Transcript Vivir con Deuda - Inter-American Development Bank

San Jose
February 9, 2007
Eduardo Borensztein
1
Why Now?
(when we have no crises)
• Precisely. Crises can be prevented (at least
some)
• Favorable Markets. Possibility of improving
debt management (local currency, new
instruments)
• Role of the IFIs is being reconsidered.
New ideas to make global finance safer
2
Outline
• Stylized Facts
• International Borrowing
• Domestic Borrowing
• Towards Safer Debt
3
Outline
• Stylized Facts
• International Borrowing
• Domestic Borrowing
• Towards Safer Debt
4
Public debt is not going
away…
Public Debt in Latin America and the Caribbean
120
80
Weighted average (right axis)
Average (left axis)
Median (right axis)
Weighted average excluding Argentina (right axis)
100
70
60
80
50
60
40
30
40
20
20
10
0
0
1991
1992
1993
1994
1995
1996
1997
1998
1999
2000
2001
2002
2003
2004
2005
Source: Authors' calculations based on Cowan et al. (2006).
Note: Countries included: Argentina, Bahamas, Barbados, Belize, Bolivia, Brazil, Chile, Colombia, Costa Rica, Ecuador, El Salvador, Guatemala, Guyana, Haiti, Honduras,
Jamaica, Mexico, Nicaragua, Panama, Paraguay, Peru, Trinidad and Tobago, Uruguay, and Venezuela.
5
..and domestic debt is
increasing
Composition of Public Debt in Emerging Latin American Countries
70
Domestic debt
External due to private creditors
External due to official creditors
60
50
40
30
20
10
0
1991
1992
1993
1994
1995
1996
1997
1998
1999
Source : Authors' calculations based on Cowan et al. (2006).
Note : Countries included: Argentina, Brazil, Chile, Colombia, Ecuador, El Salvador, Mexico, Panama, Peru, Uruguay, and Venezuela.
2000
2001
2002
2003
2004
6
Domestic Debt is also growing
in the smaller markets
0.9
0.8
Domestic
External due to Private Creditors
0.7
External due to Official Creditors
0.6
0.5
0.4
0.3
0.2
0.1
0
1
Countries
Included: Barbados, Belize, Bolivia, Costa Rica, Guatemala, Honduras, Jamaica, Nicaragua, Paraguay.
7
Official borrowing remains
important…
Composition of Public External Debt in Latin American and Caribbean Countries
with Limited Market Access
90
80
70
60
Bank loans
Bonded debt
Other external
Bilateral
Multilaterals
IMF
50
40
30
20
10
0
1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004
Source : Authors' calculations based on Cowan et al. (2006).
Note : Countries included: Belize, Bolivia, Costa Rica, Guatemala, Guyana, Honduras, Jamaica, Nicaragua, and Paraguay.
8
LAC debt levels are not very
high…
Public Debt around the World (weighted averages)
SAS
MNA
IND
SSA
LAC
2001–2005
ECA
1996–2000
1991–1995
EAP
0
10
20
Source: Authors' calculations based on Jaimovich and Panizza (2006).
30
40
50
60
70
80
90
9
…but debt structure matters
more than debt level
Standard & Poor's sovereign rating
Public Debt and Sovereign Rating (1995–2005)
AAA
Germany
United Kingdom
Switzerland
France
Austria
Norway
Spain Finland
CanadaDenmark
United
States
Ireland
New Zealand
Australia
Luxembourg
Sweden
Netherlands
?
AA
Iceland
Belgium
Portugal
Italy
Cyprus
Saudi Arabia
Malta
Botswana
Chile
Czech Republic
Korea
EstoniaLatvia
Thailand
China
Oman
?
A
Slovenia
Israel
BahrainBahamas
Malaysia
Poland
Qatar
and Tobago
Africa
Slovak Republic TrinidadSouth
Lithuania
El Salvador
Mexico
Croatia
Colombia
Kazakhstan
Peru Uruguay
Costa Rica
Guatemala
BBB
?
BB
Venezuela
Investment grade line
Barbados
Hungary
Tunisia
Egypt
Panama
India
Brazil
Russian Federation
Benin
Ukraine Belize
Paraguay
Turkey
B
Japan
Morocco
Philippines
Bulgaria
Senegal
Bolivia
Mongolia
Papua New Guinea
Grenada
Indonesia Argentina
Pakistan
Jordan
Ghana
Jamaica
Ecuador
0
10
20
30
40
50
60
70
80
90
100
110
Public debt (percentage of GDP)
Source : Jaimovich and Panizza (2006) and Standard & Poor's.
