- A new book by Jagoda Kaszowska-Mojsa challenges the suitability of traditional dynamic stochastic general equilibrium (DSGE) models for systemic risk analysis;
- Agent-Based Models (ABMs) offer superior capabilities due to their ability to capture the endogenous nature of systemic risk and the non-linear dynamics that are inherent to financial systems.
Financial Crisis driven by 'internal vulnerabilities'
Crisis and systemic risk were a significant cause of rising inequality in both the United States and the EU following the global financial crisis in 2008.
While the financial crisis in the US was driven by deregulation processes, and global current account imbalances, the crisis in EU periphery countries was predominantly driven by internal vulnerabilities - such as excessive exposure to the real estate sector, fragile banking systems, high levels of indebtedness and heightened uncertainty, which intensified cross-sector contagion.
In Financial Instability and Systemic Risk: From the Global Financial Crisis and Beyond, Dr Jagoda Kaszowska-Mojsa of the Complexity Economics programme at INET Oxford, finds that the relative role of systemic risk in EU countries was greater during the course and aftermath of the crisis than during the initial phases. This is primarily explained by the presence of feedback effects and amplification mechanisms that altered the functioning of the financial system during the crisis.
What is financial instability and systemic risk?
Financial instability arises when the financial system becomes fragile and less able to support the real economy — for example by providing credit, managing risk or absorbing shocks. Systemic risk is the danger that problems in one part of the system spread more widely, as distress in one bank, market or country is amplified through financial linkages and feedback effects.
The book argues that crises are not always triggered by external shocks. Fragility can also build up from within complex financial systems, as the decisions of banks, firms and households interact in unexpected ways. This is why agent-based models are useful: they allow researchers and policymakers to simulate how risks emerge, how they spread, and how different regulations may affect both stability and inequality.
"The agent-based modelling approach is evolving rapidly worldwide." Jagoda Kaszowska-Mojsa, Economic Expert at the Bank of Poland; Researcher at the Institute of Economics, Polish Academy of Sciences
Financial crisis 'exposed limitations' of traditional prudential approaches
A well-calibrated macroprudential stance stabilises the economy and reduces inequality, whereas a mis-calibrated stance may lead to heightened fiscal instability and increased inequality.
But macroprudential policy design involves important trade-offs. While certain policy combinations may improve financial stability, they may also generate welfare effects and substantial changes in inequality indicators, depending on their calibration and institutional context. This makes new tools such as data-driven Agent-Based Models essential for simulating how alternative regulatory choices affect both systemic risk and redistribution across the economy.
Dr. Kaszowska-Mojsa said that certain prudential regulations in EU countries may have increased systemic risk and financial instability during the period, contrary to their intended purpose.
"The outbreak and progression of the financial crisis exposed the limitations of traditional micro prudential approaches to financial regulation and policymaking.
"In this work we looked to identify the role of systemic risk in the causes, course and effects of the financial crisis in EU countries and assess the effectiveness and optimality of the regulatory and prudential measures that have been introduced within the EU to mitigate systemic risk.
"This is an under-researched area. Developing a coherent methodological framework for the study of systemic risk and applying this to evaluate regulatory and prudential policies appears essential to address practical shortcomings in financial policymaking.
"There is growing interest globally in measuring and modelling of financial sector and systemic risk and there are urgent needs for mechanisms to assist this cause. Research on macroprudential stances should rely on agent-based approaches rather than the DSGE models, which are based on the social planner framework."
A 'remarkable book' informed by 'years of hands on experience'
John Muellbauer, Co-Director of the Macroeconomics and Finance programme at INET Oxford, said that the book should be essential reading at central banks, financial regulators and university courses on finance and the economy.
"Financial Instability and Systemic Risk is a remarkable book by a remarkable woman: economist-mathematician-computer scientist, informed by years of hands-on macroprudential policy experience.
"The book opens by explaining the concepts of financial instability and systemic risk. Emphasising the importance of institutions and history, the book continues with a remarkably comprehensive account of the roots of the Global Financial Crisis in US institutions and in risky financial deregulation. The dynamics of how the GFC played out and its economic consequences are very clearly explained. Further chapters explore European versions of recent financial crises and the foundations of macroprudential policy making.
"The media and politicians often worry that tighter borrower-based macroprudential measures (BBMs) such as lower loan-to-value or debt-service-to-income ceilings on borrowing or higher amortisation requirements, might reduce access to owner-occupation, increase inequality and lower welfare. The complexity of how sectoral and microeconomic interactions operate in different countries calls for an analysis of granular data in a systems approach with a complete capture of how households – renters as well as owner-occupiers, investors, landlords, banks, firms, the fiscal authority and the central bank and/or financial regulator interact dynamically.
"The effects of macroprudential policies like BBMs are not solely tied to institutional features and market conditions but also to the personal attributes of households. Households may perceive their consumption set as being restricted by regulations, but these constraints could also contribute to smoothing consumption over the business cycle, thus resulting in moderate downturns and reduced volatility. There may be varied effects of BBMs on different groups of households with specific characteristics and in specific situations e.g., in the labour market in different phases of the economic and financial cycle. Currently, only the Agent-Based Modelling (ABM) simulation method, with fully heterogeneous agents and based on empirical data, permits analyses at such a finely disaggregated level. This method probably has the best potential for assessing the medium-run distributional and welfare consequences of different macroprudential choices under different scenarios. The last chapter of Jagoda Kaszowska-Moja’s book explains her pioneering, path-breaking work on the ABM approach to examine the distributional effects of macroprudential policies with illustrations from the Polish economy."
