Better Safe than Sorry: Options for Managing Bank Runs in the Future

Summary:
Page Title:Better Safe than Sorry: Options for Managing Bank Runs in the Future – C.D. Howe Institute
Article Title:Better Safe than Sorry: Options for Managing Bank Runs in the Future
URL:https://cdhowe.org/publication/better-safe-sorry-options-managing-bank-runs-future/
Published Date:October 1, 2024
Accessed Date:February 9, 2025

The Study in Brief

  • While Canada’s prudential framework has successfully promoted a stable banking system for many years there are some emerging trends that suggest it may not be sustainable in the future.

  • This Commentary considers four options to manage the risk of future bank failures in the event that we need to change our prudential framework. They range from some fairly minor adjustments to the status quo to more radical options that even include, at one extreme, separating the money creation and credit extension roles of deposit-taking institutions.

  • We conclude that a more stable and “resolvable” financial system may result from considering more radical options but they would result in more expensive financial services. That said, they may make it possible to contemplate a simpler and less intrusive prudential regulatory regime.

  • Regardless of the path taken there are some steps that should be taken now to make it easier to resolve a failing deposit-taking institution and reduce the run-risk that currently exists in bank deposits.

Press release

The author thanks Jeremy Kronick, Rosalie Wyonch, Charles DeLand, John Crean, and several anonymous reviewers for helpful comments on an earlier draft. The views expressed are the author's and do not reflect those of his former employers.

Introduction

Global banking history is replete with examples of bank runs not only bringing down a deposit-taking institution but – when a financial panic sets in – also seriously crippling confidence in the broader financial system and the economy more generally. The Global Financial Crisis of 2007-08 is a classic example. Unfortunately, the failures of several regional banks in the US and the demise of Credit Suisse in 2023 in Switzerland suggest that reforms to bank regulation and supervision enacted in the wake of that crisis have not been sufficient. Authorities in both countries had to step in with public money to prevent those failures from triggering broader disruptions to their financial systems and economies. As noted in a recent IMF blog and working paper (Tobias and Dobler 2024)1 and summarized in Box 1, not only were prudential standards (i.e., the safety and soundness requirements applied to deposit-taking institutions) arguably too lax for the affected institutions, but supervisors failed to require those institutions to take prompt corrective action to address the underlying vulnerabilities that ultimately led to their failure.

Those experiences suggest there is still a risk that public authorities may need to supply public funds or guarantees to a deposit-taking institution when it experiences a rapid outflow of deposits and other liabilities (a bank run), rendering it unable to access markets for fresh sources of funds or capital. This risk may continue to grow now that implementation of Basel III reforms has stalled as some jurisdictions shy away from fully implementing the final chapter of those reforms.2

Canada’s banking history, by contrast, has been far calmer. We have been blessed with a stable banking system with no failures of any significance for nearly 30 years. It would be tempting to conclude that we have successfully contained the systemic risks posed by bank runs and thus our prudential regulatory framework is working well. But for reasons that will be outlined later in this paper we should take care not to rest on our laurels. There are some emerging trends that suggest our current prudential regulatory framework’s focus on preventing bank failures may not be sustainable going forward.3

If indeed our prudential regulatory framework cannot be sustained in the future, one must ask what preventative steps can be taken to manage the systemic risks posed by bank runs, and facilitate an orderly resolution or exit of a deposit-taking institution that is no longer viable before it fails – all without calling on the public purse. While Canada has not had any serious failures in many years, it would be best to prepare for possible problems now when things are quiet, rather than in the heat of the moment.

As will become apparent in this paper, there are no simple answers to this question. Instead, we propose four broad options for policymakers to consider. They range from some fairly minor adjustments to the status quo to more radical options that even include, at one extreme, separating the money creation and credit extension roles of deposit-taking institutions. While larger measures must be weighed against the potential day-to-day impact on the cost of financial services for the broader economy, we conclude that a more stable and “resolvable”4 financial system may result from considering more radical options. The more radical options may even make it possible to contemplate a simpler and less intrusive prudential regulatory regime; a more resolvable banking system might allow policymakers to place more reliance on bank boards and management to be held accountable for the governance of their institutions.

Regardless of the option selected, this paper outlines some steps that should be taken now to make it easier to resolve a failing deposit-taking institution and reduce the run-risk that currently exists in bank deposits. These steps include:

(i) completing work at the Canada Deposit Insurance Corporation (CDIC) to speed up the payout of future deposit insurance claims when a failing institution cannot be merged with a healthier institution;

(ii) making use of the information contained in bank resolution plans and paying more attention to resolvability issues in current prudential supervision frameworks to make it easier to resolve failing institutions and reduce the run-risk that currently exists in bank deposits; and,

(iii) making sure that resolution plans incorporate the possibility that multiple institutions may encounter stress and have to be resolved at the same time.

We end the paper with some thoughts on the broader changes taking place in the financial system and the need to work with international counterparts to ensure that the current risks embedded in the banking system do not appear in other guises, in other parts of the financial system that are not subject to the same prudential safeguards in place for deposit-taking institutions.

Some Options for the Way Forward

Deposit-taking institutions are inherently more fragile than other types of financial institutions due to their large reliance on deposits, and other funding liabilities, that can run at a moment’s notice and potentially give rise to panics that destabilize the broader banking system and economy more generally. Given these institutions are meant to take risks in competitive markets, runs can be expected whenever there is a hint that a deposit-taking institution is no longer able to access funds or capital from private markets. Loss of access generally arises when there are concerns about whether the value of a deposit-taking institution’s assets remains sufficient to cover all its obligations – i.e., when the institution’s solvency is in doubt. This is especially true in a social-media age where, as the experience of SVB showed, rumours can quickly surface and become panics in the blink of an eye.

