Govt Debt and Fiscal Frameworks

PRUDENT DEBT
TARGETS AND FISCAL

FRAMEWORKS
OECD economic
© Victoria Kalinina/shutterstock.com
Policy paper
July 2015 No. 15
Prudent Debt Targets
and Fiscal Frameworks
This paper has been prepared by:

Falilou Fall
Debbie Bloch
Jean-Marc Fournier
Peter Hoeller
Authorised for publication by Catherine Mann

Chief Economist and G20 Finance Deputy
The OECD Economic Policy Paper Series is published on the responsibility of the Secretary-General of the OECD.
The opinions expressed and arguments employed herein do not necessarily reflect the official views of the
Organisation or of the governments of its member countries.
An earlier version of this paper was discussed at the Working Party No. 1 of the Economic Policy
Committee. The authors are indebted to Sebastian Barnes, Hansjoerg Bloechliger, Hans
Blommestein, Ronald Downes, Catherine Mann, Christian Kastrop, Jean-Luc Schneider, Arent
Skjaeveland, Lisa von Trapp, David Turner and Peter van de Ven for their comments. The authors
thank Mabelin Villareal-Fuentes for her important statistical contribution and Celia Rutkoski for
excellent editorial assistance.
More detailed information is contained in the following papers:
Bloch, D. and F. Fall (2015), “Government Debt Indicators: Understanding the Data”, OECD
Economics Department Working Papers, No. 1228, OECD Publishing.
Fournier, J-M. and F. Fall (2015), “Limits to Government Debt Sustainability”, OECD Economics
Department Working Papers, No. 1229, OECD Publishing.
Fall, F. and J-M. Fournier (2015), “Macroeconomic Uncertainties, Prudent Debt Targets and Fiscal
Rules”, OECD Economics Department Working Papers, No. 1230, OECD Publishing.
Series: OECD Economic Policy Papers

ISSN 2226583X
The statistical data for Israel are supplied by and under the responsibility of the relevant Israeli authorities. The use
of such data by the OECD is without prejudice to the status of the Golan Heights, East Jerusalem and Israeli
settlements in the West Bank under the terms of international law.
© OECD 2015
You can copy, download or print OECD content for your own use, and you can include excerpts from OECD
publications, databases and multimedia products in your own documents, presentations, blogs, websites and
teaching materials, provided that suitable acknowledgment of OECD as source and copyright owner is given. All
requests for public or commercial use and translation rights should be submitted to [email protected]. Requests for
permission to photocopy portions of this material for public or commercial use shall be addressed directly to the
Copyright Clearance Center (CCC) at [email protected] or the Centre français d’exploitation du droit de copie
(CFC) at [email protected].
TABLE OF CONTENTS
TABLE OF CONTENTS
Executive summary ...................................................................................................................... 6
1. Introduction ......................................................................................................................... 8
2. Gross and net debt definitions: The narrow and the broad .................................................. 8
3. The effects of debt on the economy .................................................................................. 13

3.1. Limits to sustainability ................................................................................................ 14
3.2. The debt level that maximises growth ........................................................................ 18
3.3. The effectiveness of fiscal policy in stabilising the economy....................................... 19
3.4. When does debt affect growth? .................................................................................. 19
4. Defining debt targets ......................................................................................................... 20

4.1. Finding a debt threshold: Cross-country evidence ...................................................... 20
4.2. Debt and the vulnerabilities of countries to specific factors and risks ......................... 24
4.3. Defining a prudent debt target as conveyed by stochastic simulations ....................... 26
5. Designing fiscal frameworks ............................................................................................. 30

5.1. Justifications, objectives and composition of the fiscal framework .............................. 30
5.2. Setting a debt target ................................................................................................... 30
5.3. Fiscal rules ................................................................................................................. 31
5.4. Fiscal councils............................................................................................................ 44
5.5. Medium-term budgetary frameworks .......................................................................... 47
5.6. Budgetary procedures ................................................................................................ 47
6. Political economy issues ................................................................................................... 48

