Cato Institute Forum
Argentina Is Unique - Implications for
Sovereign Debt Restructurings
Elena Duggar, Group Credit Officer - Sovereign Risk, Moody’s Investors Service
11 December 2013
Agenda
1. Why the Case of Argentina Is Unique in the Historical Context
2. Implications for Sovereign Debt Restructurings
Cato institute – 11 December 2013
2
Related Research
Sovereign Default Research: www.moodys.com/sdr
»
Sovereign Defaults Series Compendium: The Aftermath of Sovereign Defaults, October
2013
»
The Role of Holdout Creditors and CACs in Sovereign Debt Restructurings, Moody’s
Special Comment, April 2013
»
US Court Ruling on Argentina’s Debt Could Have Limited Implications for Sovereign
Debt Restructurings, Moody’s Special Comment, December 2012
Cato institute – 11 December 2013
3
Disclaimers
»
The author is not an attorney, and this presentation does not offer any legal opinions or
interpretations.
»
Neither the author nor Moody’s Investors Service advocates any particular approach or
policy with respect to sovereign debt restructurings.
Cato institute – 11 December 2013
4
1
Why the Case of Argentina Is Unique
Cato institute – 11 December 2013
5
Main Conclusion: Recent History Shows that the Case of
Argentina Is Unique - Holdout Litigation Has Not Been
an Obstacle to Sovereign Debt Restructurings
» The case of Argentina was the only one of the 34 modern era sovereign bond exchanges
that resulted in persistent litigation
» There were several unique features about Argentina’s experience:
– The economic and banking crisis at the time of default were extremely severe
– The losses involved in the debt exchange were large (about 70%)
– The debt exchange was large and very complex
– The negotiation process was contentious
– The amount of holdout debt was relatively large
– The litigation process was prolonged
» In general, sovereign bond restructurings have been resolved quickly, without severe
creditor coordination problems, and involving little litigation
Cato institute – 11 December 2013
6
Sample coverage: The 34 Sovereign Bond Exchanges
Since 1997
Capturing the rise in emerging market bond finance since the late 1990s and focusing on
bond restructurings, where creditor coordination problems were expected to be more
pronounced.
Modern Era Sovereign Bond Restructurings by Region, 1997-Q1 2013
Central America and
the Caribbean
32%
Europe
29%
Africa
15%
South America
21%
Asia
3%
Source: Moody’s
Cato institute – 11 December 2013
7
Extremely Severe Crisis
Argentina experienced extremely severe economic, banking, debt and political crises in the
2001-2002 period.
Argent ina
Argentina Debt-to-GDP (%) and Exchange Rate (%
change)
Jamaica
Debt-to-GDP (lha)
160
20
10
140
10
5
120
0
-10
100
0
Percent
Percent change
15
Exchange rate (rha)
-5
-20
80
-30
60
-40
-50
40
-10
-60
20
-15
t-3
t-2
t-1
t
t+1
t+2
Year of init ial dist ressed exchange
t+3
Percent change
Real GDP (%
change)
-70
0
-80
1998 1999 2000 2001 2002 2003 2004
Source: Moody’s
Cato institute – 11 December 2013
8
Very Large Investor Losses in the Debt Exchange
Losses Suffered by Investors and Closing
Time in the Debt Exchange
» The 2005 Argentinean foreign debt
exchange imposed losses on investors
of over 70% in NPV terms
100
» Compared to the average investor
losses in sovereign debt exchanges
since 1997 of 47%
80
70
Loss (%)
» From the 34 sovereign debt exchanges
since 1997 only 6 exchanges imposed
such large losses – Russia, Cote
d’Ivoire, Argentina, the Seychelles,
Ecuador, and Greece
90
60
50
40
30
20
10
0
0
Source: Moody’s
5
10
15
20
25
Time to negotiate (months)
Cato institute – 11 December 2013
9
Complex Debt Exchange
» Argentina’s debt exchange was large and very complex:
– Almost US$ 80bn of debt, equivalent to over 50% of GDP
– 152 different bond series
– Across 8 governing laws
– 6 different currencies
– And dispersed creditor structure
» The typical sovereign bond restructuring involved only a few bond series, across one or
two governing laws
» The average sovereign bond exchange involved US$ 16bn of debt. The only debt
