Matthew Willison

Funding liquidity risk in a quantitative model of systemic stability

We demonstrate how the introduction of liability-side feedbacks affects the properties of a quantitative model of systemic risk. The model is known as RAMSI and is still in its development phase. It is based on detailed balance sheets for UK banks and encompasses macro-credit risk, interest and non-interest income risk, network interactions, and feedback effects. Funding liquidity risk is introduced by allowing for rating downgrades and incorporating a simple framework in which concerns over solvency, funding profiles and confidence may trigger the outright closure of funding markets to particular institutions. In presenting results, we focus on aggregate distributions and analysis of a scenario in which large losses at some banks can be exacerbated by liability-side feedbacks, leading to system-wide instability.

A Network Model of Financial System Resilience

We examine the role of macroeconomic fluctuations, asset market liquidity, and network structure in determining contagion and aggregate losses in a stylised financial system. Systemic instability is explored in a financial network comprising three distinct, but interconnected, sets of agents - domestic banks, overseas banks, and firms. Calibrating the model to advanced country banking sector data, this preliminary model generates broadly sensible aggregate loss distributions which are bimodal in nature. We demonstrate how systemic crises may occur and analyse how our results are influenced by fire-sale externalities and the feedback effects from curtailed lending in the macroeconomy. We also illustrate the resilience of our model financial system to stress scenarios with sharply rising corporate default rates and falling asset prices.

Real-Time Gross Settlement and Hybrid Payment Systems: A Comparison

This paper contrasts Real-Time Gross Settlement and hybrid payment systems that are based on payment offset, using a two-period, multi-bank model. The comparison is performed according to two criteria: liquidity needs and speed of settlement. We assume that the existence of a payment is common knowledge but that the specific degree of time-criticality of a payment is the private information of the bank sending the payment. Hybrid payment systems are shown to outperform Real-Time Gross Settlement when payments are offset in the first period and when they are offset in both periods. This suggests that in a hybrid system, the offsetting facility should be in operation all day, or, at the very least, for some time after the system opens in the morning. A system in which the offsetting facility was only switched on late in the day would not necessarily be preferred to Real-Time Gross Settlement. These results are shown to be robust to changes in the transparency of the central queue of payments awaiting offset. However, this robustness may not hold with different forms of information asymmetry.

Systemic Capital Requirements

The credit risk that an individual bank poses to the rest of the financial system depends on its size, the type of exposures it has to the real economy, and its obligations to other institutions. This paper describes a system-wide risk management approach to calibrating individual banks’ capital requirements that takes into account these factors and which correspond to a policymaker’s chosen target for systemic credit risk. The optimisation strategy identifies the minimum level of aggregate capital for the system and its distribution across banks that are consistent with a chosen objective for systemic credit risk. This parameterises a trade-off between efficiency and stability.

Modelling the Cross-Border Use of Collateral in Payment Systems

Banks often rely on collateralised intraday liquidity from the central bank in order to be able to effect payments in a real-time gross settlement (RTGS) payment system. If a bank is holding insufficient eligible collateral in a particular country, and therefore cannot obtain credit from the local central bank, it may have to delay payments. This constitutes a liquidity risk to the system. Furthermore, a bank operating in multiple systems may face a mismatch between the location of its collateral holdings and liquidity needs. In this paper, we examine the extent to which the liquidity risk arising from such a mismatch may be mitigated by allowing cross-border use of collateral. We develop a two-country, two-bank model in which risk-neutral banks minimise expected costs with respect to their collateral choice in each country. In our baseline model, in which each bank faces a liquidity need in only one country, we find that liquidity risk is indeed reduced by cross-border use of collateral. This result holds despite the fact that banks may find it optimal to economise on their total holdings of collateral. However, when we extend the model to allow for the possibility that a bank faces liquidity needs in both countries simultaneously, the total quantum of collateral held is important. Indeed, when a bank finds it optimal to reduce its total holdings, there may be an increase in liquidity risk in at least one country when simultaneous liquidity demands are realised.

Using Shapley's Asymmetric Power Index to Measure Banks' Contributions to Systemic Risk

An individual bank can put the whole banking system at risk if its losses in response to shocks push losses for the system as a whole above a critical threshold. We determine the contribution of banks to this systemic risk using a generalisation of the Shapley value; a concept originating in co-operative game theory. An important feature of this approach is that the order in which banks fail in response to a shock depends on the composition of the banks’ asset portfolios and capital buffers. We show how these factors affect banks’ contributions to systemic risk, and the extent to which these contributions depend on the level of the critical threshold.

The welfare benefits of stable and efficient payment systems

The Bank of Englands second core purpose is to maintain the stability of the financial system. Payment systems, by supporting transactions, are a key aspect of this. In this paper, we examine the importance of smoothly functioning payment systems to the economy by extending a recently developed theoretical model of banks. In the model the risk of theft implies a cost to using cash. This risk can be avoided by depositing cash in banks and transferring money through an interbank payment system. However, agents are then exposed to the risk that the payment system is unreliable. Agents will use a payment system (rather than cash) to make transactions if the system is sufficiently cheap to use and/or it is sufficiently reliable. We show that the introduction of a payment system that buyers and producers choose to use unambiguously increases social welfare if it expands the number of trades occurring in the economy. This is more likely the more reliable is the payment system. When the introduction of a payment system does not increase the number of trades, social welfare may increase or decrease depending on the trade-off between the risk of using cash and the risk that the payment system is unreliable. We again show that the more reliable is the payment system, the more likely welfare is increased by its introduction and we illustrate how this benefit might be quantified.