## _wp11220

## Source details

**Canonical URL:** [_wp11220](https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2011/_wp11220.pdf)

## Other formats

- [Markdown version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2011/_wp11220.pdf.md)
- [Structured JSON version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2011/_wp11220.pdf.json)

---

### I. Introduction
- Motivation: The global crisis of 2008-2010 increased debt ratios in Central America, Panama and Dominican Republic (CAPDR), raising questions about debt sustainability and fiscal space.
- Objective: Use a revised debt intolerance approach (building on Reinhart, Rogoff and Savastano (2003) and Reinhart and Rogoff (2009)) to estimate country-specific debt targets and an index of debt intolerance using the Institutional Investor Rating (IIR).
- Methodological modifications versus prior literature:
  - Dynamic panel data framework instead of static cross-section.
  - Generalized Method of Moments estimators (Arellano and Bond (1991); Arellano and Bover (1995)).
  - Single-equation estimation for all countries (no ad-hoc IIR “clubs”).
  - Threshold IIR levels selected by correspondence with credit ratings of major rating agencies.
  - Use of general government debt dataset (Abbas et al. (2010)) instead of external debt only.

### II. Theoretical and empirical benchmarks for debt targets
- Representative threshold findings cited (preserved exactly):
  - 25 IMF (2003) Debt Sustainability (historical primary surplusses) Emerging markets
  - 25 IMF (2008, 2009) Effectivness of countercyclical fiscal policy conditional on initial level of debt Emerging markets
  - 60-75 IMF (2008, 2009) Effectivness of countercyclical fiscal policy conditional on initial level of debt Industrial countries
  - 64 Caner et al. (2010) Threshold least squares regression Developing countries
  - 77 Caner et al. (2010) Threshold least squares regression Industrial countries
  - 90 Reinhart and Rogoff (2010) Histograms relating debt to growth Industrial countries
  - 90 Reinhart and Rogoff (2010) Histograms relating debt to growth Emerging markets
  - 90 Kumar and Woo (2010) Panel growth equation All countries
  - 170-180 Ostry et al. (2010) Debt Sustainability (fiscal reaction function) Industrial countries
- Theoretical calibration examples:
  - Aiyagari et al. (1998): optimal debt at 66 percent of GDP (US).
  - Weh-Sol (2010): optimal debt at 62 percent of GDP (South Korea).

### III. Original debt intolerance approach (RRS / RR) — key elements
- Empirical observation: Some countries default or experience payment problems at relatively low external debt-to-GDP — labeled "debt intolerance."
- RRS methodology (summary):
  - Use Institutional Investor Rating (IIR) (0 to 100 scale) as measure of perceived creditworthiness.
  - Explanatory variables include initial external debt-to-GDP, history of default (1970–2009), episodes of inflation over 40 percent.
  - Two-step approach: sample division into IIR "clubs" and cross-section regression on club averages to predict country-specific external debt thresholds consistent with target IIR regions.
- RRS empirical findings reproduced:
  - Over half of observations for countries with solid credit history have external debt well below 35 percent of GDP; more than half of observations for countries with a history of default correspond to debt levels over 40 percent of GDP — suggesting external debt ratio higher than 35 percent of GDP may increase default risk for debt-intolerant countries.

