United States and the European Union from an increase in fuel efficiency and vehicle electrification results in a decrease in emission of between 188.8 and 468.1 MtCO 2 e for minimum and maximum carbon coefficients, respectively 12 . The land-use-change-induced emissions are reduced to 527.2 MtCO 2 e if EU and US biofuel use is cut in half.
The increase in maize production in Brazil represents an important contribution to the aforementioned carbon emissions. Scenario A3, 'no exports from Ukraine and 50% biofuel reduction in the European Union and the United States' , leads to an increase in the exports of US maize at the expense of exports from other countries. For example, maize exports from Brazil decrease compared to the baseline if the United States reduces its biofuel production by 50%. This changes the emission profile in Brazil notably (Extended Data Fig. 1). There are important reductions in terms of lower GHG from land use, although those are not all net gains as less biofuels will lead to an increase in fossil fuel use.
While the Russia-Ukraine grain agreement from July 2022 is a positive development, the situation in Ukraine and the status of agricultural exports remains uncertain. The attack on the port of Odesa or, potentially, mines in the Black Sea have made grain shipments expensive and far below normal (so far, less than 0.4 Mt). Our analysis presents plausible ranges from no exports to some production shortfall from Ukraine to give a sense of impacts given the unknown outcome and end of this war. However, it assumes that countries are able to respond to price signals by increasing production and trade. Yet, drought conditions in South America, the decision by major producing countries (for example, Argentina and Indonesia) to curb exports of agricultural commodities, and high fertilizer costs are exacerbating food insecurity in many poor communities. Policy for aiding vulnerable populations could include domestic food subsidies and the reduction or elimination of trade restrictions. The effect of future climate change can be mitigated by unrestricted trade, allowing a shift of comparative advantage across countries. These and other multifaceted approaches will be needed in the near and long term 13 . Although price increases are dampened by area and production expansions in other countries, this may come at the expense of potentially large carbon emissions-highlighting how trade and production disruptions in Russia and Ukraine have the double impact of compromising global food security and efforts to mitigate climate change.

Data availability
All data required to evaluate the conclusions in the paper are present in the paper and/or the Supplementary Information. The output of the agricultural trade model is available at www. github.com/foodclimate. The repository also includes the codes to generate all results, figures and GHG calculations based on the agricultural trade model output. Source data are provided with this paper.

Code availability
Details and code for the agricultural trade model are available from the corresponding author upon request.  Corresponding author(s): JEROME DUMORTIER Last updated by author(s): Aug 10, 2022 Reporting Summary Nature Research wishes to improve the reproducibility of the work that we publish. This form provides structure for consistency and transparency in reporting. For further information on Nature Research policies, see our Editorial Policies and the Editorial Policy Checklist.

Statistics
For all statistical analyses, confirm that the following items are present in the figure legend,

Data analysis
All data analyses and preparation of figures were done in Microsoft Excel and using R version 4.1.2 (2021-11-01) "Bird Hippie" Copyright (C) 2021 The R Foundation for Statistical Computing Platform: x86_64-w64-mingw32/x64 (64-bit).
For manuscripts utilizing custom algorithms or software that are central to the research but not yet described in published literature, software must be made available to editors and reviewers. We strongly encourage code deposition in a community repository (e.g. GitHub). See the Nature Research guidelines for submitting code & software for further information.

Data
Policy information about availability of data All manuscripts must include a data availability statement. This statement should provide the following information, where applicable: -Accession codes, unique identifiers, or web links for publicly available datasets -A list of figures that have associated raw data -A description of any restrictions on data availability All data required to evaluate the conclusions in the paper are present in the paper and/or the Online Material. The output of the agricultural trade model is available at www.github.com/foodclimate. The repository also includes the codes to generate all results, figures, and GHG calculations based on the agricultural trade model output.