Given bank runs are bound to happen from time to time, the question then becomes how to manage a run when it happens and prevent it from turning into a broader banking system panic that could disrupt the wider economy; and how do we do this without drawing on the public purse.13 Or, simply put, if we cannot avoid a run we need a plan to manage runs that allows for smooth resolutions of affected deposit-taking institutions, if they are not able to recover on their own.

A key ingredient for the smooth resolution of a stressed financial institution is time. One should focus on options that can buy more time during a bank run, so that either its recovery plan can be carried out or, if necessary, an orderly resolution of the stressed institution can be conducted – be it through an orderly sale to another institution, a bridge bank process or, in the case of the largest institutions, an open door process whereby the bank remains open under Canada Deposit Insurance Corporation (CDIC) control while the institution is recapitalized, and the issues that led to the resolution are addressed.

The more confident we are that stressed institutions can be resolved in an orderly fashion without causing significant harm to the broader financial system or economy, the more we can tolerate future failures. Sound resolution strategies require time and advanced planning, and may reduce the need for our current complex regulatory and supervisory systems that are primarily focused on preventing failures. In turn, perhaps we could place more responsibility on bank boards and management to manage their institution’s own affairs, knowing that if they fail to do so, the institution can be safely removed from the financial system.

Four stylized options are put forward to stimulate discussion.

Option A: Introduce More Comprehensive and Efficient Deposit Insurance

The 100 percent deposit limits introduced in 2023 by US authorities to protect depositors of failing banks, and the guarantees provided to depositors of other banks, quickly stabilized the regional banking system in the United States. That success has led to calls in Canada for higher deposit insurance limits more generally to protect depositors and reduce the risk of future bank runs. Indeed, given that deposit insurance limits are often raised when an institution experiences stress, it prompts the question: why not simply recognize the obvious and formally offer 100 percent deposit insurance now?14

Such a step would also recognize the fact that with the introduction of bail-in debt triggers for the six major banks, most of the depositors of those banks have effectively been given de facto priority over other creditors in a resolution scenario.15 Hence, raising deposit insurance limits would arguably simply level the playing field in terms of how most depositors are treated in resolution – at both the largest banks and the smaller deposit-taking institutions that do not issue wholesale bail-in-able debt.16

However, as pointed out by Kronick, Munn and Zelmer (2023), increasing deposit insurance limits may not actually eliminate the risk of a bank run as was shown in Home Capital’s deposit run in 2017 where fully insured deposits ran too. Such a step may also generate additional moral hazard in the banking system that would need to be managed. For example, larger risk-based deposit insurance premiums or additional regulatory requirements might be needed, especially for smaller deposit-taking institutions that benefit the most from deposit insurance.17 One would also need to determine how big the supporting deposit insurance funds should be.

Meanwhile, other near-term actions could be considered under this option to help manage the risk of bank runs, other than raising deposit insurance limits. An obvious one would be for CDIC to complete its payout modernization project18 so that all depositors can have greater confidence that they will be quickly reimbursed in the event that CDIC decides to close an institution and reimburse depositors.19

Another would be for CDIC and its federal partners (Department of Finance, Bank of Canada, OSFI, and Financial Consumer Agency of Canada (FCAC)) to review the resolution plans that the six major banks have prepared and consider whether they indicate a need for structural change to the corporate structure and business models of those banks, so as to facilitate a more efficient resolution process if necessary. OSFI and its provincial counterparts should also consider expanding their current prudential supervisory frameworks to include resolvability risk, and consider potential future resolution implications when reviewing transactions that require regulatory or ministerial approvals.

CDIC should also consider reviewing the assumptions underpinning these resolution plans in light of the events of 2023. For example, the assumption that CDIC would have a weekend to take control of a failing institution and commence resolution before the institution is re-opened for business under CDIC control may no longer be realistic, in light of how quickly deposits melted away from Silicon Valley Bank.

By the same token, the recent US and past UK experiences suggest that CDIC and its sister agencies at the federal level should be prepared for the possibility that more than one institution might need to be resolved at the same time, given that many deposit-taking institutions have similar business models and exposures. In such situations, certain types of shock, such as a widespread economic slowdown, could conceivably destabilize multiple institutions in a short period of time.

In the longer term, consideration could also be given to adopting the US practice of granting depositors priority ahead of other creditors in a resolution scenario, along with reintroducing a bygone practice in the trust company industry wherein guaranteed investment certificates issued by certain trust companies were backed by specific assets on the balance sheet, plus a more general claim with those of other creditors on the residual assets of the trust company. That said, steps like these may increase the risk that other creditors could become further incentivized to run in times of stress.

Option B: Introduce More Stringent Liquidity Standards

Another way to buy more time for a smooth resolution could be to encourage deposit-taking institutions to carry larger stocks of high-quality liquid assets (HQLA), which can be sold as needed to meet a bank run. That way they would be able to survive for a longer time in the event of a run. The fact that they would be carrying larger stocks of liquid assets to back their liabilities may even help to prevent a run from emerging in the first instance.20

The global financial crisis served as a good reminder of the need for prudent liquidity management. Among other things it prompted OSFI and its foreign counterparts to introduce two new liquidity requirements as part of the post-crisis Basel III reforms: the Liquidity Coverage Ratio (LCR) and the Net Stable Funding Ratio (NSFR), designed to encourage deposit-taking institutions to carry enough liquid assets to survive a 30-day stress scenario (LCR) and to reduce their reliance on short-term wholesale funding of less than one-year remaining term to maturity (NSFR). OSFI also requires larger deposit-taking institutions to adhere to a Net Cumulative Cash Flow (NCCF) supervisory metric that computes a survival horizon for each institution based on specified assumptions regarding their cash inflows and outflows, including likely deposit and funding rollovers in times of stress.21 Meanwhile, the smallest institutions are required by OSFI to file operating cash flow statements to facilitate liquidity monitoring by OSFI supervisors.22

Appendix II summarizes the salient assumptions regarding expected deposit and funding outflows in times of stress for the LCR and NCCF tests. While OSFI’s Liquidity Adequacy Guideline claims that all its liquidity tests are meant to assess a deposit-taking institution’s liquidity position in a severe but plausible stress environment, the run-off rates prescribed for these two tests clearly assume that most of an institution’s core deposit base will stay put in times of stress.

