Hiro and Robert McCauley, in this groundbreaking 2019 working paper, have meticulously assembled a dataset spanning the years 1999-2021 that delves into the currency composition of international reserves. Fueling critical research and analysis, this dataset has been referenced in various significant studies such as those by Ito and McCauley (2020), Chinn, Ito, and McCauley (2022), and Chinn, Ito, and Frankel (2024).
Key Points to Note:
- The IMF’s COFER gathers aggregate currency shares, providing a broad overview. In contrast, other datasets focusing on individual central bank reserves are often limited in scope or detail, with some reporting gross rather than net values (Iancu, et al. 2020).
- When observing individual central banks, the disparities can be quite remarkable. While Figure 13 relies on data from Iancu et al. (2020) for Sweden’s Riksbank, Figure 14 derives its data from Ito and McCauley (2022) records.
Insightful Observations:
- The variations in the datasets are influenced by coverage disparities (COFER encompasses all reporting IMF members, whereas Iancu et al. and Ito-McCauley focus on distinct sets of countries) and the methods used to handle reserves.
- Illustrating the USD shares of international reserves, Figure 1 showcases data from IMF COFER (green), Iancu et al. (red), and Ito and McCauley (blue).
- Conversely, in Figure 2, EUR shares of international reserves are displayed from IMF COFER (green), Iancu et al. (red), and Ito and McCauley (blue).
Remarkably, this dataset has been instrumental in investigating the gradual shift away from the US dollar, or de-dollarization trends, within the BRICS nations.
Additionally, Hiro Ito has conducted preliminary research on nontraditional reserves up to 2022, as reflected in Figure 4 demonstrating the geometric average of reserve holdings expressed in USD by reporting central banks.
Furthermore, while this dataset is publicly accessible in Excel or Stata formats, its coverage is limited to 63 countries, considerably fewer than the datasets utilized in studies by Chinn-Ito-McCauley (73 countries) or Chinn-Ito-Frankel (2024). Certain Latin American central banks have imposed confidentiality restrictions, restricting the dataset scope.
Future Insights:
It is worth noting that Laser, Milhailov, and Weidner (2024) have disseminated a database covering individual central bank holdings across eight currencies. However, given the current data landscape and information available in the It-McCauley dataset, it is anticipated that coverage for reserve currencies like CAD, AUD, and CNY might be scant, especially for recent years like 2022 and 2023.
In conclusion, these datasets and analyses shed light on the intricate dynamics of international reserve compositions, offering invaluable insights into global economic trends and policy considerations. Stay tuned for further developments in this crucial field of study.
Leave feedback about this