Public Debt Sustainability Analysis in Advanced and Emerging Market Economies:
The IMF’s New Framework
April 2014 International Monetary Fund
at
Eurasian Economic Commission
Until recently, focus of debt sustainability analysis was on debt trajectories….
New MAC DSA introduced in 2013
Key features of new DSA framework
MAC DSA examples
Key features of the new MAC DSA
• More analysis for countries with potentially greater vulnerabilities
Risk - based approach
• Realism of underlying assumptions
• Risks to debt level, gross financing needs, and debt profile, macro-fiscal nexus, and bank-sovereign nexus
Multifaceted analysis
• Standardized charts and tables, fan charts, heat map, write-up
Clear and transparent presentation of results
Risk assessment based on benchmarks derived from early warning model
Toolkit is flexible and can be tailored to country-specific circumstances
Risk indicators
• Risk indicators include:
– Ratio of debt to GDP
– Ratio of gross financing needs to GDP
– Bond spreads
– Gross external financing requirements
– Share of public debt in foreign currency
– Share of debt held by non-residents
– Change in short-term public debt
Public debt
burden
Debt profile
Country’s debt burden
Market perception
Rollover risks
Signal approach
Other risk considerations
• Inclusion of guarantees and quasi-fiscal operations
– DSAs are done on a gross debt basis
– Government guarantees are sometimes included
– State-owned enterprises that impact the fiscal balance are sometimes included
• Relation between domestic debt capital market and public debt sustainability
– Banks’ holdings of sovereign debt can help lower sovereign borrowing costs
– But it may have implications for public debt accumulation and fiscal adjustments
Risk-based approach
Public debt to GDP > 50/60% for EMs/AEs
Public gross financing needs (GFN) to GDP > 10/15% for EMs/AEs
Exceptional access to Fund resources
Lower scrutiny
Higher scrutiny
Judgment
Risk-based outputs
8
Basic DSA
DSA write-up
Basic DSA
Realism of baseline assumptions
Heat map, fan charts, debt profile indicators
where relevant Customized and
contingent liabilities analysis
Lower scrutiny Higher scrutiny
Basic DSA (1)
-20
-15
-10
-5
0
5
10
15
cumulative
-20
-15
-10
-5
0
5
10
15
20
2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018
Debt-Creating Flows Above
Primary deficit Real GDP growth Real interest rate Other debt-creating flows Residual Change in gross public sector debt
projection
(in percent of GDP)
As of March 26, 2013
2011 2012 2013 2014 2015 2016 2017 2018 Sovereign Spreads
Nominal gross public debt 26.2 36.8 37.4 37.9 37.8 36.4 34.8 33.5 32.2 EMBI (bp) 450
Public gross financing needs 5.0 6.7 8.1 5.6 6.4 6.5 6.5 6.9 7.3 CDS (bp) 475
Nominal net public debt 26.2 36.8 37.4 37.9 37.8 36.4 33.5 32.2
Real GDP growth (in percent) 4.0 5.2 0.2 0.0 2.8 3.5 3.5 3.5 3.5 Ratings Foreign Local
Inflation (GDP deflator, in percent) 16.2 14.4 8.0 5.0 7.2 7.6 7.8 7.8 7.8 Moody's Aa3 Aa3
Nominal GDP growth (in percent) 20.9 20.3 8.2 5.0 10.2 11.4 11.6 11.6 11.6 S&Ps AA AA-
Effective interest rate (in percent) 3/ 4.4 5.8 5.0 1.7 1.9 2.2 2.4 2.6 2.8 Fitch AA A
Debt, Economic and Market Indicators 1/
Projections
2002-2010
Actual
Basic DSA (2)
Alternative Scenarios
Composition of Public Debt
Baseline Historical Constant Primary Balance
52
54
56
58
60
62
64
66
2011 2012 2013 2014 2015 2016 2017 2018
Gross Nominal Public Debt
(in percent of GDP)
projection
0
2
4
6
8
10
12
2011 2012 2013 2014 2015 2016 2017 2018
Public Gross Financing Needs
(in percent of GDP)
projection
0
10
20
30
40
50
60
70
2002 2004 2006 2008 2010 2012 2014 2016 2018
By Maturity
Medium and long-term
