Transcript Country ratings: Initial Considerations

Vulnerability of HIPC countries to
the Financial Crisis
Daniel Ottolenghi, Chief Development Economist
Bernard Ziller, Head of Development Policy Unit
European Investment Bank
Author: Markus Berndt ([email protected])
Washington, 8 July 2009
1
Impact of the financial crisis on EIB partner countries
The global financial crisis
1000
Origin of the crisis in advanced
economies but…
J.P. M organ, EM BI Global, Composite , Inde x, All M aturitie s,
Strippe d Spre ad, USD
800
> Heightened risk aversion
Impact on HIPC countries through...
600
400
Financial market links:
- Deleveraging and heightened risk aversion
200
0
01/01/07
01/07/07
01/01/08
01/07/08
01/01/09
108
107
106
forecast of
October 2008
Output of
advanced
econom ies
forecast of
April 2009
(index 2005=100)
> Global recession forecast
105
104
103
102
2007
250
230
210
190
170
150
130
110
90
70
50
2008
2009
Commodoty price indice s (2005 =100)
Real economy links:
- Global slowdown / recession
→ lower trade,
→ falling commodity prices
→ decline in remittances
M e tals
> Tumbling commodity prices
2007M11
→ financing on international markets more difficult
and costly
→ a reversal of short-term capital more likely
Oil
Indirect impact on
-
2008M5
2008M11
bank credit risk/quality of loan portfolio
government revenue and fiscal balance
2008M11
2
Analytical framework
Domestic crisis risk: Potential
for a global financial crisis leading
to a surfacing of pre-existing
weaknesses in the domestic
financial system, measured by
recent credit growth and credit to
private sector
Fiscal risk: Potential for fiscal
strains due to the need of bailing out
or providing stimulus, measured by
fiscal deficit and debt service
Higher risk
Direct liquidity risk: Direct exposure of the banking
system to the global financial markets, measured by shortterm liquid external assets over liquid external liabilities
Capital/remittances outflow
risk: Exposure to a reversal or
drying out of volatile capital inflows,
measured by dependence on
portfolio inflows and remittances
Export slump risk: Exposure to
slowing of import demand from
crisis-hit advanced economies,
measured by dependence on exports
to advanced economies
Insufficient reserves risk: Potential for balance of
payment crises due to insufficient initial reserves to
smooth crisis related swings in the current account,
measured by reserves coverage of imports
3
Risk categories
3 Risk categories:
High:
above empirically determined threshold
Medium:
above 50% of threshold
High
Medium
Low
Low:
below 50% of threshold
Threshold determined by
optimizing the fit of
probit regressions explaining
- banking,
- currency and
- debt crises
of a panel of
161 countries,
1970-2007
4
Estimated thresholds
Dependent Vulnerabilities Explanatory variables
variables
Banking
crisis
Currency
crisis
Direct
Domestic
Liquid external assets over liabilities
0.805
Credit to private sector (% of GDP)
187 %
Prior average credit growth
20.2%
Reserves
Reserves (months of imports)
Exports
Dependence on exports to adv. econ (% of GDP)
22.3 %
Dependence on portfolio investments (% of GDP)
7.3 %
Dependence on remittances (% of GDP)
7.6 %
Prior government deficits (% of GDP)
3.6 %
Prior debt service (% of exports)
9.2 %
Flows
Debt
crisis
Estimated
Thresholds
Fiscal
1.6
5
Results: HIPCs compared to all LICs
HIPC countries as a group show a similar
risk profile as all LICs in most dimensions
including potentially vulnerable financial
sectors due to excessive credit growth during
the recent commodity price boom.
However, HIPC countries tend to have
more vulnerable fiscal and reserve
positions than an average LIC.
6
Notes: Averages weighted by GDP. Afghanistan, Kyrgyz Republic, and Somalia not included.
Results: Differences within HIPC
Reserves are the lowest in the
pre-decision point countries
Decision point countries
face greater external
vulnerabilities than
completion point countries
7
Results: country by country
Completion point countries 1/3
8
Results: country by country
Completion point countries 2/3
9
Results: country by country
Completion point countries 3/3
Decision
point
countries
1/2
10
Results: country by country
Decision point countries 2/2
Pre-decision
point
countries
1/1
11
Conclusion
Impact of the crisis on HIPC countries
– Direct external liquidity risk limited in most HIPC countries’ financial
sectors
– Reduction in remittances a concern in Latin American HIPCs but also
in Senegal, Sierra Leone and Uganda
– Impact of recession on exports expected to be high in Guyana,
Honduras, Mauritania, Chad and the Republic of Congo.