6.1. The credibility of the consolidation path matters for avoiding fiscal fatigue ................. 48
6.2. Political economy of fiscal consolidation..................................................................... 48
BIBLIOGRAPHY ........................................................................................................................ 50
APPENDIX................................................................................................................................. 56
OECD ECONOMIC POLICY PAPERS, NO. 15 © OECD 2015 3

TABLE OF CONTENTS
Tables
Table 1. Long-term interest rate determinants ....................................................................................13
Table 2. Estimations of growth maximising long-run debt-to-GDP ratio..............................................19
Table 3. Fiscal performance and risk indicators for emerging and lower income economies .............23
Table 4. Strengths and weaknesses of existing fiscal rules ................................................................32
Table 5. Synthesis of existing fiscal rules ............................................................................................33
Table 6. Estimations of the effectiveness of fiscal rules ......................................................................34
Table 7. Synthesis of the effects of rules with respect to fiscal discipline and stabilisation ................37
Table 8. Simulation results: Grouping of countries and fiscal rules .....................................................44
Table 9. The remit of fiscal institutions in OECD countries .................................................................46
Table A.1. Fiscal reaction function ..........................................................................................................56
Figures
Figure 1. The Standardised National Accounts framework for general government debt ......................9
Figure 2. Calculating general government net worth ............................................................................10
Figure 3. Beyond gross financial liabilities: Accounting for contingent and implicit liabilities ...............11
Figure 4. Gross and net debt concepts .................................................................................................12
Figure 5. Determination of the debt limit ...............................................................................................15
Figure 6. Debt limits drivers ..................................................................................................................16
Figure 7. Debt limits ..............................................................................................................................17
Figure 8. General government gross debt and its external component ................................................25
Figure 9. General government debt by currency denomination in selected OECD countries ..............26
Figure 10. Public debt ranges under a prudent scenario ........................................................................28
Figure 11. The effect of larger shocks ....................................................................................................29
Figure 12. The trade-off between counter-cyclicality and hitting the debt target ....................................35
Figure 13. The pre-crisis overestimation of structural balances .............................................................36
Figure 14. Illustration of a budget balance rule combined with a spending rule: The US example ........37
Figure 15. Comparison of fiscal rules by country group .........................................................................43
Boxes
Box 1. Debt data: Understanding the terminology ....................................................................................... 9

Box 2. Public debt limits .............................................................................................................................14
Box 3. Simulating the effect of shocks on public debt ...............................................................................27
Box 4. Simulated fiscal rules ......................................................................................................................40
4 OECD ECONOMIC POLICY PAPERS, NO. 15 © OECD 2015

PRUDENT DEBT TARGETS AND FISCAL FRAMEWORKS
Abstract/Résumé
Prudent debt targets and fiscal frameworks
The sharp rise in debt experienced by most OECD countries raises questions about debt indicators and
the prudent government debt level countries should target. It also raises questions about the fiscal
frameworks needed to reach the prudent debt level and to accommodate cyclical fluctuations along the
convergence path towards a prudent debt target. The objective of this paper is to define long-run prudent
debt targets for OECD countries and country-specific fiscal rules. The paper presents a comprehensive
analysis of government liabilities and assets and formulates recommendations for debt indicators. It also
reviews the different linkages between government debt and the economic activity. The lessons from these
analyses are combined with an assessment of the uncertainties surrounding the development of
macroeconomic variables to define a prudent debt target. Different fiscal rules are compared with regard
their impact on fiscal discipline and the risk of recession for country-specific fiscal rules recommendations.
JEL classification codes: E27; E61; E62; H62; H68
Keywords: Macroeconomic uncertainties, debt, fiscal policy, fiscal rules
******
Cibles de dette prudentes et cadres budgétaires
La montée rapide de l’endettement public dans la plupart des pays de l'OCDE soulève des questions sur le
niveau prudent de dette que les pays doivent cibler. Il soulève également des questions sur les cadres
budgétaires nécessaires pour les atteindre et qui fournissent une marge de manœuvre suffisante pour
faire face à des chocs négatifs. Ce document examine les dettes publiques explicites et implicites, ainsi
que les actifs qui sont utiles dans l'évaluation des risques budgétaires et la viabilité budgétaire à long
terme. Il souligne également les liens positifs et négatifs entre la dette publique et l'activité économique.
Les leçons tirées de ces analyses sont combinées avec une évaluation des incertitudes
macroéconomiques pour définir une cible prudente de dette. Différentes règles budgétaires sont
comparées à l'égard de leur impact sur la discipline budgétaire et le risque de récession, conduisant à des
recommandations pour les règles budgétaires qui tiennent compte des spécificités de chaque pays.
Classification JEL : E27 ; E61 ; E62 ; H62 ; H68
Mots-clés : incertitudes macroéconomiques, dette, politique budgétaire, règles budgétaires

