Three Ways a Central Bank Opens the Liquidity Channel
Central bank lending facilities all move liquidity through a balance sheet, but they do not all become available in the same way. This chapter focuses on the trigger and initiator of the operation—whether a channel is standing, discretionary, or a repo—and gives the reader a way to classify an unfamiliar facility by its timing and control rather than by its name.
The trigger is part of the facility
A useful starting point is to separate two questions that are often merged. When is the channel available? And who initiates the transaction? Those questions produce three operating forms in the BIS classification: standing lending facilities, discretionary lending facilities, and repurchase agreements used as open market operations.
A standing lending facility is available on any business day. The counterparty initiates the drawing. That does not mean an institution borrows continuously, or that the central bank has promised to meet every request on any terms. It means the channel exists before a particular liquidity need appears, subject to the facility's eligibility, collateral, and other conditions. A bank can therefore treat the facility as an option in its liquidity planning rather than wait for the central bank to announce a new program.
The sequence is straightforward. An eligible counterparty identifies a need for reserves or short-term funding. It requests a loan from the central bank. The central bank applies the facility's rules to the borrower and the collateral, advances reserves for the agreed term, and receives repayment later. The central bank has made the channel available in advance; the borrower decides whether and when to use it. That allocation of initiative is economically meaningful. It lets the facility respond to changing demand without requiring the central bank to predict the exact day or institution that will need liquidity.
The Federal Reserve discount window is a concrete example of this standing-channel logic. Its current overview says that depository institutions and U.S. branches and agencies of foreign banks must execute legal agreements and pledge collateral before borrowing. Once a generally sound institution has been set up for primary credit, the borrower normally supplies the amount and term requested. The rate and maturity still matter, and the lending Reserve Bank must be satisfied with the collateral, but the institution—not a new public announcement—starts the request.
A discretionary lending facility begins differently. The central bank first activates the channel for a defined period, typically in response to market stress. Only after that activation do counterparties initiate drawings. The individual transaction may still look like a counterparty request for reserves against collateral, but the availability condition has changed: the central bank has decided that a particular tool should be open now.
This two-stage structure gives the central bank a way to respond to a type of stress without keeping every specialized facility permanently active. Activation can define the circumstances, period, or market segment for which the facility is available. Once the channel is open, the counterparty can draw under the announced rules. The central bank retains control over whether the tool exists at that moment, while the counterparty decides whether its own liquidity need warrants a drawing under those rules.
That distinction also explains why “emergency facility” is not a complete mechanical description. A discretionary facility can be used in a crisis, but the defining feature in this classification is not the emotional intensity of the event. It is the sequence: central-bank activation first, counterparty drawing second. A standing facility may be used during stress, and a discretionary facility may be designed with terms that make its transactions resemble ordinary secured lending. Availability is the key variable.
Repurchase agreements, or repos, shift the initiative again. The BIS treats them as a type of open market operation in which the central bank provides reserves against collateral, with the central bank initiating the recurring or ad hoc operation. The counterparties still supply collateral and receive reserves, but the transaction begins as a central-bank decision to conduct an operation rather than as a borrower deciding to exercise an always-available loan option.
A repo can therefore reach the market through a different route from a standing facility. In a standing facility, the central bank has built a channel and waits for an eligible counterparty to request a draw. In a repo operation, the central bank decides to operate and invites or transacts with eligible market participants under the operation's terms. Both arrangements can adjust reserves and both are collateralized, but the control point is different. One is primarily demand-initiated; the other is supply-initiated.
The distinction is useful even when real-world designs are less tidy. A central bank may use several operations at once. A facility may have recurring windows, special activation conditions, or multiple functions. The classification is not a claim that every operation fits one pure box. It is a way to locate the first causal event: an option already available, a central-bank activation, or a central-bank-initiated market operation.
The diagram shows why the three forms should not be confused merely because each can put reserves into the financial system. A standing facility's first event is the counterparty's request. A discretionary facility's first event is the central bank's activation. A repo's first event is the central bank's decision to conduct the operation. The later exchange of collateral and reserves does not erase that difference.
One channel can serve different policy purposes
The trigger tells us how the operation begins. Its policy function tells us why it exists. The BIS groups lending facilities and open market operations into three broad purposes: monetary-policy implementation and transmission, financial stability, and payment-system support. These are not three names for the same objective.