### IV. Revised approach to debt intolerance (methodology)
- Data and sample:
  - 120 countries, developed and developing, 1989 to 2009 (unbalanced panel).
  - Observations averaged into four periods: 1989-1994, 1995-1999, 2000-2004, 2005-2009 (5-year averages; first period noted as 6 years).
  - Debt measure: General Government Debt/GDP (Abbas et al. (2010)).
- Empirical pattern:
  - Scatter indicates a C-shaped relationship between IIR and debt-to-GDP with a marked difference at IIR ≈ 65: above 65 debt tolerance is higher; below 65 debt tolerance is lower (Japan example: 2000-2004 IIR 85 with debt-to-GDP 160; 2005-2009 IIR 88 with debt-to-GDP 197).
- Estimated specification (dynamic panel error-correction fixed effects with lagged IIR):
  - IIR_it = α + β1 IIR_i,t−1 + β2 D_it + β3 D_it^2 + δ d_it + γ CGDP_it + T τ_t + v_i + u_it
  - Definitions:
    - IIR_it: Institutional Investor Rating
    - D_it: General Government Debt/GDP
    - d_it: contemporaneous inflation (>10 percent) and default dummies
    - CGDP_it: Per capita GDP
    - τ_t: time trend
    - v_i: country-specific fixed effects
- Estimation methods and choices:
  - Arellano-Bond difference GMM and Arellano-Bover/System GMM (including forward orthogonalized version).
  - Instruments number controlled and kept at 32 for GMM estimations.
  - Four-period panel used for main estimations; countries with IIR below 25 dropped (sample reduced from 120 to 102).

### V. Estimation results (selected exact coefficients and diagnostics)
- Key coefficient estimates (preserved as reported across estimators):
  - Lagged IIR:
    - 0.555 (Arellano-Bond 1-step)
    - 0.535 (System GMM)
    - 0.522 (System GMM orth)
  - Debt/GDP coefficient (negative and significant in GMM estimations):
    - -0.337 ABond 1-step
    - -0.318 System GMM
    - -0.347 System GMM orth
  - Debt/GDP squared (positive and significant in System GMM specifications):
    - 0.00116* ABond
    - 0.00114*** System GMM
    - 0.00129*** System GMM orth
  - Inflation dummy:
    - -2.247*** System GMM
    - -2.412*** System GMM orth
  - Default dummy:
    - -1.487*** ABond
    - -1.462*** System GMM
    - -1.106** System GMM orth
  - Per capita GDP:
    - 0.000502*** ABond
    - 0.000550*** System GMM
    - 0.000594*** System GMM orth
- Diagnostics and sample details:
  - Observations: 271
  - Number of ifs (countries): 102
  - No. of instruments: 32
  - Hansen test p-values: 0.607 (OLS/FEs), 0.518 (System GMM)
  - Arellano-Bond AR(1) test p-values around 0.016–0.024
- Robustness checks (selected exact outcomes):
  - Excluding EU countries (sample to 88) did not qualitatively change results.
  - Annual data estimation: lagged IIR coefficient 0.78; debt coefficient -0.2; inflation/default/per-capita GDP similar.
  - External debt used instead of total debt produced a very small coefficient: 0.02.
  - CAPDR regional dummy interacting with debt had a small positive significant coefficient.
  - Instruments number for GMM estimations kept at 32.