Prudent Debt Targets and Fiscal Frameworks
Executive summary
During the recent crisis, government debt has increased sharply in most OECD countries, raising
questions about what government debt level countries should target and how fiscal frameworks should be
designed to bring debt down to a prudent level.
Debt targets can serve as a fiscal policy anchor to ensure the sustainability of fiscal policy and that
there is sufficient policy room to cope with adverse shocks. Prudent debt targets provide the commitment
tool that reassures markets and thereby diminishes risk premia not only for government debt, but also
lowers the cost of capital for the whole economy.
This policy paper starts by analysing different debt concepts, which are key for analysing fiscal risks
and sustainability. Government gross debt serves to monitor contractual liabilities and should include
explicit future liabilities such as civil servant pension liabilities. Gross debt should be tracked separately
from government assets, as sharp swings in asset prices can mask underlying debt developments. Net
debt, the difference between government gross debt and assets, is relevant for solvency analysis, in
particular when governments hold a sizeable amount of liquid assets. Implicit and other off-balance sheet
liabilities should also be estimated and monitored to assess fiscal risks.
To define a prudent debt target, it is necessary to establish a threshold beyond which debt has
adverse effects on economic activity or debt developments become unsustainable. Debt sustainability
largely depends on economic growth, the interest rate and the capacity of governments to run primary
balances that ensure that the government meets its liabilities. Debt default limits are currently not binding
in most countries thanks to very low interest rates. But such limits are state dependent. At high debt levels,
countries can lose market confidence and see their borrowing rates increase steeply. Countries should
thus endeavour to steer clear of their default limit. The currently high default limits cannot be an anchor for
setting the prudent debt target as harmful effects of debt on economic activity are likely to kick in well
before.
The anchor of the prudent debt target should therefore be based on the analysis of the impact of debt
on the economy. An assessment of the effect of debt on economic activity suggests that beyond a debt
threshold, government debt can undermine economic activity and the ability to stabilise the economy.
Different channels through which debt can affect the economy are assessed. Empirical evidence gathered
from the literature shows that: i) high government debt levels are associated with lower growth (above 80
to 100% of GDP), though causality is probably running both ways; ii) the ability to stabilise the economy
decreases at debt ratios beyond around 75% of GDP; iii) when a specific role for government debt in
financing public infrastructure is taken into account, estimations find a positive but limited “optimal”
government debt level at 50-80% of GDP; and iv) government debt also provides a safe asset in a very
liquid market, thus easing liquidity constraints. Therefore, low levels of debt are welfare enhancing.
In addition, the empirical cross-country evidence suggests different debt thresholds, defined as the
turning point at which negative effects of debt on the economy kick in, for three groups of countries:
• For higher-income countries, a debt threshold range of 70 to 90% of GDP.

• For euro area countries, the debt threshold is lower, as they do not control monetary policy.
Given the no-bail-out clause, the absence of debt pooling, a higher dependency on foreign
financing and difficulties in adjusting to shocks, the debt threshold is 50-70%.

