Systemic Risk and Interconnectedness for Banks and Insurers Mary A. Weiss, Ph.D. SAFE-ICIR Workshop Goethe University Frankfurt May 2014
What is interconnectedness? Working definition of interconnectedness: Interconnectedness is an assessment of the potential impact of a company s financial distress on the broader economy Interconnectedness arises from actual and perceived complex webs of contract relationships across financial institutions.
Approaches to Measuring Systemic Risk and Interconnectedness Two sets of tools used in crosssectional dimension: 1.Network analysis 2.Market-based indicators
Network Analysis Allen and Gale (2000) model of financial networks liquidity shocks having domino effect Interbank deposits are primary mechanism of liq. Shock Complete vs Incomplete Networks Moral of lesson: Diversification Gai, Haldane and Kapadia (2010) Shocks propagate in network structures in which some financial institutions more interconnected than others. Contagion more likely at higher levels of connectivity shock transmitters vs shock absorbers Regulators care about shock transmitters.
Network Analysis: Unexplored Issues 1. Focus on nodes and not on edges of financial networks 2. Detailed and comprehensive data requirements 3. Determinants of network formation (in first place) are unknown. 4. Don t know how networks change in event of new information or stress events 5. Don t know the conditions that lead to formation of fragile/robust networks.
Market-Based Indicators Two types of Market-Based Indicators: One set concerned with size of financial distress Other set concerned with analysis of market information econometrically
Market-Based Indicators: Potential Size of Financial Distress I Several factors related to likelihood of major financial dislocation: 1. Degree of correlation among holdings of financial institutions 2. How sensitive institutions are to changes in market prices and economic conditions 3. How concentrated the risks are among the financial institutions 4. How closely linked institutions are with each other and the rest of the economy
Market-Based Indicators: Potential Size of Financial Distress II Three primary measures used to measure systemic linkages: Adrian and Brunnemeier(2010) Conditional Value at Risk: (CoVaR) and ΔCoVaR Acharya et al. (2011) Systemic Expected Shortfall (SES) Huang, Zhou, and Zhu (2011) Distress Insurance Premium (DIP)
Market-Based Indicators: Potential Size of Financial Distress III Measures can be implemented with data publicly available on a high frequency basis Limits need to rely on detailed (confidential) supervisory data Forward-looking and reflect investors assessment of the financial health of specific institution Reflect domestic and global policy actions to contain risk.
Market-Based Indicators: Observations and Issues 1. Measures cannot be used to determine causability 2. Do not provide a scale for interpreting results as high, medium, or low systemic risk 3. Not clear how financial distress can be mapped into outcomes for broader economy (such as decrease in GDP). 4. Don t consider whether failure of firm can be absorbed by rest of economy 5. Don t indicator whether competitors can fill in void 6. Don t measure how important services of the financial institution (e.g., insurers) are to rest of the economy 7. Market based measures cannot be used if stock price data unreliable or institution not publicly traded 8. Have short horizon as early warning indicator of financial distress
Market-Based Measures: Econometric Approaches Billio et a. (2012) hedge funds, banks, brokers, and insurers Use principal components analysis used to estimate the number and importance of common factors driving returns of financial institutions Use Granger causality pairwise Granger causality tests used to identify network of statistically significant Granger-causal relations among institutions. Suffer from most of same drawbacks as first set of market-based indicators, but appear to have good out of sample properties. Chen et al. (2013) also use Granger causality
Overall Comments about Systemic Risk and Interconnectedness Measurement Don t know how complementary these measures are to each other. Given confidentiality domestic financial network studies, almost no empirical work has been done to study relationship between network and pricebased measures.
Insurance-Banking Interconnectedness I Quantitative studies of insurance-banking interconnectedness Institutional studies of insurance-banking interconnectedness
Insurance-Banking Interconnectedness: Quantitative Studies of Interconnectedness I Billio et al. (2012) Hedge funds, banks, insurers, and brokers become more interrelated over recent years Banks and insurers more important to interconnectedness than brokers and hedge funds By insuring financial products, writing CDS and engaging in derivatives and investment management, insurers became more part of interconnected system Chen et al. (2013) Use spread on CDS for 11 insurers and 12 banks Dominating influence is that of banks affecting insurance companies.
Insurance-Banking Interconnectedness: Quantitative Studies of Interconnectedness II Acharya et al. (2010) Inference is that if insurer has large systemic expected shortfall (SES) when other financial institutions do, it is interconnected. Insurers least systemically risky compared to depository institutions and securities dealers and brokers. AIG more systemic than Berkshire-Hathaway Top 3 insurers in terms of systemic risk were heavily involved in providing financial guarantees for structured products (Genworth, Ambac, and MBIA)
Insurance-Banking Interconnectedness: Quantitative Studies of Interconnectedness III Baluch, Mutanga, and Parsons (2011) Significant correlation between banking and insurance sectors and finds correlation increased during crisis period. Greatest impact of crisis on: 1. specialist finance guarantee insurers 2. insurers heavily engaged in capital market activities 3. bancassurers 4. credit and liability insurers (to lesser extent). Conclude that systemic risk is lower in insurance than banking but grown due to increasing linkages with banks and non-traditional insurance activities.