Short-term
projection
(in percent of GDP)
0
10
20
30
40
50
60
70
2002 2004 2006 2008 2010 2012 2014 2016 2018
By Currency
Local currency-denominated
Foreign currency-denominated
projection
(in percent of GDP)
Realism of baseline assumptions
Cross-country tools to help identify potential optimism
Forecast track record
Projected fiscal adjustment Boom-bust
Realism of baseline assumptions
-10
-8
-6
-4
-2
0
2
4
6
2004 2005 2006 2007 2008 2009 2010 2011 2012
Year
Real GDP Growth
Interquartile range (25-75)
Median
Italy forecast error
-1.46
7% Has a percentile rank of:
Italy median forecast error, 2004-2012:
Distribution of forecast
errors: 1/
(in percent, actual-projection)
Source: Italy 2013: Article VI Consultation Staff Report, September 2013
Forecast track record
pess
imis
tic
op
tim
isti
c
Realism of baseline assumptions
1/ For Italy the bulk of the adjustment already occurred in 2012 and the pace of consolidation slows to 1 percent of GDP in 2013. Source: Italy 2013: Article VI Consultation Staff Report, September 2013
0
2
4
6
8
10
12
14
Less
-4
-3
-2
-1
0
1
2
3
4
5
6
7
8
Distribution 4/
Italy
3-Year Adjustment in Cyclically-Adjusted
Primary Balance (CAPB) 1/ (Percent of GDP)
Mo
re
Projected fiscal adjustment
Real GDP
Growth Shock
Primary
Balance Shock
Change in the
Share of Short-
Term Debt
Exchange Rate
Shock
Contingent
Liability shock
Debt level
Foreign
Currency
Debt
Real Interest
Rate Shock
Exchange Rate
Shock
Contingent
Liability Shock
Primary
Balance Shock
Real Interest
Rate Shock
Market
Perception
Gross financing needs
Public Debt
Held by Non-
ResidentsDebt profile
External
Financing
Requirements
Real GDP
Growth Shock
Heat map
Risks to debt level and GFN
• Capture interactions between key macro variables
• Designed to balance standardization and tailoring country-specific circumstances
Macro-fiscal stress tests
15
pass-through to higher inflation
pass-through to higher interest rate
Primary balance shock
primary balance deteriorates,
interest rate increases,
inflation declines
Real GDP growth shock
Real interest rate shock
Real exchange rate shock
• Triggered automatically for countries with banking
sector vulnerabilities
• May also be relevant to assess risks related to
natural disasters, PPPs, SOEs
Contingent liabilities shocks
•Shock of 10 percent of banking system assets
•Primary balance deteriorates, interest rate increases, inflation decreases
Financial sector CL shock
Triggers AEs EMs
Private sector credit-to-
GDP (3-year cumulative
level change)
30% 15%
Loan-to-deposit ratio 1.5 1.5
Real Exchange Rate Shock
Combined Macro-Fiscal Shock
Additional Stress Tests
Baseline Contingent Liability Shock
Reform Fatigue Deep recession and confidence shock
Italy Public DSA - Stress Tests
Macro-Fiscal Stress Tests
Baseline Primary Balance Shock
Real GDP Growth Shock
Real Interest Rate Shock
110
115
120
125
130
135
140
145
150
2013 2014 2015 2016 2017 2018
Gross Nominal Public Debt(in percent of GDP)
240
250
260
270
280
290
300
310
2013 2014 2015 2016 2017 2018
Gross Nominal Public Debt
(in percent of Revenue)
15
17
19
21
23
25
27
29
31
2013 2014 2015 2016 2017 2018
Public Gross Financing Needs
(in percent of GDP)
110
120
130
140
150
160
170
2013 2014 2015 2016 2017 2018
Gross Nominal Public Debt(in percent of GDP)
240
250
260
270
280
290
300
310
320
330
340
2013 2014 2015 2016 2017 2018
Gross Nominal Public Debt
(in percent of Revenue)
15
20
25
30
35
40
2013 2014 2015 2016 2017 2018
Public Gross Financing Needs
(in percent of GDP)
Real GDP
Growth Shock
Primary
Balance Shock
Change in the
Share of Short-
Term Debt
Exchange Rate
Shock
Contingent
Liability shock
Debt level
Foreign
Currency
Debt
