Equity market and crisis

1
What has made thefinancial crisis truly global?
Michael Ehrmann, Marcel Fratzscher and Arnaud
Mehl European Central Bank
Tsinghua Columbia Workshop in International
Economics Beijing, 26-27 June 2009 Views are
those of the authors, and not necessarily those
of the ECB.
2
Severity of the 2007-09 financial crisis
3
yet high dispersion across countries
portfolios
4
Symmetry in collapse of growth
Real GDP growth in EMEs and Advanced Economies
(y-o-y change)
Source Haver Analytics and ECB Staff
Calculations. The projections are based on the
April IMF WEO. NoteLast observation refers to
2009Q1. Advanced economies are the weighted
average of UK, euro area, US, Japan and Canada.
The weights are based on 2008 IMF WEO GDP PPP
weights.
5
Equity markets EMEs hit hardest particularly
via capital flows retrenchment and flight to
safety
Total EME equity flows and prices (since Jan
2004, cumulative monthly flows in bn USD and
total MSCI return index)
Stock market developments in selected EMEs (total
return since 15/9/2008, in pp contribution)
Source Haver Analytics and ECB Staff
Calculations Note Last observation refers to 8
June 2009. A indicates a country for which the
stock price level is meanwhile above that of
15/9/2008.
Source EPFR and Bloomberg. Note Last
observation refers to end April 2009.
6
FX markets the elusive US dollar crisis
Total flows by country (since 15/9/2008,
cumulative weekly flow, of AUM)
Exchange rate developments in selected EMEs
(vis-à-vis USD, in pp contribution)
Source EPFR Note Last observation refers to 3
June 2009.
Source Haver Analytics and ECB Staff
Calculations Note Last observation refers to 5
June 2009.
7
1. Motivation

• What explains transmission and discrimination?
• Even markets with little financial leverage
  affected
• Was the spread of the crisis predictable or
  different?
• Three hypotheses
• Financial integration (with US markets)
• transmission reflecting usual mechanism ?
  predictable
• Macro risk vs. micro risk
• macro policies fundamentals vs. risk at firm
  level (leverage, financial constraints, etc.)
• Contagion
• above and beyond 1. and 2.

8
Motivation

• Mechanisms and potential implications
• retrenchment or repatriation hypothesis role
  of financial globalisation and integration
• global de-leveraging hypothesis macro risk
  calls for macroprudential analysis and
  surveillance, also globally
• liquidity squeeze hypothesis micro risk
  stresses role for microprudential regulation and
  supervision
• Two parts of the analysis
• Unconditional explaining cross-sectional
  heterogeneity ? which of the three factors
  explains the transmission?
• Conditional conditioning on common US shocks
  ? has the transmission of specific shocks
  changed?

9
Motivation

• Main findings
• Financial integration (US beta) plays key role in
  transmission ? crisis transmission to that extent
  predictable
• Macro/country risk key explanations portfolios
  in countries with strong fundamentals declined by
  one third less than those with weak fundamentals
• Financial policies esp. debt and deposit
  guarantees played major role in shielding
  economies
• 2007-09 crisis fundamentally different from past
  TMT boom-bust cycle and more tranquil periods ?
  role of macro/country risk related to policy
  interventions?

10
Role of integration with US markets for crisis
severity
11
Role of country-level macro fundamentals for
crisis severity
12
2. Literature

• Current financial crisis
• Focus on US policy responses (Calomiris 2008,
  Taylor 2009)
• Role of liquidity (Heider, Hoerova and Holthausen
  2009)
• Financial constraints rather than demand in US
  (Tong Wei 2008)
• Little on global transmission IMF (2009) on
  financial stress transmission, Fratzscher (2009)
  on global FX markets
• Crisis and role of contagion for transmission
• Contagion and related channels (Bae et al. 2003,
  Karolyi 2003, De Gregorio and Valdes, 2001,
  Dungey et al. 2004)
• Transmission channels (Forbes and Rigobon 2002,
  Bekaert, et al. 2005)
• Time-varying global market integration (Bekaert
  Harvey 1995, 2000)

13
Literature

• International transmission of specific
  events/shocks
• Financial linkages and trade account for part of
  cross-country equity returns (Forbes and Chinn
  2004)
• Transmission of specific monetary policy or macro
  news shocks to global equity and FX markets
  (Hausman and Wongswan 2006, Wongswan 2006,
  Ehrmann and Fratzscher 2006, Ammer, Vega and
  Wongswan 2009)

14
3. Data 64 countries, 2000 firms, 455 portfolios
15
Portfolio equity returns and their dispersion
16
Summary stats macro micro determinants
17
4. Integration with US markets -- beta from CAPM
(Fama-French)

• Ri,t return of country-sector portfolio i on
  date t
• Rust return of US stock market on date t
• RRGt return of regional stock market on date t
• (Bekaert, Hodrick and Zhang, 2009)

18
Evolution of time-varying betas their
dispersion
Average beta and beta dispersion
19
Evolution of time-varying betas their
dispersion
Dispersion within vs. across countries
20
Distribution of time-varying beta
Before the crisis (Q2 2007)
21
Distribution of time-varying beta
During the crisis (Q4 2008)
22
Decomposition of cross-section return dispersion
Dispersion in betas vs. idiosyncratic component
23
4. Empirical benchmark specification

