The Emerging UN Debate on Food Security and Nutrition Data
The Committee on World Food Security (CFS) is an international platform that brings together UN member states, and international agencies and bodies with a mandate on food security and nutrition, civil society, international agricultural research systems, financial institutions, private sector associations, and philanthropic foundations for concerted and coordinated action for the progressive realization of the right to food for all. Three UN agencies jointly support the work of the CFS: the Food and Agriculture Organization (FAO), the International Fund for Agricultural Development (IFAD), and the World Food Programme (WFP). In addition to core funding from the UN system, the CFS also relies on voluntary contributions. The CFS meets every year to shape global policy directions for long-term food security and nutrition, and in the current context – the realization of the Sustainable Development Goal 2: ending hunger. To aid this process, the CFS has established a High Level Panel of Experts on Food Security and Nutrition who provide evidence-based analysis and policy guidance.
Understandably, for quite some time, the work of the CFS has been tracked by civil society organizations active in the domains of health and nutrition, food sovereignty, sustainable agriculture, and indigenous people’s rights.
In its most recent multi-year program of work, the CFS approved a thematic workstream on data collection and analysis tools for food security and nutrition. In January 2023, an Open-Ended Working Group on Data was kicked off. Over the next six months, through periodic deliberations and consensual decision-making, this working group will develop a set of concrete policy recommendations for leveraging the data revolution to transform global food systems and present the same to the upcoming 51st Plenary of the CFS. The working group is relying upon the September 2022 Report of the High Level Panel of Experts on Food Security and Nutrition as a starting point for informed dialogue.
The main problem at hand, according to the September 2022 Report, is that currently, much of the data that is needed for effective decision-making in the food, nutrition, and agriculture domain is being generated outside of national statistical systems through Big Data and AI technologies. The report recognizes the risks in an economy of global data flows, that “data collected in one country are processed in cloud-based facilities operated by other countries or private companies, creating dependencies and risks for data privacy and data access”. It also notes the challenges for data sharing because private data collectors and processors assetize food security and nutrition data (FSN data) and ‘subtract’ such data from the public domain. The solution to this is seen as a new global institutional governance framework that treats FSN data as a global public good, akin to what the WHO has been proposing for health data. Towards this, the report also recommends the establishment of a Global Food Security and Nutrition Data Trust Fund through which multistakeholder cooperation for FSN data innovation can be undertaken.
The main problem according to the September 2022 Report, is that currently, much of the data that is needed for effective decision-making in the food, nutrition, and agriculture domain is being generated outside of national statistical systems through Big Data and AI technologies.
At first glance, the ‘FSN data as a global public good’ approach put forward by the September 2022 Report of the High Level Panel of Experts seems an exciting path forward in overturning the inequities of the current data paradigm. But when we dig into the details, its deficits become increasingly evident.
Data Public Goods: Moving Beyond a Simplistic ‘Open by Default’ Approach
The FSN expert panel’s report makes a specific suggestion:
“Data-governance mechanisms, including institutional mechanisms, must recognize the contributions of all stakeholders – those who provide, collect, process, share, and use data – and regulate their rights, while fostering cooperation between them. Such mechanisms must uphold privacy rights, protect personal information and intellectual property, and establish codes of conduct.”
This ignores the elephant in the room – the numerous ways in which the existing Intellectual Property Rights (IPR) regime on confidential business information – more specifically, trade secrets – is utilized by data processors to enclose valuable insights from the knowledge commons of data and monopolize all its potential downstream uses.
For the openness of data to address the issues of inequality that the report underlines strongly, such openness should be grounded in a framework for economic governance of data that can create, nurture, and sustain a dynamic and vibrant data ecosystem. Balancing rights and freedoms with duties and obligations to ensure multiple interests in the ecosystem is necessary. While data sovereignty is crucial for the inalienable rights of the data-producing individuals and communities, data accessibility too is vital in order to enable innovation. Further, there is also the question of safeguards (individual and collective) for potential harms in data use/s and the claims (individual or collective) over the value generated from data. The Civil Society and Indigenous People’s Mechanism (CSIPM) has underscored the need for a new governance approach to FSN data that guarantees individual and collective rights in data while accounting for the deep imbalances of power between corporate actors producing frontier digital technologies and farmers, peasants, local communities, and other key actors in the food system from whom data is collected. The CSIPM has further argued that rather than an ‘open by default’ approach to governing FSN data, inclusive mechanisms that preserve the data ownership and control claims of local communities that can enable the flourishing of local knowledge systems for food security is the appropriate way forward.
While data sovereignty is crucial for the inalienable rights of the data-producing individuals and communities, data accessibility too is vital in order to enable innovation.
The United Nations Declaration on the Rights of Indigenous Peoples (UNDRIP) (A/RES/61/295) and the International Covenant on Economic, Social and Cultural Rights provide a robust basis for the CSIPM’s call – to respect the sovereignty of indigenous peoples and their claims over their knowledge resources. But would this be a full solution?