10
Composition of Sovereign Debt, 2000-04
(in percent of GDP)
Source: Faria and Tolosa, forthcoming IMF Occasional Paper
11
Structure of External Debt
2001–04
,
Total PPG debt
GROUP I
Total PPG debt
Concessional
Private
Official
Total PPG debt
Private
Official
Privately-held PPG debt
Bank
loans
Other
Bonds
Nonconcessional
GROUP II
Total PPG debt
Nonconcessional
Privately-held PPG debt
Other
Concessional
Bank loans
Group I consists of Costa Rica, Dominican Republic, El Salvador, Guatemala, and Panama. Group II includes Honduras and Nicaragua.
Source: Faria and Tolosa, forthcoming IMF Occasional Paper
12
Debt and Deficits
• Discussions on how to control the growth
public debt focus on the fiscal deficit
• This is not surprising as:
DEBT GROWTH = DEFICIT + SF
• We expect SF (stock-flow reconciliation) to be
a small adjustment
• But this is not the case
13
SF around the world
The Stock- Flow Adjustment
(percentage of GDP)
Sub-Saharan Africa
Latin America and the Caribbean
Middle East and North Africa
All Countries
Eastern Europe and Central Asia
East Asia and Pacific
All observations
Excluding outliers
South Asia
Advanced economies
0
1
2
3
4
5
6
7
8
9
10
Source : Campos, Jaimovich, and Panizza (2006).
14
SF around the world
Decomposition of Debt Growth in Different Regions of the World
Percentage of GDP
15
Inflation
GDP growth
0
Primary
balance
Interest
expenditure
Stock flow
adjustment
-15
IND
SAS
Caribbean
Source: Campos, Jaimovich, and Panizza (2006).
EAP
ECA
MNA
LA
SSA
15
Outline
• Facts
• International Borrowing
• Domestic Borrowing
• Towards Safer Debt
16
The emerging bond market
today
• Private international lending revived in the
1970s in the form of syndicated bank loans
• This new wave of sovereign lending ended
with the debt crisis of the 1980s
• The Brady deals restarted the market for
emerging market bonds and gave birth to a
new asset class
17
The International Market:
Large but Volatile…
• Capital flows to emerging markets tend to be
volatile and procyclical
• External factors explain a large share of this
volatility
• External debt is almost all in foreign currency
18
Mark Twain on procyclicality
A banker is a fellow who lends you his
umbrella when the sun is shining and
wants it back the minute it begins to
rain.
19
Sudden Stops and
Contagion
• Episodes of “market closure” and jump in
spreads that reverse within 2 years
• Spreads closely connected with global risk
appetite
• Effects extend to countries with little
economic link to the original event
• Uninformed investors may herd and/or chase
an index
20
The international market is
large but volatile
Emerging Markets and Latin American Spreads
2,000
1,800
1,600
Tequila
crisis
September
11 attacks
1,200
Russian crisis
1,000
800
600
Brazil
elections and
400
200
Composite
Aug-06
Oct-05
Dec-04
Feb-04
Apr-03
Jun-02
Aug-01
Oct-00
Dec-99
Feb-99
Apr-98
Jun-97
Aug-96
Oct-95
Dec-94
Feb-94
0
Apr-93
Basis points
1,400
Latin America
Source: Authors' calculations based on JPMorgan (2006) from Bloomberg (2006). Spreads are a combination of EMBI and EMBI plus indices. The Latin America index comprises the four largest debtors: Argentina, Brazil, Mexico, and
Venezuela. Weights are adjusted to consider the structural break resulting from the Argentine default.
21
Has volatility abated?
• Spreads are at record low levels; investors
show appetite for local currency bonds
• Weak or no contagion from Brazil’s election
anxiety and Argentina’s and Turkey’s crises
• Less cross correlation among EMBI assets
22
Has Contagion Disappeared?
Average Correlations among Sovereign Emerging Market Bonds and
Industrial Sector Indices of US High Yield Bonds
0.9
0.8
Tequila
crisis
Russian crisis
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0.0
Jul-94
Oct-95
Jan-97
Apr-98
Jul-99
JP Morgan EMBI Plus Index
Oct-00
Jan-02
Apr-03
Jul-04
Oct-05
Merrill Lynch U.S. High Yield Index
Source: Authors' calculations based on JPMorgan (2006) and Merril Lynch (2006) from Bloomberg (2006).