• For the emerging economies the debt threshold is even lower at 30 to 50% of GDP as they are
exposed to capital flow reversals.
These debt thresholds are used to anchor prudent debt targets. Prudent debt targets should be set to
avoid an overshooting of the debt thresholds in the case of large adverse shocks. Prudent debt targets
take into account uncertainties surrounding macroeconomic variables and are thus country-specific.
Stochastic simulations, which take into account past macroeconomic shocks and fiscal behaviour, indicate
prudent debt targets of 15 percentage points of GDP below the debt threshold on average for the
advanced OECD as well as the euro area countries, with a considerable cross-country variation.
A well-designed fiscal framework ensures that prudent debt targets are adhered to over the cycle. The
fiscal framework should have two objectives: promoting fiscal discipline and permitting stabilisation
policies. Five complementary components of the fiscal framework are considered: a debt target, a fiscal
rule, fiscal councils, budgetary processes and medium-term budgeting.
In designing the fiscal framework, the prudent debt target should serve as the reference point to
define numerical fiscal rules. Fiscal rules should be based on observable variables thus reducing
uncertainties, when setting policy. Rules based on estimated variables, such as potential output, are
subject to measurement errors that can mislead policy-making. A combination of a budget balance rule
and an expenditure rule seems to suit most countries well. A budget balance target encourages hitting the
debt target. And, well-designed expenditure rules appear decisive to ensure the effectiveness of a budget
balance rule and can foster long-term growth. Moreover, to provide flexibility in the face of large shocks
that cause a considerable slippage from the budget balance target, a mandatory minimum debt reduction
rule (debt break) can be introduced to guarantee that measures will be taken to offset unexpected
increases in the debt level.
Simulations show that the capacity of fiscal policy to mitigate shocks and their adverse consequences
on debt trajectory uncertainty varies substantially across countries. In particular, highly indebted countries
have less potential to counteract large adverse shocks. However, stronger growth loosens the constraints
on fiscal policy that aims at lowering debt to the prudent debt level.
Fiscal councils can underpin transparency and thereby foster fiscal discipline and the credibility of
fiscal rules. The adoption of fiscal rules, in particular of complex rules, increases the need for transparency.
A fiscal council can also help address the common-pool problem, in particular in federal countries, by
pointing out the externalities of fiscal policies generated by sub-central governments that may benefit them
at the expense of others. The effectiveness of fiscal rules with regard to the stabilisation objective is
enhanced when rules are embedded in a medium-term budgeting framework.
The timing of consolidation, a clear mandate to reform and strong majority support are among the
political economy elements that can help overcome “fiscal consolidation fatigue”.

1. Introduction
Most OECD countries have seen their government debt rise sharply during the global financial crisis.
The OECD weighted average gross financial liability-to-GDP ratio increased from 73% of GDP in 2007 to
111% in 2013. It is the highest ratio since the aftermath of the Second World War. Such high debt levels
raise questions about their sustainability, and some euro area countries have entered into adjustment
programmes in an attempt to control their spiralling debt. The adjustment programmes include important
welfare state reforms, and fall hard on the most vulnerable, as illustrated by the explosion of poverty rates
in Greece, Portugal and Spain since 2007.
The sharp rise in debt in most OECD countries raises questions about the prudent debt level
countries should target. It also raises questions about the fiscal frameworks needed to reach them and to
accommodate cyclical fluctuations along the path towards a prudent debt target, creating fiscal space to
react to future shocks and taking into account countries' specificities.
In Section 2 of the paper, the pros and cons of using different government debt indicators are
reviewed. Section 3 examines why governments should set prudent debt targets. The issue of how to set
government debt targets is analysed through three lenses: the sustainability of government debt, the
impact of government debt on the economy, and finally the prudent debt level exploring the effects of
macroeconomic shocks and other fiscal risks (Section 4). In Section 5, the design and the components of
fiscal frameworks are discussed. In particular, fiscal rules, institutions and budgeting processes are
studied. In Section 6, the political economy issues that can help overcome “fiscal consolidation fatigue” are
analysed.
2. Gross and net debt definitions: The narrow and the broad
Government gross financial liabilities and gross debt are the headline indicators of government
indebtedness. But such liabilities provide only a partial view of all government liabilities and fiscal risks. For
a better assessment of the government fiscal position, the gross debt indicator should be complemented
by explicit and probable future liabilities and by defining broader indicators that take into account other
liabilities depending on their features (contractual, implicit or contingent). Financial and non-financial
assets add further information for a more complete view of current and future government debt
sustainability.
Debt data are often a source of confusion, as various definitions, coverage and valuation methods
exist. Bloch and Fall (2015) explore the various issues that arise in defining and measuring debt, and
examines other data which should be considered, both within and beyond general government debt, to
better track and analyse fiscal risks and sustainability issues. Box 1 summarises the main features of
various government debt measures.