exchange that was larger than Argentina in terms of the amount of debt involved was
Greece’s March 2012 restructuring of US$ 273bn of debt
Cato institute – 11 December 2013
10
Unlike the Case of Argentina, the Typical Sovereign
Bond Restructuring Was Resolved Relatively Quickly
Time from First Offer to Closing of Debt
Exchange
» On average sovereign bond
restructurings closed 10 months after
the government had announced its
intention to restructure
» Only 4 exchanges took longer than an
year to resolve – Dominican Republic,
Russia, Argentina, and Cote d’Ivoire
» Delays were related to the parallel
restructuring of official sector and
commercial loan debt, and the civil
conflict in Cote d’Ivoire
80%
70%
% of exchanges
» And 7 months after the first offer or the
start of negotiations with creditors
90%
60%
50%
40%
30%
20%
10%
0%
Closed within 2 Closed within 4 Closed within 10
months
months
months
Source: Moody’s
Cato institute – 11 December 2013
11
Unlike Argentina, High Level of Participation In
Sovereign Debt Restructurings Was The Norm Outcome
Participation Rates in Sovereign Bond
Exchanges
» Creditor participation averaged 95%
» All but 2 cases had 90% or higher
participation rate
100%
» Moreover, 74% of exchanges had
95% or higher participation rate
80%
» However, later on, participation rates
increased to 93% in Argentina and
close to 100% in Dominica
% of exchanges
» In only two exchanges, holdout
creditors represented more than 10%
immediately after the exchange Argentina (76% participation rate)
and Dominica (72% participation
rate)
90%
70%
60%
50%
40%
30%
20%
10%
0%
Participation rate at least Participation rate at least
90%
95%
Source: Moody’s
Cato institute – 11 December 2013
12
Unlike Argentina, Holdout Litigation Has Not Been an
Obstacle
» Only 1 of the 34 modern era sovereign bond exchanges resulted in persistent litigation
– The case of Argentina.
» Only a few other court cases have been filed over the years and they have generally not
represented an obstacle to the conclusion of debt exchanges:
– 1 lawsuit in the case of Ecuador in 2001 by a commercial bank
– 1 lawsuit in the case of Dominica in 2005 by the Export-Import Bank of Taiwan
– 1 lawsuit filed in the case of Grenada in 2006 by the Export-Import Bank of Taiwan
– 1 lawsuit in the case of Pakistan in 1999 by one small creditor.
» “Runs to the courthouse” have been the exception rather than the rule in sovereign debt
crises
Cato institute – 11 December 2013
13
2
Implications for Sovereign Debt
Restructurings
Cato institute – 11 December 2013
14
In Practice Implications for Future Sovereign Debt
Restructurings Could Prove Limited
» The experience of Argentina with respect to its debt restructuring had several unique
features and was an outlier in the historical context
» Pari passu clauses in sovereign bond contracts are not all equal
» CACs are prevalent in sovereign bond contracts
» Exit consents can be used to amend bond contracts during debt exchanges
Cato institute – 11 December 2013
15
Pari Passu Clauses Are Not All Equal (1)
» Pari passu clauses in sovereign bond contracts exist in three different formulations:
– “Low risk” formulation, commonly used before 1990 provides that “the bonds rank pari passu with
all External Indebtedness” -- generally considered not readily susceptible to the ratable payment
interpretation.
– “Medium risk” formulation might state that “the bonds will rank pari passu in priority of payment and
in rank of security”
– “High risk” formulation adds “and shall be paid as such” to the “rank equally” -- the last two versions
are more susceptible to the ratable payment interpretation as they explicitly require equal treatment
at the moment of payment.
» Empirical evidence suggests that sovereign bonds increasingly incorporated the two
“more risky” versions only in the 1990s and 2000s
» About two-thirds of sovereign bonds issued in the 1990s and almost half of the bonds
issued in the 2000s contained the “low risk” version of the pari passu clause.