These assumptions paint a sunnier picture than the reality experienced in practice by Home Capital in 2017 and Silicon Valley Bank in 2023, raising the question of whether the run-off rates prescribed in the liquidity stress tests should be raised significantly to be more consistent with those 2017 and 2023 deposit-run experiences.23

Table 1 presents a summarized version of the LCR liquidity disclosures for the six major banks as a whole for the three months ending October 31, 2023. Recall that the LCR summarizes a bank’s capacity to withstand liquidity stress for a 30-day horizon using the rather lenient run-off rate assumptions prescribed by OSFI and its international counterparts. If one simply doubles the run-off rate assumptions for less stable retail deposits (from 10 to 20 percent), the last column of Table 1 suggests that net cash outflows would increase by $118 billion to $235 billion (i.e., doubling the $117,724 million weighted-average shown in Table 1 for that line item). If one also assumes that all non-operational unsecured wholesale deposits would run, the last column of Table 1 suggests that net cash outflows for that line item would increase by another $372 billion ($852,100 million – $479,843 million). Together, this would result in a $490 billion increase in net cash outflows in the LCR test, equivalent to about 1/3 of those banks’ $1,472 billion holdings of High-Quality Liquid Assets (HQLA), or roughly 7 percent of their $7.4 trillion holdings of loans and investments as of October 31, 2023.24 The impact on the supply of credit extended by these banks could thus be rather significant if no countervailing actions were taken to mitigate the impact of the more stringent (though still rather lenient) recalibrations.

This simple analysis suggests that, all things being equal, this approach – expecting banks to self-insure themselves against the risk of bank runs by carrying enough high-quality liquid assets to meet a major run and survive for at least 30 days – could be very costly to society both in terms of revenue forgone by the banks by having to carry more lower-yielding liquid assets, and in terms of the potential reduction in the supply of credit to households and businesses served by those banks.

Table. Title: "Table 1: Six Major Banks, Liquidity Coverage Ration (LCR) August 1, 2023 to October 21, 2023 (Daily Average) (in $millions)

Of course, all things are not equal. If they were faced with such an outcome. it would be reasonable to expect the banks to take other steps to contain the increase in the size of their required liquidity holdings. They could, for example, reduce their need for liquid assets by following the lead of EQ Bank and routinely enforce notice requirements for savings deposit withdrawals, in return for offering higher rates on those deposits to entice customers to voluntarily accept such a restriction.25 That would have the salutary effect of shifting those deposits beyond the 30-day horizon of the LCR test, thereby reducing the amount of HQLA that banks need to carry, given those deposits could no longer run at short notice.

If combined with not paying interest on accounts that permit immediate withdrawals without sufficiently stringent notice requirements, it could help to encourage both retail and wholesale customers to constrain the flightiness of their deposits.

All told, this option could help generate a larger pool of high-quality liquid assets that would be quick to mobilize in case of a bank run. But it would likely come at a high cost in terms of the potential impact on the price and supply of credit to the private sector. Thus, expecting banks to fully protect themselves against the risk of a bank run by carrying more HQLA may not be feasible in practice. Indeed, that is why most jurisdictions including Canada have had central banks for many decades. A key role of central banks is to be prepared to offer emergency liquidity assistance to solvent institutions that are experiencing stress. Some suggestions to enhance their function are offered in the next option.

There remain some important unanswered questions. Even if it is practicable, would such an option buy enough time to facilitate an orderly resolution of a stressed institution? Could breaching the minimum thresholds of liquidity standards in times of stress, as assets are sold to meet a run, actually make things worse? Could publicly disclosing an LCR ratio below 100 percent trigger further runs, even though OSFI’s liquidity requirements and their Basel counterparts explicitly allow for drawdowns of liquid assets in times of stress? Such a path would no doubt require international agreements to protect the competitive position of Canadian banks relative to their international counterparts.

Option C: Redesign the Current Liquidity Requirements and Expand the Scope of Central Bank Liquidity Support

The idea in this option would be to replace the current liquidity standards with a new standard focused on facilitating a smooth resolution. This new standard would require banks to carry enough high-quality liquid assets and other assets, such as loan portfolios, that can be quickly pledged (i.e., posted as collateral) to the Bank of Canada in exchange for cash on short notice. This would ensure that any deposit, or other funding claim likely to run within the timeframe needed for an orderly resolution, can be quickly met. The holders of instruments that could run within that timeframe would have first claim on those assets, giving them the comfort of knowing that, even if the institution is at risk of becoming insolvent, there is no risk that their obligations will not be honoured. That way depositors and other creditors that can run within the time required to commence the resolution process would have less incentive to run in the first place. And if they did run, their claims would be promptly honoured.