Box 1. Debt data: Understanding the terminology

Differences in headline debt indicators can be confusing. Studies often refer seamlessly to gross debt or gross
financial liabilities, net debt or net financial liabilities, without clarifying the terms being used. However, the details do
matter, and can affect analyses, comparisons and conclusions. Following the latest Standardised National Accounts
framework, Figure 1 highlights the relationships between gross and net debt, and between gross and net financial
liabilities as reported in the government financial balance sheet.
Figure 1. The Standardised National Accounts framework for general government debt
General
General government gross General government gross government
- = net financial
financial liabilities financial assets
liabilities
General government Gr
gross debt debt
 Monetary gold and SDRs
 Monetary gold and SDRs  Currency and deposits
 Debt securities General


Currency and deposits
Debt securities
-  Loans government
 Loans  Insurance pension and net debt
 Insurance pension and standardised guarantees
standardised guarantees  Other accounts payable
 Other accounts payable
Non-debt instruments Non-de

bt instruments
 Financial derivatives and  Financial derivatives and
employee stock options employee stock options
 Equity and investment fund  Equity and investment fund
shares/unitsnd shares/unitsnd
Note: The dotted line highlights an alternative definition of general government net debt, used by the OECD, which takes into account
all financial assets, and not only those related to debt instruments, as all financial assets may be used to redeem debt.
The following definitions can be distinguished:

• General government gross financial liabilities: all financial liabilities reported in the National Accounts
financial balance sheet for the general government sector.
• General government gross debt: generally defined as all financial liabilities reported on the National
Accounts financial balance sheet for the general government sector except “non-debt” instruments (financial
derivatives and employee stock options, equity and investment funds).
• General government gross financial assets: all financial assets reported in the National Accounts financial
balance sheet for the general government sector.
• General government net financial liabilities: the difference between general government gross financial
liabilities and general government gross financial assets.
• General government net debt: the difference between general government gross debt and general
government gross financial assets, excluding financial assets pertaining to financial derivatives, employee
stock options, equity and investment funds. Alternatively, general government net debt may exclude all
financial assets.
• General government net financial worth: the inverse of general government net financial liabilities (general
government gross financial assets less general government gross financial liabilities).

Box 1. Debt data: Understanding the terminology (cont.)
Within the National Accounts framework, it should be possible to go a step further to calculate general
government net worth, taking into account not only financial assets but also the non-financial assets of the general
government (Figure 2). In practice, these data are currently only available for a handful of OECD countries. Using flow
data, a calculation of the change in government net worth over the past 20 years shows that net worth has declined in
many countries and in some by a substantial amount (Bloch and Fall, 2015).
Figure 2. Calculating general government net worth
General General General

government government government General
non-financial + gross financial - gross financial = government
assets assets liabilities net worth
Produced assets
debt
 Fixed assets =
 Intellectual
property product General
 Inventories government
 Valuables net financial worth
Non-produced
assets
 Natural resources
 Intangible non-
produced assets
The terminology can therefore be expanded to include:
• General government non-financial assets: produced and non-produced assets of the general government
sector as reported in the National Accounts balance sheet for non-financial assets.
• Produced non-financial assets: includes tangible fixed assets such as buildings and machinery, as well as
intellectual property, inventories and valuables.
• Non-produced non-financial assets: includes natural resources such as patents, leases and purchased
good-will.
Finally, beyond the National Accounts framework, there are a number of future liabilities which are not
necessarily accounted for, but which are important for discussions of future pressures on government finances.
Contingent liabilities, for example, are of increasing interest to analysts and policy makers. These uncertain liabilities
are either explicit – those that are defined by contract or law, or implicit – those for which there is a moral or historical
obligation (Figure 3).

Box 1. Debt data: Understanding the terminology (cont.)