• Ri portfolio i total return during crisis
• X portfolio i country risk and sector-level risk
  (incl. value and size factors as Fama-French
  controls)
• bUS portfolio i US beta
• Cross-section 455 country-sector
  portfolios
• Crisis period 7 August 2007 31 March 2009

24
Role of macro risk and integration in crisis
25
Role of country risk in the financial crisis
Macro fundamentals
26
Role of integration with US markets for crisis
severity
27
Role of country-level macro fundamentals for
crisis severity
28
Role of country-level macro fundamentals for
crisis severity
29
Role of micro risk / sector-level variables in
the financial crisis
30
Role of firm-level characteristics for crisis
severity
31
Testing the three hypotheses

• Financial integration (US beta) important
• Macro risk important, but not micro/firm-level
  risks
• Interesting switch in sign of coefficients before
  vs. during crisis
• increase in leverage across institutions
• high return to risky assets
• Is the current crisis different?
• Yes key role of macro/country risk
• Robustness

32
Is it the beta? Explaining beta by country risk
33
Robustness endogeneity of country risk and beta
34
Robustness excluding financial sector stocks
35
6. Role of financial policies

• What has made the crisis different why is
  country risk priced and micro risk less so?
• Significant extensions introductions of
  financial policies (Pi)
• Debt guarantees
• Deposit guarantees
• Capital injections
• Not all countries implemented such policies, and
  important differences across countries

36
Empirical benchmark specification

• Pi indicator for financial policy during crisis
• Three types of financial policies debt
  guarantees, deposit guarantees, capital
  injections
• Size versus presence of policy intervention
• Sources National sources, Bloomberg, BIS
• Cross-section 455 country-sector
  portfolios
• Crisis period 7 August 2007 31 March 2009

37
Role of financial policies deposit guarantees
38
Role of financial policies

• Findings emphasise dominant role of risk
  reduction from financial policies
• Capital injections less relevant for aggregate
  equity markets than wider debt and deposit
  guarantees
• Note of caution plenty of caveats
• Endogenous character of financial policies
• Are they more likely to occur in countries with
  weak or strong fundamentals?
• Quality and precision of data on financial
  policies.
• announced financial policy measures what
  matters for effectiveness is perceived
  credibility and government commitment hard to
  measure, but should be related to announcements

39
7. Conditioning on common shocks

• How are common shocks transmitted?
• Does this confirm our previous findings?
• Is this transmission any different during the
  crisis?
• Two sets of common US-specific shocks
• Key crisis events
• US macroeconomic news (comparison with
  pre-crisis)
• Advantages
• Clean identification high/daily frequency of
  shocks
• Difference-in-difference analysis

40
Six key US crisis-related events
4. 30/09/08 FDIC Seeks Authority to Raise
Deposit Insurance Limit 5. 13/10/08 US Treasury
announces investment in 9 major US banks 6.
23/03/09 U.S. Treasury Announces 1 Trillion
Plan to Buy Distressed Debt
1. 16/03/08 Bear Stearns collapse and sale to
JPMorgan 2. 15/09/08 Lehman Brothers declares
bankruptcy 3. 29/09/08 U.S. House Rejects 700
Billion Financial-Rescue Plan
41
US macroeconomic news shocks
42
Modelling the global transmission of common US
shocks

• Scrisis vector of 6 key US crisis-related
  shocks
• Snews vector of US macro news
• R , X and Z as before but at daily frequency

43
Global transmission of US shocks
44
Estimating the global transmission of US shocks
(difference-in-difference approach)

• Dt 1 during the crisis, 0 otherwise

45
Global transmission of US shocks causality
(difference-in-difference results for
beta)
46
Global transmission of US shocks causality
(difference-in-difference results for FX
reserves)
47
Global transmission of US shocks causality
(difference-in-difference results for
current account)
48
Confirmation from conditional analysis

• The crisis is different! ? 3- to 4-fold increase
  in the strengths of the shock transmission
• Yet also US returns have become more sensitive to
  a given shock during crisis
• ? explains why US beta from CAPM has been rather
  stable, yet return dispersion has increased
• Confirmation of role of financial integration as
  transmission channel for crisis
• Role of idiosyncratic, country-specific shocks key

49
8. Summing up

• Focus of paper on channels of crisis transmission
  (not crisis severity per se)
• Financial integration/globalisation important
• What makes the crisis different?
• Macro / country risk important
• but not micro/firm-level risks
• Role of financial policies (esp. deposit and debt
  guarantees) key

50
Implications for policy

• Focus on reform of international financial
  architecture
• Emphasis on microprudential reforms regulation
  and supervision of institutions and markets
• But findings suggest that macroprudential risks
  dwarfed micro / firm-level risks during crisis
  at least when analysing global transmission of
  crisis
• Role of macroprudential analysis

51
APPENDIX
52
Estimation of exchange rate interest rate
exposures

• Ri,t return of country-sector portfolio i on
  date t
• Rust return of US stock market on date t
• Si,t bilateral exchange rate change vs. USD on
  date t
• ri,t change in domestic 3-month interest rate
  on date t
• (Dominguez and Tesar, 2001 2006 Amer et al.
  2009)

53
Decomposition of cross-section return dispersion
Dispersion in beta-fitted returns vs.
idiosyncratic component
54
Role of country risk in the financial crisis
External exposure
55
Role of country risk in the financial crisis
fundamentals external exposure
56
Linking equity returns to beta and fundamentals