Four teams of data experts recently competed in the EITI’s first datathon, where they used extractive sector data to tackle challenges that resource-dependent countries may face as the global energy transition gathers pace. The winning prototype tool, PREDICT, combines various datasets to estimate future extractive revenues under different commodity price scenarios.
In this blog, the creator of PREDICT, Marco Zaplan, a self-described “data ninja”, answers questions about his prototype, his hopes for the future, and how data can empower stakeholders.
What is PREDICT and how do you envision it being used?
PREDICT is short for Projecting Revenues from the Extractives for the Development and Investment in Communities Tool. The tool aims to lower the barrier for more people to understand the flow and volatility of resource revenues by visualising resource revenues, revenue allocations and changes caused by economic factors like commodity prices, production, and tax rates.
PREDICT could be used by national and local economic planning agencies for budget scenario planning. It could also be used by EITI implementation countries as a capacity building and communications tool, and by civil society organisations who advocate for fiscal policy reforms in the extractives sector.
What do you hope is the future of the PREDICT?
My hope is for more resource-rich countries and stakeholders to use PREDICT in ways that are relevant to their contexts. PREDICT can do so much more than estimating resource revenues. With more granular data, PREDICT can visualise and analyse a wide range of information, including data relevant for environmental, social and governance (ESG) reporting.This could include comparing expected and actual payments by companies, visualising how revenues are shared with local governments,, tracking social expenditures, and even monitoring environmental impacts of extractive operations, such as water consumption and greenhouse gas emissions.
My hope is for more resource-rich countries and stakeholders to use PREDICT in ways that are relevant to their contexts.
What attracts you to working with data projects specific to extractives?
I find extractives data very exciting because it is continuously evolving. I started working with extractives data in 2015, and at the time we were mostly analysing revenue data. Today, there is a lot more to work with including beneficial ownership data, social and environmental data, national oil companies data, contracts data, procurement data, and so on. It’s never boring, and I think there is a lot more that we can work on to make data easier to access, analyse and apply.
What are the challenges of fiscal modelling?
One challenge is that fiscal modelling is never 100% accurate because there are many variables that are difficult to predict. Furthermore, existing tools are often too complicated to use, and spreadsheets are not visually appealing or an inspiring medium for communicating findings. Financial modelling can be very powerful, but there is still a way to go in terms of making this type of analysis accessible to a non-technical audience.
What is the benefit of modelling and visualising extractives data?
Fiscal modelling helps inform policy debates, raise public awareness and build trust among stakeholders. By modelling extractives data, we are able to better equip stakeholders during planning and policy discussions. We can use data visualisation to show trends, highlight key data points and analyse information more efficiently and effectively.
How can stakeholders be empowered to use data to shape decisions on natural resource management?
We should make extractives data more accessible and understandable to more stakeholders. This means designing tools that are user-friendly, straightforward and presentable. There are many opportunities to demonstrate how data can be used to inform decision-making, which can help stakeholders understand the potential of extractives data. We should also invest in building capacity of data users and design tools such as PREDICT to make data use even easier.
Funding for the datathon was provided by USAID.