Note: Figure plots the average six-month correlation in 78 country pairs of the JPMorgan EMBI plus index and 465 sector pairs of the Merrill Lynch U.S. High Yield Index. Correlation coefficients are based on daily returns of indices.
23
A new phase?
• Policies are stronger: fiscal surpluses, reserve
accumulation, current account surpluses
• Debt structure changing to local currency
• Investor base widening; EM assets more
mainstream
• Or, in fact, are spreads too low?
24
Will the good times last?
Spreads are lower than predicted
1400
1200
1000
800
600
400
200
0
1993
1994
1995
1996
1997
1998
1999
Predicted Spreads
2000
2001
2002
2003
2004
2005
Actual Spreads
25
Predicted Spreads would be lower
with average external conditions
1400
1200
1000
800
600
400
200
0
1993
1994
1995
1996
1997
1998
1999
Spread for average External Conditions
2000
2001
2002
2003
2004
2005
Actual Spreads
26
Possibly important changes
in global markets
• Impact of the growth of hedge funds
• Impact of the growth of credit derivatives
• Is the appetite for local currency instruments
permanent?
27
Outline
• Facts
• International Borrowing
• Domestic Borrowing
• Towards Safer Debt
28
The Domestic Markets:
More Stable but Still Small
• Natural habitat of local currency instruments
• “Spare tire” for the banking system
• High volatility in LA makes bank finance a
very short-run proposition
• “Captive audience” of domestic institutional
investors (pension funds, banks)
29
Domestic bond markets in LAC
are growing but are still small
Percentage of GDP, simple average
140
120
Corporate issuers
100
Financial institutions
80
Governments
60
40
20
0
Latin America 1994
Latin America 2004
East Asia 1994
East Asia 2004
Advanced 1994
Advanced 2004
30
But in fact it is the whole
financial sector that is small
Bond Markets relative to Domestic Bank Credit
80
70
60
Corporate issuers
50
Financial institutions
40
Governments
30
20
10
0
Latin America 1994
Latin America 2004
East Asia 1994
East Asia 2004
Advanced 1994
Advanced 2004
31
Plausible Determinants of
Bond Market Development
• Scale of the market, in turn related to size
of the economy and saving rates
• Scale of firms that are potential issuers
• Market liquidity
• Institutional strength: creditor rights,
transparency, etc.
• Market microstructure
32
Other Important Factors with
Less Definite Effect
• Large domestic government debt: market
development or crowding out?
• Well established banking system:
competitor or complement?
33
Scale of the Markets
.4
Corporate bond market as a function of country size
.3
MYS
.2
KOR
TWN
THA
.1
CHL
ARG
SGP
HKG
PER
0
COL
PHL
2
4
IDN
6
Ln GDP
MEX
BRA
CHN
8
10
34
Scale of the Markets
1
Market capitalization as a function of saving rates
MYS
.8
KOR
.6
BRA
.4
CHL
PHL
COL
THA
CHN
HKG
.2
MEX
SGP
TWN
ARG
0
PER
0
.2
.4
Domestic Savings over GDP
.6
35
Scale of Firms
.4
Corporate Bond Market as a Function of Firm Size
.2
.3
KOR
CHL
.1
ARG
THA
PER
0
IDN
CHN
-.05
MEX
SGP
BRA
HKG
COL
PHL
0
.05
.1
Adjusted Firm Size
36
Investor Protection
G3 AV
Malaysia
Korea
ASIA AV
Chile
EM AV
India
Brazil
Mexico
Thailand
LAC AV
Colombia
Argentina
0
1
2
3
4
5
6
7
Source: IMF, Global Financial Stability Report (2005)
8
9
37
Effects of Large
Government Debt
• Government bonds provide a reference yield
curve
• Larger markets are needed for an efficient
microstructure
• Crowding out?
38
Share of corporate bonds in total domestic credit to private sector
Interaction between
government and private
bond market
Domestic Government Bonds and Corporate Bonds
0.04
0.03
0.02
Latin America
Other emerging markets
Fitted values
0.01
0
-0.01
-0.02
-0.03
-0.04
-0.4
-0.3
-0.2
-0.1
0
0.1
0.2
Share of domestic government bonds in total public debt
0.3
0.4
39
Investor Surveys in Argentina,
Brazil and Mexico
Do you agree or disagree with the following statements?
