Information about Asia and the Pacific Asia y el Pacífico
Journal Issue

India: Selected Issues

International Monetary Fund. Asia and Pacific Dept
Published Date:
February 2017
  • ShareShare
Information about Asia and the Pacific Asia y el Pacífico
Show Summary Details

Assessing Safe Debt Levels for India1

The Fiscal Responsibility and Budget Management (FRBM) Review Committee constituted in May 2016 is examining the pros and cons behind altering the fiscal rules attached to 2003 FRBM Act. Analysis of fiscal rules used across the world points to rising popularity of a debt rule. Based on the debt intolerance approach, staff assess a safe level of debt for India to be in the range of 60-65 percent of GDP, allowing a buffer for contingent liabilities.

1. Different types of fiscal rules are currently used across countries in order to ensure fiscal sustainability. The popularity of fiscal rules has substantially increased since 2000, and countries mostly target the budget balance, debt, and expenditures. A debt-to-GDP target is the most-widely used, though it is often used in combination with other fiscal targets. The majority of countries target a public debt-to-GDP ratio of 60 percent.

Figure 1.Fiscal Rules: Cross-Country Experience

Source: IMF Fiscal Rules Dataset.

2. A number of approaches could be used to estimate the appropriate debt target. The approaches vary from estimation of optimal public debt in agent models, to deriving a benchmark level of public debt based on the relationship between the primary deficit and the debt ratio, to determining a benchmark level of public debt based on a country’s ability to access capital markets. The various approaches focus on different motives countries could have for debt issuance and the challenges they could face. Agent models derive optimal public debt levels by optimizing the utility of agents with precautionary saving motives and borrowing constraints and are usually calibrated for advanced economies. They provide a wide range of results depending on the underlying assumptions. On the other hand, emerging markets with often relatively less developed domestic investor bases need to attract foreign investors to their debt securities.

3. The approach which focuses on debt intolerance is often used for the assessment of safe debt levels in emerging markets. Debt intolerance is the inability of emerging markets to manage levels of external debt that are manageable by advanced countries (Reinhart et al, 2003). Debt intolerance has been found to be explained by a relatively small number of variables: countries’ default and inflation histories. The non-monotonic relationship between debt intolerance and public debt-to-GDP ratios suggests the existence of country-specific debt thresholds, at which a country switches from having access to capital markets to not having such access (or the necessity to pay relatively large interest on its debt to achieve the placement).

4. Debt intolerance focuses on the ability of a country to attract investors. It is proxied by the Institutional Investor rating (IIR). It shows a country’s attractiveness for investment, with 100 indicating lowest chance for default. We group countries into three clubs: (i) countries with continuous market access (Club A); (ii) countries with intermittent market access (Club B); and (iii) countries with no market access (Club C). The division between clubs is made based on their average rating over the 2000-2014 sample period. The cutoff for any particular country to belong to Group A is a rating above the sample mean of ratings plus one standard deviation, and the cutoff to belong to Group C is a rating below the sample mean of ratings minus one standard deviation. Any country with a rating in between the abovementioned cutoffs belongs to Group B.

Country Clubs and Market Access

Note: StD denotes one standard deviation.

5. India’s attractiveness for investors as measured by the IIR has improved as public debt has declined. In the early 2000s, India’s IIR worsened with an increase in the public debt-to-GDP ratio. However, the subsequent fiscal consolidation contributed to an improvement in the country’s attractiveness to investors. Compared to other countries, India’s IIR has broadly moved in sync with the mean rating for the emerging markets, except that it has been slightly below the mean in recent years. India’s IIR has moved to the Club BI of countries with an IIR above the mean in 2004, and has stayed there since. However, India’s public debt-to-GDP level is relatively high compared to other countries in this club. In our view, as roll-over risks in India are somewhat mitigated by long average maturities and limited exposure to non-residents, these considerations improve the country’s attractiveness for investors.

India: Institutional Investor Rating and Government Debt, 2000-2014

Sources: Institutional Investor, Inc.; IMF, World Economic Outlook.