### VI. Application to CAPDR — operationalization and country-specific targets
- Operational equation for IIR adjustment (first differences, holding inflation, default, per-capita GDP, trend and fixed effects constant):
  - ΔIIR_t = β̂1 ΔIIR_{t−1} + β̂2 (D_t − D_{t−1}) + β̂3 (D_t^2 − D_{t−1}^2)
- Benchmarking:
  - Calculations start from 2010 IIR and debt levels.
  - Mapping IIR to credit-ratings (three major agencies) yields thresholds:
    - Upper threshold IIR for unambiguous investment grade: 58.7
    - Lower threshold IIR for unambiguous non-investment grade: 51.3
    - Countries with IIR between 58.7 and 51.3 are “borderline.”
- CAPDR specific 2010 levels and target debt-to-GDP (exact values from Table 6):
  - Panama: IIR 56.9 (2010); target Investment grade IIR 58.7; Debt 2010 40.0 -> Target Debt 32.7 (% of GDP)
  - Costa Rica: IIR 55.1 (2010); target Investment grade IIR 58.7; Debt 2010 37.5 -> Target Debt 25.4 (% of GDP)
  - El Salvador: IIR 45.5 (2010) Non-inv. grade -> Target Borderline IIR 51.3; Debt 2010 51.7 -> Target Debt 34.4 (% of GDP)
  - Guatemala: IIR 45.3 (2010) Non-inv. grade -> Target Borderline IIR 51.3; Debt 2010 24.1 -> Target Debt 11.4 (% of GDP)
  - Dominican Republic: IIR 40.8 (2010) Non-inv. grade -> Target Borderline IIR 51.3; Debt 2010 36.9 -> Target Debt 14.3 (% of GDP)
  - Honduras: IIR 30.9 (2010) Highly speculative -> Target Speculative IIR 38.6; Debt 2010 26.1 -> Target Debt 8.6 (% of GDP)
  - Nicaragua: IIR 23.9 (2010) Substantial risk -> Target Highly speculative IIR 25.8; Debt 2010 66.3 -> Target Debt 31.3 (% of GDP)
- Debt intolerance index (debt required to reach an IIR of 50 in 2010 for CAPDR — exact values from Table 8):
  - Panama: 55.64
  - Costa Rica: 51.08
  - El Salvador: 45.31
  - Guatemala: 37.35
  - Dominican Republic: 36.32
  - Nicaragua: 28.38
  - Honduras: 21.10
- Observations on heterogeneity and operational caveats:
  - Similar IIRs can correspond to very different debt-to-GDP levels (example: Guatemala and El Salvador have almost identical IIR in 2010 but very different debt ratios).
  - Countries with limited domestic debt markets and high concessional external debt (e.g., Honduras, Nicaragua) require large relative debt reductions to reach the next credit-ratings rung; concessionality of debt may require adjustment when comparing thresholds.

### VII. Conclusions and policy implications
- Methodological contributions:
  - Dynamic panel GMM estimation of a debt intolerance equation using general government debt and a large cross-country sample (120 countries, 1989–2009).
  - Mapping IIR to credit ratings to set operational IIR targets tied to sovereign rating categories.
  - Proposal of a debt-intolerance index: the debt level required to reach a given IIR (illustrated with IIR = 50).
- Policy-relevant findings for CAPDR (exact required reductions as reported):
  - Panama and Costa Rica: reducing 2010 debt-to-GDP by 7 and 12 percentage points respectively would move them into unambiguous investment grade by the IIR/credit-rating mapping, all else equal.
  - El Salvador, Guatemala, Dominican Republic: to reach “borderline” would require reductions of:
    - Dominican Republic: 22.6 percentage points
    - El Salvador: 17.3 percentage points
    - Guatemala: 12.7 percentage points
  - Honduras and Nicaragua: require substantial debt reduction to access limited non-concessional markets; adjustments needed to account for concessionality of existing debt.
- Caveats and recommended policy priorities:
  - Estimates hold other determinants of IIR fixed (inflation/macroeconomic stability, default history, per-capita GDP).
  - Policy efforts to improve institutional capacity, macroeconomic stability, and growth are advisable alongside debt reduction.
  - The debt-intolerance index enables cross-country comparisons, but further work is needed to explain differences, identify peer groups, and isolate key drivers of debt intolerance.

*Source — IMF staff summary of _wp11220._*

### References .............................................................................................................

### _wp11220 - References

### I. Introduction
- Motivation: The global crisis of 2008-2010 increased debt ratios in Central America, Panama and Dominican Republic (CAPDR), raising questions about debt sustainability and fiscal space.
- Objective: Use a revised debt intolerance approach (building on Reinhart, Rogoff and Savastano (2003) and Reinhart and Rogoff (2009)) to estimate country-specific debt targets and an index of debt intolerance using the Institutional Investor Rating (IIR).
- Key methodological modifications versus prior literature:
  - Use a dynamic panel data framework rather than static cross-section.
  - Use Generalized Method of Moments estimators (Arellano and Bond (1991); Arellano and Bover (1995)).
  - Estimate a single equation for all countries instead of dividing into ad-hoc IIR “clubs”.
  - Select threshold IIR levels by correspondence with credit ratings of major rating agencies.
  - Use a general government debt dataset (Abbas et al. (2010)) instead of external debt only.

