Hernán Muñóz
Data Ecosystem Advisor
h.munoz@cepei.org
The 2030 Agenda posed a series of challenges regarding data needs which, in turn, represented an opportunity to make —often neglected— innovations and improvements in the supply of official information. Using non-traditional sources such as financial transaction data, mobile networks, satellite images, sensors, or internet platforms can substantially increase the quality and reduce the cost of generating the information required for monitoring and implementing the Sustainable Development Goals (SDGs). In addition, “traditional” statistical surveys and censuses can be complemented with new sources of information. This trend gained momentum during the COVID-19 pandemic, when data producers faced sudden interruptions of traditional collection sources and had to develop new methodologies for using alternative sources, thus spontaneously extending the data ecosystem. [1]
In this context, the SDG Acceleration Roadmap initiative, financed by IDRC and coordinated by Cepei and LIRNEasia, mapped private sector data initiatives [2] in Latin America that contribute to strengthening data capacities to monitor and achieve the SDGs. Thus, to learn what the private sector is doing to ensure that there are more and better data for the SDGs in Latin America, we identified a series of key initiatives in which the private sector participates.
We were particularly interested in those that met the criteria of relevance, replicability, scalability, and co-creation. We identified 42 data initiatives distributed in 14 Latin American countries, some of which are large multinational companies with reach throughout the region.
Most of the initiatives identified are related to data sharing, either directly or through actions such as data analysis, capacity building, or impact assessment reports that include data sharing.
The data revolution implied not only a quantitative leap in the data supply but also a series of new unavoidable requests: Greater relevance (data on new phenomena), timeliness (real-time data), and coverage, including granularity (new data sources allow for high levels of detail), in addition to the reduction of costs and response burden. Cepei’s analysis of the region’s data capacities reveals the SDGs have not had the expected catalytic effect. Our mapping reinforced this conclusion, as the private sector’s participation in Latin American data ecosystems is still low.
In this regard, no initiatives are systematically included in the national SDG data ecosystems, but rather specific and isolated actions of companies with a greater or lesser degree of association with the public sector. One recommendation in this regard is to articulate data sharing on a sectoral basis to aggregate companies’ data (e.g., for cell phone data to be handy, data from all operating companies in a country must be available). The previous has not been observed in any of the cases. Still, there is potential for this, for example, in certain specific industrial branches or the mining sector. Likewise, efforts to incorporate non-traditional data should focus on available and valuable data, such as satellite imagery, web crawlers, price readings, mobility, bank consumption data, etc.
In addition, the use of non-traditional sources and the inclusion of private sector data is still in the exploratory stage. Generating reliable and sustainable information within data ecosystems requires an enabling environment, understood as all the resources, processes and systems, legislation, incentives, and information on the actors involved in the information generation processes. Current challenges need incentives and legislation that promotes public-private partnerships.
Some of the conclusions arising from this mapping exercise are presented below.
They point to two significant aspects affecting private sector participation in the data agenda for the SDGs, such as a possible loss of momentum and interest in the 2030 Agenda by the private sector, as well as the lack of an enabling environment for public-private initiatives:
- There is data and initiatives, but there is a lack of data initiatives: Companies, in general, are leaning toward reporting their impact on the SDGs through data collection and analysis. But this does not necessarily imply strengthening the SDGs’ data ecosystem.
- In some cases, companies in the industrial and mining sectors disseminate reports on their environmental impact accompanied by databases. Consolidating this information could contribute to those dimensions where the 2030 Agenda presents the most prominent data gaps.
- A large part of the data initiatives identified, although they impact the SDGs, have not been designed as contributions to the 2030 Agenda but rather to sustainable development in general. It seems that greater visibility of the 2030 Agenda and positioning in the private sector is required.
- In the case of public-private partnerships, it is helpful to distinguish between partnerships that seek to identify the potential of an alternative data source to measure specific phenomena [2] and partnerships that aim to measure them regularly .
- Data initiatives that aim to measure phenomena regularly require sustainable collaboration. These must be supported by binding obligations from companies and legislation that enables private data to be reused. In this regard, legal frameworks in the region need to evolve to support this practice. The development of laws promoting private data reuse is a crucial issue for the future of public statistics. [4]
[1] Data ecosystem is the name commonly given to the place where the processes of production, consumption, and transfer of data are generated, beyond the national statistical systems.
[2] Capacity building and skills sharing, data collection, data sharing, data analysis, data infrastructure, data governance, data mapping, and funding, among other actions.
[3] The analysis of mobility data during the COVID-19 pandemic or during natural disasters in the case of Telefónica is an example of a one-time initiative to address a specific phenomenon. These cases, however, may present replicable models.
[4] In the identified case of DymaxionLabs, the reuse of satellite image analysis models developed for the agricultural sector to identify slums is an example of how the private sector can enhance regular statistical operations.