Domestic Stability Buffer Review - Summary Note - December 2022

Summary

• The Domestic Stability Buffer (DSB) is an adjustable buffer that is applied to Canada’s largest banks, known as Domestic Systemically Important Banks, or D-SIBs. It requires D-SIBs to build capital as vulnerabilities grow so that it can be used to absorb losses and support lending in times of stress.
• Since its introduction in June 2018, the DSB has moved within a range of 0% to 2.5% of D-SIBs’ total risk-weighted assets. After being reduced to 1.0% during the COVID-19 pandemic, it was raised to 2.5% in June 2021 in response to rising systemic vulnerabilities.
• While overall the DSB has been effective in ensuring the stability of the financial system, the risk environment has become increasingly complex since 2018. In addition, both domestic and international vulnerabilities have become more pronounced. As promised in 2021, OSFI has reviewed the design and range of the DSB to ensure it remains effective given these trends.
• As a result of this review, OSFI increased the DSB’s range to 0% – 4% on the premise that DSIBs should be able to absorb losses based on a range of severe but plausible scenarios, while maintaining lending.
• The current design will be maintained to allow the DSB to be responsive to changes in vulnerabilities while acting as a stabilizer - or shock absorber - to absorb losses.
• OSFI has enhanced transparency around the calibration of the DSB through the provision of indicators that contribute to the semi-annual decision. Moreover, supervisory judgment will continue to play a strong role in future DSB decisions.

DSB background

The Domestic Stability Buffer (DSB) is an important macroprudential tool that helps protect the Canadian financial system against vulnerabilities and risks. It applies to all D‑SIBs and takes the form of a usable buffer that is based on a percentage of D‑SIBs’ total risk-weighted assets. In June 2018, OSFI launched the DSB with a range of 0 to 2.5 percent and initially set the level at 1.5 percent. Since then, the DSB’s level has been re-examined and where necessary adjusted twice a year, in June and December, with a special announcement in March 2020 due to the onset of the COVID-19 pandemic being an exception.

The DSB was designed to cover vulnerabilities that are measurable, material, cyclical and systemic. Specific vulnerabilities addressed by the DSB since inception have included: (i) Canadian household indebtedness, e.g debt-to-income ratio (ii) Canadian asset imbalances such as house price-to-income ratios, and (iii) Canadian institutional indebtedness, i.e., the level of debt held by corporations. More recently, OSFI has included geopolitical and external vulnerabilities that potentially have greater spillover effects to the Canadian financial system. 

As a result of continued growth in vulnerabilities and increasing complexity in the risk environment, OSFI announced its intention to review the DSB in December 2021 to ensure it remains relevant, appropriate and fit for purpose. This report provides a summary of key findings and decisions of this review. Going forward, OSFI intends to conduct periodic reviews of the DSB to ensure it continues to meet its objectives and contributes to the long-term effectiveness of the capital regime.

Changing vulnerabilities in an uncertain risk environment

OSFI’s review of the DSB looked at the scope of risks and vulnerabilities. At inception, the DSB primarily covered domestic vulnerabilities as outlined in the previous section. More recently, as noted in OSFI’s 2022-23 Annual Risk Outlook, there has been a shift in the global risk environment given the rising cost of debt and geopolitical tensions against a backdrop of elevated systemic debt levels and inflation and monetary policy tightening. These new considerations along with an increasingly complex risk environment justify an expanded scope for the DSB to account for external systemic vulnerabilities as the fourth category of vulnerabilities covered by the DSB.

In addition, OSFI has used stress-testing – that is, modelling a variety of severe yet plausible scenarios and their possible effects on the Canadian economy and financial system – to evaluate the DSB range.  Among the scenarios considered were recent stress tests and sensitivity analysis, including Bank of Canada and IMF top-down stress tests. Results of these stress tests indicate that under certain hypothetical scenarios, capital levels could fall sharply and, in the more severe scenarios, approach or pass beyond the 8% Capital Conservation Buffer threshold, indicating that a wider range for the DSB would be appropriate should vulnerabilities grow, to support lending and absorb potential credit losses in a variety of economic conditions^{Footnote 1}.

