Session 28 Systemic Risk of Banks & Insurance. Richard Nesbitt, CEO Global Risk Institute in Financial Services

Transcription

1 Session 28 Systemic Risk of Banks & Insurance Richard Nesbitt, CEO Global Risk Institute in Financial Services

2 Our Mission GRI is the premier risk management institute, that defines thought leadership in risk management for financial institutions globally. It brings together leaders from industry, academic and government to draw actionable insights on emerging risks globally.

3 GRI Member institutions As at February 2017

4 Systemic Risk Systemic risk is the global interconnectedness of today s financial systems, and the numerous channels along which distress can propagate and affect other economic sectors. Arises from 2 Main Reasons: Counterparty Risk - institutions share risk through derivatives trading and interbank loans, Common balance sheet holdings - forced sale of illiquid assets done by institutions who need to meet redemption requests or satisfy regulatory requirements may depress prices. Source Systemic Risk, Policies and Data Needs Agostino Capponi 4

5 Why is Systemic Risk Important? Contagion Severe economic impacts Stakeholders harmed who are innocent bystanders Possibility of needing government intervention 5

6 Financial Crisis 2008 and Systemic Risk The 2008 financial crisis resulted in substantial public funds used to bail out banks in US and Europe Too Big to Fail was coined and became a political football Living Wills are now a regular part of the regulatory regime for banks 6

7 Regulatory Focus on Systemic Risk Companies deemed by regulators as systemically important are designated Global SIFI or Domestic SIFI (sometime called SIB ) In Canada 6 largest banks 3 insurance companies and 1 Credit Union are regarded as D-SIFI. No G-SIFI in Canada. Those designated must keep additional capital and are subject to more stringent regulatory regime. 7

8 Insurance Companies and Systemic Risk Are Insurance Companies systemically important? Subject to significant debate. MetLife disputed in court the US designation of it as a G- SIFI. Given President Trump's stance on regulation, MetLife, Prudential and AIG may lose their SIFI designation. 8

9 Measuring Systemic Risk Various models are used Most information is private and not available to public so accurate measurement is not possible Can systemic risk be approximated using public data? Much debated in the case of insurance companies. 9

10 Systemic Risk Measure: Critical Analysis Paper by Thomas F. Coleman, Alex LaPlante, Alexey Rubitsov

11 SRISK: An Introduction SRISK 1 : A systemic risk measure that is defined as the expected capital shortfall of an institution conditional on a financial crisis (i.e. the capital a firm is expected to need, in addition to reserves, in a crisis) A function of the institution s size, leverage, and expected equity loss conditional on a specified market decline Is calculated using publicly available data 1 Brownlees, C., Engle, R. (2016) SRISK: A Conditional Capital Shortfall Measure of Systemic Risk, RFS 11

12 SRISK: Crisis Scenarios Financial crisis scenarios are defined by a prolonged market decline below a threshold over a time horizon which are specified by analyst Example: A market decline of 40% (crisis threshold) over a time horizon of 6 months One must also choose the market which is represented by an index Example: MSCI World Index, S&P/TSX Composite Index, etc. 12

13 SRISK formulation SRISK = Prudential Capital Stressed MV of Equity = k Debt + Stressed MV of Equity Stressed MV of Equity = k(debt) (1 k)(stressed MV of Equity) = k(debt) 1 k 1 LRMES Equity SRISK is a modification of the Capital Shortfall equation: CS = k(debt) 1 k (Equity) k is the Prudential capital ratio and set to 8% for US and Canadian companies and 5.5% for European ones LRMES (Long-Run Marginal Expected Shortfall) is the expected percentage loss of a firm s equity value during a crisis scenario Can be evaluated using different return models including CAPM-beta and the bootstrap approach 13

14 Why SRISK? Straightforward calculation Uses non-proprietary, publically available data; Topdown approach Has been shown to be a good indicator of distress: SRISK measure was a predictor of the Fed capital injections Aggregate SRISK (the sum of SRISK across an entire country) was found to be a predictor of macroeconomic distress 14

15 Main Research Questions 1. Is SRISK overly sensitive to choice of: a) Prudential Capital Ratio? b) Crisis Scenario and Market Index? c) Market Return Model? 2. How should SRISK be interpreted for Canadian institutions? 3. Is SRISK appropriate for all financial service industries? This study focuses on the banking and insurance industries 15