That said, this option is unlikely to prevent the kind of insured deposit run issue that took place at Home Capital. As noted previously, third parties helping clients manage deposits would likely still shun deposit-taking institutions experiencing stress. They would want to avoid the potential reputational consequences of having to explain why clients’ money was placed with a failing institution, even if the clients’ funds are fully protected by the additional liquid assets available under this option.

Clearly, this option comes with some issues that would need to be addressed.

All creditors and shareholders, including the deposit insurer/resolution authority, would need to understand from the outset that a deposit-taking institution’s high-quality liquid assets – and other assets that can be pledged to the Bank of Canada for emergency liquidity assistance – will be reserved, in the first instance, to satisfy depositors and other creditors that have claims that can run on demand or that are maturing within a pre-set timeframe. To put it simply, those creditors would be granted at least a de facto priority over other creditors from a resolution perspective.

Putting this into practice is feasible but likely complicated from a legal perspective. It would likely raise the cost of financial intermediation in the economy as those creditors who don’t benefit from the priority claim would likely demand higher yields to compensate them for the additional risk they would bear going forward. By the same token, deposit insurers are unlikely to be thrilled because, in the event of a run, the best assets are likely to be liquidated first, making resolutions more expensive at the end of the day.

The Bank of Canada would also need to be ready to manage the broader market consequences of a stressed institution having to mobilize its assets quickly to meet any runs. That would be especially true if the run involved one of the major deposit-taking institutions. Fortunately, the Bank of Canada has developed a range of facilities that could be used for such a purpose.26

Former Bank of England Governor Mervyn King has proposed having deposit-taking institutions pre-pledge assets at the central bank so that the central bank could quickly provide liquidity support on demand to any solvent institution experiencing a run (King 2016). He suggests that the introduction of such a “pawnbroker for all seasons” service could alleviate the need for a large segment of the current prudential regulatory and supervisory apparatus.

Indeed, the Bank of Canada encourages members of the Lynx payment system to pre-pledge collateral with it, to facilitate quick access to its liquidity facilities. However, stressed deposit-taking institutions may be loathe to access the Bank’s liquidity facilities on an emergency basis due to stigma concerns – i.e., for fear that it would signal that they have lost access to markets, further exacerbating a run.27

To combat this stigma, it might be worth considering allowing distressed deposit-taking institutions to sell the assets needed to meet a run on an outright basis to the Bank of Canada, which could then either hold them to maturity or resell them to the market on an orderly basis over time, to reduce the risk that markets might be disrupted in the short term. In effect, this would expand the Bank of Canada’s current emergency lending facilities to include an outright purchase of assets option. No stigma should arise from accessing such a facility because no credit extension is involved. (There are no requirements for an affected deposit-taking institution to disclose the counterparty when it sells assets on an outright basis). Alternatively, the Bank of Canada could stand ready to intervene in the affected markets as necessary, to address any disruptions caused by a distressed institution selling assets to the marketplace instead of the central bank.

Tucker (2018) cautions against outright facilities for emergency lending purposes, arguing that an outright purchase is a one-shot game, exposing the central bank purchaser to market and default risk, whereas a secured liquidity advance – or repo – enables a central bank to demand additional collateral or a different type of collateral if the original pledged asset has fallen in value. While a legitimate concern, one can argue that it is a risk that can be managed if the central bank applies suitably conservative discounts in its purchase price and limits the range of assets that can be purchased to tradable securities, which are easier to value quickly than illiquid loan portfolios whose values are more likely to be dependent on the deposit-taking institution’s own credit risk models. In any event, this Rubicon has already been crossed. Witness the Bank of Canada and other central banks’ purchases of private-sector-issued securities on an outright basis during the pandemic.

One must also question whether Governor King is correct in saying that most prudential requirements could be set aside under this kind of option – that greater onus could be placed on bank boards and management to govern themselves, subject to the oversight of the marketplace.

Option D: Split Deposit-taking Institutions into Narrow and Broad Institutions

Our fourth and final option would entail a major restructuring of deposit-taking institutions so that their money creation and credit extension roles are split into separate entities. This is not a novel idea. In the 1930s, a large number of leading US macroeconomists supported a proposal for fundamental monetary reform that later became known as the Chicago Plan, after its strongest proponent, professor Henry Simons of the University of Chicago. It was also supported and well summarized by Irving Fisher of Yale University (Fisher 1936).

The key feature of this Plan was the call for 100 percent backing of deposits by government-issued money, and also by ensuring that the financing of new bank credit could only take place through earnings that have been retained in the form of government-issued money, or through the borrowing of existing government-issued money from non-banks – not the traditional way, through the creation of new deposits by banks. (Each time a bank makes a loan, it simultaneously creates a matching deposit, thereby creating new money.)

Fisher claimed four major advantages for this Plan. First, preventing banks from creating their own funds during credit booms and then destroying these funds during subsequent contractions would allow for much better control of credit cycles, which were perceived at the time to be the major source of business cycle fluctuations. Second, 100 percent reserve backing would completely eliminate bank runs. Third, allowing the government to issue money directly at zero interest, rather than borrowing that same money from banks at interest, would lead to a reduction in the interest burden on government finances and to a dramatic reduction of (net) government debt, given that irredeemable government-issued money represents equity in the commonwealth rather than debt. Fourth, given that money creation would no longer require the simultaneous creation of mostly private debts on bank balance sheets, the economy could see a dramatic reduction not only of government debt but also of private debt levels.

Benes and Kumhof (2012) studied these claims by embedding a comprehensive and carefully calibrated model of the banking system in a dynamic stochastic general equilibrium model of the US economy. They found support for all four of Fisher’s claims. Further, their model suggested that adoption of the plan could generate output gains approaching 10 percent, and that steady state inflation could drop to zero without posing problems for the conduct of monetary policy.

