Sovereign Credit Ratings: Definition, Role, Methodologies

The credit ratings for sovereign debt as well as the private sector have been under intense scrutiny since the outbreak of the financial crisis that occurred in 2008. Credit Rating Agencies (or CRAs) have received critique for not being able to predict the risk associated with mortgage-backed securities and for only being reactive to, and not predictive of, the mentioned securities’ ratings.

A sovereign credit rating is essentially an opinion expressed about the likelihood that a government is in default. This is based on a perception of whether the government is able and willing to meet its financial responsibilities.

Significance of Credit ratings and why rating agencies are important

It is widely known that Credit Rating Agencies’ impact on the cost of funding of sovereign issuers is significant. Therefore, credit ratings are one of the main concerns for financial stability.

CRAs have an immense influence on debt issuer survival due to their ability to affect issuers’ access to funding markets and their funding costs - or so is the perception. Simultaneously, there have been claims that the CRAs only produce a lag effect instead of a lead effect in the sense that they only react to market events instead of predicting market movements. If the latter is true, CRAs have an important role from a financial stability viewpoint. However, if the former is true, opinions of CRAs are essentially rendered irrelevant.

Individual CRA methodologies

The three main techniques to determine a rating grade are a quantitative, qualitative and combined approach. Qualitative approaches are primarily used by Standard and Poor’s and Moody’s, which implies that they use knowledge of credit risk experts rather than classical statistical approaches. There are however disadvantages of the qualitative approach. The main disadvantages are that it is very expensive and time consuming. This means that a qualitative approach cannot be used to rate large numbers of small and medium sized debt issuers.

Some banks adopt statistical approaches to determine a rating from historical data only. Other banks use a combined approach between purely statistical methods to get a quantitative based rating grade and opinions from experts that results in an improved rating.

The primary disadvantage of using only the quantitative approach is that there is a shortage of judgement from experts who might add significant contributions in the forecasts. It is important for banks to be able to lean and depend on the system of credit rating forecasting, because the resulting default probability is an essential input parameter to credit risk portfolio models,

To be able to analyse the ratings’ accuracy we need to understand exactly what the ratings resemble and measure.

Default formal definition

It is important to obtain a consistent and structured definition of ‘default’ to be able to analyse the default data consistently and accurately. According to David Beers and Jamshid Mavalwalla, a default has taken place when due debt is not paid on the determined date or within a grace period subsequently determined, when guarantee payments are not made timeously as determined or where material losses are made by creditors who are holders of sovereign debt in certain circumstances. These mentioned circumstances include for example, where governments purchase debt at material discounts to their par value or where sovereign debt payments are taxed in retrospect.

This formal definition is in line with the definition of sovereign default in a journal by Juan J. Cruces and Christoph Trebesch, and most other literature studies on this topic.

Factors influencing Credit ratings

To understand the workings of the Credit Rating Agencies, an extensive study of the factors influencing the ratings is required.

PAGET-BLANC contains information used by Fitch Ratings, Moody’s and Standard and Poor - the three biggest rating agencies - on what determinants were used for the credit ratings produced. To determine which factors are common in influencing these ratings, a principal component analysis is done. These common factors’ variables are then evaluated for the impact they have on ratings by using an ordered logistic model. These factors are normally identified as government income, exchange rate changes in real terms, inflation and per capita earnings. Another big contributing factor of a country’s rating is corruption as it reflects on a country’s economic developments and the standard of a country’s management.

Credit ratings as a Probability of default: appropriate vs not appropriate

Some of the literature suggests that the quantitative measure (probability of default) is not an appropriate representation of a rank ordering of a credit rating. A Report by the bank of Canada, Database of Sovereign Defaults, 2017, suggests the reasons why using default probabilities to validify the credit ratings are not appropriate in the journal: Database of Sovereign Defaults. Banks, however, require these quantitative measures for their internal rating systems.

What are the reasons

Credit ratings essentially serve as a rank ordering of credit risk. This is dependent on downgrade and default risks. This credit risk rank ordering takes into consideration probabilities of default, expected losses, severity of losses and warnings. It is explained that there is not a direct, one to one correlation between any of these quantitative measures (like default probabilities) and credit ratings.

The majority of the literature takes the opposite view implying that a Credit rating is in fact in essence a probability of default:

Vito Polito & Mike Wickens implies that credit ratings express the default probability and that to measure the probability of default, it is required to map the default probability into a credit rating.

The main advantage of interpreting a credit rating as a probability of default is to produce measures of credit ratings that are independent, timely and transparent. Transparency refers to the ease of the financial sector to access, reproduce and to assess sovereign credit ratings. It also refers to the public’s ability to make its own conclusions and judgements about their validity.