16 Prudential Capital Ratio

17 Selecting Prudential Capital Ratio 300,000 SRISK as a function of prudential capital ratio, k (in million USD) June 30, , , , ,000 50, % 7.5% 7.0% 6.5% 6.0% 5.5% 5.0% BOA Citigroup JPMorgan Morgan Stanley Goldman Sachs 17

18 Historical Capital Ratio 2 In Brownlees and Engle (2016), the choice of k=8% is based on the average capital ratio (CR) maintained by large, well-managed financial institutions in normal times (summer spring 2011). Historical Capital Ratio (CR) American Banks Canadian Banks Wells Fargo JPM TD Bank CIBC 24% 19% 14% 9% 4% 14% 12% 10% 8% 6% 4% 12% 11% 10% 9% 8% 7% 6% 5% 4% 12% 11% 10% 9% 8% 7% 6% 5% 4% CR 10% CR 7% CR 9% CR 7% 2 CR = Market Value of Equity Debt+Market Value of Equity

19 Accounting standards: US GAAP vs. IFRS Under US GAAP, net derivatives are reported and thus derivatives represent a negligible fraction of the assets Brownlees & Engle (2016) employ k = 8% for firms using US GAAP Under IFRS, gross derivatives are reported Engle et al. (2015) report that under IFRS the assets of large US banks would increase by 40-60%. To account for this difference in accounting standards, k=5.5% is used for European firms. 19

20 Derivatives to Total Assets 44% 39% 34% 29% 24% 19% 14% 9% European banks (IFRS) 14% 12% 10% 8% 6% 4% 2% Canadian banks (IFRS) Deutsche Bank RBS HSBC TD RBC CIBC Based on the differences in Canadian derivatives holdings to both US and European firms, we find that a k = 7.5% is most appropriate 20

21 PCR: Insurance There is evidence that suggest the prudential capital ratio of 8% is too conservative for some insurance companies (Acharya & Richardson, 2014) This is especially true for P&C insurance companies as their market value of equity covers a higher fraction of their total assets Historical Capital Ratio analysis for the insurance industry suggests: U.S. insurance companies: k 6.5% Canadian insurance companies: k 14% Note: Historical Capital Ratio analysis may be less suitable for insurance companies resulting in the overestimation of k 21

22 PCR: Conclusions SRISK can be very sensitive to the choice of prudential capital ratio Despite Canadian banks being listed under IFRS, our work suggests that using k = 7.5% is more appropriate than k = 5.5% as proposed by Engle et al. (2015) for the European banks that also operate under IFRS. Careful consideration needs to be given when selecting the prudential capital ratio (k): Historical capital ratios can differ between industries and institutions Accounting standards as well as the size balance sheet constituents should be taken into account when selecting k 22

23 Market Index, Crisis Scenario & Market Choice

24 Market Index, Crisis Scenario & Model Choice Market Index As one would expect, SRISK values were higher for a local crisis (40% decline of S&P/TSX composite) than for a global crisis (40% decline of MSCI World) Crisis Scenario SRISK values were found to fluctuate significantly when different crisis thresholds and lengths were specified Example: Reducing the crisis threshold from 40% to 35% decreased TD s SRISK by 20% (3 billion USD) Model Choice Similarly, SRISK values varied substantially when different return modeling approaches were used Example: SRISK for Wells Fargo was: 7,119 million USD using CAPM-beta & 38,758 million USD using the bootstrap approach Values obtained from NYU Stern s V-Lab

25 Interpretation

26 Interpretation (US Institutions) Brownlees and Engle (2016) show that: SRISK was a significant predictor of the capital injections carried out by the Fed during the financial crisis SRISK delivers useful rankings of systemically risky firms Example: the SRISK rankings identified Fannie Mae, Freddie Mac, Morgan Stanley, Bear Stearns and Lehman Brothers as top systemic contributors as early as Q Aggregate SRISK provides early warning signals of worsening macroeconomic conditions Brownlees and Engle (2016) show that an increase in SRISK predicted future declines in industrial production and increases in the unemployment rate, and that the predictive ability of aggregate SRISK is stronger at longer horizons 26

