A global database of emerging market debt statistics, created by leading development banks, is set to incorporate artificial intelligence to comprehensively analyse risk for investors. The goal is to reduce borrowing costs for developing nations. UK-based AI firm Galytix is developing a framework to process data in the Global Emerging Markets Database (GEMs). Galytix aims to attract more private investment into developing countries by addressing the gap between perceived and actual risks. Galytix is an AI firm that specialises in data analytics and risk management solutions for the financial sector. GEMs was created in 2009 by the World Bank Group and the European Investment Bank.
Raj Abrol, co-founder and CEO of Galytix, stated that the new model could help potential investors understand risk and offer more competitive financing, potentially leading to increased private capital flowing into these markets. GEMs contains data on debt defaults, recovery rates, and other relevant metrics from emerging market companies and countries. Originally intended for information exchange between banks, some data has been shared publicly in response to demands for more detailed information from private sector investors.
The integration of AI is particularly timely, as developed countries reduce aid spending and bilateral finance, diminishing capital sources for some developing countries. This comes as their needs for infrastructure, climate change mitigation, and social programs intensify. The database gathers information from numerous development banks globally, anonymising it to provide sector- and country-level risk statistics.
According to Gregor Cigüt, head of the GEMs secretariat, the partnership aims to transform decades of risk knowledge into actionable market intelligence to facilitate increased investment. While the data is now public, Abrol noted that it remains challenging for investors to utilise effectively. The algorithms compiling the figures will be subject to continuous human oversight, and statistics will not be produced if data gaps or flaws are excessive for any country or sector.
