Envisioning ethical data governance in agriculture. With the rising use of digital technologies on farms, there is also growing attention to the agricultural data they use and produce. There are many potential benefits from increased data collection, sharing, and analysis for farmers.
This data can support on-farm decision-making, climate change adaptation, and evidence-based policy. However, There are still concerns about who controls the data, who owns it, and who can profit from it. And this can limit the use and sharing of data.
In our research, we asked, what matters most to farmers when it comes to managing and sharing agricultural data? And what practices or approaches support ethical data governance? To find out, we led over three years of outreach and research with farmers and others working with agricultural data.
This is what we learned. First, we asked them about their challenges and concerns related to data governance. Here are the most common. Agricultural data is varied and diverse.
Not all agricultural data needs to or can be handled in the same way. Some information, like production and sales data, might be more sensitive than others, like weather data. Collecting useful agricultural data can be challenging and expensive. Beyond these technical obstacles, there are important ethical issues for data governance.
Data agreements and privacy policies involving farmers, researchers, agricultural service providers, and others often lack transparency. It can be difficult to understand the regulations and terms and conditions, and there is usually no opportunity to negotiate. Without transparency, There is often also a lack of trust. There are extreme imbalances in access and use of agricultural data and the valuable knowledge it can inform, which produces unequal distributions of benefits.
Consider the differences between an individual farmer and a large corporate enterprise using agricultural data in terms of the resources and capabilities to analyze and derive benefit from the data. In response to these challenges, we identified four best practices. Open data or open access, a principle and movement around the belief that data, software, research, and other forms of knowledge should be as open as possible, where people can use and share available data freely. Voluntary codes of conduct, like AgData Transparent, and guiding principles, like the FAIR Principles for Data Management, and care principles for Indigenous data governance, increasing privacy protection and data rights, like the right to access or the right to be forgotten, and securing data ownership for farmers.
While these approaches may be beneficial or even necessary for addressing some of the data governance challenges, they are not sufficient. Open data can help address practical challenges and allow more people, including farmers, to access useful information. But open data is not necessarily more just. Much of the value of data comes from its aggregation and the analysis of massive collections of data to produce insights.
Again, large businesses have the resources to benefit most from increased access to data, especially compared to individual farmers. Voluntary codes of conduct and guiding principles can improve trust and encourage more collective benefits. However, they can be difficult to enforce, and there can be tensions and trade-offs between the different approaches, like fare and care.
Data rights can offer the subjects and creators of data more control over how it can be used and shared. In practice, the burden is on the individual. It can be difficult to know what data rights are protected in your jurisdiction, and even more challenging to enforce them. Most people are unlikely to have the time, resources, or expertise required to do so.
Further, privacy protection and common data rights can exist alongside the imbalance in access and ability to benefit from data. Farmers can regain some power in data governance when their ownership of the data is respected and protected. But data ownership does not guarantee control over the data.
Like privacy protections and data rights, Data ownership usually focuses on the individual, and there are cases where farmers own their data without that preventing others from benefiting from it. Ethical data governance will require more. We think this calls for a data justice-centered approach. So what could data justice look like in agriculture? Based on our research, this could build upon accepted best practices and include laws and regulatory changes, new governance structures, building capacity, and solidarity and collaboration.
Data justice in agriculture may require expanded legal protections of data rights across jurisdictions. But it can't stop there. These interventions must also get to the structural reasons why there are such extreme imbalances in access, use, and benefits from agricultural data. This could include legal mechanisms to prevent further corporate concentrations of companies that produce data tools for agriculture. Currently, much agricultural data is collected and controlled by very few powerful companies.
There are, however, alternative data governance structures. For example, data cooperatives are decentralized arrangements to share data, where the people who share the data control it both individually and collectively. In data trusts, a central organization like a government agency, a research institute, or a non-profit gathers and governs data from multiple people and organizations for a collective benefit.
Building capacity means that farmers and others working with agricultural data have the time, resources, knowledge, and skills to take action to shape data governance. Farmers and farming organizations are already building capacity from the ground up. For example, farmers are sharing knowledge and tools and groups like FarmHack and OpenTeam. Farmers can also collaborate with consultants and researchers to have access to other skills and resources.
Data justice in agriculture will also need to enable individuals and groups to make their own decisions about what is right for them in their context. These are significant changes to the status quo. Addressing the challenges for ethical data governance for farmers, farming organizations, researchers, and governance will require collaboration and cooperation across groups. A truly ethical model of data governance must not reinforce unjust practices in agriculture. We encourage you to think about engaging and centering the voices and priorities of other affected groups, including farm workers and Indigenous data sovereignty groups.
This is a wonderful opportunity for collaboration and solidarity. Interested to learn more? Take a look at the discussion guide and glossary in our Toolkit for Ethical Data Governance in Agriculture.