The Digital Markets Act comes into force today. It is probably one of the most complex pieces of legislation in the European Union to date, alongside the Digital Services Act, as you can read in the News. And there is already much discussion about the next massive project. The AI Act is intended to define the game rules for the use of artificial intelligence in the future. There is still a long way to go before the trilogue, as Corinna Visser reports. But the Council under the Czech presidency sets a rapid pace.
The sale of Twitter to multi-billionaire entrepreneur Elon Musk is completed. But what will happen to the platform now? What rules will Musk impose? And what does this mean for the communication channels of the European authorities? Falk Steiner analyzed what could happen next.
The election victory of Luiz Ignacio Lula da Silva in Brazil fuels hope in Europe: German Chancellor Olaf Scholz is not the only one hoping for a revival of the trade agreement with the South American Mercosur states. Read the details in the News.
Finally, we take a look at nature: According to a recent report by the EU Commission, the 2021 wildfire season in Europe was the second worst in the history of reporting, which began in 2006. More than 5500 square kilometers of land in the European Union were burned – more than twice the area of Luxembourg. This year could be even more destructive: “After 2021 and 2017, in 2022, we are again experiencing one of the most dramatic wildfire seasons ever recorded,” said Marija Gabriel, EU research commissioner.
Work on the Artificial Intelligence Act (AI Act) has made good progress. But it is still a long way to the trilogue. While those involved, particularly in Parliament, would prefer to deliberate a little longer and more thoroughly, the Czech Council Presidency is pushing the pace. It has presented the fourth version of its compromise text.
The Parliament, on the other hand, sees the first half of the road behind it. The USA participates in the discussion and has sent a non-paper to Brussels. The definition of artificial intelligence (AI), the classification into risk classes and the treatment of general purpose AI are still controversial.
From informed circles, it is heard that the new compromise text of the Czech Council Presidency met with great approval in the Telecommunications Working Group on Oct. 25. It looks as if agreement can be reached on this compromise. This would also justify the speed at which the Council wants to move forward. The working group will meet on Nov. 8. On Nov. 18, the Permanent Representatives will meet, and on Dec. 6, the Ministers. They are then expected to reach a general approach for the trilogue.
Among other things, the fourth version contains the following changes:
The Parliament’s work is delayed by the very fact that there are two lead committees and thus twice the number of rapporteurs: for IMCO it is Brando Benifei (S&D) and for LIBE Dragoș Tudorache (Renew). They presented their latest draft on Oct. 21.
Among the key innovations in the text: The Commission is given new powers to expand the list of prohibited AI practices, AI applications that must meet enhanced transparency requirements, and areas of high-risk AI applications following an annual review. It will also be assisted by a newly created AI Board or Office. What form this oversight body will take is apparently still unclear.
Committee members discussed the draft in a joint IMCO-LIBE meeting last week. Tudorache tweeted afterwards, “we are making good progress“.
There is another way of looking at it. At the political level, there have only been two exchanges so far. And so there are hardly any final decisions, even though the working groups meet at the technical level, in some cases three times a week. For better processing, the working groups have divided the articles of the regulation into seven groups (batches).
On the status of the deliberations, Tudorache told the committee that he believed he had reached an agreement on Batch 1. This is about Articles 30 to 39 on notification and notifying authorities. “We’ve added a few tweaks to the Commission’s proposal to reduce the burden on startups, ensure cybersecurity and confidentiality, ensure impartiality of staff and conformity assessment bodies.”
On standards, conformity assessments, certificates, and registration (Batch 2), Tudorache said, “We’re still discussing fundamental rights and fundamental rights-related standards, and there are positions in some of our groups that want a different regime.”
On the other batches, too, it was said that much had been optimized and refined, but that discussions were still ongoing. Only on Batch 7 – secrecy, penalties, as well as final provisions – Tudorache said, “We just made an important breakthrough this week on penalties and fines, and that’s supported by a majority of us.”
A person familiar with the negotiations in parliament told Europe.Table that there were still many critical points that had not even been discussed. In the committee meeting, almost every speaker praised the constructive atmosphere in the negotiations. Nevertheless, each and every one of them still had their own wishes that they wanted to be taken into account.