Cato institute – 11 December 2013
16
Pari Passu Clauses Are Not All Equal (2)
» Further, Argentina’s case is unique in that Argentina passed the “Padlock Law” in 2005
– It forbade the government to settle with holdout creditors who had refused to participate in the
restructuring
– The law effectively granted preferential status to one group of creditors
– Should the Courts interpret the ruling in a narrow sense related to the Padlock Law, rather than in a
broader sense related to the pari passu clause, the implications of the ruling for other sovereigns
could be limited
» Finally, new sovereign bond contracts are currently being modified to specify that the
pari passu clause should not be interpreted to mean “ratable payment”
Cato institute – 11 December 2013
17
CACs and Exit Consents Have Played a Significant Role
in Sovereign Bond Exchanges
CACs and Exit Consents in Sovereign
Bond Exchanges
» 35% of sovereign debt
exchanges relied on using CACs
or exit consents to bind a larger
share of creditors in the
restructuring
» Exit consents allow a majority
group of creditors to change the
non-financial terms of the old
bonds
35%
30%
% of exchanges
» CACs allow a supermajority of
creditors to amend the debt
instrument’s payment terms
40%
25%
20%
15%
10%
5%
0%
CACs
Exit consents CACs and/or exit
consents
Source: Moody’s
Cato institute – 11 December 2013
18
CACs Are Prevalent in Sovereign Bond Contracts
» CACs set a lower threshold for the completion of a debt restructuring, limiting the impact
of the court ruling
» Empirical evidence shows that the vast majority of foreign-law sovereign bond contracts
contain CACs:
– CACs are commonly included in almost all New York law issuances after 2003
– English law bonds typically contain modification clauses
» In local-law bonds, CACs can be retroactively inserted by act of legislation, as was done
in the case of Greece
» Thus, the pari passu clause risk is more applicable to New York law bonds issued before
2003, which contain “high risk” pari passu clauses but no CACs
» Finally, it can be possible in future restructurings to legally subordinate holdout bonds by
using exit consents which require lower level of bondholder approval (for example, to
waive the pari passu and negative pledge clauses on old debt)
Cato institute – 11 December 2013
19
Appendix
Cato institute – 11 December 2013
20
More Details (1)
Sovereign Bond Exchanges Since 1997
Time to Closing of Exchange (Months)
From Initial
From
From Start of
Default Announcement
Negotiations
Initial Default
Date
Country (NR = not rated
at the time)
Distressed Exchange Details
Announcement of
Restructuring (or
Missed Payment)
Start of
Negotiations/
First Offer
Final
Exchange
Offer
Distressed
Exchange
Date
Aug-1998
Russia
LC debt (GKO and OFZ)
Aug-98
Aug-98
Mar-99
May-1999
10
10
10
yes
Russia
FC debt (MIN FIN III)
May-99
Nov-99
Jan-00
Feb-2000
19
10
4
yes
Russia
FC debt (PRIN and IAN)
Dec-98
May-99
Feb-00
Aug-2000
25
21
16
yes
Ukraine
LC T-bills held domestically
Aug-98
Aug-98
Aug-98
Sep-1998
n.a.
2
2
no
Ukraine
LC T-bills held by non-residents
Aug-98
Sep-98
Sep-98
Sep-1998
n.a.
2
1
no
Ukraine
FC Chase-Manhatt an loan
Aug-98
Aug-98
Sep-98
Oct-1998
1
3
3
no
Ukraine
FC ING bond and Merrill Lynch bond
May-99
May-99
Jul-99
Aug-1999
12
4
4
yes
yes
Sep-1998
In Default During
the Bond
Exchange?
Ukraine
FC Eurobonds
Jan-00
Jan-00
Feb-00
Mar-2000
19
3
3
Dec-1999
Pakistan
Eurobonds
Jan-99
Nov-99
Nov-99
Dec-1999
n.a.