India: Institutional Investor Rating’s Developments

Sources: Institutional Investor, Inc. and IMF World Economic Outlook.

Note: Confidence intervals (shaded) show the 10th and 90th percentile of rankings.

6. An estimate of the relationship between the debt intolerance and debt level was conducted. Given the non-monotonic nature of this relationship, the coefficients are allowed to vary among different clubs. In addition, the regression has an indicator for high inflation over the sample period and a dummy variable for India (as in the equation below, where i denotes a country). Regressing the IIR on inflation and the three club’s debt levels indicates that the debt intolerance rises with high inflation and debt:

7. The analysis suggests that a safe level of debt for India is in the range of 60-65 percent of GDP. Simulation of the IIR based on different debt-to-GDP ratios suggests that India would stay above the mean of the IIR distribution in a club of countries with relatively good market access (club BI), if its debt-to-GDP ratio is lower than the threshold which lies in the range of 65-70 percent of GDP. Inclusion of some buffer to account for uncertainty (including for the materialization of contingent liabilities) suggests a safe level of debt at 60-65 percent of GDP. That said, results of the debt intolerance approach depend on the specific regression and sample period used, and confidence intervals around these point estimates.

Public Debt and Rating, Cross-Country Perspective

Sources: IMF World Economic Outlook; Institutional Investor, Inc.; and author’s calculations.

8. The result is largely in line with the estimates for other emerging market economies. Using the above approach, thresholds for India’s safe public debt-to-GDP ratio were previously estimated in the range of 40-45 (using a sample of 54 countries) and 70-75 percent (sample of 142 countries). (Topalova and Nyberg, 2010). For Kenya, Everaert (2008) estimated thresholds for its safe public debt-to-GDP ratio in the range of 35-40 percent. Most recently, thresholds for South Africa’s safe public debt-to-GDP ratio have been estimated in the range of 50-60 percent (Saxegaard, 2014).

Table 1.Simulation of Debt Benchmark
Debt (in percent of GDP)Predicted IIRClub
4054.67Club BI
4553.93Club BI
5053.20Club BI
5552.47Club BI
6051.74Club BI
6551.01Club BI
Source: Author’s calculations.
Source: Author’s calculations.
Table 2:Debt Intolerance and Debt 1/
Model 1Model 2Model 3Model 4
Debt x Club A0.312***0.300***0.311***0.299***
Debt x Club B−0.146***−0.163***−0.150***−0.169***
Debt x Club C−0.365***−0.363***−0.367***−0.364***
Inflation 2/−15.104***−0.869***−15.006***−0.871***
Adjusted R20.5750.5660.5720.565
Source: Author’s calculations.

Numbers in square brackets are standard errors. *** indicates significance at the 1 percent level, ** at the 5 percent level, and * at the 10 percent level.

Models 1 and 3 have inflation defined as a dummy showing if average inflation exceeds the 75th percentile. Models 2 and 4 use average inflation.

Source: Author’s calculations.

Numbers in square brackets are standard errors. *** indicates significance at the 1 percent level, ** at the 5 percent level, and * at the 10 percent level.

Models 1 and 3 have inflation defined as a dummy showing if average inflation exceeds the 75th percentile. Models 2 and 4 use average inflation.


    EveraertG.2008Public Debt Thresholds for Kenya,Selected Issues Paper Kenya Article IV Consultation.

    ReinhartC.K.Rogoff and M.Savastano2003Debt IntoleranceBrookings Papers on Economic Activity Vol. 1 pp. 174.

    SaxegaardM.2014Safe Debt and Uncertainty in Emerging Markets: An Application to South Africa,IMF WP/14/231 (International Monetary Fund: Washington).

    • Search Google Scholar
    • Export Citation

    TopalovaP. and D.Nyberg2010What Level of Public Debt Could India Target,?IMF WP/10/7 (International Monetary Fund: Washington).

    • Search Google Scholar
    • Export Citation

Prepared by Svitlana Maslova.

Other Resources Citing This Publication