Introduction
Good afternoon. It’s a pleasure to be in New York among so many familiar faces from across the financial community.
I am looking forward to a great discussion today. Before we get to the conversation, I’d like to take a few minutes to update you on two areas of work at the Bank of Canada that will be of particular interest to this audience.
First, I’ll talk about what we’re doing to support resilience and efficiency in repurchase agreement (repo) markets, specifically by improving market infrastructure.
Second, I’ll explain how our monetary policy implementation framework has evolved since quantitative tightening, or QT, ended last year. To maintain effective control over our policy interest rate, we’ve made some modest adjustments to our framework to better reflect evolving dynamics in repo markets.
Strengthening repo-market infrastructure
As this audience knows well, repo markets are a critical part of the financial system. They support liquidity in government bond markets by allowing participants to borrow or lend cash or securities. And they play a key role in transmitting monetary policy.
This central role in the financial system means stress in repo markets could spread quickly.
Governments around the world are issuing a lot of sovereign debt these days, including to fund infrastructure and defence spending. And some of the biggest buyers of this debt are hedge funds and other investors that use leverage—including money borrowed in repo markets—to fund their purchases.1
The growing role of leveraged investors in repo and government bond markets has improved liquidity in these markets. But their reliance on leverage can become a vulnerability in times of stress.2 If repo funding suddenly became constrained, leveraged investors could be forced to sell their bond holdings, which, in the extreme, could lead to a liquidity spiral.
This is why governments, central banks and market participants are focused on ensuring that repo markets are resilient to stress.
A key part of that effort is strengthening repo-market infrastructure. And the Bank has been working closely with industry and other stakeholders for years to advance that goal.
In March, we announced plans to join a tri-party platform operated by the TMX—the Canadian Collateral Management Service (CCMS). Today I can confirm the Bank will start using this infrastructure for our repo operations in the first quarter of 2027.
Participants in the CCMS essentially outsource the post-trade management of their repo transactions. They also gain much greater flexibility in how they manage collateral, which supports market liquidity. The tri-party platform will help further develop the term repo market and attract new participants looking to park cash, such as large corporate treasuries.
For the Bank, using the CCMS will make settling our operations more efficient, making it easier to scale up these operations when needed.
The Bank and industry are also working to update and enhance Canada’s repo clearing infrastructure. The aim is for most repo activity involving major Canadian buy- and sell-side participants to be centrally cleared, in line with what is happening in other major markets.3
Central clearing infrastructure and services in Canada are being modernized, a process that is expected to be largely complete in 2028. Once everything is ready, we plan to centrally clear our own repo operations through the Canadian Derivatives Clearing Corporation.4
Generally, central clearing can reduce counterparty credit risk by guaranteeing settlement. In the case of repo, central clearing also frees up capacity on dealers’ balance sheets by allowing offsetting repo transactions to be netted. This gives dealers more room to absorb liquidity shocks and continue providing repo funding when markets come under stress.
Once we move to central clearing, these netting benefits will reduce the balance sheet costs our counterparties incur when using our repo operations, particularly at the end of a quarter or year. That should make our operations more attractive to market participants and, therefore, even more effective as tools for implementing monetary policy.
Taken together, these improvements should make repo markets more resilient while ensuring that we can implement monetary policy as intended.
Updating the Bank’s implementation framework
With that, let me turn to how the Bank’s implementation framework has evolved.
Before I go further, an important feature to know is that we target a secured rate: the overnight repo rate. When we set our policy interest rate, it’s really our target for the rate that we want banks to charge each other for overnight funding.
Targeting a secured rate makes us different from many other central banks that target an unsecured rate.5 And it means that repo markets play a big role in how we implement monetary policy. Funding pressures in these markets—and how we respond to those pressures—can affect how well our policy interest rate transmits to other interest rates across the economy. So we pay close attention to imbalances between funding supply and demand in repo markets, not just to the level of our reserves, which we call settlement balances in Canada.6
A key way that we monitor conditions in repo markets is through CORRA, the Canadian Overnight Repo Rate Average.7 When funding pressures push CORRA away from our policy rate, we have tools to limit such deviations and guide CORRA back toward the target.
Since QT ended, CORRA has often been modestly above target, prompting market participants to ask how we decide when to use our tools. So let me clarify.
Overnight repo operations, ORs, are our main tool for responding to upward pressure on CORRA. They provide liquidity on an ad hoc basis.
Offering ORs on a discretionary basis means we tolerate some deviations in CORRA. That’s deliberate—we want to maintain incentives for market participants to trade and manage their day-to-day liquidity needs. We also don’t want to react to every small deviation because they can provide helpful market signals.
Don’t get me wrong: we tolerate only so much upward pressure on CORRA. Large deviations can disrupt the transmission of our monetary policy. ORs are there to help us ensure CORRA doesn’t stray too far from target.
But before using ORs to add funds to the overnight repo market—and implying we see excessive pressure on CORRA—we carefully assess the size, breadth, persistence and source of the pressure, along with overall conditions in the market.
In October 2025, we made an important adjustment to our toolkit, adding two-week term repo operations.
These weekly operations have become an effective complement to our ORs. We’ve adjusted how we use them so they can do more than simply respond to pressures. We now regularly increase the size of our two-week repo operations to get ahead of expected or seasonal pressures around quarter-end reporting dates for Canadian banks or large government bond maturities. Supplying extra term cash in the repo market in this way helps reduce the likelihood of upward CORRA pressure.8
Using two-week term repos more actively can impact our reserves. It could mean they sometimes exceed our best estimate of steady-state demand for reserves, which remains $50 billion to $70 billion.
But keeping the supply of reserves within this range is not an objective in itself. What matters most is deploying our operations as needed to control our policy rate when imbalances in the repo market shift and push repo rates up or down, even if doing so temporarily pushes reserves outside the range.
We also get a lot of questions about our deposit rate, which reinforces the floor of our operating band by paying interest on reserves held at the Bank. It’s currently set at 5 basis points below our policy rate, a change we made in January 2025 to encourage institutions to redistribute reserves rather than hold on to them.
We only change where we set the deposit rate when persistent or structural pressures warrant it. Examples of these pressures include long-lasting shifts in repo market imbalances, or structural shifts in the demand for reserves across banks.
Finally, I’ll say a word about our Standing Liquidity Facility, or SLF for short. The SLF plays a supporting role by providing a ceiling for overnight rates.9
We know that financial institutions are hesitant to use the SLF for overnight liquidity because they worry that using it could send the wrong signal. They shouldn’t worry. It’s a routine facility. Any member of Lynx, Canada’s high-value wholesale payment system, should feel comfortable using the SLF for unexpected shortfalls in end-of-day liquidity instead of holding extra reserves just to avoid SLF draws.
Today, we issued a joint statement with the Office of the Superintendent of Financial Institutions to make it clear that we view an institution’s use of the SLF for overnight liquidity as a normal part of day-to-day liquidity management, not as an exceptional event or a sign of stress.
I invite you to read our new background paper on monetary policy implementation for more details.10 Our priority is to ensure our framework remains effective as market conditions evolve—and to be transparent about any adjustments we make.
Thank you. I’d be happy to take some questions.
I would like to thank Philippe Muller and Kaetlynd McRae for their help in preparing this speech.