In its non-paper, while the US starts by praising the EU for “developing a thoughtful architecture that addresses many of our shared AI goals,” it does have a number of “suggestions for improvement” for the AI Act. For example, in terms of intergovernmental cooperation. The US points out that it has developed its own AI standards. The planned EU legislation in its current form (the non-paper still refers to the third version of the Czech Council presidency from September) makes cooperation more difficult, they say. The US would like to see more concessions.
But the US also has suggestions for better, clearer, wording, as well as a recommendation to “clarify the relationship between the law and other existing and proposed EU legislation, such as the General Data Protection Regulation (GDPR), the EU Medical Devices Regulation, and others“.
Above all, however, the Americans would like to see “a coherent, globally applicable definition of the term” artificial intelligence. In addition, they criticize the definition of “artificial intelligence system” for still including systems “that are not sophisticated enough to merit special attention in the context of AI-focused legislation”. Therefore, the US is proposing a definition that captures the spirit of the OECD definition. Incidentally, this is also what many EU parliamentarians call for.
One thing is virtually certain: When the Digital Services Act becomes applicable law on April 17, 2024, Twitter will fall under it – and then be subject to the EU Commission’s DSA supervision. This is because the platform currently meets all the criteria to be subject to the rules of the DSA in the future as a very large platform in the EU, especially when it comes to moderating content that is relevant under criminal law. As became known on Monday, Musk had again assured Commission representatives that Twitter would abide by the rules.
But it was precisely in the area of moderating content relevant to criminal law that the new owner Elon Musk announced changes: “The bird is free,” he tweeted. What exactly Musk means by free remains unclear for the time being. Musk himself proved over the weekend that conspiracy theory nonsense could be part of it.
He posted a tweet after the alleged attempted assassination of House Speaker Nancy Pelosi, alleging that Pelosi’s husband, who was injured in the attack, was drunk. Musk deleted his tweet after a few hours, but it may have been a foretaste of what will happen on the platform in the future.
Instead of being outlawed, Twitter must abide by rules in the EU, digital commissioner Thierry Breton tweeted in Musk’s direction, reminding him of what he had promised after a meeting in May: He wanted to do with his companies anything that was good for Europe.
But for the EU, the problem is more complex than the issue of content moderation. If Twitter were to become a toxic environment for communication, the EU Commission, Council, Parliament, and many subordinate authorities would have to grapple with whether Twitter is the right place for them to communicate. The liberal Guy Verhofstadt, for example, is already publicly thinking about this – but would actually want something else.
Twitter has been the preferred communication platform for many political actors: The EU Commission’s account was created in 2010, the Parliament’s account in 2009, and the Council’s press account in 2010. Since then, the institutions and their actors have been using the platform intensively to bridge the gap between the public in Brussels and the public in the member states.
Despite the increased relevance of EU politics, the weights are still unevenly distributed: The german newspaper Bild-Zeitung, for example, has only one correspondent in Brussels, FAZ three, and Deutsche Presse-Agentur four. By comparison, dpa’s Berlin office has 23 correspondents, FAZ 14, and Bild 19.
This has long since ceased to reflect the political weighting between Berlin and Brussels – and yet the German media are still better represented than those of many other member states. Twitter is a way for the EU institutions to reach out to national publics. In addition, almost all relevant associations, think tanks and civil society actors are also present there.
And so EU representatives have also been looking for alternatives for some time – as have many users. The fediverse is a particular focus of attention: Services such as Mastodon, which are not identical to Twitter but are similar in some respects, are operated in a decentralized manner and are interoperable with one another via standardized protocols.
According to the Federal Ministry of Digital Affairs and Transport, which is responsible for the Digital Services Act in Germany, fediverse instances with less than 45 million users in Germany will also fall under the oversight. The driving force behind many of the instances set up by public institutions such as the Commission and ministries are, interestingly enough, the data protection commissioners – in Germany and Europe.
This could become an exciting oversight issue: After all, the data protection authorities act independently by virtue of their mandate. With their servers, however, as platform operators they would in turn fall under the supervisory authority of the DSA supervisors. “The scope of application of the DSA does not differentiate between public and private entities,” the BMDV informs us upon request.