Figure 3. Beyond gross financial liabilities: Accounting for contingent and implicit liabilities
General government gross

General government contingent and implicit liabilities
financial liabilities
General government Explicit contingent Implicit liabilities

gross debt
liabilities
 Monetary gold and
 Obligations for future
SDRs Other social security
 Currency and deposits Guarantees explicit benefits 3
 Debt securities contingent  Future bailouts (of
 Loans liabilities public enterprises,
 Insurance, pension1 and
financial institutions,
standardised guarantees
subnational
 Other accounts payable One-off  Legal governments, strategic
guarantees claims private firms)
 Loan and against  Future natural disaster
other debt the state2 relief
Non-debt instruments instrument  Insurance  Future environmental
guarantees schemes2 cleanup
 Financial derivatives and  Other  Other  Other implicit
employee stock options one-off contingent liabilities
 Equity and investment guarantees
fund shares/units nd
1. Includes explicit unfunded or underfunded civil servant pension liabilities.

2. Refers only to explicit contingent liabilities. In the case of legal claims against the state, this could be claims relating to
privatisations, liquidation of agencies, or personnel management. In the case of insurance schemes, this could refer to flood or
crop insurance, for example (Cebotari , A.(2008), “Contingent Liabilities: Issues and Practice”, IMF Working Paper, No. 08/245.).
3. Excludes explicit unfunded or underfunded civil servant pension liabilities.
The key terms related to these future liabilities follow:

• Contingent liabilities: liabilities which only exist if a certain event occurs. They are recorded if the probability
of occurrence is high.
• Explicit contingent liabilities: contingent liabilities defined by contract or law, such as guarantees, deposit
1
insurance schemes, and other legal claims which might be exercised against the state.
• Guarantees: generally related to a loan, refers to an agreement whereby the government agrees to repay
any amount outstanding in the event of default. Government guarantees may also refer to price guarantees,
exchange rate or revenue guarantees, particularly in relation with public-private partnerships. The probability
of standardised guarantees being called is generally calculable, and should be estimated and reported in
the National Accounts framework within the financial balance sheet.
• One-off guarantees: guarantees which are particular in nature, and not possible to assess in terms of the
risk being called. They are not included in the financial balance sheet.
• Implicit liabilities: future potential liabilities, which are not expressly codified, but where a moral obligation or
a historical expectation for the government to act exists. Implicit liabilities include future social security
benefits, future bailouts (not covered by current guarantee agreements) or future natural disaster relief.
Currently, only a few countries report their contingent liabilities fully, although there are efforts underway by
2
statistical offices and fiscal councils to provide a fuller picture of future fiscal risks.
________________________
1. Earlier National Accounts methodology allows for unfunded or under-funded public employee pension liabilities to be treated as contingent.
However, six OECD countries (Australia, Canada, Iceland, New Zealand, Sweden and the United States) record these liabilities on their
balance sheets, resulting in debt positions which are overstated relative to those countries where large unfunded or under-funded liabilities for
public employee pensions are not yet recorded.
2. See for example, Barnes and Smyth (2013) for Ireland and Eurostat (2015).

The difference between gross financial liabilities, which include non-debt instruments such as shares
and financial derivatives, and gross debt is generally small, while the difference between net financial
liabilities and net debt can be significant for some countries (Figure 4). The breakdown of debt and asset
instruments also allows for calculating specific indicators as needed, for example to measure debt in
relation to only the most liquid assets (Figure 4).
Gross and net financial debt measures remain currently the key indicators for fiscal analysis and
surveillance. The gross debt indicator serves to monitor government contractual liabilities, and should be
tracked separately from government assets, as sharp swings in asset prices can mask underlying debt
developments. The net debt indicator is relevant for solvency analysis, in particular when the government
holds liquid assets. Beyond these two headline indicators, comprehensive balance sheets are necessary to
evaluate net worth developments (Bloch and Fall, 2015), but are currently available for only a few
countries. Implicit and other off-balance-sheet liabilities should also be estimated and monitored to assess
fiscal risks (Kopits, 2014), though fiscal risk assessment is still in its infancy (Bloch and Fall, 2015).
Figure 4. Gross and net debt concepts