– Reserves are low in many of those HIPC countries that have not
reached the completion point, yet.
– The Fiscal positions are challenging in many HIPCs including those
that have already reached the completion point
– High risk of domestic vulnerabilities in commodity producers
12
Methodology: Mapping past risks and past crises
Data source: IMF working paper by
Laeven and Valebcia (2008)
Banking crisis
- Large number of defaults in corporate
and financial sector
- Large number of difficulties to repay
contracts on time
Currency crisis
-
Nominal depreciation or devaluation of
at least 30% and 10% higher than the
previous year’s depreciation or
devaluation
Debt crisis
-
Sovereign debt default
Sovereign debt restructuring
13
Methodology: Target regressions to determine ‘optimal’ thresholds
Dependent variables
explained by*
Banking crisis dummy =
F ( Prior average credit growth,
Credit to private sector,
Liquid external assets over liabilities )
Currency crisis dummy =
F ( Gross portfolio investments +
Gross remittances,
Exports to advanced economies,
Reserve cover of imports
)
Debt crisis dummy =
F ( Prior government deficits,
Debt service in % of exports
)
14
* Dummies of explanatory variables are used to explain the probability of a crisis starting in the next year. Crisis and two post-crisis observations are excluded.
Methodology:
Example: Searching for optimal thresholds to explain banking crises
Banking crisis dummy =
F ( Prior average credit growth,
Credit to private sector,
Liquid external assets over
liabilities
Credit growth dummy = 1
if credit growth > threshold_cg
Credit to private dummy
= 1 if credit
to priv.dummy
> threshold_cp
Banking
crisis
=
)
Liquidity dummy = 1
F ( Prior average credit growth,
to private
if Credit
liquidity
< sector,
threshold_li
Liquid external assets over liabilities )
Search function
Iterative search to identify of
those ‘optimal thresholds’
that maximize the fit of the
target probit regression with
the respective dummies
15
Methodology:
Example: Searching for optimal thresholds to explain banking crises
.01
0
-.01
-.02
.061
.062
.063
Marg. effect (right axis)
.064
.02
.065
Threshold for explaining a banking crisis: (1) Avg. credit growth in last two years: 20.15
0
10
20
Threshold
Pseudo R2 (left axis)
30
40
Marg. effect (right axis)
16
Methodology:
Example: Searching for optimal thresholds to explain banking crises
.02
.015
.01
.054
.056
.058
.06
Marg. effect (right axis)
.062
.064
.025
Threshold for explaining a banking crisis: (2) Liquid external liabilities over assets: 124%
50
100
Threshold
Pseudo R2 (left axis)
150
Marg. effect (right axis)
17
Methodology:
Example: Searching for optimal thresholds to explain banking crises
.1
.05
.06
0
.061
.062
.063
Marg. effect (right axis)
.064
.065
.15
Threshold for explaining a banking crisis: (3) Credit to the private sector, % of GDP: 187
100
120
140
160
Threshold
Pseudo R2 (left axis)
180
200
Marg. effect (right axis)
18
Methodology:
Example: Regressions for optimal thresholds to explain banking crises
Banking crisis regressions
VARIABLES
credgro_d
liab_d
cred_d
credgro_2y
liabcover_l1
cred_l1
real_growth
Constant
Pooled
(1)
banking
0.242
(0.158)
0.320***
(0.124)
1.422***
(0.549)
0.000631
(0.00332)
0.000170
(0.000166)
-0.00291*
(0.00167)
-0.0499***
(0.00955)
-1.749***
(0.0927)
2091
0.0741
Observations
2
Pseudo R
Number of id
Standard errors in parentheses
*** p<0.01, ** p<0.05, * p<0.1
(2)
banking
(3)
banking
(4)
banking
0.255*