### II. Background: debt targets in theory and practice
- Theoretical benchmarks and examples:
  - Aiyagari et al. (1998) calibrate optimal debt at 66 percent of GDP (US).
  - Weh-Sol (2010) calibrates optimal debt at 62 percent of GDP (South Korea).
- Empirical/practical approaches and representative thresholds:
  - IMF (2003): average sustainable debt for emerging markets around 25 percent of GDP.
  - IMF (2008, 2009): fiscal policy effectiveness declines above 60-75 percent of GDP (industrial countries) and above 25 percent of GDP (emerging markets).
  - Ostry et al. (2010): debt limit for industrial countries ranges from 170 to 180 percent of GDP.
  - Reinhart and Rogoff (2010): median growth falls when debt rises above 90 percent of GDP.
  - Caner et al. (2010): threshold level 77.1 percent (industrial countries) and 64 percent (developing countries).
  - Kumar and Woo (2010): threshold where debt negatively affects growth around 90 percent of GDP.
- Table 1 (empirical studies of debt thresholds) entries preserved:
  - 25 IMF (2003) Debt Sustainability (historical primary surplusses) Emerging markets
  - 25 IMF (2008, 2009) Effectivness of countercyclical fiscal policy conditional on initial level of debt Emerging markets
  - 60-75 IMF (2008, 2009) Effectivness of countercyclical fiscal policy conditional on initial level of debt Industrial countries
  - 64 Caner et al. (2010) Threshold least squares regression Developing countries
  - 77 Caner et al. (2010) Threshold least squares regression Industrial countries
  - 90 Reinhart and Rogoff (2010) Histograms relating debt to growth Industrial countries
  - 90 Reinhart and Rogoff (2010) Histograms relating debt to growth Emerging markets
  - 90 Kumar and Woo (2010) Panel growth equation All countries
  - 170-180 Ostry et al. (2010) Debt Sustainability (fiscal reaction function) Industrial countries

### III. The original debt intolerance approach (RRS / RR)
- Empirical observation: Some countries default or experience payment problems at relatively low external debt-to-GDP—termed "debt intolerance".
- RRS methodology summary:
  - Use Institutional Investor Rating (IIR) (0 to 100 scale) as measure of perceived creditworthiness.
  - Explanatory variables: initial external debt-to-GDP, history of default (1970–2009), episodes of inflation over 40 percent.
  - Two-step approach:
    1. Divide sample into "clubs" based on average IIR and standard deviation (Club A, B, C; Club B further into Types I–IV based on external debt/GNP).
    2. Cross-section regression using averages across time periods to explain IIR as function of debt, default history, and inflation; then predict country-specific debt thresholds consistent with target IIR regions.
- Key RRS empirical findings reproduced in paper:
  - Over half of observations for countries with solid credit history have external debt well below 35 percent of GDP; more than half of observations for countries with a history of default correspond to debt levels over 40 percent of GDP — from which RRS suggest external debt ratio higher than 35 percent of GDP may increase default risk for debt-intolerant countries.
- Table 2 (example): Predicted IIR and Debt Intolerance Regions for Argentina and Malaysia (selected values preserved, e.g. Argentina External debt/GNP 0 -> Institutional Investor Rating 51.4 I; Malaysia Example points retained as presented).

