Integrated Risk Management in a Financial Conglomerate

1
Integrated Risk Management in aFinancial
Conglomerate

• Til Schuermann
• Federal Reserve Bank of New York
• World Bank Risk Management Workshop
• Cartagena, Colombia
• February 17, 2004

Any views expressed represent those of the
author only and not necessarily those of the
Federal Reserve Bank of New York or the Federal
Reserve System.
2
What Is a Financial Conglomerate?

• Joint Forum definition (2001)
• Any group of companies under common control
  whose exclusive or predominant activities
  consists of providing significant services in at
  least two different financial sectors (banking,
  securities, insurance)
• Virtually all of the large, internationally
  active multinational financial institutions are,
  to some degree, financial conglomerates
• Strict 3 of 3 or weaker 2 of 3 definitions

3
Market Context

• Rapid growth in scope of large, multi-line
  financial institutions
• Consolidation
• Financial deregulation
• Globalization
• Not just bigger, also (much) more complex
• Major advances in risk measurement and capital
  management practices across the industry
• Capital regulation still largely based around
  single business lines or silo approach

4
What Is the Regulatory Issue?

• Banks, securities firms and insurance companies
  all conduct trading business with
• Many of the same instruments and
• Many of the same counterparties, but . . .
• . . . subject to very different regulatory
  capital charges
• Differences are profound and pervasive
• Differences in regulatory objectives
• Differences in definition of regulatory capital
• Differences in regulatory capital charges

5
Philosophical Differences About What Should Count
as Capital

• Differing assumptions about how to deal with a
  faltering firm
• Securities regulators liquidate without loss to
  customers or recourse to bankruptcy proceedings,
  emphasis on subordinated claims
• Bank regulators want time to detect and
  remediate, emphasis on patient money
• Insurance regulators ring-fence for protection
  of customers, emphasis on adequacy of technical
  reserves
• Evident in capital ratios
• Securities firms 5
• Banks 10
• Insurers
• Life 8
• PC 25

6
Differing Definitions of Capital

• Net Worth similarities more apparent than real
• Mark to market accounting in securities firm
• Mix of mark to market book value in banks
• Statutory accounting in insurance companies

7
Example of Different Treatments

• Consider a credit exposure to an A rated
  counterparty

8
Key Questions and Approach

• 1. How should assessments of capital adequacy
  take into account diversification or
  concentration of activities within a
  conglomerate?
• 2. What are the implications for regulating the
  solvency of a multi-line financial conglomerate?
• Our Approach
• Adopt a top-down economic perspective
• Focus on unique problems of risk aggregation
  within a conglomerate
• Initially, make simplistic assumption that all
  risk types have multivariate normal distribution
• Risk is taken to be 99 VaR

9
Risk Types and Distributions
How to Aggregate?
10
Risk Types and Modeling Approaches

• There is a large variety of measurement and
  modeling approaches

11
Risk Management in a Financial Conglomerate
12
Levels of Risk Aggregation in a Financial
Institution
13
Diversification Size of Portfolio and Degree
of Correlation
14
Diversification Across Risk Types
15
Diversification Across Risk Types Financial
Conglomerate
16
Summary of Results so Far

• Diversification effects typically decrease at
  successive levels in an organization Level 1 gt
  Level 2 gt Level 3
• Provided standalone risks are correctly measured,
  incremental diversification benefits across
  banking and insurance fall into an expected range
  of 5-10
• Diversification effects are greatest when
  businesses are of similar size
• Combining a bank with a PC company produces the
  greatest diversification benefit because PC and
  credit risks predominate and are uncorrelated

17
Alternative Interpretations
Risk Aggregation at Level 3 Can Only Be as Good
as the Standalone Measures on Which It Is Based
18
Level 2 A More Sophisticated Approach

• Goal was to model market, credit and operational
  risk of a typical large, internationally active
  bank
• Market and credit risk distributions from market
  data
• Operational risk distribution from industry
  (proprietary) database of operational risk events
• Compare different ways of computing total risk
  distribution
• Add-VaR add-up marginal VaR to arrive at total
• Effectively BIS 2
• Normal-VaR assume joint normality
• Copula-VaR use copulas to arrive at total risk
  distribution
• Hybrid approach assume elliptical distribution
  (not as strict as joint normal but almost as
  easy)
• Risk is taken to be 99.9 VaR

19
Marginal Risk Distributions
20
Characteristics of Risk Distributions

• Market
• Credit
• Operational

• Highest volatility
• Lowest skewness
• Slightly fat tails
• Moderate to high volatility
• Skewed
• Moderately fat-tailed
• Low volatility
• Very skewed
• Very fat-tailed

21
Impact of Correlation at 99.9 VaR

• ?(market,credit) 50 ? vary ? to operational
  risk

22
Bottom Line

• Consistent ordering of approaches
• Add-VaR gt Hybrid-VaR gt Copula-VaR gt Normal-VaR
• Add-VaR biggest imposes perfect inter-risk
  correlation
• Normal-VaR smallest since it imposes thinnest
  tails
• The Hybrid approach is strikingly close to
  copula-VaR
• Use volatility multiples from marginals
• Incorporates correlation
• Diversification benefits at 99.9 VaR can be
  substantial
• Depending on correlations, 10 to 35
• As business mix or correlation shifts towards
  operational risk (very fat-tailed and skewed),
  99.9 VaR increases dramatically
• Normal-VaR fails especially here
• Hybrid approach can handle this well (sensitive
  to tail shape of marginals)

23
What Has Been the Market Response?
24
(No Transcript)
25
Thank You! http//nyfedeconomists.org/schuermann/
26
Limitations of Silo Regulation
Inconsistency
Aggregation
Incompleteness

• Capital requirements dependent on where risk is
  booked
• Boundaries breaking down due to product
  innovation
• Increasing demand/potential for regulatory
  arbitrage

• Concentration of risks across legal boundaries
• Diversification across risk classes within a
  legal entity
• Diversification of risks across business
  activities and operating companies

• Capital requirements of unlicensed operating
  companies
• Capital requirements/funding structure of
  holding company
• Strategic investments in non-financial companies