Another consideration is D‑SIBs’ exposure to the systemic vulnerabilities and overall asset growth over time relative to the economy. Total D‑SIB assets to GDP has risen to about 261% as of Q2 2022, up from 237% in 2018 and 186% in 2012, and have generally expanded their international operations and become more complex and exposed to global risks. 

In the context of the foregoing, an increased upper limit of the DSB would help to address recent trends and potential for further growth in vulnerabilities to D‑SIBs.

Review decision #1

The range of the DSB has been increased to 0 - 4% to account for a potential increase in systemic vulnerabilities and uncertainty in the risk environment, while allowing D‑SIBs to maintain lending and absorb losses across a range of a severe but plausible risk events.

Flexibility to be responsive to changes in vulnerabilities

The DSB is designed to ensure the Canadian financial system remains resilient in the face of a number of possible risk events at any point in time, scaled to OSFI’s assessment of current environment. This approach is consistent with OSFI’s supervisory mandate and provides flexibility to be responsive to changes in vulnerabilities and risks.

Until this review, the DSB’s design involved semi-annual DSB level-setting announcements based on the analysis of a range of vulnerabilities and risks, as well as supervisory judgement by OSFI staff. To date, increases to the DSB level were based on OSFI’s assessment of a trend increase in vulnerabilities. In many cases, this upward trend was set against an environment where risks to DSIBs over the next six months, were seen to be low and/or stable.

Increases to the DSB (to build up capital buffers) have traditionally been accompanied by a phase-in period. This “advance notice” allows DSIBs time to acquire the requisite capital and acknowledges the benefit of predictability and stability in capital requirements, to the extent possible.

Decisions to decrease the DSB level – or, put more accurately, the release of capital preserved through the DSB – take an immediate effect and would be responsive to crises, risk events or adverse economic conditions, the COVID-19 pandemic being the sole example to date.

Such releases, whether in part or whole, are ordered when OSFI believes that risks to the DSIBs and the financial system are important enough to warrant regulatory intervention to ensure that lending to households and businesses is maintained.

In exceptional circumstances, OSFI may make and announce adjustments to the buffer in-between scheduled review dates.

Review decision #2

OSFI maintains the existing design of the DSB which provides flexibility to respond to changes in vulnerabilities and risks, while reaffirming the role of supervisory judgment.

Increased transparency supports public confidence in the financial system

OSFI believes that in addition to making any announced change to the DSB, such decisions should be accompanied by a supporting rationale. In the context of OSFI’s refocused mandate of strengthening public confidence in Canada’s financial system, OSFI has enhanced transparency around decisions on the range, design and level of the DSB, while still emphasizing the role that supervisory judgement plays in these decisions.

To further this goal, OSFI has shared some indicators related to vulnerabilities it monitors, as part of every DSB announcement (see Appendix). These indicators provide insight into some of the considerations that inform the calibration of the DSB but should not be taken as an exclusive or exhaustive list of factors as they may evolve as the economic and risk environment changes.

Further, it should be noted that there is no automatic or mechanical link between any indicator and changes to the DSB. These caveats aside, monitoring a set of indicators provides insight into how OSFI sees potential risks to the financial system and, consequently, how it makes adjustments to the DSB.

Review decision #3

OSFI has enhanced transparency around the calibration of the DSB through the provision of indicators that contribute to the semi-annual decision. This supports OSFI’s mandate to promote confidence in the Canadian financial system.

Conclusion

On balance, OSFI’s review found that the DSB has performed well as a tool for systemic stability that requires D‑SIBs to build up capital buffers during relatively stable or positive periods in the economic cycle, and ready for release as needed during volatile or negative periods. With the few adjustments identified as a result of OSFI’s 2022 fall review, the DSB’s suitability, functionality and visibility will be enhanced, thus furthering public confidence in Canada’s financial system.