Real Interest
Rate Shock
Exchange Rate
Shock
Contingent
Liability Shock
Primary
Balance Shock
Real Interest
Rate Shock
Market
Perception
Gross financing needs
Public Debt
Held by Non-
ResidentsDebt profile
External
Financing
Requirements
Real GDP
Growth Shock
Heat map
Risks to debt profile
Debt Profile Vulnerabilities Upper early warning Lower early warning
20
60
87%
1 2
200
600
356
bp
1 2
5
15
17%
1 2
0.5
1
- 4.5%
1 2
EMBI External Financing
Requirement
Annual Change in Short - Term Public
Debt
Public Debt in Foreign Currency
(in basis points) 4/ (in percent of GDP) (in percent of total) (in percent of total)
15
45 24%
1 2
Public Debt Held by Non - Residents
(in percent of total)
Fan charts
Evolution of Predictive Densities of Gross Nominal Public Debt
(in percent of GDP)
100
105
110
115
120
125
130
135
140
145
2011 2012 2013 2014 2015 2016 2017 2018
10th-25th 25th-75th 75th-90thPercentiles:Baseline
Symmetric Distribution
100
105
110
115
120
125
130
135
140
145
2011 2012 2013 2014 2015 2016 2017 2018
Restricted (Asymmetric) Distribution
no restriction on the growth rate shock
no restriction on the interest rate shock
0 is the max positive pb shock (percent GDP)
no restriction on the exchange rate shock
Restrictions on upside shocks:
Source: Italy 2013: Article VI Consultation Staff Report, September 2013
DSA write-up
• Reflects staff’s overall assessment of debt sustainability risks
• Highlights vulnerabilities and country-specific circumstances that mitigate or amplify risks
• Includes background and key assumptions, including the realism of the baseline
Example - Mexico
Gross debt levels in Mexico, at 45 percent of GDP projected by end-2013, remain moderate. The broad institutional coverage of Mexico’s public debt…provides reassurance that gross liabilities of the public sector are well captured… Given Mexico’s debt structure, the pass-through of either interest rates or exchange rate shocks to the budget is relatively small over the projection period… While public debt profile indicators are below early warning benchmarks, there is a large share of debt held by non-residents—about 48 percent of total debt.
Example - Romania
The main risks arise from underperformance of GDP growth and a banking sector contingent liability, which could push the public debt ratio to nearly 50 percent of GDP…however, …banks are well capitalized with limited exposure to short-term external debt. External vulnerabilities persist due to relatively high external debt and rollover needs…
Example - Italy
Italy
Source: IMF staff.
4/ An average over the last 3 months, 21-Mar-13 through 19-Jun-13.
Real GDP
Growth Shock
2/ The cell is highlighted in green if gross financing needs benchmark of 20% is not exceeded under the specific shock or baseline, yellow if exceeded under specific shock
but not baseline, red if benchmark is exceeded under baseline, white if stress test is not relevant.
Italy Public DSA Risk Assessment
1/ The cell is highlighted in green if debt burden benchmark of 85% is not exceeded under the specific shock or baseline, yellow if exceeded under specific shock but not
baseline, red if benchmark is exceeded under baseline, white if stress test is not relevant.
Real Interest
Rate Shock
External
Financing
Requirements
Real GDP
Growth Shock
Heat Map
Upper early warning
Evolution of Predictive Densities of Gross Nominal Public Debt
(in percent of GDP)
Debt profile 3/
Lower early warning
400 and 600 basis points for bond spreads; 17 and 25 percent of GDP for external financing requirement; 1 and 1.5 percent for change in the share of short-term debt; 30
and 45 percent for the public debt held by non-residents.