(0.144)
0.330***
(0.109)
0.872*
(0.454)
0.142
(0.139)
0.334***
(0.105)
0.863*
(0.457)
0.187
(0.129)
-0.0500***
(0.00929)
-1.833***
(0.0769)
2122
0.0652
-1.956***
(0.0718)
2157
0.0218
(5)
banking
(6)
banking
0.277***
(0.102)
0.891**
(0.353)
-0.0311***
(0.00640)
-1.817***
(0.0517)
3092
0.0265
-0.0363***
(0.00780)
-1.767***
(0.0664)
2269
0.0380
-0.0319***
(0.00585)
-1.770***
(0.0458)
3310
0.0343
(7)
banking
dF / dx
0.0199
(0.0131)
0.0248***
(0.00919)
0.123
(0.106)
-0.00328***
(0.000605)
2122
0.0652
Panel with random effects
(9)
(8)
lnsig2u
banking
0.255*
(0.144)
0.330***
(0.109)
0.872*
(0.454)
-0.0500***
(0.00929)
-1.833***
(0.0769)
2122
-14.79
(26.68)
2122
144
144
19
Methodology: Definition of fit
Definition of the maximized Pseudo R2 (McFadden, 1973)
ˆ

ln
L
M Full 
2
R  1
ln Lˆ M Intercept 
Mfull = Model with predictors
Mintercept = Model without predictors
L̂ = Estimated likelihood
20
Selected References
Berg, Nadrew, Eduardo Borentzstein and Catherine Patillo « Assessing Early Warning Systems:
How have they worked in practice?, IMF Working Paper, WP/04/52
Borio, Claudio (2003) “Towards a macroprudential framework for financial supervision and
regulation?”, BIS Working Paper No 128
Duttagupta, Rupa and Paul Cashin (2008) “The Anatomy of Banking Crises”, IMF Working Paper,
WP/08/93
Kaminssky, Graciela L., Saúl Lizondo, and Carmen M. Reinhart (1998) “Leading Indicators of
Currency Crises”, IMF Staff Papares, Vol. 45, Issue 1 (march), pp.1-48
Kraay, Aart and Vikram Nehru (2004) “When is External Debt Sustainable?”, World Bank Policy
Research Paper 3200
Laeven, Luc and Fabian Valencia (2008) “Systemic Banking Crises: A New Database”, IMF
Working Paper, WP/08/224
21
Backup
Countries in panel 1970-2007:
Albania, Algeria, Angola, Argentina, Armenia, Australia, Austria, Azerbaijan, Bangladesh, Barbados, Belarus,
Belgium, Belize, Benin, Bolivia, Bosnia and Herzegovina, Botswana, Brazil, Brunei, Bulgaria, Burkina Faso,
Burundi, Cambodia , Cameroon, Canada, Cape Verde, Cayman Islands , Central African Rep., Chad, Chile,
China, P.R., Colombia, Comoros , Congo, Dem. Rep. of, Congo, Rep. of, Costa Rica, Côte d’Ivoire, Croatia,
Czech Republic, Denmark, Djibouti, Dominica, Dominican Republic, Ecuador, Egypt, El Salvador, Equatorial
Guinea, Eritrea, Estonia, Ethiopia , Fiji , Finland, France, Gabon, Gambia, The, Georgia, Germany, Ghana,
Gibraltar , Greece, Grenada, Guatemala, Guinea, Guinea-Bissau, Guyana , Haiti , Honduras, China, P.R.:
Hong Kong, Hungary, Iceland, India, Indonesia, Iran, I.R. of, Ireland, Israel, Italy, Jamaica, Japan, Jordan,
Kazakhstan , Kenya, Korea, Kuwait, Kyrgyz Republic , Lao People’s Dem. Rep., Latvia, Lebanon, Lesotho,
Liberia, Libya , Lithuania, Luxemburg, Macedonia, Madagascar, Malawi, Malaysia, Maldives , Mali,
Mauritania, Mauritius, Mexico, Moldova , Mongolia , Morocco, Mozambique, Myanmar, Namibia , Nepal,
Netherlands, New Zealand, Nicaragua, Niger, Nigeria, Norway, Pakistan, Panama, Papua New Guinea,
Paraguay, Peru, Philippines, Poland, Portugal, Romania, Russia, Rwanda, São Tome and Principe, Saudi
Arabia, Senegal, Serbia, Republic of, Sierra Leone, Singapore, Slovak Republic, Slovenia, South Africa,
Spain, Sri Lanka, Sudan, Suriname, Swaziland, Sweden, Syrian Arab Republic, Switzerland, Tajikistan,
Tanzania, Thailand, Togo, Trinidad and Tobago, Tunisia, Turkey, Turkmenistan, Uganda, Ukraine, United
Kingdom, United States, Uruguay, Uzbekistan, Venezuela, Vietnam, Yemen, Zambia, Zimbabwe
22