Determining Sovereign Credit Risks Using News Articles

Alternative data is becoming increasingly popular in the financial world. While some traditional data sources such as earnings and economic indicators are still relied upon heavily, alternative data is providing new insights that can help investors make better-informed decisions.

There are a variety of sources for alternative data, including social media, satellite imagery, and even wearables. This data can be used to track everything from consumer sentiment to economic activity.

Reference [1] explored the use of news articles for determining sovereign default risks. Default risk is the probability that a country will be unable to repay its debt obligations. The study found that news articles could be used to predict default risk,

This paper develops a novel framework for quantifying sovereign default risk in real-time using machine learning and natural language processing techniques. Our framework relies on two building blocks: topic modelling for dimensionality and noise reduction and text similarity for quantifying default concerns in the news media. Applying our framework to a corpus of 10 million news articles, we obtain the News-implied Sovereign Risk Index (NSRI) for a broad cross- section of countries.

Our analyses provide convincing evidence that NSRI is strongly associated with CDS spreads contemporaneously and predicts future CDS spread dynamics and rating downgrades. Importantly, we find that NSRI contains additional sovereign default risk information not fully reflected in CDS spreads. For example, controlling for CDS spreads, NSRI significantly predicts rating downgrades. Conversely, CDS spreads have no incremental information on top of NSRI about future rating downgrades. These results imply that NSRI can complement CDS spreads in tracking sovereign default risk at a high frequency for countries with liquid CDS contracts. More importantly, for countries without liquid CDS contracts and other daily market-based proxies of default risk, NSRI can serve as a reliable stand-alone high- frequency proxy for sovereign default risk.

In short, news articles can be used to determine countries’ credit risks. And more importantly, this method works even for countries for which no liquid credit default swaps are traded.


[1] Dim, Chukwuma and Koerner, Kevin and Wolski, Marcin and Zwart, Sanne, News-implied Sovereign Default Risk (2021). Available at SSRN:

Further questions

What's your question? Ask it in the discussion forum

Have an answer to the questions below? Post it here or in the forum

Leave a Reply