BLOG — Jun 9, 2025
Navigating Uncertainties with the Early Warning Signals (EWS) Framework
Over the past year, the global macroeconomic landscape has seen a cautious recovery from previous challenges. According to S&P Global Ratings, the number of global corporate defaults ticked lower in 2024[1]. While inflation is easing, the global economy still faces ongoing uncertainty surrounding US tariffs. Under these circumstances, proactively monitoring credit risk becomes increasingly important for businesses as they navigate the ever-changing business environment.
S&P Global Market Intelligence’s EWS framework is designed to provide timely warning signals for early detection of entities with genuine risk of defaulting, expressed in an intuitive, traffic light color scale.[2] This blog delves into a test dataset, spanning from January 1, 2024, to December 31, 2024, to showcase the real-world effectiveness of the EWS framework.
The test dataset mainly covers S&P Global Ratings' rated entities, as well as key development defaulters.[3] Specifically speaking, defaulter sample includes debt defaults, bankruptcies, and corporates rated as CC or below,[4],[5] while non-defaulter sample consists of corporates rated as CCC- or above. There are in total 178 default flags and 2671 unique non-defaulters included in this analysis.
As shown in Figure 1 below, the EWS Framework exhibits good and robust performance in detecting default events that happened in 2024. It consistently achieves more than 80% hit rates on a daily basis, meaning it correctly captured most actual defaults, with relatively low false alarm rates below 20%, during a 2-year period[6].
Figure 1: Daily Hit Rates and False Alarm Rates
Source: S&P Global Market Intelligence as of May 14, 2025. For illustrative purposes only.
Drilling down to regional levels, we can observe high hit rates exceeding 80% across all regions, and relatively low false alarm rates as shown in Figure 2 below.
Figure 2: EWS Performance by Region[7]
Source: S&P Global Market Intelligence as of May 14, 2025. For illustrative purposes only.
A breakdown by selected industries is also outlined in Figure 3.[8] Most industries are observed with more than 90% hit rates, with false alarm rates below 30%.
Figure 3: EWS Performance by Selected Industries
Source: S&P Global Market Intelligence as of May 14, 2025. For illustrative purposes only.
In conclusion, this out-of-sample test showcases the reliability of the EWS Framework in identifying default events in 2024, for different regions and industries, globally. Incorporating both absolute and relative PD levels, we gain valuable insights into the level and trend of creditworthiness. Amid global trade uncertainty, risk managers and investors can navigate challenges with greater agility and foresight, by adopting an automated and effective EWS framework.
If you want to learn more about EWS, please click here.
Appendix
Additionally, Table 1 below illustrates how early the EWS framework generated high-risk red signals for defaulters in 2024. It is worth mentioning that more than 90% defaulters were flagged at least 6 months in advance and nearly 75% defaulters were signaled at least 2 years prior to defaults, leaving enough time for businesses to perform due diligence, and potentially reduce exposures.
Table 1: EWS Red Signals Prior to Defaults
Source: S&P Global Market Intelligence as of May 21, 2025. For illustrative purposes only.
1 Default, Transition, and Recovery: 2024 Annual Global Corporate Default and Rating Transition | S&P Global Ratings (spglobal.com)
2 Please refer to the Early Warning Signals Framework 1.0 White Paper.
3 Source: The S&P Capital IQ® Platform.
4 Key development defaulters and corporates rated as CC or below are included to increase the sample size.
5 Only the first default event for companies with multiple defaults during the analysis period was selected.
6 Two years prior to default date for defaulters, and two years prior to December 31, 2024 for non-defaulters.
7 Based on the whole period aggregated results. ASEAN (except Singapore) and APAC Developed are merged together due to the limited number of defaulters.
8 Based on the whole period aggregated results. Only industries with greater than or equal to 10 defaulters are plotted for illustrative purposes.