Citi Research has released a detailed report combining its own econometric model with the output from Euromoney Country Risk to yield a new, more powerful tool for identifying default probabilities and relative value for sovereign issuers.
The combined approach overcomes many of the shortcomings associated with applying other, previously published, model-based strategies to sovereigns, and is a novel means for incorporating the crowd-sourcing consensus technique embodied in the Euromoney Country Risk (ECR) approach in such a practical and comprehensive returns-seeking environment.
The survey regularly asks more than 400 experts worldwide for their assessments of sovereign risk spanning 186 countries based on 15 political, economic and structural risk factors and other qualitative inputs. Although domiciled worldwide, ECR experts are not confined to the financial sector, but spread among finance, industry, government and academia to prevent any biasing of the results.
Incorporating the survey data into Citi’s 'Hybrid probability of default model' – a structural-statistical concept developed by Citi to estimate corporate default probabilities – confirms that ECR score trends lead upgrades and downgrades in sovereign credit ratings, much as regular users of the ECR survey believed and other evaluative studies had suggested.
Both of these findings suggest a powerful portfolio investment strategy tool for marrying risk and return, enabling portfolio losses, credit momentum and potential asset value to be measured more accurately across countries on a daily basis.
Predicting ratings actions
The usefulness of ECR scores in this process is telling. As Citi concludes from its detailed, statistical approach: “ECR scores begin to reflect changes in sovereign credit quality as early as 10 months prior to agency rating changes”, though this feature is more common where downgrades rather than upgrades are concerned for reasons that Citi also attempts to explain.
The research, moreover, points to the improved ability of ECR score trends to better predict Fitch ratings actions than other agencies, which “suggests that Fitch ratings actually lag those of S&P and Moody’s, particularly for credit downgrades.”
The new model-based approach is used to demonstrate how ECR scores and default probabilities can be usefully applied to analyze portfolio risk and relative value options for a broad cross-section of sovereign credit default swaps.
Adjusting the weights attached to each of the economic, political and structural risk sub-factors comprising the ECR scores can, furthermore, improve the predictive capabilities of the model, which provides default probabilities on a more-timely basis (i.e. daily) than the lagged reaction function that the ratings agencies tend to employ.
While further analysis may be required concerning the model’s robustness and adaptability to individual portfolios, this detailed paper undoubtedly represents a major advancement in applying Euromoney’s Country Risk Survey to multi-asset risk-return opportunities using existing model-based frameworks, backed up by positive simulation results using long/short trading strategies.