12
1
no
Aug-1999
Ecuador
External debt
Aug-99
Jun-00
Jul-00
Aug-2000
13
13
3
yes
Ecuador
FC domestic bonds
Sep-99
Aug-00
Aug-00
Aug-2000
13
12
1
yes
Mar-2000
Cote d'Ivoire (NR)
Brady bonds
Apr-08
Apr-08
Sep-09
Apr-2010
122
25
25
yes
Nov-2001
Argent ina
Domestic debt
Nov-01
Nov-01
Nov-01
Nov-2001
1
1
1
no
Argent ina
External debt
Nov-01
Sep-03
Jan-05
Feb-2005
40
40
18
yes
Jun-2002
Moldova
Eurobond
Jun-02
Jun-02
Aug-02
Oct-2002
5
5
5
yes
Jan-2003
Paraguay (NR)
Domestic debt due in 2003-06
Oct-03
Oct-03
Nov-03
Jul-2004
19
10
10
yes
May-2003
Uruguay
LT FC bonds (external and domestic)
Mar-03
Mar-03
Apr-03
May-2003
n.a.
3
3
no
Jul-2003
Nicaragua
CENI bonds FC-denom. payable in LC
Jun-03
Jun-03
Jul-03
Jul-2003
n.a.
2
2
no
Nicaragua
CENI bonds FC-denom. payable in LC
Apr-08
Apr-08
Jun-08
Jun-2008
60
3
3
(yes) [1]
Jul-2003
Dominica (NR)
LC bonds (domestic and external)
Dec-03
Dec-03
Apr-04
Jun-2004
12
7
7
(yes) [2]
H2-2004
Cameroon (NR)
Domestic debt
H1-2005
12
Dec-2004
Grenada (NR)
Global bond and domestic debt
Oct-04
Dec-04
Sep-05
Nov-2005
12
14
12
yes
May-2005
Dominican Rep.
International bonds
Apr-04
Apr-04
Apr-05
May-2005
1
14
14
no [3]
Dec-2006
Belize
Privat e ext ernal debt
Aug-06
Aug-06
Dec-06
Feb-2007
3
7
7
no
Jul-2008
Seychelles (NR)
External debt
Oct-08
Mar-09
Dec-09
Jan-2010
19
16
11
yes
Dec-2008
Ecuador
Global bonds
Nov-08
no neg.
Apr-09
May-2009
6
7
no neg.
yes
Feb-2010
Jamaica
Domestic debt
Jan-10
Jan-10
Jan-10
Feb-2010
n.a.
2
2
no
Jan-2011
Cote d'Ivoire (NR)
Treasury bills (short-term)
Jan-11
Oct-11
Oct-11
Dec-2011
12
12
3
yes
Cote d'Ivoire (NR)
Eurobond coupon
Jan-11
Oct-12
Nov-12
Nov-12
23
23
1
yes
St. Kitts and Nevis (NR)
Domestic bonds and external debt
Jun-11
Jul-11
Feb-12
Mar-2012
5
10
9
yes
St. Kitts and Nevis (NR)
Domestic loans (debt-land swap)
Jun-11
Jul-11
Apr-12
Apr-2012
6
11
10
yes
Mar-2012
Greece
Greek and foreign law bonds
Jul-11
Jul-11
Feb-12
Mar-2012
n.a.
9
9
no
Sep-2012
Belize
2029 Superbond
Aug-12
Aug-12
Feb-13
Mar-13
7
8
8
yes
Feb-2013
Jamaica
Domestic debt
Feb-13
Feb-13
Feb-13
Feb-13
n.a.
1
1
no
18
10
7
Nov-2011
Exchange Average
yes
Source: Moody’s, IMF country reports, Sturzenegger and Zettelmeyer (2005), and Diaz-Cassou, Erce-Dominguez and Vazquez-Zamora (2008).
Notes: Time is rounded to the month. [1] Payments suspended due to legal investigation. [2] Bonds under legal dispute. [3] In default on loans
Cato institute – 11 December 2013
21
More Details (2)
Creditor Participation Rates and Legal Features of Sovereign Bond Exchanges Since 1997
Country (NR = not
rated at the time)
Russia
Russia
Russia
Ukraine
Ukraine
Ukraine
Ukraine
Ukraine
Pakistan
Ecuador
Ecuador
Cote d'Ivoire (NR)
Argent ina
Argent ina
Distressed
Exchange
Date
May-1999
Feb-2000
Aug-2000
Sep-1998
Sep-1998
Oct-1998
Aug-1999
Mar-2000
Dec-1999
Aug-2000
Aug-2000
Apr-2010
Nov-2001
Feb-2005
Jun-2002
Jan-2003
May-2003
Moldova
Paraguay (NR)
Uruguay
Oct-2002
Jul-2004
May-2003
Jul-2003
Nicaragua
Nicaragua
Dominica (NR)
Cameroon (NR)
Grenada (NR)
Dominican Rep.