Appendix

Sources: Statistics Canada; Bank of Canada; BIS

Graph description - Selected indicators related to Canadian vulnerabilities

The periods covered for the Global Financial Crisis (GFC) are from Q3-2008 until Q2-2009. For the Covid crisis, the peak of the crisis was from March to September 2020.

Canadian Household Debt-to-Income Ratio

Date	Debt-to income
June 2006	140%
September 2006	139%
December 2006	141%
March 2007	142%
June 2007	147%
September 2007	151%
December 2007	152%
March 2008	153%
June 2008	154%
September 2008	154%
December 2008	157%
March 2009	160%
June 2009	162%
September 2009	162%
December 2009	164%
March 2010	162%
June 2010	167%
September 2010	166%
December 2010	166%
March 2011	167%
June 2011	169%
September 2011	170%
December 2011	168%
March 2012	169%
June 2012	170%
September 2012	170%
December 2012	170%
March 2013	168%
June 2013	169%
September 2013	170%
December 2013	170%
March 2014	170%
June 2014	171%
September 2014	170%
December 2014	172%
March 2015	170%
June 2015	171%
September 2015	172%
December 2015	175%
March 2016	181%
June 2016	181%
September 2016	181%
December 2016	181%
March 2017	183%
June 2017	182%
September 2017	181%
December 2017	181%
March 2018	183%
June 2018	184%
September 2018	185%
December 2018	183%
March 2019	183%
June 2019	181%
September 2019	182%
December 2019	181%
March 2020	181%
June 2020	162%
September 2020	173%
December 2020	177%
March 2021	175%
June 2021	176%
September 2021	179%
December 2021	185%
March 2022	180%
June 2022	182%

Canadian House Price to Income Index

Date	House Price to Income
June 2006	1.46
September 2006	1.46
December 2006	1.46
March 2007	1.47
June 2007	1.53
September 2007	1.57
December 2007	1.59
March 2008	1.59
June 2008	1.56
September 2008	1.50
December 2008	1.45
March 2009	1.42
June 2009	1.44
September 2009	1.52
December 2009	1.60
March 2010	1.60
June 2010	1.62
September 2010	1.58
December 2010	1.58
March 2011	1.63
June 2011	1.68
September 2011	1.67
December 2011	1.64
March 2012	1.63
June 2012	1.64
September 2012	1.62
December 2012	1.60
March 2013	1.59
June 2013	1.61
September 2013	1.64
December 2013	1.68
March 2014	1.71
June 2014	1.71
September 2014	1.70
December 2014	1.74
March 2015	1.73
June 2015	1.75
September 2015	1.79
December 2015	1.83
March 2016	1.97
June 2016	2.03
September 2016	2.01
December 2016	1.97
March 2017	2.01
June 2017	2.07
September 2017	2.02
December 2017	1.95
March 2018	1.97
June 2018	1.95
September 2018	1.92
December 2018	1.91
March 2019	1.89
June 2019	1.86
September 2019	1.90
December 2019	1.95
March 2020	1.98
June 2020	1.72
September 2020	1.88
December 2020	2.08
March 2021	2.19
June 2021	2.27
September 2021	2.36
December 2021	2.46
March 2022	2.46
June 2022	2.49

Canadian Non-Financial Corporate Debt to GDP

Date	Non-Financial Corporate debt to GDP
June 2006	43.23%
September 2006	43.72%
December 2006	44.98%
March 2007	46.06%
June 2007	45.02%
September 2007	45.86%
December 2007	46.75%
March 2008	47.86%
June 2008	46.65%
September 2008	46.62%
December 2008	50.51%
March 2009	52.55%
June 2009	53.15%
September 2009	51.95%
December 2009	49.86%
March 2010	48.55%
June 2010	48.41%
September 2010	48.63%
December 2010	47.79%
March 2011	47.84%
June 2011	47.56%
September 2011	49.18%
December 2011	48.12%
March 2012	49.01%
June 2012	50.16%
September 2012	50.95%
December 2012	51.22%
March 2013	51.78%
June 2013	53.26%
September 2013	52.99%
December 2013	53.34%
March 2014	54.83%
June 2014	55.30%
September 2014	56.45%
December 2014	57.27%
March 2015	62.26%
June 2015	62.18%
September 2015	62.86%
December 2015	63.99%
March 2016	62.43%
June 2016	64.68%
September 2016	65.14%
December 2016	64.15%
March 2017	64.49%
June 2017	64.89%
September 2017	65.17%
December 2017	65.59%
March 2018	66.41%
June 2018	66.94%
September 2018	66.31%
December 2018	68.28%
March 2019	68.88%
June 2019	69.18%
September 2019	69.99%
December 2019	69.27%
March 2020	73.33%
June 2020	86.50%
September 2020	77.78%
December 2020	76.02%
March 2021	72.59%
June 2021	72.73%
September 2021	72.28%
December 2021	71.10%
March 2022	68.39%
June 2022	65.19%