The EU Commission account on Mastodon is followed by a good 13,000 other users on Monday. It was created at the end of April, when the first takeover rumors emerged. This is likely to increase significantly after the popularity of the fediverse instances in recent days. And other users who have been active primarily on Twitter must now also consider whether and to what extent they will now become active on other platforms.
But the alternatives differ considerably from Twitter in terms of their functionalities. And whether they are successful in the medium term depends heavily on whether the user base reaches the so-called critical mass, above which communication develops independently. This is also likely to depend on how many users use Twitter and other services in parallel, causing social media consultants quite some headache: Should they be early adopters and already build up a user base – or should they wait and save the effort?
However, should Elon Musk actively steer Twitter in a direction that is unacceptable from an EU perspective and want to exercise freedom of expression as he personally sees fit, the question could become louder as to whether EU institutions and politicians can remain on the platform and thus contribute to its success. In turn, abandoning Twitter would mean significantly reduced visibility. The coming months are also likely to be anxious ones for many communications officers and PR agencies.
Today, the Digital Markets Act (DMA) comes into force. The new regulation is intended to put an end to unfair practices by companies acting as gatekeepers in the online platform economy. The Commission had proposed the regulation in December 2020. In March 2022, the European Parliament and the Council adopted it.
There are three main criteria for classification as a gatekeeper in the DMA:
First, it is a matter of determining who belongs to the gatekeepers. From May 2, 2023, potential gatekeepers will have to notify the Commission when their central platform services reach the thresholds set out in the DMA.
The Commission then has until early September 2023 to designate the gatekeepers covered by the act. Once designated, gatekeepers have six months to comply with the requirements of the DMA.
To prepare for the enforcement of the DMA, the Commission will hold a series of technical working meetings. The first working meeting will be held on Dec. 5, 2022, and will focus primarily on the preferential treatment of gatekeepers’ own products and services. The Commission also works on an implementing regulation that will include provisions on procedural aspects. The Act gives the Commission the authority to conduct market investigations. In this way, the Commission can ensure that the established obligations are kept up to date.
Together with the DMA, the Digital Services Act (DSA) aims to create a safer digital space. In this space, the fundamental rights of users will be protected and a level playing field for companies will apply. The Digital Services Act will be directly applicable throughout the EU and it will apply from Jan. 1, 2024 at the latest. However, for very large online platforms and very large online search engines, the Digital Services Act will apply earlier, as early as four months after their designation as gatekeepers. vis
Between rising hunger and the prospect of still more supply shortages, dark clouds are hovering over the global food system. Not only has the war in Ukraine limited access to, and raised prices for, food and fertilizer, but extreme weather events are disrupting production, and economic downturns have diminished people’s ability to afford adequate and nutritious diets. Climate change is fueling droughts and exacerbating water scarcity, adding to the list of threats facing agricultural production.
Progress in tackling hunger and malnutrition was already stalling before the COVID-19 pandemic. Now, the World Food Program estimates that the number of people facing, or at risk of, acute food insecurity (defined as those who “lack regular access to enough safe and nutritious food for normal growth and development and an active and healthy life”) has increased by more than 200 million since 2019.
Key global objectives like the second United Nations Sustainable Development Goal – which commits countries to end hunger by 2030, and to improve nutrition and promote sustainable agriculture – are significantly off track.
Trade is vital for achieving food security, as the World Trade Organization’s 164 members emphasized at our 12th ministerial conference (MC12) this past June. Apart from its basic role as a conveyor belt for getting food from where it is abundant to where it is needed, trade supports jobs, livelihoods, and incomes. It is a key factor in the sustainable and efficient use of scarce global resources.
But all too often, international markets for food and agriculture function poorly, owing to problems like trade-distorting subsidies and high levels of protection. In many regions, persistent underinvestment in research, extension services, and market-linking infrastructure has led to low and stagnating agricultural productivity. Crop yields across much of Africa are particularly small, and have not matched the increases achieved in other developing regions.
WTO members have taken some important steps to address these problems. In 2015, they agreed to eliminate agricultural export subsidies – payments that cover the difference between domestic costs and international market prices – which are widely seen to distort production and harm farmers in other countries. The use of such subsidies has fallen dramatically, from almost $7 billion in 1999 to under $12 million in 2020.
But OECD data show that total public support to the farm sector in 54 advanced economies and 11 major emerging markets reached $817 billion in 2019-21. And only $207 billion of that was for general services such as research, training, or payments to consumers.