Foreign
Currency
Debt
Public Debt
Held by Non-
Residents
(Indicators vis-à-vis risk assessment benchmarks)
Market
Perception
Gross financing needs 2/ Primary
Balance Shock
Real Interest
Rate Shock
Exchange Rate
Shock
Contingent
Liability Shock
Primary
Balance Shock
Debt Profile Vulnerabilities
3/ The cell is highlighted in green if country value is less than the lower risk-assessment benchmark, red if country value exceeds the upper risk-assessment benchmark,
yellow if country value is between the lower and upper risk-assessment benchmarks. If data are unavailable or indicator is not relevant, cell is white.
Lower and upper risk-assessment benchmarks are:
Change in the
Share of Short-
Term Debt
Exchange Rate
Shock
Contingent
Liability shock
Debt level 1/
1 2
Not applicable
for Italy
400
600
273
bp
1 2
17
25
51%
1 2
1.0
1.5
0.8%
1 2
Bond Spread over
German Bonds
External Financing
Requirement
Annual Change in
Short-Term Public
Debt
Public Debt in
Foreign Currency
(in basis points) 4/ (in percent of GDP) (in percent of total) (in percent of total)
100
105
110
115
120
125
130
135
140
145
2011 2012 2013 2014 2015 2016 2017 2018
10th-25th 25th-75th 75th-90thPercentiles:Baseline
Symmetric Distribution
100
105
110
115
120
125
130
135
140
145
2011 2012 2013 2014 2015 2016 2017 2018
Restricted (Asymmetric) Distribution
no restriction on the growth rate shock
no restriction on the interest rate shock
0 is the max positive pb shock (percent GDP)
no restriction on the exchange rate shock
Restrictions on upside shocks:
30
45
30%
1 2
Public Debt Held
by Non-Residents
(in percent of total)
At close to 130 percent of GDP, Italy’s public debt ratio is the second highest in the euro area and financing needs are large… Given the debt structure (average maturity around 7 years), the direct interest pass-through to the budget is relatively slow… The country’s high total external financing requirements point to continued vulnerability to changes in market perceptions.
Appendix
Indicators AEs EMs
3-year cumulative primary balance adjustment (percent of GDP) 2 2
Coefficient of variation of growth 1 1
Bond yield spreads or EMBI global spreads (basis points) 600 600
External financing requirements (percent of GDP) 25 15
Public debt held by non-residents (share of total) 45 45
Public debt in foreign-currency (share of total) n.a. 60
Annual change in the share of short-term public debt at
original maturity 1.5 1.0
Lower scrutiny
Higher scrutiny
Indicators for additional analysis
Macro-fiscal shocks in detail
26
Risk Size and Duration of Shocks Default Interaction
Primary
balance
Minimum shock of 50 percent of planned
cumulative adjustment or baseline minus
half of the 10-year historical standard
deviation, whichever is larger.
Additional borrowing leads to increase in interest rate of 25 basis
points per 1 percent of GDP worsening of the deficit.
Real GDP
growth
Real GDP growth is reduced by 1
standard deviation for 2 consecutive
years.
Primary balance deteriorates (the revenue-to-GDP ratio remains the
same as in the baseline, but the ratio of non-interest expenditures to
GDP increases as the level of spending is kept the same as in the
baseline). Deterioration in primary balance leads to higher interest
rate (see above).
Decline in growth leads to lower inflation (25 percentage points per 1
percent of GDP growth).
Interest
rate
Nominal interest rate increases by the
difference between the maximum real
interest rate over history (last 10 years)
and the average real interest rate level
over projection, or by 200bp, whichever
is larger.
Size of shock can be adjusted if risks are high (gross financing needs
are higher).
Exchange
rate
Estimate of real exchange rate
overvaluation, or maximum historical
movement of exchange rate over 10
years, whichever is highest.
Pass-through to inflation with default elasticity of 0.25 for EMs and
0.03 for AEs.