Insurance-Banking Interconnectedness: Quantitative Studies of Interconnectedness IV Park and Xie (2013) Focus on interconnectedness between U.S. primary insurers and their reinsurers. They look at whether a reinsurer downgrade is associated with a primary insurer downgrade (yes it is) whether a reinsurer downgrade is associated with primary insurer reduction in stock price (yes it is) the likely impact of major global reinsurer insolvencies on the U.S. property-casualty insurance industry. Even under extreme assumption of 100% reins. recoverable default by one of the top three global reinsurers, only about 2% of insurers would be downgraded, and one percent would become insolvent.
Insurance-Banking Interconnectedness: Institutional Studies of Interconnectedness Cummins and Weiss (2013); Geneva Association (2010) Argument that insurance-banking interconnectedness studies should focus on business activities of insurers Business activities may be core insurance activities or non-core (possibly banking) activities Distinction important from regulatory perspective. It is important to distinguish among business activities otherwise regulatory arbitrage through the migration of risky business activities from highly regulated institutions to less regulated institutions would be likely.
Future Research Topics 1. How can regulation be designed so that systemic risk is mitigated? 2. Does new regulation such as Solvency II contribute to systemic risk as discussed in academia and practice? 3. How can regulatory arbitrage be avoided practically? 4. Even if systemic risk relatively low in traditional insurer activities, indirect contagion risk such as reputational risks have not been considered. 5. What is the contribution of derivatives and other innovative products from the field of alternative risk transfer? 6. Need explanation of how regression variables used to explain systemic risk measures are related to reduced output in economy. 7. Is it sufficient to rely on stock price information to measure an interconnection?
Conclusion From Billio et al., p. 555 As long as human behavior is coupled with free enterprise, it is unrealistic to expect that market crashes, manias, panics, collapses, and fraud will ever be completely eliminated from our capital markets. The best hope for voiding some of the most disruptive consequences of such crises is to develop methods for measuring, monitoring, and anticipating them. By using a broad array of tools for gauging the topology of the financial network, we stand a better chance of identifying black swans when they are still cygnets.
. Thank you!
References Acharya, V.V., L. H. Pederson, T. Phillipon, and M.P. Richardson (2012). Measuring systemic risk, working paper, New York University, New York. Adrian, T., and M.K. Brunnermeier (2011). CoVaR, working paper, Federal Reserve Bank of New York, New York, NY. Allen, Franklin and Douglas Gale (2000). Financial contagion, Journal of Political Economy, vol 108 (February) pp. 1-33. Arregio. M., M. Norat, A. Pancorbo, and J. Scarlata(2013) Addressing interconnectedness: concepts and prudential tools, IMF Working Paper. Baluch, R., S. Mutenga and C. Parsons (2011). Insurance, systemic risk and the financial crisis, Geneva Papers on Risk and Insurance Issues and Practices 36 (1):126-163. Billio, M., M. Getmansky, A.W. Lo, and L. Pelizzon(2012). Econometric measures of systemic risk in the finance and insurance sectors, Journal of Financial Economics 104: 535-559. Campbell, S., Research Agenda for Measuring Interconnectedness. Powerpointpresentation. Chen, H., J.D. Cummins, K.S. Viswanathan, and M.A. Weiss (2013a) Systemic risk and the interconnectedness between banks and insurers: An econometric Analysis, Journal of Risk and Insurance advance online publication March 14, doi 10.1111/j.1539-6975.2012.0153.x. Chen, H., J.D. Cummins, K.S. Viswanathanand M.A. Weiss (2013b) Systemic risk measures in the insurance industry: a copula approach, working paper, Temple University, Philadelphia, PA. Cummins, J.S. and M. A. Weiss (2009). Convergence of insurance and financial markets: Hybrid and securitized risk-transfer solutions, Journal of Risk and Insurance 73 (3); 493-545. Cummins, J.D. and M. A. Weiss (2013) Systemic risk and the U.S. insurance sector, forthcoming in Journal of Risk and Insurance. Eling, M. and D. Pankoke(2014). Systemic Risk in the Insurance Sector-What Do We Know? ed. By HatoSchmeiser, University of St. Gallen, working paper. Gai, Prasanna, Andrew Haldane, and SujitKapadia (2010). Complexity, concentration and contagion, Journal of Monetary Economics, vol. 58 (July), pp. 453-470. Geneva Association (2010) Systemic risk in insurance An analysis of insurance and financial stability(march). Huang, S., H. Zhou, and H. Zhu (2009). A framework for assessing the systemic risk of major financial institutions, Journal of Banking and Finance 33 (11): 2036-2049. IAIS (2011) Insurance and Financial Stability(November). IAIS (2012). Reinsurance and Financial Stability(July). IMF (2007). Chapter II: Do market risk management techniques amplify systemic risks? Global Financial Stability Report(October). Kessler, D. (2013) Why (re)insurance is not systemic, Journal of Risk and Insurance advance online publication 21 July, doi: 10.1111/j.1539-6975.2013.12007.x. Park, S.C. and X. Xie(2014). Reinsurance and systemic risk: the impact of reinsurer downgrading on property-casualty insurers. Forthcoming in The Journal of Risk and Insurance. Weiss, G.N.F., D. Bostandzicand F. Irresberger(2013) Catastrophe bonds and systemic risk, working paper, TechnischeUniversitat, Dortmund, Dortmund.