Interbank lending can help banks cope with unexpected withdrawals by allowing connected institutions to share liquidity when it is needed. But these connections can also introduce vulnerabilities. If banks reduce their own reserves because they expect to rely on their partners, financial stress at one institution may spread across the wider network.
In a new study published in Risk Sciences, researchers developed a theoretical model to explore why banks form interbank credit networks and how the size of those networks affects market efficiency. Their findings suggest that bigger networks do not necessarily produce better outcomes, because the advantages of sharing risk can eventually be outweighed by strategic behaviour among participating banks.
The model considers two closely related decisions made by banks. First, each bank determines how much money to keep in reserve, balancing the potential profits from lending more money against the need to withstand unexpected liquidity shocks. Second, banks decide whether participating in an interbank network would leave them better off than operating independently.
The researchers found that membership in a network creates both cooperation and competition. Banks benefit from being able to share liquidity risk with their partners, which can provide protection when unexpected withdrawals occur. At the same time, individual banks may have an incentive to reduce their own reserves and depend more heavily on the liquidity held by other institutions in the network.
The researchers describe this strategic behaviour as a “free-riding” effect. Although each bank can benefit individually from holding fewer reserves and putting more funds into potentially profitable lending, widespread free-riding can weaken the network as a whole. Lower reserves can reduce banks’ ability to survive liquidity shocks and may ultimately decrease their expected profits.
Network size therefore plays an important role. In smaller interbank networks, the advantages of sharing liquidity risk tend to outweigh the negative effects of free-riding. However, as additional banks join, free-riding becomes increasingly significant. The researchers found a rise-and-fall relationship between expected profits and network size, suggesting that relatively small networks can sometimes be Pareto optimal—where no participating bank can be made better off without making another worse off.
The study also considers networks containing banks of different sizes. Under certain conditions, smaller and larger institutions may have incentives to establish connections even when their deposit sizes differ considerably. The findings offer a theoretical explanation for core-periphery structures commonly observed in banking systems, where a relatively small number of highly connected institutions interact with a much larger group of smaller banks.
The researchers also considered the implications for financial regulation. Implicit government guarantees may encourage institutions to take greater risks and become excessively interconnected because they expect support during periods of financial distress. The findings suggest that appropriately designed capital requirements could help counteract free-riding, discourage excessive interconnectedness, and improve market efficiency in larger banking networks. Overall, the study highlights an important trade-off: interbank connections can strengthen financial institutions through risk-sharing, but expanding those networks too far may create incentives that undermine the very benefits they are intended to provide.
More information: Tongkui Yu et al, Interbank network and market efficiency, Risk Sciences. DOI: 10.1016/j.risk.2026.100059
Journal information: Risk Sciences Provided by KeAi Communications Co., Ltd.