 

Sources: OECD; BIS

Graph description - Selected external and systemic macro-financial indicators

The periods covered for the Global Financial Crisis (GFC) are from Q3-2008 until Q2-2009. For the Covid crisis, the peak of the crisis was from March to September 2020.

Inflation(CPI): G-7 Countries
Total, Annual growth rate (%)

Date	Inflation (CPI)- G7
September 2006	2.53
December 2006	1.67
March 2007	1.93
June 2007	2.07
September 2007	1.88
December 2007	3.07
March 2008	3.27
June 2008	3.60
September 2008	4.37
December 2008	1.78
March 2009	0.43
June 2009	-0.54
September 2009	-1.16
December 2009	0.79
March 2010	1.66
June 2010	1.40
September 2010	1.05
December 2010	1.28
March 2011	1.89
June 2011	2.70
September 2011	2.99
December 2011	2.72
March 2012	2.44
June 2012	1.80
September 2012	1.59
December 2012	1.65
March 2013	1.37
June 2013	1.19
September 2013	1.46
December 2013	1.24
March 2014	1.32
June 2014	1.95
September 2014	1.70
December 2014	1.24
March 2015	0.31
June 2015	0.23
September 2015	0.23
December 2015	0.42
March 2016	0.74
June 2016	0.64
September 2016	0.75
December 2016	1.34
March 2017	1.99
June 2017	1.62
September 2017	1.69
December 2017	1.78
March 2018	1.90
June 2018	2.20
September 2018	2.29
December 2018	1.96
March 2019	1.42
June 2019	1.61
September 2019	1.46
December 2019	1.60
March 2020	1.71
June 2020	0.36
September 2020	0.76
December 2020	0.67
March 2021	1.37
June 2021	3.36
September 2021	4.00
December 2021	5.25
March 2022	6.38
June 2022	7.50
September 2022	7.60

Global Non-Financial Debt to GDP Ratio

Date	Global Non-Financial Debt to GDP
March 2006	202%
June 2006	206%
September 2006	205%
December 2006	206%
March 2007	206%
June 2007	205%
September 2007	210%
December 2007	211%
March 2008	216%
June 2008	209%
September 2008	198%
December 2008	202%
March 2009	205%
June 2009	223%
September 2009	236%
December 2009	232%
March 2010	224%
June 2010	218%
September 2010	229%
December 2010	228%
March 2011	229%
June 2011	228%
September 2011	219%
December 2011	217%
March 2012	220%
June 2012	220%
September 2012	228%
December 2012	227%
March 2013	224%
June 2013	222%
September 2013	226%
December 2013	226%
March 2014	229%
June 2014	231%
September 2014	222%
December 2014	220%
March 2015	219%
June 2015	226%
September 2015	229%
December 2015	232%
March 2016	243%
June 2016	246%
September 2016	247%
December 2016	236%
March 2017	240%
June 2017	245%
September 2017	247%
December 2017	247%
March 2018	247%
June 2018	236%
September 2018	233%
December 2018	235%
March 2019	240%
June 2019	243%
September 2019	241%
December 2019	245%
March 2020	246%
June 2020	267%
September 2020	278%
December 2020	291%
March 2021	281%
June 2021	275%
September 2021	267%
December 2021	264%
March 2022	261%