### IV. Revised approach to debt intolerance (methodology)
- Data and sample:
  - 120 countries, developed and developing, from 1989 to 2009 (unbalanced panel).
  - Observations averaged into four periods: 1989-1994, 1995-1999, 2000-2004, 2005-2009 (5-year averages; first period noted as 6 years).
  - Debt measure: General Government Debt/GDP (Abbas et al. (2010)).
- Empirical relationship:
  - Scatter plot suggests a C-shaped relationship between IIR and debt-to-GDP with a marked difference at IIR ≈ 65: above 65 debt tolerance is higher; below 65 debt tolerance is lower. Example: Japan 2000-2004 IIR 85 with debt-to-GDP 160; 2005-2009 IIR 88 with debt-to-GDP 197.
- Debt intolerance equation (estimated specification):
  - Dynamic panel one-way error correction fixed effects model with lagged IIR:
    - IIR_it = α + β1 IIR_i,t−1 + β2 D_it + β3 D_it^2 + δ d_it + γ CGDP_it + T τ_t + v_i + u_it
  - Variables:
    - IIR_it: Institutional Investor Rating
    - D_it: General Government Debt/GDP
    - d_it: matrix of contemporaneous inflation (>10 percent) and default dummies
    - CGDP_it: Per capita GDP
    - τ_t: time trend
    - v_i: country-specific fixed effects
- Estimation method:
  - Use Arellano-Bond difference GMM and Arellano-Bover/System GMM (including forward orthogonalized version) to address endogeneity and fixed effects.
  - Instruments number controlled (kept at 32 for GMM estimations).
  - Four-period panel (four 5-year averages) used for main estimations; countries with IIR below 25 dropped (sample reduced from 120 to 102).

### V. Estimation results (Table 4 findings preserved)
- Common coefficient signs and key point estimates (from System GMM / System GMM orth columns):
  - Lagged IIR: 0.555 (Arellano-Bond 1-step), 0.535 (System GMM), 0.522 (System GMM orth) — positive and significant, indicating persistence.
  - Debt/GDP coefficient: negative and significant in GMM estimations (example values: -0.337 ABond 1-step; -0.318 System GMM; -0.347 System GMM orth).
  - Debt/GDP squared: positive and significant in System GMM specifications (example values: 0.00116* ABond; 0.00114*** System GMM; 0.00129*** System GMM orth).
  - Inflation dummy: negative and significant (e.g., -2.247*** System GMM; -2.412*** System GMM orth).
  - Default dummy: negative and significant in GMM (e.g., -1.487*** ABond; -1.462*** System GMM; -1.106** System GMM orth).
  - Per capita GDP: positive and significant (e.g., 0.000502*** ABond; 0.000550*** System GMM; 0.000594*** System GMM orth).
  - Trend and constant reported as in Table 4.
- Diagnostics:
  - Observations: 271; Number of ifs: 102; No. of instruments: 32.
  - Hansen test p-value: 0.607 (OLS/FEs), 0.518 (System GMM) — suggesting instruments appropriate.
  - Arellano-Bond AR(1) test p-values around 0.016–0.024 (reject no AR(1) as expected); AR(2) test not presented for four-period panel.
- Sensitivity and robustness checks:
  - Excluding EU countries (sample to 88) did not qualitatively change results.
  - Annual data estimation: lagged IIR coefficient higher (0.78), debt coefficient lower (-0.2), inflation/default/per-capita GDP similar.
  - External debt used instead of total debt produced a very small coefficient (0.02) in alternative specification.
  - Revenue/GDP proxy and regional dummies tested; CAPDR regional dummy interacting with debt had small positive significant coefficient.