A modern version of the Chicago Plan would be to have all liabilities to the public – demand and notice deposits and redeemable term deposits that are to be perfectly liquid and redeemable at par – 100 percent backed by central bank reserves or high-quality liquid assets, while credit extension would be funded by non-redeemable term deposits, marketable debt and bank capital. An even more radical version of this option would see the public holding central bank digital currencies for payment purposes, with the payment system operated by technology companies, and credit intermediation handled by investment funds and other institutional investors.28

This option would likely be seen as too radical for most people to contemplate at this time. That said, some aspects of this option are already emerging as the shadow banking system expands, relative to the regulated banking system. For example, private debt funds are playing a rapidly growing role in financing businesses by raising money from investors. Meanwhile, on the other side of the intermediation spectrum, various types of money market funds offer the public the opportunity to place some of their savings in funds backed by a variety of short-term assets, with the ability to withdraw their money at par on short notice. Disintermediation of the banking system could accelerate further if the payment system is opened up to a broader range of participants in the future.

Weighing the Options

Table 2 assesses the four options against four key criteria:29 (i) the extent to which the option might reduce the risk of a bank run; (ii) whether the option would help facilitate an orderly resolution if a bank run happened; (iii) the impact on cost of financial intermediation; and finally, (iv) whether the current web of regulatory and supervisory requirements could potentially be reduced.

The rankings in the table are basic relative rankings of each of the four options against the four criteria. Not surprisingly, there is no dominant option.

Table. Title: "Table2: The Options and Their Outcomes

Option A’s more efficient deposit insurance regime offers some elements that should be pursued to facilitate smoother resolutions of distressed institutions in the future. But it is unlikely to reduce the risk of bank runs, as shown by the Home Capital 2017 experience, discussed previously, where even insured depositors were quick to run when there were concerns about that institution. This option also raises the question of whether larger risk-based deposit insurance premiums and even more regulation, at least for smaller institutions, might be needed if higher deposit insurance limits are introduced, given the greater moral hazard that would be expected with higher deposit insurance limits.

The more stringent liquidity requirements of Option B may offer some resolution benefits, such as allowing more time to carry out a smoother resolution of a troubled institution. But this would likely come at the expense of more costly financial intermediation, due to the need for deposit-taking institutions to carry more liquid assets.

Option C’s redesign of liquidity requirements offers the hope of reducing the risk of bank runs and facilitating smoother (though potentially more costly) resolutions if a run occurs, but much depends on whether it would actually prevent the types of runs experienced by Home Capital when third-party intermediaries are involved. Much also depends on whether an outright purchase facility would actually reduce the stigma of obtaining emergency support from the central bank, and whether easier access to central bank emergency facilities would in fact allow for reduced regulatory burden as suggested by Governor King. But it, too, would likely increase the cost to the economy of financial intermediation due to the preferential treatment that would be accorded to deposits and other claims that could run prior to the commencement of resolution.

Finally, restructuring deposit-taking institutions under Option D offers the greatest potential benefits across each of the criteria, but this would come at the expense of a major re-engineering of the banking system, bound to be costly at least in the short run. But, as noted previously, this may well be the option that emerges over the longer run in practice, as competition drives many banking services into the hands of entities that operate outside of the prudentially regulated banking system. Indeed, by raising the cost of financial intermediation, all of the options presented in this paper may serve to hasten our path to an end-state of an even larger shadow banking system.

Discussion and Policy Recommendations

The events of 2023 have shown us that more work needs to be done if we want to reduce the risk of bank runs that could undermine confidence in the banking system and the functioning of the economy more generally. To stimulate discussion on how best to move forward, this paper has sketched out four high-level options for possible reforms, ranging from some relatively modest adjustments of current practices to a radical rethink of how the business of banking should be structured.

None of the four options is a clear panacea because, while one can envision ways to reduce the risk and consequences of bank runs, it is clear that they would likely be accompanied by higher costs in financial intermediation – costs that would be borne by Canadian households and businesses. That said, there are some steps that should be taken now to move forward.

• CDIC should move expeditiously to complete its payout modernization project so that insured depositors can be confident that they will be reimbursed quickly in the event that a failing institution is closed.

• CDIC and its federal partners should take stock of the resolution plans that the six major banks have prepared, and consider whether there is need for structural changes to corporate structure and business models – in the interest of facilitating a more efficient resolution process, should the need arise. OSFI and its provincial counterparts should also consider expanding their current prudential supervisory frameworks to include resolvability risk, and consider potential future resolution implications when reviewing transactions that require regulatory or ministerial approvals.

• CDIC and its federal partners should also review the assumptions underpinning the major bank recovery and resolution plans, in light of the events of 2023. They should also be prepared for the possibility that more than one institution might need to be resolved at the same time, given many deposit-taking institutions have similar business models and exposures.

• While it is tempting to offer full deposit insurance coverage – seeing that deposit insurance limits are often raised anyway when an institution is failing, and given that most depositors of the major banks now have added protection following the introduction of bail-in debt – past experience suggests such a step may not actually significantly reduce the risk of bank runs. Plus, that path may require higher risk-adjusted deposit insurance premiums and more regulation. This could be a topic worth exploring in the federal government’s planned review of the deposit insurance framework announced in its 2024 Budget.

• Rather than tightening up current liquidity requirements, OSFI and its international counterparts should instead consider what changes could be introduced to reduce the run risk in existing deposits, such as routinely enforcing notice requirements for withdrawals from savings accounts in exchange for higher interest rates on those accounts.