27 Interpretation (Canadian Institutions) At the end of 2008, the expected capital shortfall of the top five Canadian banks was about 80 billion CAD, more than 5% of Canadian GDP this is somewhat at odds with what was actually observed in Canada during the crisis Although Canadian FIs did not require a government bailout, some argue that the Bank of Canada s liquidity provision helped the sector avoid the need for any bailouts Overall, there seems to be no real evidence suggesting that the SRISK value of a given Canadian bank is equal to its expected capital shortfall Perhaps, SRISK for Canadian banks should be regarded as the propensity of a bank to suffer severe losses during a crisis Example: During the onset of the financial crisis, CIBC had high values of SRISK, whereas TD s SRISK was rather negligible CIBC experienced large losses during the crisis; TD s write-downs were smaller Based on our analysis of Canadian Institutions, the following interpretation seems more appropriate: The upward (downward) trend in SRISK values usually indicates an increase (decrease) in systemic risk Financial institutions that ranked the most systemically risky, based on their SRISK values, can be viewed as more prone to severe losses during a financial crisis compared with those at the bottom of the ranking (smaller SRISK values) 27

28 US vs. Canada Comparison 500, , , , ,000 0 SRISK for US banks, k=8% (in million USD) Goldman Sachs Morgan Stanley JPMorgan 100,000 50,000 0 SRISK for Canadian Banks, k=7.5% (in million USD) TD Bank Scotiabank RBC BMO CIBC 28

29 Application to Insurance Companies

30 Insurance Industry Concerns SRISK does not consider the presence of segregated funds in insurer balance sheets, which leads to a large miscalculation of systemic risk in insurance companies SRISK assumes that the expected value of debt does not change during a financial crisis 1 Insurance companies have variable annuity guarantees included in their liabilities. If not properly hedged, these guarantees will change in value during a financial crisis. Currently SRISK will miss this. 1 This is also true for banks 30

31 Segregated Funds On the balance sheet, segregated funds are both an asset and a liability However, the value of the segregated funds is the value of the underlying mutual fund and is distinct from the actuarial liability of the policyholder s guarantee on the segregated fund SRISK s inclusion of segregated funds as a liability: Gives a misleading impression of high leverage in insurance companies Translates into a substantial overestimation of SRISK values due to their significant size To address this issue we propose an adjusted SRISK value Adjusted SRISK: SRISK value calculated using Debt net of the segregated funds Note that if the guarantees are not well hedged there may be significant debt dependence on the market 31

32 Segregated Funds US insurance companies (in million USD) - June 30, 2016 SRISK Adjusted SRISK Seg. Funds Debt Prudential 43,835 21, , ,730 Metlife 54,754 30, , ,463 Lincoln National 16,928 7, , ,060 Hartford 7, , ,381 Genworth 6,725 6,115 7,624 91,200 Canadian insurance companies (in million CAD) - June 30, 2016 SRISK Adjusted SRISK Seg. Funds Debt Manulife 34,739 12, , ,000 Great-West Life 7, , ,714 Sun Life 3, , ,721 Intact ,362 32

33 Segregated Funds: Canadian Insurance 60,000 40,000 20,000 0 SRISK for Canadian insurance companies (in million USD) Manulife Great-West Life 50,000 30,000 10,000-10,000 Adjusted SRISK for Canadian insurance companies (in million USD) Manulife Great-West Life 33

34 Conclusions on SRISK Is SRISK sensitive to parameter and model choice? The SRISK measure cam be very sensitive to the choice of model and parameter values. Consequently, the margin of error due to model and parameter uncertainty may be substantial. How should SRISK be interpreted for Canadian institutions? Instead of expected capital shortfall it would be more suitable to interpret SRISK values as propensity of financial institutions to have large losses during a financial crisis. Is SRISK appropriate for all financial service industries? When applying SRISK to Insurance companies, the segregated funds should be removed; their inclusion can lead to a substantial overestimation of SRISK values. There may be a significant debt dependence on the market if the guarantees are not well hedged. This has not been accounted for. 34

35 Systemic Risk of Banks & Insurance Richard Nesbitt, CEO Global Risk Institute in Financial Services