### VI. Application to CAPDR (operationalization and results)
- Operational equation for IIR adjustment (first differences, holding inflation, default, per-capita GDP, trend and fixed effects constant):
  - ΔIIR_t = β̂1 ΔIIR_{t−1} + β̂2 (D_t − D_{t−1}) + β̂3 (D_t^2 − D_{t−1}^2)  (Equation presented as in text).
- Benchmark: calculations start from 2010 IIR and debt levels.
- Determining IIR thresholds via mapping to credit ratings (three major agencies: Moody’s, Fitch, S&P):
  - Upper threshold IIR for unambiguous investment grade: 58.7
  - Lower threshold IIR for unambiguous non-investment grade: 51.3
  - Countries between 58.7 and 51.3 are “borderline” with mixed ratings.
- CAPDR country-specific results (Table 6 preserved):
  - Panama: IIR 56.9 (2010); target Investment grade IIR 58.7; Debt 2010 40.0 -> Target Debt 32.7 (% of GDP)
  - Costa Rica: IIR 55.1 (2010); target Investment grade IIR 58.7; Debt 2010 37.5 -> Target Debt 25.4 (% of GDP)
  - El Salvador: IIR 45.5 (2010) Non-inv. grade -> Target Borderline IIR 51.3; Debt 2010 51.7 -> Target Debt 34.4 (% of GDP)
  - Guatemala: IIR 45.3 (2010) Non-inv. grade -> Target Borderline IIR 51.3; Debt 2010 24.1 -> Target Debt 11.4 (% of GDP)
  - Dominican Republic: IIR 40.8 (2010) Non-inv. grade -> Target Borderline IIR 51.3; Debt 2010 36.9 -> Target Debt 14.3 (% of GDP)
  - Honduras: IIR 30.9 (2010) Highly speculative -> Target Speculative IIR 38.6; Debt 2010 26.1 -> Target Debt 8.6 (% of GDP)
  - Nicaragua: IIR 23.9 (2010) Substantial risk -> Target Highly speculative IIR 25.8; Debt 2010 66.3 -> Target Debt 31.3 (% of GDP)
- Observations on heterogeneity:
  - Similar IIRs can correspond to very different debt-to-GDP levels (e.g., Guatemala and El Salvador have almost identical IIR in 2010 but very different debt ratios).
  - Countries with limited domestic debt markets and high concessional external debt (e.g., Honduras, Nicaragua) require large relative debt reductions to reach next credit-ratings rung; concessionality of debt may require adjustment when comparing thresholds.
- Debt intolerance index (Table 8): level of debt required to reach an IIR of 50 in 2010 for CAPDR (Debt required to reach an IIR of 50 in 2010):
  - Panama: 55.64
  - Costa Rica: 51.08
  - El Salvador: 45.31
  - Guatemala: 37.35
  - Dominican Republic: 36.32
  - Nicaragua: 28.38
  - Honduras: 21.10

### VII. Conclusions and policy implications
- Methodological contribution:
  - Dynamic panel GMM estimation of debt intolerance equation using general government debt and a large cross-country sample (120 countries, 1989–2009).
  - Mapping IIR to credit ratings provides operational IIR targets tied to sovereign rating categories.
  - Proposed index of debt intolerance: the debt level required to reach a given IIR (example used: IIR = 50).
- Policy-relevant findings for CAPDR:
  - Panama and Costa Rica: reducing 2010 debt-to-GDP by 7 and 12 percentage points respectively would move them into unambiguous investment grade by the IIR/credit-rating mapping, all else equal.
  - El Salvador, Guatemala, Dominican Republic: to reach “borderline” (candidate for upgrade) would require reductions for Dominican Republic (22.6 percentage points) and El Salvador (17.3 percentage points); Guatemala requires 12.7 percentage points.
  - Honduras and Nicaragua: require substantial debt reduction to access limited non-concessional markets; adjustments needed to account for concessionality of existing debt.
- Caveats and recommendations:
  - Estimates hold other determinants of IIR fixed (inflation/macroeconomic stability, default history, per-capita GDP). Policy efforts to improve these determinants (institutional capacity, macro stability, growth) are advisable alongside debt reduction.
  - The proposed debt-intolerance index enables cross-country comparisons, but further work needed to explain differences and identify peer groups and key drivers of debt intolerance.

*Italic: Source — IMF staff summary of _wp11220 - References (extracted content)._

### REFERENCES

### _wp11220 - REFERENCES

### References

- Abbas, S. Ali, Nazim Belhocine, Asmaa ElGanainy, and Mark Horton, 2010, “A Historical Public Debt Database,” Working Paper 10/245 (Washington, International Monetary Fund).