Other support programs frequently incentivize unsustainable production and consumption patterns and undermine fair competition in global markets. While average tariffs on agricultural products were around 6 percent in 2020, many countries maintained prohibitively high tariffs on certain products, with peaks reaching 1,000 percent.
At the beginning of this year, food prices had already reached high levels compared to recent years, and international agencies had begun to warn of growing hunger. Then the war in Ukraine made the situation much worse.
Prices spiked to record highs, and WTO monitoring found that many governments responded by restricting food exports, potentially easing strains at home but exacerbating shortages elsewhere in the world.
Against this backdrop, the multilateral cooperation that we saw at MC12 in June was most welcome. WTO members collectively committed to exempt World Food Program humanitarian purchases from export restrictions – saving the agency time and money in providing relief to millions of vulnerable people. They pledged to keep agri-food trade open, and to minimize the trade distortions resulting from emergency measures.
And they agreed on a binding accord to curtail $22 billion in annual public spending on harmful fisheries subsidies, which will contribute to food security and livelihoods by improving the health of marine habitats.
But over the past month, export restrictions on food, feed, and fertilizers have ticked back up, reversing the earlier downward trend. The total number of such measures introduced since February – and still in force – is now up to 52. The WTO is working hard through monitoring, transparency, and jawboning to bring this number down, but it remains a cause of global concern.
Ensuring that trade contributes to a more sustainable and equitable global food system will require greater efforts to align trade, environmental, and food policies. A holistic approach to managing how food is produced, distributed, and consumed is necessary to maintain sufficient supply, address climate and other environmental risks, ensure sustainable land use, and manage scarce water resources.
A long-overdue update of the WTO rulebook for agricultural trade is needed more urgently than ever, because we need to create a better incentive framework for investment in the sector.
To find solutions to growing food insecurity, we must expand the dialogue and explore the systemic drivers of the problem. With a stronger evidence base, WTO members will have a better understanding of how trade can be used to the fullest possible extent to enhance access to food for everyone.
In cooperation with Project Syndicate, 2022.
The Digital Markets Act comes into force today. It is probably one of the most complex pieces of legislation in the European Union to date, alongside the Digital Services Act, as you can read in the News. And there is already much discussion about the next massive project. The AI Act is intended to define the game rules for the use of artificial intelligence in the future. There is still a long way to go before the trilogue, as Corinna Visser reports. But the Council under the Czech presidency sets a rapid pace.
The sale of Twitter to multi-billionaire entrepreneur Elon Musk is completed. But what will happen to the platform now? What rules will Musk impose? And what does this mean for the communication channels of the European authorities? Falk Steiner analyzed what could happen next.
The election victory of Luiz Ignacio Lula da Silva in Brazil fuels hope in Europe: German Chancellor Olaf Scholz is not the only one hoping for a revival of the trade agreement with the South American Mercosur states. Read the details in the News.
Finally, we take a look at nature: According to a recent report by the EU Commission, the 2021 wildfire season in Europe was the second worst in the history of reporting, which began in 2006. More than 5500 square kilometers of land in the European Union were burned – more than twice the area of Luxembourg. This year could be even more destructive: “After 2021 and 2017, in 2022, we are again experiencing one of the most dramatic wildfire seasons ever recorded,” said Marija Gabriel, EU research commissioner.
Work on the Artificial Intelligence Act (AI Act) has made good progress. But it is still a long way to the trilogue. While those involved, particularly in Parliament, would prefer to deliberate a little longer and more thoroughly, the Czech Council Presidency is pushing the pace. It has presented the fourth version of its compromise text.
The Parliament, on the other hand, sees the first half of the road behind it. The USA participates in the discussion and has sent a non-paper to Brussels. The definition of artificial intelligence (AI), the classification into risk classes and the treatment of general purpose AI are still controversial.
From informed circles, it is heard that the new compromise text of the Czech Council Presidency met with great approval in the Telecommunications Working Group on Oct. 25. It looks as if agreement can be reached on this compromise. This would also justify the speed at which the Council wants to move forward. The working group will meet on Nov. 8. On Nov. 18, the Permanent Representatives will meet, and on Dec. 6, the Ministers. They are then expected to reach a general approach for the trilogue.