Collective rights translated into data stewardship mechanisms are an important strategy to counter corporate data extractivism. But there is a real risk that such mechanisms may just end up being co-opted by corporate data monopolies. Stewardship entities may enter into so-called collaborations with corporations for a promise of a share in the benefits, but without any access to or control over the algorithmic code/digital intelligence and other techno-material infrastructures central to value generation. Corporations entering into community-based partnerships on the other hand, can externalize their regulatory burden (putting the onus on the community stewardship entity to meet data rights obligations), reduce costs, and minimize reputational risks, while having a pipeline to the community’s ‘raw’ data resources that they can then deploy towards patented commercial products or even share for secondary use.
Unless such enclosure of data is dismantled through reform of the current global IPR regimes, ideas of community-driven data management may not add up to distributive and redistributive justice in the data economy.
The Urgent Imperative to Restrict the Ambit of IPRs in Data Resources
The dream of data as a global public good enabling a just, equitable, and inclusive socio-economic order will come to pass only if we overturn the current day restrictive IP regime that enables data processors (particular companies adopting the platform business model) to enclose data and data-based intelligence in perpetuity. To extend this specifically to the current FSN data debate – without a concrete challenge to the IP rights of agri-tech companies, a public goods approach is only likely to entrench existing inequalities by furthering data extractivism by powerful digital and agricultural corporations, enabling them to maraud more territories with lesser effort. It is useful to take a leaf out of the critiques of ‘health data as a public good’ approach of the WHO. In global health policy debates, developed countries have pushed for the idea that “WHO should develop a mechanism for State Parties to automatically share real-time emergency information, including genomic sequencing, needed by WHO for risk assessment, that builds on relevant regional and global digital systems”. But this call for free flows of health data across borders is not accompanied by concomitant attention to equitable distribution of the value of the resultant data-based innovation (which will be unfree and propertized by pharma IP claims). Against this backdrop, free cross-border flows of health data and the creation of global health data public goods is only likely to exacerbate free riding by Big Pharma.
We need to challenge the expansive extension of trade secrets protection by data processors to the aggregate, anonymized data commons (non-personal data, to use the language of mainstream data regulation). As legal scholar, Tomasso Fia, argues in a recent paper, there is a need to steer courts and legislatures towards applying a more restrictive understanding of IP forms in order to limit and contain the propertization of new intangibles, such as aggregations of non-personal data. Trade secret protection of data aggregations needs to be counterbalanced effectively against social considerations; for instance, fundamental rights of freedom of information and expression, freedom of academic research, freedom to conduct a business, and freedom of services in order to prevent data enclosure that are antithetical to data innovation for social good.
In our own work at IT for Change, we have argued for such an approach, calling attention to the need for a radical departure from assuming that openness provides the necessary and sufficient condition. Rather, we argue that without a new normative baseline that ensures non-exclusive accessibility of the infrastructural resource of networked data (including the data already held by companies) cannot be prevented. This means the boundary question between private and public claims in data must start from the legal recognition of data as inappropriable social commons with commensurate freedom of open use with limited privileges for data producers and variegated rights and responsibilities for different types of economic actors. This also means that a priori aspects of data governance are about techno-institutional visions of how data as a means of production will build a society and economy that is egalitarian .
Just as we need to overhaul the existing IP regime, we also need to account for the fact that less powerful states in the global digital economy have a right to control the flows of their citizens’ data resources to other territories, to preserve their citizens’ data sovereignty as an extension of the right to development. From the principle that data can flow freely across borders as long as there is trust in privacy and security safeguards, we need to shift to a new principle of data flows with rights, with the sovereign right to economic self-determination of individuals and collectivities. Sovereignty, here, implies avoiding dependencies on digital transnational corporations and strengthening the domestic digital economy sector for local value generation and distribution.
From the principle that data can flow freely across borders as long as there is trust in privacy and security safeguards, we need to shift to a new principle of data flows with rights, with the sovereign right to economic self-determination of individuals and collectivities.
From this perspective, we have the following specific recommendations to make to Recommendation 5 of the September 2022 Report of the High Level Panel of Experts on improving data governance.
Rather than restrict the ambit of improvements in FSN data governance to “promoting inclusiveness to recognize and enhance agency among data users and data generators”, the Committee on Food Security’s 51st Plenary should call for the urgent adoption of a “rights-based data governance paradigm for FSN data grounded in the recognition of the data sovereignty of all peoples and a legal acknowledgment of aggregate non-personal data as inappropriable social commons”.
Towards this, the following sub-recommendations should also be adopted.
Firstly, all government data that refer to agriculture and FSN should be treated as “inappropriable social knowledge commons” with clear safeguards against free-riding and capture by powerful players, particularly to prevent their over-broad exercise of IP rights. They need to be governed on the basis of FAIR and CARE principles for equitable generation and distribution of data value while eliminating the risk of individual and collective harms.
Secondly, there should be clarity on the membership, functioning, and regulatory norms of the global FSN data trust that has been proposed to be set up through the CFS, especially in relation to the governance of the cross-border flows of such data. The policy space of governments – especially those in the Global South – to restrict flows of such data to protect the interests of vulnerable sections of their population from the ills of data extractivism and regulate the access of foreign digital service companies to strategically important FSN data at the national level – must be held as inviolate in all operations of this proposed data trust.
In the final analysis, demands for safeguards against public good-washing of food security and nutrition data comprise an important agenda for action. Without a robust global governance regime that legitmizes the role of the data commons as a mechanism for distributive justice, the historical rights of people in food sovereignty and security cannot be met.