• The Bank of Canada may wish to explore whether there is merit in modifying the structure of its emergency liquidity facilities to make access easier for deposit-taking institutions in times of stress, and whether introducing an outright purchase option as part of these facilities could help reduce the stigma for institutions wishing to access those facilities.

Finally, while the Chicago Plan is likely too radical for most stakeholders to consider at this time, it may emerge in practice over time. Institutional investors are becoming more active in extending credit and supplying equity capital to firms, and households and firms are increasingly looking to various types of investment funds as repositories for their savings, instead of bank deposits. A key question is whether this will result in an inherently more stable financial system in the future, as some adherents believe, or if we will simply see the same fragilities emerge in a less prudentially regulated system.

Policymakers have been attuned to this risk. As a result, there is work underway led internationally by the Financial Stability Board, and international standard setters like the International Organisation of Securities Commissions (IOSCO), to enhance the resilience of non-bank financial intermediation in the global financial system.30 The main focus of this work so far has been to reduce excessive spikes in the demand for liquidity that can emerge from liquidity mismatches at open-ended investment funds (e.g., mutual funds and exchange-traded funds), and to enhance the margin requirements in capital markets. We encourage Canadian securities and prudential regulators to continue actively supporting this policy work and implementing its recommendations in Canada on a timely basis, to help foster a more resilient non-bank financial intermediation system here at home.

Conclusion

Preventing bank runs from turning into panics that disrupt the financial system, and the welfare of the public, has been a longstanding challenge in global banking history. The failures of some US regional banks and Credit Suisse in Switzerland in 2023 suggest that reforms to bank regulation and supervision enacted in the wake of the global financial crisis have not been sufficient to contain risk. While Canada has not had any banking failures for many years, we should not be complacent. There are some emerging clouds on the horizon that suggest there is risk that our current well-performing prudential regulatory framework may not be sustainable in the future. If so, the best time to step back and consider how best to manage problems is while they remain in the future, and not in the moment they descend upon us.

To initiate discussion, this paper summarizes some recent trends in the banking system, and society more generally, and how they may impinge on the conduct of future prudential regulation and supervision. They suggest that it might make sense to consider paying more attention at the margin to facilitating a smooth resolution when a failure is imminent, rather than simply preventing failures from occurring in the first place.

With that in mind, the paper presents for discussion four broad high-level options for how to move forward on this front. The options range from some fairly minor adjustments to the status quo to more radical options that even include, at one extreme, separating the money creation and credit extension roles of deposit-taking institutions. We conclude that while a more stable and resolvable financial system may result from considering more radical options, that needs to be weighed against the potential day-to-day impact on the cost of financial services for the broader economy. Having said that, the more radical options may make it possible to contemplate a simpler and less intrusive prudential regulatory regime, because a more resolvable banking system might allow for more reliance on bank boards and management to be held accountable for the governance of their institutions.

Regardless of the decision going forward – and there was no clearly dominant winner among our four options – we have outlined some steps that should be pursued now. They range from moving expeditiously to complete CDIC’s payout modernization project, to making use of the resolution plans prepared for the six major banks, to considering how best to structure the system going forward to make all deposit-taking institutions easier to resolve, and their deposit liabilities less likely to run in the first place.

Finally, we encourage policymakers to bear in mind the broader changes that have been taking place in the financial system over the past 15 years. Institutional investors are becoming more active in extending credit and equity capital to firms through various private credit and private equity vehicles, and households and firms are increasingly shifting their savings from bank deposits to various types of debt and equity investment funds, and other investment products. This could accelerate with open banking, as fintech firms and other non-bank entities seek to help Canadians conduct payments and manage their financial affairs outside of the prudentially regulated banking system. Whether the end result will be a more resilient financial system less prone to bank runs is still an open question. We thus encourage Canadian policymakers and regulators to continue participation in the work led by the Financial Stability Board and IOSCO on this topic, while thinking through what will be necessary for the specifics of the Canadian economy.

Appendix I: Why Canadian Prudential Regulation and Supervision May Not be Sustainable in the Future

Canadian prudential regulation stands out from the regulatory frameworks in most jurisdictions for its reliance on informal principles-based guidelines to set expectations for financial institutions. This allows prudential supervisors in Canada to apply more judgment than most of their foreign counterparts in tailoring the application of those guidelines to the specific circumstances of individual institutions. The outcome has been a safe and sound financial system that has served Canadians well over many years. But we should not rest on our laurels. This appendix sets out some reasons why our prudential framework needs to evolve to meet the changes taking place in the financial system and society more broadly.

Three trends may combine in the future to undermine our current approach to prudential oversight.

Changing structure of the financial system suggests financial institutions may become less willing to accept informal guidance from prudential authorities

Canada’s approach to prudential regulation has worked well because our banking system is quite profitable and dominated by a few large institutions. The latter have been willing to work cooperatively with the regulators because everyone has a stake in promoting a stable system that offers lucrative returns.

But this could change. There are many players outside the regulated sphere that would like to offer banking services without being regulated like banks. Consider open banking and fintech firms that are keen to offer payment services, financial advice, and other products and services to the Canadian public. Plus, we have institutional investors in private debt and equity markets that are increasingly active in extending credit and supplying equity capital to firms here and abroad. Many of them are not subject to prudential oversight by either federal or provincial agencies.

Growing competitive pressures can give rise to a less harmonious relationship between banks and their regulators. Consider the United Kingdom. The long-standing informal moral suasion afforded by the eyebrows of the Bank of England Governor vanished the moment US banks arrived in London and started bringing their lawyers to meetings with regulators.