Among other things, the fourth version contains the following changes:
The Parliament’s work is delayed by the very fact that there are two lead committees and thus twice the number of rapporteurs: for IMCO it is Brando Benifei (S&D) and for LIBE Dragoș Tudorache (Renew). They presented their latest draft on Oct. 21.
Among the key innovations in the text: The Commission is given new powers to expand the list of prohibited AI practices, AI applications that must meet enhanced transparency requirements, and areas of high-risk AI applications following an annual review. It will also be assisted by a newly created AI Board or Office. What form this oversight body will take is apparently still unclear.
Committee members discussed the draft in a joint IMCO-LIBE meeting last week. Tudorache tweeted afterwards, “we are making good progress“.
There is another way of looking at it. At the political level, there have only been two exchanges so far. And so there are hardly any final decisions, even though the working groups meet at the technical level, in some cases three times a week. For better processing, the working groups have divided the articles of the regulation into seven groups (batches).
On the status of the deliberations, Tudorache told the committee that he believed he had reached an agreement on Batch 1. This is about Articles 30 to 39 on notification and notifying authorities. “We’ve added a few tweaks to the Commission’s proposal to reduce the burden on startups, ensure cybersecurity and confidentiality, ensure impartiality of staff and conformity assessment bodies.”
On standards, conformity assessments, certificates, and registration (Batch 2), Tudorache said, “We’re still discussing fundamental rights and fundamental rights-related standards, and there are positions in some of our groups that want a different regime.”
On the other batches, too, it was said that much had been optimized and refined, but that discussions were still ongoing. Only on Batch 7 – secrecy, penalties, as well as final provisions – Tudorache said, “We just made an important breakthrough this week on penalties and fines, and that’s supported by a majority of us.”
A person familiar with the negotiations in parliament told Europe.Table that there were still many critical points that had not even been discussed. In the committee meeting, almost every speaker praised the constructive atmosphere in the negotiations. Nevertheless, each and every one of them still had their own wishes that they wanted to be taken into account.
In its non-paper, while the US starts by praising the EU for “developing a thoughtful architecture that addresses many of our shared AI goals,” it does have a number of “suggestions for improvement” for the AI Act. For example, in terms of intergovernmental cooperation. The US points out that it has developed its own AI standards. The planned EU legislation in its current form (the non-paper still refers to the third version of the Czech Council presidency from September) makes cooperation more difficult, they say. The US would like to see more concessions.
But the US also has suggestions for better, clearer, wording, as well as a recommendation to “clarify the relationship between the law and other existing and proposed EU legislation, such as the General Data Protection Regulation (GDPR), the EU Medical Devices Regulation, and others“.
Above all, however, the Americans would like to see “a coherent, globally applicable definition of the term” artificial intelligence. In addition, they criticize the definition of “artificial intelligence system” for still including systems “that are not sophisticated enough to merit special attention in the context of AI-focused legislation”. Therefore, the US is proposing a definition that captures the spirit of the OECD definition. Incidentally, this is also what many EU parliamentarians call for.
One thing is virtually certain: When the Digital Services Act becomes applicable law on April 17, 2024, Twitter will fall under it – and then be subject to the EU Commission’s DSA supervision. This is because the platform currently meets all the criteria to be subject to the rules of the DSA in the future as a very large platform in the EU, especially when it comes to moderating content that is relevant under criminal law. As became known on Monday, Musk had again assured Commission representatives that Twitter would abide by the rules.
But it was precisely in the area of moderating content relevant to criminal law that the new owner Elon Musk announced changes: “The bird is free,” he tweeted. What exactly Musk means by free remains unclear for the time being. Musk himself proved over the weekend that conspiracy theory nonsense could be part of it.
He posted a tweet after the alleged attempted assassination of House Speaker Nancy Pelosi, alleging that Pelosi’s husband, who was injured in the attack, was drunk. Musk deleted his tweet after a few hours, but it may have been a foretaste of what will happen on the platform in the future.
Instead of being outlawed, Twitter must abide by rules in the EU, digital commissioner Thierry Breton tweeted in Musk’s direction, reminding him of what he had promised after a meeting in May: He wanted to do with his companies anything that was good for Europe.
