Data and scripts used while developing my economics thesis about sovereign rating differences between credit rating agencies.
Besides the macro variables, the panel has six variables calculated from the individual ratings given by each of the three credit rating agencies considered. Diff_UP_SM represents the positive (UP) rating difference between S&P (S) and Moody's (M), while Diff_DW_SM represents the negative rating difference between the same two agencies.
Example:
- Diff_UP_SM = 2 for a given pair (country, year), meaning S&P gives a rating 2-notches higher than Moody's to this specific country;
- Diff_DW_SM = 1 for a given pair (country, year), meaning S&P gives a rating 1-notch lower than Moody's to this specific country;
The six dependent variables are:
- Diff_UP_SM, higher rating from S&P when compared with Moody's rating;
- Diff_DW_SM, lower rating from S&P when compared with Moody's rating;
- Diff_UP_MF, higher rating from Moody's when compared with Fitch rating;
- Diff_DW_MF, lower rating from Moody's when compared with Fitch rating;
- Diff_UP_SF, higher rating from S&P when compared with Fitch rating;
- Diff_DW_SF, lower rating from S&P when compared with Fitch rating.
From the 1735 rows of the CondensedPanelData:
- Diff_UP_SM <> NULL in 1048 rows (<> 0 in 311 rows)
- Diff_DW_SM <> NULL in 1120 rows (<> 0 in 383 rows)
- Diff_UP_MF <> NULL in 866 rows (<> 0 in 285 rows)
- Diff_DW_MF <> NULL in 802 rows (<> 0 in 221 rows)
- Diff_UP_SF <> NULL in 1109 rows (<> 0 in 236 rows)
- Diff_DW_SF <> NULL in 1135 rows (<> 0 in 262 rows)
The Default indicators were obtained from the Bank of Canada CRAG database (debt in default for several countries from the 70s until 2015) and after that, they were reshaped into panel data. The variable DefaultThisYear was created, and it would be 1 if a country had debt in default in a given year, otherwise it would be 0. This data can be found in the SovereignDefaults_History_crag-database-2016_WithDefaultsLast1_2_5_10Years.csv file.
In the mentioned CSV file, if a row doesn't have CountryCode, it means it doesn't belong to the set of countries analysed in the scope of this thesis.