To most people, all faster payment systems seem pretty similar since they all make it possible for payers and payees to see their transactions reflected in their accounts in near real time.1 What’s more, they all enable payees to use the funds they receive immediately. For example, as a payer or payee, you might have used, or in the future might be using, a number of these bank-centric services2 somewhat interchangeably: Visa’s Visa Direct, Mastercard’s Mastercard Send™, Early Warning Service’s Zelle®, The Clearing House’s RTP® network, and/or the Federal Reserve Banks’ FedNowSM Service.3,4
However, there are important differences in the way these systems transfer funds among participating financial institutions. This article provides an overview of clearing and settlement – two critical payment system functions that facilitate interbank funds transfers – and then discusses some of the risk implications of these differences for the various types of faster payment systems (including instant payment systems).
Clearing starts with financial institutions sending payment messages through the payment network; the payment network routes these messages and other related information for the participating financial institutions to correctly process payment instructions. In faster payment systems, these messages flow on a transaction-by-transaction basis, and by agreement among the participants, the clearing process enables faster payments to be credited to payees’ accounts in near real time.
In clearing, network operators validate financial institutions’ routing information (for example, “Routing Transit Numbers (Off-site),” or RTNs) to ensure the information is correct and that the receiving financial institution is a participant in that network. Clearing may also involve the validation of other payment details, including whether a given individual’s identifying information, such as an account number, mobile number or email address, is associated with a participating financial institution. In addition, during clearing, the participating financial institutions validate and take action on the payment messages they receive, including by rejecting or accepting the messages.
These actions will trigger obligations among the participating financial institutions, and between them and their customers. As an example, when a payee’s financial institution accepts a payment order from the payer’s financial institution, the payer’s financial institution is obliged to pay the payee’s financial institution for the amount of the transaction. In turn, the payee’s financial institution is obliged to pay its payee customer. As between the two financial institutions and their respective customers, they will typically settle by debiting and crediting their respective customers’ accounts they maintain. But the financial institutions involved in the transaction will also need to settle among themselves.
The second key function is the interbank settlement process, which results in the discharge of the financial obligations that arise as a result of the clearing process between payers’ and payees’ financial institutions. One way that interbank settlement may take place is through debits and credits to accounts they maintain with a common correspondent bank. This is essentially how the Reserve Banks effect settlement: they maintain settlement accounts, referred to as “master accounts,” for financial institutions that participate in their payments systems.
Settlement generally falls into two broad categories. Real-time settlement is when settlement between participants in the payment system occurs more or less concurrently with clearing the payment message(s). In contrast, deferred settlement generally occurs at the end of a predefined settlement cycle or at an agreed point in time (e.g., at the end of the business day), after the payment message(s) have been cleared. Deferred settlement systems often involve netting, which offsets payment obligations between or among participating financial institutions, generally reducing the amount of funds required to settle.
In the United States, real-time and deferred settlement tend to be associated with whether the network settles on a transaction-by-transaction (called gross settlement) or on a net basis. In a netting5 arrangement, the network tallies all the payments that a given financial institution should receive from or owe to other financial institutions in the network. These tallies determine an aggregate net amount that each financial institution owes or should receive at a settlement time.6
We can now put all of this together. In real-time gross settlement, a payer’s financial institution pays the payee’s financial institution at the time and for the amount of each of its customers’ individual transactions. Among faster payment systems, all of which require participating financial institutions to make funds available to payees in near real time on a 24x7x365 basis, only the RTP Network and the FedNow Service operate (or will operate (Off-site)) with real-time gross settlement. The Federal Reserve uses the term “instant payment system” to highlight this distinction vis-a-vis other types of faster payment systems. By contrast, in deferred net settlement, participating financial institutions settle their net obligations to one another periodically.
Network operators design their systems to create efficiencies and reduce risk for their participants, and network design will have implications for the risks involved in the settlement process. In particular, real-time gross settlement and deferred net settlement introduce tradeoffs in how they handle liquidity risk and credit risk.
Let’s start with liquidity risk, or the risk that a financial institution can’t settle at the designated time because it has insufficient funds available to it and can’t readily obtain funding from other sources. A deferred net settlement structure helps to optimize liquidity (and reduce liquidity risk), as each financial institution’s total settlement obligation is reduced by the amount owed to it by the other participants in the network. In addition, by settling only at predesignated times that are typically within the operating hours of the intraday credit markets (Off-site) and/or the Federal Reserve’s Discount Window (Off-site), financial institutions can typically access additional liquidity (if needed).
By contrast, real-time settlement requires that financial institutions ensure they maintain adequate liquidity to settle the gross value of transactions at all times. In the event a financial institution does not maintain adequate liquidity, it may be unable to settle its own transactions and create a ripple effect, or liquidity “trap,” for other participants in the network to the extent that they become unable to settle transactions as a result. To avoid these risks, certain real-time settlement systems can and do include protections to mitigate liquidity risk between participants. For instance, the Reserve Banks will provide access to intraday credit to participants of the FedNow Service during the Federal Reserve’s normal operating hours, and a liquidity-management tool to help participating financial institutions conduct interbank transfers during hours when the Federal Reserve’s normal liquidity services are not open (e.g., on weekends). These liquidity tools will not eliminate risk altogether – as an example, a Reserve Bank or other correspondent bank that provides intraday credit to a participant will take on credit risk to its account holder – but will help participants in managing risk to other participants in the system.
While deferred net settlement can simplify some liquidity issues, it retains credit risk between participants up to the point of settlement. In this case, all other financial institutions owed funds by the defaulting financial institution bear a risk of loss because faster payment system rules require the receiving financial institutions to make final payments to their customers even if they never receive payment from the paying financial institution. Credit risk between participants is reduced in real-time settlement systems because payments are settled between participants on a transaction-by-transaction basis before or concurrent with the payee’s financial institution crediting the payee’s account.
Deferred net settlement systems are able to mitigate credit risk in a variety of ways. Examples include introducing more frequent settlement windows, adding prefunding or collateral requirements to back intraday obligations, and/or monitoring each participating financial institution’s balance against its net obligations to ensure it maintains the ability to pay at settlement time.
This article described key functions involved in the transfer of funds among financial institutions participating in a faster payment network. Below is a list of the key takeaways:
1 According to the Bank for International Settlements (BIS), a faster payment is "… a payment in which the transmission of the payment message and the availability of ‘final’ funds to the payee occur in real time or near-real time on or as near to a 24-hour and seven-day (24/7) basis as possible." (Committee on Payment and Settlement Systems (2016), Fast payments: Enhancing the speed and availability of retail payments, Bank for International Settlements, November).
2 In the context of this article, the term “bank” refers to all depository institutions, such as commercial banks, thrift institutions and credit unions, and the term “interbank” is meant to include the interrelated activities among them.
3 “FedNow” is a service mark of the Federal Reserve Banks. Other service marks noted in this article belong to the organizations listed.
4 Paypal®, Venmo and Square’s Cash App are other examples, but these use nonbank, closed-loop structures that this article doesn’t address.
5 There are two forms of netting. In Bilateral Net Settlement, financial institutions settle with each other on a financial institution to financial institution basis. In Multilateral Net Settlement, the network operator provides each financial institution one sum to pay or receive based on all of its customer transactions with all of the other participating financial institutions. Multilateral net settlement is more typical and is the implied netting approach in this article.
6 Some countries use real-time net settlement, where transactions net every few seconds, essentially in real time. Some countries also use deferred gross settlement. However, these two systems are uncommon, particularly in the United States, so this article doesn’t address them.