But for the EU, the problem is more complex than the issue of content moderation. If Twitter were to become a toxic environment for communication, the EU Commission, Council, Parliament, and many subordinate authorities would have to grapple with whether Twitter is the right place for them to communicate. The liberal Guy Verhofstadt, for example, is already publicly thinking about this – but would actually want something else.
Twitter has been the preferred communication platform for many political actors: The EU Commission’s account was created in 2010, the Parliament’s account in 2009, and the Council’s press account in 2010. Since then, the institutions and their actors have been using the platform intensively to bridge the gap between the public in Brussels and the public in the member states.
Despite the increased relevance of EU politics, the weights are still unevenly distributed: The german newspaper Bild-Zeitung, for example, has only one correspondent in Brussels, FAZ three, and Deutsche Presse-Agentur four. By comparison, dpa’s Berlin office has 23 correspondents, FAZ 14, and Bild 19.
This has long since ceased to reflect the political weighting between Berlin and Brussels – and yet the German media are still better represented than those of many other member states. Twitter is a way for the EU institutions to reach out to national publics. In addition, almost all relevant associations, think tanks and civil society actors are also present there.
And so EU representatives have also been looking for alternatives for some time – as have many users. The fediverse is a particular focus of attention: Services such as Mastodon, which are not identical to Twitter but are similar in some respects, are operated in a decentralized manner and are interoperable with one another via standardized protocols.
According to the Federal Ministry of Digital Affairs and Transport, which is responsible for the Digital Services Act in Germany, fediverse instances with less than 45 million users in Germany will also fall under the oversight. The driving force behind many of the instances set up by public institutions such as the Commission and ministries are, interestingly enough, the data protection commissioners – in Germany and Europe.
This could become an exciting oversight issue: After all, the data protection authorities act independently by virtue of their mandate. With their servers, however, as platform operators they would in turn fall under the supervisory authority of the DSA supervisors. “The scope of application of the DSA does not differentiate between public and private entities,” the BMDV informs us upon request.
The EU Commission account on Mastodon is followed by a good 13,000 other users on Monday. It was created at the end of April, when the first takeover rumors emerged. This is likely to increase significantly after the popularity of the fediverse instances in recent days. And other users who have been active primarily on Twitter must now also consider whether and to what extent they will now become active on other platforms.
But the alternatives differ considerably from Twitter in terms of their functionalities. And whether they are successful in the medium term depends heavily on whether the user base reaches the so-called critical mass, above which communication develops independently. This is also likely to depend on how many users use Twitter and other services in parallel, causing social media consultants quite some headache: Should they be early adopters and already build up a user base – or should they wait and save the effort?
However, should Elon Musk actively steer Twitter in a direction that is unacceptable from an EU perspective and want to exercise freedom of expression as he personally sees fit, the question could become louder as to whether EU institutions and politicians can remain on the platform and thus contribute to its success. In turn, abandoning Twitter would mean significantly reduced visibility. The coming months are also likely to be anxious ones for many communications officers and PR agencies.
Today, the Digital Markets Act (DMA) comes into force. The new regulation is intended to put an end to unfair practices by companies acting as gatekeepers in the online platform economy. The Commission had proposed the regulation in December 2020. In March 2022, the European Parliament and the Council adopted it.
There are three main criteria for classification as a gatekeeper in the DMA:
First, it is a matter of determining who belongs to the gatekeepers. From May 2, 2023, potential gatekeepers will have to notify the Commission when their central platform services reach the thresholds set out in the DMA.
The Commission then has until early September 2023 to designate the gatekeepers covered by the act. Once designated, gatekeepers have six months to comply with the requirements of the DMA.
To prepare for the enforcement of the DMA, the Commission will hold a series of technical working meetings. The first working meeting will be held on Dec. 5, 2022, and will focus primarily on the preferential treatment of gatekeepers’ own products and services. The Commission also works on an implementing regulation that will include provisions on procedural aspects. The Act gives the Commission the authority to conduct market investigations. In this way, the Commission can ensure that the established obligations are kept up to date.
