Default statistics: Private and sub-sovereign lending 2001-2019

FOREWORD 
The Global Emerging Markets Risk Database consortium (GEMs consortium) was established in 2009 
as a joint initiative between the European Investment Bank (EIB) and the International Finance 
Corporation (IFC) to pool data on credit defaults. Since then, the GEMs consortium has grown to include 
25 members, mainly multilateral development banks and development finance institutions, covering 
emerging markets and developing economies.
 Based on data from 11 member institutions, this is the first GEMs publication on default statistics for 
private and sub-sovereign lending. We are proud to share our results building on more than ten years 
of successful cooperation among GEMs members. 
Data in this report cover the period from 2001 to 2019, unless otherwise indicated. Built by the 
consortium to support technical cooperation between international institutions, the GEMs risk database 
features sub-investment grade observations, and explains the GEMs methodological framework and the 
various applications of GEMs statistics. 
Compiling the report required extensive collaboration and technical harmonisation between the 
participating multilateral development banks and development finance institutions – this in itself is one 
of the consortium’s major achievements. We will continue building and refining both methodology and 
processes that make our statistics robust and valuable. Our aim is to support the scaling-up of 
investments where most needed.
 GEMs is pleased to make this publication freely available. We look forward to receiving feedback from 
our readers and working with our stakeholders to improve the focus and content of our future reports.

GEMs AT A GLANCE
 GEMs collects de-identified data from multilateral development banks and development finance 
institutions on their performing and non-performing exposures and, in return, the member 
institutions gain access to GEMs output statistics on default rates by geography, sector and time period.
 GEMs is based on the principle of keeping underlying data strictly confidential, and the consortium has 
jointly developed an anonymised data submission and collection process that protects confidentiality. 
Because GEMs counterparties are encrypted and cannot be identified, the aggregated credit risk 
statistics are anonymous.
 GEMs output statistics enable members to calibrate and benchmark internal models, assess capital 
requirements and make better informed investment decisions. The compilation of GEMs risk statistics 
relies on a sound and shared methodological framework.
 Bringing together data and expertise from 25 public development finance institutions, GEMs is a:
 •  Harmonised dataset reflecting more than 30 years of experience in investing in emerging 
markets.
 •  Framework for multilateral development bank and development finance institution 
collaboration to improve risk management practices. 
•  Trusted data collection and processing platform hosting a scalable data model that can 
be expanded to cover other contract-level data on investment projects. 
•  Statistics resource for member multilateral development banks and development 
finance institutions – GEMs outputs, default rates, recovery rates and rating migrations 
enable better informed investment decisions.
 •  Risk data hub of contract-level information for private, sovereign, sub-sovereign 
lending and sovereign-guaranteed lending. It is one of the largest datasets for emerging 
markets and developing economies. 

Default statistics: Private and sub-sovereign lending 2001-2019