What liquidity risk means in real-time payments

Liquidity risk is the danger that a payment provider won't have enough money on hand to settle a transaction right now, even though the money will arrive later. In real-time payment systems like FedNow (launched by the Federal Reserve in 2023) or The Clearing House's RTP network, banks and payment companies promise to move money instantly or within seconds. That speed creates a timing problem: a provider might send out $10 million in payments before it receives the incoming funds to cover them.

If a provider runs short, it can't complete transactions, customers get rejected payments, and trust in the system breaks. Payment providers solve this by managing how much money they hold in reserve, how fast they can borrow if needed, and how they spread risk across their network.

Key Takeaways

  • Payment providers hold reserve accounts at the Federal Reserve or correspondent banks so they have cash ready to send payments before customer funds arrive.
  • Intraday credit lines let providers borrow money during the day to cover the gap between outgoing and incoming payments, then repay by end of business.
  • Providers use real-time monitoring to watch cash flows second by second and adjust how many payments they process if reserves drop too low.
  • Payment networks spread transactions across many participants so no single provider carries all the risk of a large payment failing.
  • Collateral requirements and fee structures discourage providers from taking on more payment volume than they can safely handle.

Reserve accounts and prefunding

The simplest way a payment provider reduces liquidity risk is by keeping money in a reserve account. Banks and payment companies maintain accounts at the Federal Reserve or at larger correspondent banks. These accounts hold cash that the provider owns or has borrowed specifically to cover real-time payments.

When a customer initiates a payment, the provider draws from this reserve to send the money immediately. As incoming payments arrive from other customers or other banks, those funds replenish the reserve. The provider's job is to keep the reserve large enough that it never runs dry between inflows.

The size of the reserve depends on the provider's transaction volume and the pattern of money flowing in and out. A provider that sends more payments than it receives in a given hour needs a larger reserve than one with balanced flows. Providers calculate their reserve needs based on historical data and stress tests—scenarios where outflows spike or inflows slow down.

Intraday credit and borrowing arrangements

Reserves alone are not always enough. A provider might face an unexpected surge in outgoing payments or a delay in receiving funds. To bridge that gap without rejecting payments, providers use intraday credit—short-term borrowing that lasts only hours, not days.

The Federal Reserve offers intraday credit to banks that participate in FedNow and other Fed payment systems. A bank can borrow during the day to cover a temporary shortfall, then repay the loan by the end of the business day when all settlements are complete. The Fed charges a small fee for this service, typically a fraction of a percent of the borrowed amount.

Private payment networks and correspondent banks also offer intraday credit arrangements. A smaller bank might have a credit line with a larger correspondent bank, allowing it to borrow intraday funds at an agreed rate. These arrangements are negotiated based on the smaller bank's creditworthiness and transaction history.

Real-time monitoring and flow controls

Payment providers now use software that watches their cash position second by second. These systems track every outgoing payment, every incoming deposit, and the current reserve balance in real time. If the reserve drops below a safety threshold, the system can automatically slow down or temporarily hold outgoing payments until more funds arrive.

This is not the same as rejecting a payment permanently. The system queues the payment and releases it as soon as reserves recover. For the customer, the delay might be seconds or minutes, depending on how quickly incoming funds arrive. For the provider, this buffer prevents the reserve from hitting zero.

Some providers also use flow controls—rules that limit how many payments can be sent in a given time window. For example, a provider might allow no more than $50 million in outgoing payments per minute. This spreads the drain on reserves and gives incoming payments time to arrive and replenish the account.

Collateral and capital requirements

Regulators and payment networks require providers to hold collateral—assets that back up their ability to cover losses. For a bank, this might be Treasury securities, cash, or other highly liquid assets held in a segregated account. If a provider fails to settle a payment, the collateral can be used to cover the loss.

Collateral requirements force providers to think carefully about how much payment volume they can handle. A provider with $100 million in collateral cannot safely process $1 billion in daily payments, because the collateral would not cover a major failure. This natural limit discourages providers from overextending.

Capital requirements work similarly. Regulators require banks to hold a minimum amount of capital (shareholder equity) relative to their assets and liabilities. A bank that processes large payment volumes must hold more capital to absorb potential losses. This makes it expensive to take on too much payment risk.

Network design and risk distribution

Real-time payment networks are designed so that no single provider bears all the risk of a large transaction. In FedNow and RTP, payments move through multiple banks and intermediaries. If one provider in the chain runs short, the network has rules to pause the transaction and alert other participants, rather than letting one provider's shortfall cascade into failures across the system.

Payment networks also use multilateral netting—a process where all transactions between participants are summed up at the end of the day, and only the net difference is settled. For example, if Bank A sends $10 million to Bank B and Bank B sends $8 million to Bank A, only the $2 million difference needs to settle. This reduces the total amount of money that has to move through the system and lowers liquidity risk for all participants.

Some networks use a central settlement account where all participants deposit funds. The network then distributes those funds to cover all outgoing payments. This pooling spreads the risk: if one participant's incoming payments are delayed, the pool can still cover its outgoing payments using funds from other participants.

Fee structures and incentives

Payment providers charge fees for real-time payments, and those fees are partly designed to cover liquidity risk. A provider that processes high-volume real-time payments charges higher fees than one processing low volume, because the high-volume provider faces greater liquidity risk and must hold larger reserves and collateral.

Fees also create an incentive for providers to manage their liquidity carefully. If a provider's liquidity risk is high, it pays more in fees to the network and more in interest on borrowed funds. This cost pressure encourages the provider to balance its incoming and outgoing payments, reduce unnecessary transactions, or negotiate better terms with customers.

Some networks charge variable fees based on the time of day. Payments sent during peak hours (when liquidity risk is highest) cost more than payments sent during off-peak hours. This incentivizes customers to spread their payments throughout the day, which smooths out the flow of money and reduces the peak liquidity demand on providers.

Frequently Asked Questions

What happens if a payment provider runs out of money mid-transaction?

The payment is queued and held until the provider's reserves recover. The customer sees a delay of seconds to minutes, not a permanent failure. If the delay extends beyond a set time (usually 30 seconds to a few minutes), the payment may be rejected and the customer is notified. The money is not lost—it remains in the customer's account.

Can the Federal Reserve run out of money to lend for intraday credit?

No. The Federal Reserve can create money and has no liquidity limit. However, the Fed does charge fees for intraday credit and can adjust those fees to discourage excessive borrowing. The real constraint is cost, not availability.

Why do real-time payments cost more than regular bank transfers?

Real-time payments require providers to hold larger reserves and collateral, and to monitor their cash position constantly. These costs are passed to customers through higher fees. A regular ACH transfer takes one to two days, so providers can use that time to manage their liquidity. Real-time payments demand immediate settlement, which is more expensive to operate.

Do payment providers share liquidity with each other?

Yes, through correspondent banking relationships and payment network arrangements. A smaller bank can borrow intraday funds from a larger correspondent bank. Payment networks also use pooled settlement accounts where multiple providers deposit funds that are then distributed to cover all outgoing payments.

What is the difference between liquidity risk and credit risk?

Liquidity risk is the risk of not having cash on hand right now, even though money is coming later. Credit risk is the risk that money never comes—that a customer or counterparty defaults. Providers manage liquidity risk through reserves and borrowing. They manage credit risk through underwriting and collateral requirements.