Please use this identifier to cite or link to this item: https://hdl.handle.net/10419/297392 
Year of Publication: 
2023
Series/Report no.: 
Bank of Canada Staff Working Paper No. 2023-7
Publisher: 
Bank of Canada, Ottawa
Abstract: 
This paper explores whether different funding structures-including the source, instrument, currency, and counterparty location of funding-affected the extent of financial stress experienced in various countries and sectors during the Covid-19 spread in early 2020. We measure financial stress using a new dataset on changes in credit default swap spreads for sovereigns, banks, and corporates. Then we use country-sector and country-sector-time panels to assess if these different funding structures mitigated-or amplified-the impact of this risk- off shock. A higher share of funding from non-bank financial institutions (NBFI) or in US dollars was correlated with significantly greater stress, while a higher share of funding in debt instruments (instead of loans) or cross-border (instead of domestic) did not significantly impact resilience. The results suggest that macroprudential regulations should broaden their current focus to take into account exposures to NBFI and dollar funding, giving less priority to regulations focused on residency (i.e., capital controls). After the sharp increase in financial stress in early 2020, policy responses targeting these structural vulnerabilities (i.e., US $ swap lines and policies focused on NBFIs) were more effective at mitigating stress related to these funding structures than policies supporting banks, even after controlling for macroeconomic policy responses.
Subjects: 
Coronavirus disease (COVID-19)
Exchange rates
International topics
Financial institutions
Financial stability
Financial system regulation and policies
JEL: 
E44
E65
F31
F36
F42
G18
G23
G38
Persistent Identifier of the first edition: 
Document Type: 
Working Paper

Files in This Item:
File
Size





Items in EconStor are protected by copyright, with all rights reserved, unless otherwise indicated.