(1 = strongly agree,…, 5 = strongly disagree)
The yield curve provided by public bonds is
crucial for pricing corporate bonds
A large stock of public sector bonds is
important for the development of the corporate
bond market
If the yield on government bonds w ere to
increase significantly and that of private bonds
remained constant I w ould sell private bonds \
Government and corporate bonds are
substitutes in your portfolio
0
Argentina
1
2
Brazil
3
4
Mexico
40
Banks vs. Bonds
• Conventional sequence:
1) Banks
2) Bond Markets
3) Equity Markets
• But interest groups can affect this evolution, e.g.
Banks can prevent markets from developing
(Rajan-Zingales)
41
Banks and Bonds:
Substitutes or Complements?
• Banks contribute to market infrastructure: bridge
finance, distribution channels, primary dealer network.
• Banks contribute to secondary-market liquidity
• Banks often are major issuers of domestic bonds and
structured securities
• Rather than being a political force against markets,
banks and bonds seem to be held back by the same
reasons in Latin America
42
Going Forward:
New Investors, New Instruments
• New investors needed. Savings are low and markets are
small
• Institutional investors, especially pension funds, are
starting to provide volume (but not liquidity).
• Foreign investors provide more liquidity (but also
volatility). Less averse to long-term nominal instruments
(see Mexico, Brazil). Capital account restrictions must be
removed.
43
Institutional Investors
• Institutional investors have been growing and
they can play a key role in developing a
stable demand for local currency bonds
• But they could become victim of their own
success, especially at times of crisis
44
Institutional Investors
Assets of Mutual Funds and Pension Funds
(percentage of GDP)
50
45
40
140
Emerging markets (left axis)
Latin America (left axis)
Advanced economies (right axis)
120
100
35
30
80
25
60
20
15
40
10
20
5
0
1997
1998
1999
2000
2001
2002
2003
0
45
Institutional Investors
IND
Other EM
LAC
Government Bonds as a Share of Pension Fund Assets
Mexico
Argentina
Uruguay
Colombia
Peru
Chile
Brazil
Singapore
Hungary
Poland
Bulgaria
Czech Republic
Slovenia
Estonia
Korea
Thailand
Austria
Italy
Netherlands
Spain
Canada
United Kingdom
United States
Germany
0
10
20
30
40
50
60
70
80
90
100
46
Institutional Investors
Banks' Exposure to Public Sector, 2003–2005
Latin American countries
Mature
markets
(percentage of total domestic credit)
Euro area
United States
Argentina
Mexico
Brazil
Colombia
Peru
Uruguay
Chile
0
10
20
30
40
50
60
47
International Investors
• Saving is low in Latin America and this affects the size of
bond markets
• Institutional investors like pension funds do not provide
liquidity
• Foreign investors display less aversion to long-term local
currency instruments. But less stability. Need to open
capital accounts to some extent.
• In Mexico, foreign investors hold more than 50% of the
ten-year bond and more than 80% of the 20-year bond.
48
Percent of Foreign Holdings
of Domestic Bonds
USA
Uruguay
Hungary
Poland
Mexico
Turkey
Argentina
Malaysia
Japan
Brazil
Thailand
Peru
Indonesia
Korea
Bulgaria
0
5
10
15
20
25
30
35
40
45
50
55
Sources: FMI (GFSR, 2005) y Takeuchi (“Study of Impediments…” 2005, asianbondsonline.adb.org)
49
Going Forward:
New Instruments
• New Instruments: Asset Backed Securities (mortgages,
receivables, consumer loans, commercial paper)
• Less complicated enforcement of creditor rights (by
recourse to collateral)
• Can overcome firms’ small scale problem
• Strong growth in Mexico, Brazil, Chile, Argentina, but
from a very small base
• Successful securitizations for working capital to SMEs
Can structured instruments also help SMEs get longterm, investment finance?