For monetary-policy implementation, an operation helps control short-term interest rates, manage reserves, and transmit a policy decision through market conditions. A standing facility can act as an alternative source of funding when the market rate is higher than the facility's rate, limiting the extent to which trading rates move above the intended policy level. A repo operation can add reserves through the market on a recurring or ad hoc basis. In both cases, the operation is part of the framework used to keep short-term conditions aligned with the policy stance.
For financial stability, the immediate problem is a liquidity pressure that can spread through markets or institutions. A facility may help stabilize market conditions and reduce the chance that a temporary funding shock becomes a broader disruption. The operation still needs a counterparty, collateral, and terms; the difference is the problem those arrangements are intended to address. The central bank is not only steering an overnight rate. It is trying to keep a liquidity shock from forcing damaging sales or interrupting credit intermediation.
For payment-system support, the timing problem is especially direct. Participants may need central-bank liquidity during the day to settle payments even when their broader balance sheets are not in distress. The BIS notes that payment-system tools address timing mismatches and help avoid disruptions. A very short-lived operation can therefore support the plumbing of settlement without carrying the same purpose as a term facility intended to stabilize a market or signal a policy rate.
These functions can coexist. The BIS notes that a single tool can have multiple functions, and most jurisdictions allocate at least one tool to monetary-policy implementation. The important point is not to force a facility into one label when the evidence says it serves more than one purpose. It is to ask which purpose is dominant in the design and what that purpose implies for access, collateral, price, and maturity.
Consider the same secured flow of reserves viewed through the three lenses. If a bank can borrow when the market rate rises above a standing facility's rate, the channel can limit unusually high short-term rates. If a market-wide liquidity shock threatens fire sales, broader access or collateral may be justified by the stability objective, though it increases risk and operating demands. If payment participants need intraday funding to complete settlement, the maturity and price may be designed around that narrow timing need. The balance-sheet movement is similar; the policy problem is not.
This is why the trigger and function must be read together. A standing facility for policy implementation is not equivalent to a discretionary facility opened during a market disruption, even if both accept collateral from banks. A repo initiated by the central bank may support routine rate control, while a discretionary lending channel may be reserved for a particular stability problem. Names such as “liquidity facility” conceal more than they reveal unless the reader asks who opens it, who draws, and what the operation is trying to accomplish.
Why some systems use many tools and others use fewer
The number of facilities is another design choice, not a scorecard. In the selected jurisdictions summarized by the BIS, the number of liquidity tools ranges from one to nine, with an average of four. Some central banks use a relatively large number of instruments with specific purposes. Others use a smaller set of tools that serve several objectives.
More tools can create useful separation. A central bank can tailor eligibility to the institutions most relevant to a market, accept different collateral for different risks, set distinct prices, or offer different tenors. That specialization can make the policy objective clearer and prevent a facility designed for one problem from automatically carrying the terms of another. It can also allow a central bank to activate a narrow response instead of changing the conditions of every liquidity channel.
The cost is not only administrative inconvenience. Each additional tool must be understood and operated by the central bank and its counterparties. More eligibility rules, collateral schedules, settlement arrangements, and pricing formulas create more interfaces where a mistake or delay can occur. A complex framework can become opaque to less sophisticated participants, even if every rule is publicly available. The BIS also notes that complexity may disproportionately benefit participants with greater resources and expertise.
A smaller toolkit has a different advantage. The central bank may adjust the modalities of a few facilities—such as their pricing, collateral, or tenor—depending on circumstances. Fewer channels can be easier for counterparties to understand and for the central bank to operate consistently. But consolidation can also blur policy functions or force one facility to carry risks and eligibility choices that would be separated in a more specialized design.
There is no evidence here that one architecture is universally better. The BIS attributes the choice to domestic legal and market structures, experience with past stress, and the central bank's assessment of effectiveness and cost. A system with a large securities market, a prominent non-bank sector, or specialized payment infrastructures may face different design needs from a system where banks dominate intermediation. The relevant judgment is conditional: does the number and type of tools let the central bank reach the needed counterparties, against manageable collateral, at usable terms, without creating unnecessary complexity?
That question brings the three operating forms back together. A standing facility solves one problem by making a channel continuously available for counterparty-initiated use. A discretionary facility preserves central-bank control over when a specialized channel opens. A repo lets the central bank initiate a market operation to adjust reserves or conditions. The choice among them is not a matter of terminology. It is a decision about timing, information, control, and the policy function the liquidity is meant to serve.
The next design question follows naturally. Once a central bank has chosen how a channel opens, it must decide who is allowed to use it. That access rule determines how far liquidity can travel through the financial system—and what credit and market risks travel back to the central bank.