Abstract: This paper examines systemic risk and contagion in the core global banking system. Using BIS immediate-counterparty banking statistics, we construct a Core Global Banking System (CGBS) network and apply a spectral eigen-pair approach to assess system fragility. The Spectral Systemic Risk Index (SRI) reveals the inherent structural instability and amplification capacity of the global banking system. Systemically important and vulnerable banking systems are identified, with the US and UK highlighted as key transmitters of systemic shocks. The analysis further maps structural fragility to realised systemic distress by introducing two outcome-based indicators of system-wide losses and institutional failures. The results show that realised systemic outcomes are nonlinear in amplification capacity and state-dependent under comparable structural conditions. All three measures exhibit early-warning signals ahead of major crisis episodes. Overall, the findings shift the focus of systemic risk monitoring beyond structural fragility to the final stage of contagion, where the systemic repercussions become visible.
Firm Liquidity and the Origins of Aggregate Fluctuations in a Network Economy, University of Essex, Economics Discussion Paper No. 42809 (2026). https://doi.org/10.2139/ssrn.6275359
Abstract: This paper investigates how microeconomic origins of liquidity shocks at the firm level influence aggregate output and financial stability in a network economy with inter-sectoral linkages. We use firm-level balance sheet data to construct two liquidity measures: the quick ratio, capturing firms’ internal short-term liquidity, and a network-based measure, capturing inter-sectoral network structure of liquidity flows derived from receivables and payables. Using sector-level aggregates, we apply a Structural Vector Autoregression (SVAR) model to examine the dynamic responses of GDP, the repo rates, the quick ratio, and the network measure to liquidity shocks. We further decompose forecast error variance to assess the relative contribution of each shock to business-cycle fluctuations. The main results indicate that the network effect is the dominant driver of business-cycle fluctuations, followed by the quick ratio, with both outweighing the remaining endogenous variables. The findings are relevant for macroprudential oversight, highlighting the importance of monitoring firms’ liquidity imbalances and network structure for financial stability and economic resilience.
Documentation of the UKMOD tax–benefit microsimulation model, analysing UK tax and welfare policies and their distributional effects on household incomes, poverty, and inequality (2019–2025).
Is There Long-Term Memory in Tehran Stock Market Returns? (with Farhad Ghaffari and Teimur Mohammadi), 2017, Quarterly Iranian Journal of Applied Economics. [In Farsi] Draft here.
Abstract: This study investigates the presence of long-term memory in the returns of the petrochemical industry index in the Tehran Stock Market. Using daily data from June 2005 to August 2015, we examine whether return dynamics exhibit persistent dependence structures beyond short-term autocorrelation. To this end, ARIMA and ARFIMA models are estimated and compared. The ARFIMA specification explicitly accounts for fractional integration and long-memory behaviour, whereas the ARIMA model captures only short-run dynamics. Empirical results indicate evidence of long-term memory in index returns.
Fiscal Incentives for Health Improvement: Repurposing Consumption Taxes on Food (FINCH), (with Franco Sassi, Matteo Richiardi, and Agathe Simon). Project page.
Examining whether indirect taxes on food can be repurposed and restructured to create incentives for healthy food consumption, without increasing the tax burden on households and without adverse distributional impacts relative to the current taxation system.
Abstract: This paper utilizes quarterly balance sheet data from cross-border banking systems to examine the network of the Core Global Banking System (CGBS) across 19 reporting BIS countries from 2005 to 2020. This study focuses on assessing systemic risk related to liquidity in the core international banking system through the application of the network-based eigen-pair method. We use the liquid assets and short-term liabilities on the balance sheets to investigate liquidity issues within the global banking system. Our systemic risk index not only provided an early warning signal for the Global Financial Crisis (GFC) of 2007-2008 but also revealed the persistent instability of the system throughout the sample period, emphasizing the urgent need for measures to stabilize the system. Additionally, this paper updates the ranking of systemically important and vulnerable banking systems using new variables. Our results continue to highlight the significant impact of the US and UK banking systems on systemic stability. (coming soon)