50
Outline
• Facts
• International Borrowing
• Domestic Borrowing
• Towards Safer Debt
51
Debt sustainability
Traditional approach: Debt to GDP ratio to stay
constant
s = (r – g) d
The primary surplus must be enough to cover
interest (adjusted by growth) on the Debt to
GDP ratio
52
Actual and “Required” Primary
Surpluses
2004 Real interest rates
1992-2004 Median real interest rates
12
7
6
10
5
Jamaica
Brazil
Argentina Pre
8
Required primary surplus
Required primary surplus
Jamaica
4
3
Belize
2
Colombia
Uruguay
Panama
Ecuador
Venezuela
Dominican Republic
Costa Rica Paraguay
Trinidad and Tobago
Peru
Mexico
Guatemala
El Salvador
1
0
Uruguay
6
Argentina Post
Paraguay
4
Colombia
Brazil
2
Venezuela
Ecuador
MexicoEl Salvador
Costa Rica
Dominican Republic
Chile
Guatemala
Trinidad and Tobago
Panama
Peru
Belize
0
-1
-2
-2
-2
-1
0
1
2
Actual primary surplus
3
4
5
6
7
-2
0
2
4
6
8
10
12
Actual primary surplus
53
Sustainability Analysis under
Uncertainty
• Variables that enter debt projections have
large variance: interest rate, exchange rate,
growth, primary surplus.
• The more volatile these variables are, the
lower the Debt/GDP ratio that is sustainable
“for sure” (95% statistical confidence, for
example)
54
Required Primary Surpluses
and Credit Ratings
14
Mexico
Trinidad and Tobago
El Salvador
12
Costa Rica
Panama
Colombia
Guatemala
10
Credit rating
Peru
Jamaica
Brazil
8
Belize
Venezuela
Uruguay
Dominican Republic
6
Ecuador
Paraguay
4
y = ?0.087x + 8.6707
(0.7689)
2
R2 = 0.0064
0
-7
-6
-5
-4
-3
-2
-1
0
1
2
Required minus observed primary surplus (percentage points of GDP)
3
4
55
Sustainability Analysis under
Uncertainty
• Many sources of risk
– GDP volatility
– Terms of trade shocks
– Natural disasters
• A better debt structure can reduce risks
56
Debt Management can reduce
the risk of sovereign finance
Debt-to-GDP Ratio Distribution
0.7
Debt-to-GDP ratio
0.6
0.5
0.4
0.3
0.2
2000
2001
Foreign
currency
2002
2003
2004
Foreign currency
–local currency
2005
2006
2007
2008
2009
Foreign currency–
local currency–linked to GDP
2010
57
Lowering the risks of
sovereign finance
• Should countries avoid government debt
altogether?
• Useful purposes: investment in infrastructure
and human capital, coping with natural
disasters, crises
• But there are political economy distortions
and market failures
58
Lowering the risks of
sovereign finance
• Controlling the flow of debt
• Managing the inherited stock of debt
• Improving the international financial
architecture
59
Controlling the flow of debt
• Fiscal rules
– Structural balance target
– Debt/GDP target
• Budget institutions
– Hierarchical rules
– Transparency Rules
60
Managing the inherited
stock of debt
• Self-insurance policies
– Stabilization funds
– International reserves
• Contingent contracts
– Cat bonds, commodities, GDP-linked
• Develop the domestic bond market
– More use of local currency debt
61
Debt and
Monetary/Exchange Rate
Systems
• Exchange Rate System => Debt Structure
– Fixed rates may promote foreign currency debt
– Higher inflation discourages
long duration, local currency
Debt Structure
instruments
– There is no “local currency” instrument in dollarized
economies (except inflation-linked bonds)
• Debt Structure => Exchange Rate System
– Harder to use exchange rate flexibility in partially dollarized
economies
– Do inflation-linked bonds discourage inflation?
62
Depreciation and Debt
Sustainability
• Depreciations are important source of large
unexpected increases in Debt/GDP ratios
(SFs)
• But the effect of depreciation is not
necessarily negative. It depends on:
– Net effect of the exchange rate on taxes and
spending (including indirect effects)
– Size of Debt/GDP ratio
– Currency composition of Debt
63
Debt Management in
(fully) Dollarized Economies
• Use contingencies!
• Contingencies: natural disasters, oil prices,
economic growth, domestic inflation
• It can be achieved through derivatives or
insurance policies (cat bonds) and use
payoffs to service debt.
64
Improving the international
financial architecture
• Rollover risk. Create a “Country Insurance”
facility. Support pooling of reserves
• Contagion risk. “Emerging Market Fund”
• Currency and real economy risks. Help
develop local currency and contingent debt
instruments. “De-dollarize” multilateral
lending
65
…and much more
• Latin American sovereign debt in 1820-1913
• The political economy of public debt
• Public debt and public investment in human
and physical capital
• Debt relief, has it worked?
• The cost of default
66