- Aiyagari, S. Rao, and Ellen R. McGrattan, 1998, “The Optimum Quantity of Debt,” Journal of Monetary Economics, Vol. 42 (3), pp. 447–469.

- Arellano, M., and S. Bond, 1991, “Some Tests of Specification for Panel Data: Monte Carlo Evidence and an Application to Employment Equations,” Review of Economic Studies, Vol. 58, pp. 277297.

- –––––, and O. Bover, 1995, “Another Look at the Instrumental Variables Estimation of Error Components Models,” Journal of Econometrics, Vol. 68, pp. 29–51.

- Baltagi, B. H., 2005, Econometric Analysis of Panel Data, (New York: John Wiley and Sons Ltd)

- Caner, Mehmet, Thomas Grennes, and Fritzi Koehler-Geib, 2010, “Finding the Tipping Point-When Sovereign Debt Turns Bad,” World Bank Policy Research Working Paper No. 5391, World Bank, Washington DC.

- Di Bella, Gabriel, 2008, “A Stochastic Framework for Public Debt Sustainability Analysis,” Working Paper 08/58 (Washington, International Monetary Fund).

- Everaert, Greetje, 2008, “Kenya: Selected Issues Paper, IMF Country Report No. 08/337 (Washington: International Monetary Fund).

- Floden, Martin, 2001, “The Effectiveness of Government Debt and Transfers as Insurance,” Journal of Monetary Economics 48 (2001) pp. 81–108.

- Hansen, Bruce E., 2000, “Sample Splitting and Threshold Estimation,” Econometrica, Vol. 68 No. 3 (May, 2000), pp 575–603.

- International Monetary Fund, 2003, “Public Debt in Emerging Markets, Is it too high?” World Economic Outlook, September 2003, Chapter 3,World Economic and Financial Surveys (Washington)

- –––––, 2008, “Fiscal Policy as a Countercyclical Tool,” World Economic Outlook, October 2008, Chapter 5, World Economic and Financial Surveys (Washington)

- –––––, 2009, “From Recession to Recovery: How soon and how strong?” World Economic Outlook, April 2009, Chapter 3, World Economic and Financial Surveys (Washington)

- Kumar, Manmohan S., and Jaejoon Woo, 2010, “Public Debt and Growth,” Working Paper 10/174 (Washington, International Monetary Fund).

- Ostry, Jonathan D. Atish R. Ghosh, Jun I Kim, and Mahvash S. Qureshi, 2010, “Fiscal Space.” IMF Staff Position Note 10/11 (Washington: International Monetary Fund).

- Reinhart, Carmen M., Kenneth S. Rogoff, and Miguel A. Savastano, 2003. “Debt Intolerance,” Brookings Papers on Economic Activity, I:2003 (Washington: Brookings Institution).

- Reinhart, Carmen M., and Kenneth S. Rogoff, 2009, This Time is Different; Eight Centuries of Financial Folly, Princeton University Press.

- –––––, 2010, “Growth in a Time of Debt,” prepared for the American Economic Review Papers and Proceedings.

- Roodman, David, 2006, “How to do xtabond2: An Introduction to “Difference” and “System” GMM in Stata,” Working Paper No. 103, Center for Global Development, Washington.

- Saint-Paul, Gilles, 2005, “Fiscal Policy and Economic Growth: the Role of Financial Intermediation,” Review of International Economics, 13(3), pp. 612-629.

- Shin, Yongseok, 2006, “Ramsey Meets Bewley: Optimal Government Financing with Incomplete Markets,” Department of Economics, University of Wisconsin.

- Topalova, Petia, and Dan Nyberg, 2009, “What Level of Public Debt Could India Target?” Working Paper 10/7 (Washington: International Monetary Fund).

- Weh-Sol, Moon, 2010, “Korea’s Optimal Public Debt Ratio,” SERI Quarterly, April 2010, Samsung Economic Research Institute, Seoul.

*Source: _wp11220 - REFERENCES*

---


_Source: https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2011/_wp11220.pdf_