Belize
Seychelles (NR)
Ecuador
Jamaica
Cote d'Ivoire (NR)
Cote d'Ivoire (NR)
St. Kitts and Nevis (NR)
St. Kitts and Nevis (NR)
Greece
Belize
Jamaica
Jul-2003
Jun-2008
Jun-2004
H1-2005
Nov-2005
May-2005
Feb-2007
Jan-2010
May-2009
Feb-2010
Dec-2011
Nov-12
Mar-2012
Apr-2012
Mar-2012
Mar-13
Feb-13
Initial Default
Date
Aug-1998
Sep-1998
Dec-1999
Aug-1999
Mar-2000
Nov-2001
Jul-2003
H2-2004
Dec-2004
May-2005
Dec-2006
Jul-2008
Dec-2008
Feb-2010
Jan-2011
Nov-2011
Mar-2012
Sep-2012
Feb-2013
Exchange Average
Governing Law
(Main)
Local law
Local law
English law
Local law
Local law
Luxembourg and German law
English law
NY law
Local law
NY law
Local law
8 governing laws
English law
Local law
Local law most, NY law, English
law, and Japanese law
Local law
Local law
English law
Local law
NY law and local law
NY law
NY law
English law
NY law
Local law
Local law
NY law
Local law
Local law
Local law and some Foreign law
NY law
Local law
Creditor
Structure
Dispersed
Dispersed
Dispersed
Dispersed
Dispersed
Concentrated
Conc. for ING bond; Disp. for other
Concentrated for majorit y of bonds
Concentrated
Concentrated
Concentrated
Dispersed
Dispersed
Concentrated
Dispersed
Dispersed
Concentrated
Concentrated
Dispersed
Dispersed
Concentrated
Concentrated
Dispersed
Concentrated
Concentrated
Concentrated
Concentrated
Dispersed
Concentrated
Concentrated
Participation
Rate
95% for residents, 88.5% for non-resident s
90%
99%
100%
100% (ING bond) and 50% (other)
99%
99%
97%
very high
99.98%
very high
76.2% in 2005, plus 69.5% in 2010, tot aling
92.6% (96% for domestic bondholders)
100%
96%
93% (98.8% domest ic and 89.2% nonresident)
very high
very high
72% (by 2012, reached close t o 100%)
94% for external
97%
98.1%
100%
91%
99%
96%
100%
100%
almost universal
96.9% (100% for domestic)
100% (CAC t riggered aft er 86.2% part.)
99%
95%
no
partly
CACs
Used in
Exchange?
no
no
no
no
no
no
no
yes
no
no
no
no
no
no
yes
partly
yes
no
yes
Exit
Consents
Used?
no
no
no
no
no
no
no
no
no
yes
no
yes
no
no
yes
yes
n.a.
yes
yes
yes,
voluntary
partly (ext ernal bonds) no
no
no
no
no
no
yes
yes
yes
yes
no
no
no
no
yes
no [1]
yes
yes
yes
yes
n.a.
no
no
yes
no
no
n.a.
no [2]
yes
yes
n.a.
ret roactively insert ed
yes
no
n.a.
no
no
n.a.
no
no
no
Included in Original
Bonds?
no
no
yes
partly
yes
no
yes
yes
n.a.
yes
yes
no
Included in
New Bonds?
no
no
no
yes
yes
no
n.a.
yes
yes
no
Source: Moody’s, IMF country reports, Sturzenegger and Zettelmeyer (2005), Diaz-Cassou, Erce-Dominguez and Vazquez-Zamora (2008), and Andritzky (2006).
Notes: [1] Each series of new bonds carried a “mandatory debt management” feature that required Dominica to retire from the market a specified percentage of the original principal
amount of that series in each year. [2] Early redemption clause triggered.
Cato institute – 11 December 2013
22
Elena Duggar
Group Credit Officer for Sovereign Risk
Credit Policy Group
Moody’s Investors Service
elena.duggar@moodys.com
Cato institute – 11 December 2013
23