Canadians are becoming less deferential towards public officials

Canadian prudential regulators have traditionally been granted a large amount of discretion in how they set expectations and apply them to the institutions they oversee. The exercise of supervisory judgment, in particular, comes with very little public oversight. Public disclosure of prescribed supervisory information is actually prohibited by law, making it difficult for the public to monitor how supervision is conducted in practice. The bar is also high to legally challenge the actions of prudential regulators. Canadian courts tend to be fairly deferential to the judgment exercised by public officials.

A quick glance at social trends reveals that public acceptance of government officials exercising judgment in an opaque manner with limited accountability is on the wane. The public increasingly expects such discretion to be constrained where possible, and that officials be transparent in the conduct of their duties.

Canada often follows social trends in the US with a lag. While both countries prohibit the public disclosure of supervisory information, it is worth noting that at the regulatory level US banks are becoming more willing to challenge public regulatory guidelines when they perceive those guidelines go beyond the requirements articulated in laws or official regulations. This resistance may intensify now that the US Supreme Court has decided to dispense with a 40-year-old legal doctrine, known as Chevron Deference, whereby US courts previously generally deferred to government agencies’ interpretation of rules and laws enacted by Congress.

One cannot help but wonder if Canada might follow the US down a similar path in the future. If so, it raises the question of whether our own prudential regulatory practices can be sustained.

Growing pressures on the ability of prudential regulators to credibly interact with regulated institutions

Anyone leading a business knows how hard it is these days to attract and retain highly qualified professionals. And the technology needed to conduct effective prudential regulation is becoming more sophisticated and expensive with every passing day. Public agencies like OSFI are not well placed to compete for that talent, nor to make those investments, given the constraints imposed by a multitude of government staffing and procurement requirements.

This, combined with growing public skepticism of people moving back and forth between regulators and the entities they regulate, may make it harder for regulators to continue to be respected as credible formulators of prudential guidance, and exercisers of the discretion provided by that guidance.

Appendix II: A Summary of Run-Off Assumptions in OSFI Liquidity Requirements

This appendix summarizes the run-off assumptions for deposits, wholesale funding, and other projected cash outflows in OSFI short-term liquidity stress tests known as the Liquidity Coverage Ratio (LCR) and the Net Cumulative Cash Flow (NCCF) supervisory metric.31

The deposit run-off assumptions for OSFI’s LCR and NCCF tests are summarized in Tables A1 and A2. Interestingly, the deposit run-off rates in both these tests are rather mild, considering that the two tests are supposed to measure a deposit-taking institution’s liquidity condition in a severe but plausible liquidity stress scenario. For example, the LCR test (Table A1) prescribes run-off rates ranging from 3 to 10 percent for retail deposits that are either redeemable on demand or mature within 30 days, and a 40 percent run-off rate for wholesale deposits held by non-financial corporates that are not covered by deposit insurance. Similarly, mild assumptions are used in the NCCF test (Table A2). While these run-off rates may have been calibrated to be broadly consistent with those experienced in the 2007-2008 financial crisis, they are certainly well below the deposit run-off rates experienced by Home Capital in 2017 and Silicon Valley Bank in 2023.

Table. Title: "Table A1: Liquidity Coverage Ratio Cash Outflow 30-Day Run-Off Rate AssumptionsTable. Title: "Table A1 Continued"Table. Title: "Table A2: Net Cumulative Cash Flow Weekly Run-Off Rate Assumptions

References

Angus Reid Institute. 2024. “Five-Year Decline: Canadians Growing More Critical of Their Provincial Governments as Unresolved Issues Linger.” Available at https://angusreid.org/provincial-government-performance-health-care-danielle-smith-eby-ford-legault/

Assuris. “How Am I Protected?” Assuris website. Available at https://assuris.ca/how-am-i-protected/

Balluck, Kushal, Artus Galiay, Gerardo Ferrara, and Glenn Hoggarth. 2016. “The Small Bank Failures of the Early 1990s: Another Story of Boom and Bust.” Bank of England Quarterly Bulletin 2016. Available at https://www.bankofengland.co.uk/-/media/boe/files/quarterly-bulletin/2016/the-small-bank-failures-of-the-early-1990s-another-story-of-boom-and-bust.pdf

Bank of Canada. “Framework for Market Operations and Liquidity Provision.” Bank of Canada website. Available at https://www.bankofcanada.ca/markets/market-operations-liquidity-provision/framework-market-operations-liquidity-provision/

BMO Financial Group. 2023. Annual Report 2023. Available at https://www.bmo.com/ir/archive/en/bmo_ar2023.pdf

Basel Committee on Banking Supervision. 2013. “Basel III: The Liquidity Coverage Ratio and Liquidity Risk Monitoring Tools.” Available at https://www.bis.org/publ/bcbs238.pdf

______________. 2014. “Liquidity Coverage Ratio Disclosure Standards.” (Rev. March 2014). Available at https://www.bis.org/publ/bcbs272.pdf

Benes, Jaromir, and Michael Kumhof. 2012. “The Chicago Plan Revisited.” IMF Working Paper WP/12/202. Available at https://www.imf.org/en/Publications/WP/Issues/2016/12/31/The-Chicago-Plan-Revisited-26178

Bouey, Gerald. 1982. “Monetary Policy – Finding a Firmer Place to Stand.” The 1982 Per Jacobsson Lecture.

Bourque, Paul, and Gherardo Caracciolo. 2024. A Roadmap for Better Financial Regulation. Commentary 664. Toronto: C.D. Howe Institute. July.