Together with the DMA, the Digital Services Act (DSA) aims to create a safer digital space. In this space, the fundamental rights of users will be protected and a level playing field for companies will apply. The Digital Services Act will be directly applicable throughout the EU and it will apply from Jan. 1, 2024 at the latest. However, for very large online platforms and very large online search engines, the Digital Services Act will apply earlier, as early as four months after their designation as gatekeepers. vis
Between rising hunger and the prospect of still more supply shortages, dark clouds are hovering over the global food system. Not only has the war in Ukraine limited access to, and raised prices for, food and fertilizer, but extreme weather events are disrupting production, and economic downturns have diminished people’s ability to afford adequate and nutritious diets. Climate change is fueling droughts and exacerbating water scarcity, adding to the list of threats facing agricultural production.
Progress in tackling hunger and malnutrition was already stalling before the COVID-19 pandemic. Now, the World Food Program estimates that the number of people facing, or at risk of, acute food insecurity (defined as those who “lack regular access to enough safe and nutritious food for normal growth and development and an active and healthy life”) has increased by more than 200 million since 2019.
Key global objectives like the second United Nations Sustainable Development Goal – which commits countries to end hunger by 2030, and to improve nutrition and promote sustainable agriculture – are significantly off track.
Trade is vital for achieving food security, as the World Trade Organization’s 164 members emphasized at our 12th ministerial conference (MC12) this past June. Apart from its basic role as a conveyor belt for getting food from where it is abundant to where it is needed, trade supports jobs, livelihoods, and incomes. It is a key factor in the sustainable and efficient use of scarce global resources.
But all too often, international markets for food and agriculture function poorly, owing to problems like trade-distorting subsidies and high levels of protection. In many regions, persistent underinvestment in research, extension services, and market-linking infrastructure has led to low and stagnating agricultural productivity. Crop yields across much of Africa are particularly small, and have not matched the increases achieved in other developing regions.
WTO members have taken some important steps to address these problems. In 2015, they agreed to eliminate agricultural export subsidies – payments that cover the difference between domestic costs and international market prices – which are widely seen to distort production and harm farmers in other countries. The use of such subsidies has fallen dramatically, from almost $7 billion in 1999 to under $12 million in 2020.
But OECD data show that total public support to the farm sector in 54 advanced economies and 11 major emerging markets reached $817 billion in 2019-21. And only $207 billion of that was for general services such as research, training, or payments to consumers.
Other support programs frequently incentivize unsustainable production and consumption patterns and undermine fair competition in global markets. While average tariffs on agricultural products were around 6 percent in 2020, many countries maintained prohibitively high tariffs on certain products, with peaks reaching 1,000 percent.
At the beginning of this year, food prices had already reached high levels compared to recent years, and international agencies had begun to warn of growing hunger. Then the war in Ukraine made the situation much worse.
Prices spiked to record highs, and WTO monitoring found that many governments responded by restricting food exports, potentially easing strains at home but exacerbating shortages elsewhere in the world.
Against this backdrop, the multilateral cooperation that we saw at MC12 in June was most welcome. WTO members collectively committed to exempt World Food Program humanitarian purchases from export restrictions – saving the agency time and money in providing relief to millions of vulnerable people. They pledged to keep agri-food trade open, and to minimize the trade distortions resulting from emergency measures.
And they agreed on a binding accord to curtail $22 billion in annual public spending on harmful fisheries subsidies, which will contribute to food security and livelihoods by improving the health of marine habitats.
But over the past month, export restrictions on food, feed, and fertilizers have ticked back up, reversing the earlier downward trend. The total number of such measures introduced since February – and still in force – is now up to 52. The WTO is working hard through monitoring, transparency, and jawboning to bring this number down, but it remains a cause of global concern.
Ensuring that trade contributes to a more sustainable and equitable global food system will require greater efforts to align trade, environmental, and food policies. A holistic approach to managing how food is produced, distributed, and consumed is necessary to maintain sufficient supply, address climate and other environmental risks, ensure sustainable land use, and manage scarce water resources.
A long-overdue update of the WTO rulebook for agricultural trade is needed more urgently than ever, because we need to create a better incentive framework for investment in the sector.
To find solutions to growing food insecurity, we must expand the dialogue and explore the systemic drivers of the problem. With a stronger evidence base, WTO members will have a better understanding of how trade can be used to the fullest possible extent to enhance access to food for everyone.
In cooperation with Project Syndicate, 2022.