Canada Deposit Insurance Corporation. 2023. Annual Report. Available at https://www.cdic.ca/wp-content/uploads/cdic-2023-annual-report.pdf

CBC News. 2023. “Home Capital Group Says Another $290M Pulled Out of Savings Accounts.” April 28. Available at https://www.cbc.ca/news/business/home-capital-friday-1.4089781

CIBC. 2023. Annual Report 2023. Available at https://www.cibc.com/content/dam/cibc-public-assets/about-cibc/investor-relations/pdfs/quarterly-results/2023/ar-23-en.pdf

Financial Stability Board. 2023. “Enhancing the Resilience of Non-Bank Financial Intermediation: Progress Report.” September 6. Available at https://www.fsb.org/wp-content/uploads/P060923-1.pdf

Fisher, I. 1936. “100% Money and the Public Debt.” Economic Forum, Spring Number, April-June 1936: 406–420.

Government of Canada. “Bank Recapitalization (Bail-in) Conversion Regulations.” Available at https://laws-lois.justice.gc.ca/eng/regulations/SOR-2018-57/FullText.html

Government of Canada. “Supervisory Information (Banks) Regulations and their Analogues for Federally Incorporated Trust and Mortgage Loans Companies and Insurance Companies.” Available at https://laws-lois.justice.gc.ca/eng/regulations/SOR-2001-59/page-1.html

Home Capital. 2017. “Management’s Discussion and Analysis, Q2 2017.” Available at https://www.sedarplus.ca/csa-party/records/document.html?id=8a543044be43ce070bb34d6a992dcf04b131a9ec7e2f51c162780bb8778378ba

Huertas, Thomas F. 2015. “A Resolvable Bank.” In Making Failures Feasible: How Bankruptcy Reform Can End ‘Too Big to Fail.’ Kenneth E. Scott, Thomas H. Jackson, and John B. Taylor (eds.). Hoover Institution. Available at https://www.hoover.org/sites/default/files/research/docs/makingfailurefeasible-ch6.pdf

International Monetary Fund. 2020. “United States: Financial Sector Assessment Program – Technical Note – Banking Supervision and Regulation.” Available at https://www.imf.org/en/Publications/CR/Issues/2020/08/07/United-States-Financial-Sector-Assessment-Program-Technical-Note-Banking-Supervision-and-49657

King, Mervyn. 2016. The End of Alchemy: Money, Banking and the Future of the Global Economy. W.W. Norton & Company.

Kronick, Jeremy, Duncan Munn, and Mark Zelmer. 2023. “The Big Questions Surrounding the Future of Deposit Insurance.” Intelligence Memo. Toronto: C.D. Howe Institute. May 1.

National Bank. 2023. Annual Report 2023. Available at https://www.nbc.ca/content/dam/bnc/a-propos-de-nous/relations-investisseurs/assemblee-annuelle/2024/na-annual-report-2023.pdf

Office of the Superintendent of Financial Institutions Act. Available at https://laws-lois.justice.gc.ca/eng/acts/o-2.7/index.html

Office of the Superintendent of Financial Institutions. “Interest Rate Risk Management Guideline B-12.” Available at https://www.osfi-bsif.gc.ca/en/guidance/guidance-library/interest-rate-risk-management-guideline-2019

_____________. “Liquidity Adequacy Requirements Guideline, Chapter 4 – Net Cumulative Cash Flow.” Available at https://www.osfi-bsif.gc.ca/en/guidance/guidance-library/liquidity-adequacy-requirements-lar-2023-chapter-4-net-cumulative-cash-flow#toc-id-0

Royal Bank of Canada. 2023. Annual Report 2023. Available at https://www.rbc.com/investor-relations/_assets-custom/pdf/ar_2023_e.pdf

Scotiabank. 2023. Annual Report 2023. Available at https://www.scotiabank.com/content/dam/scotiabank/corporate/quarterly-reports/2023/q4/Annual_Report_2023_EN.pdf

TD Bank. 2023. Annual Report 2023. Available at https://www.td.com/content/dam/tdcom/canada/about-td/pdf/quarterly-results/2023/2023-annual-report-e.pdf

Tobias, Adrian, and Marc Dobler. 2024. “More Work is Needed to Make Big Banks Resolvable: Last Year's Turmoil Showed That Further Progress is Required in a Number of Areas to Ensure Banks Aren’t Too Big to Fail.” March 18. Available at https://www.imf.org/en/Blogs/Articles/2024/03/18/more-work-is-needed-to-make-big-banks-resolvable

Adrian, Tobias, Marina Moretti, Ana Carvalho, Hee Kyong Chon, Katharine Seal, Fabiana Melo, and Jay Surti. 2023. “Good Supervision: Lessons from the Field.” International Monetary Fund Working Paper WP 23/18, September. Available at www.imf.org

Tucker, Paul. 2018. Unelected Power: The Quest for Legitimacy in Central Banking and the Regulatory State. Princeton University Press.

Zelmer, Mark. 2014. “Preventing Financial Institution Failures in Canada: Importance of Corporate Governance.” Remarks by OSFI Deputy Superintendent Mark Zelmer to the Risk Oversight Program of the Global Risk Institute, Toronto, September 29.

__________. 2024. “Let’s Not Rush New Bank Rules on Capital and Lending.” Intelligence Memo. Toronto. C.D. Howe Institute. June 13.

Zelmer, Mark, and Jeremy Kronick. 2018. International Prudential Standards in a World of Growing Nationalism and Protectionism. Commentary 517. Toronto. C.D. Howe Institute.

Membership Application

Interested in becoming a Member of the C.D. Howe Institute? Please fill out the application form below and our team will be in touch with next steps. Note that Membership is subject to approval.

"*" indicates required fields

Please include a brief description, including why you’d like to become a Member.

Member Login

Not a Member yet? Visit our Membership page to learn more and apply.