2012
Per cent of GDP Net financial liabilities
250
Net debt
Net debt based on liquid assets
200
Gross debt
Gross financial liabilities
150
100
50
Per cent of GDP

100
50
-50
-100
-150
12http://dx.doi.org/10.1787/888933223995
Note: Data for Chile, Japan and Korea are non-consolidated. Data for Mexico refer to 2009 and data for Switzerland to 2011. Data for
Iceland may include financial derivatives which were not reported in the National Accounts as such. Data include unfunded pension
liabilities for Australia, Canada, Iceland, New Zealand, Sweden and the United States.
Source: OECD (2014), "Detailed National Accounts: Financial balance sheet, consolidated and Financial balance sheet, non-
consolidated", OECD National Accounts Statistics (database), accessed 17 October 2014; OECD Economic Outlook 95 database,
June 2014.
Markets do not seem to react more to net financial liabilities than to gross financial liabilities. The
market reaction is investigated with regressions determining the long-term interest rate (Table 1).

Coefficients associated with gross and net debt are very close to each other (Table 1, Columns 1 and 2)
after controlling for the inflation rate, past levels of the growth rate, the short-term interest rate and country
fixed effects. However, estimations suggest that markets react more to gross debt in euro area countries,
while markets are relatively more sensitive to net debt in other OECD countries (Table 1, Columns 3 and
4). This result still holds when looking at the effect of both indicators on the real long-term interest rate
(Table 1, Columns 5 and 6).
Table 1. Long-term interest rate determinants
(1) (2) (3) (4) (5) (6)

10-year 10-year 10-year 10-year 10-year real 10-year real
Dependent variable
interest rate interest rate interest rate interest rate interest rate interest rate
Core inflation 0.079 0.078 0.084 0.080
[0.057] [0.058] [0.057] [0.058]
Short-term interest rate 0.55*** 0.55*** 0.55*** 0.55***
[0.037] [0.035] [0.037] [0.035]
Short-term real 0.63*** 0.62***
interest rate [0.039] [0.037]
Lag (Real growth) -0.088*** -0.090*** -0.081*** -0.088*** -0.084*** -0.096***
[0.022] [0.023] [0.022] [0.023] [0.024] [0.024]
EMU -0.28 -0.27 -1.95*** -0.49* -1.65** -0.016
[0.31] [0.30] [0.70] [0.29] [0.76] [0.33]
Lag (Gross public debt) 0.013** 0.006 0.020***
[0.005] [0.004] [0.005]
Lag (Gross public debt)
0.023** 0.025**
* EMU
[0.010] [0.010]
Lag (Net public debt) 0.016*** 0.014*** 0.018***
[0.004] [0.004] [0.005]
Lag (Net Public debt) * 0.006 0.009
EMU [0.006] [0.006]
Sample 1985-2013 1985-2013 1985-2013 1985-2013 1985-2013 1985-2013
Year fixed effects no no no no no no
Country fixed effects yes yes yes yes yes yes
N 624 618 624 618 672 670
Rho 0.53 0.50 0.55 0.51 0.59 0.56
R2 0.676 0.688 0.680 0.687 0.487 0.480
Note: Panel regression with AR(1) Prais-Winsten correction and panel heteroskedasticity-robust standard errors. Standard errors are
in parenthesis. *** significant at 1% level, ** significant at 5% level and * significant at 10% level.
Source: OECD calculations.
In the following analyses, the focus is on gross financial liabilities as they offer the most complete
dataset in terms of country and time coverage and government primary balances have reacted more to
gross financial liability changes than to net financial debt variations (Fournier and Fall, 2015). Also, gross
financial liabilities are more comparable across countries than financial assets due to valuation and
accounting issues.
3. The effects of debt on the economy
This section looks at the cross-country evidence on the limits to sustainability, the nexus between
debt and growth, the effect of debt-financed public investment on growth and the consequences of debt on
the effectiveness of fiscal policy to stabilise the economy. Theory provides little guidance, as the "optimal"
debt level depends on model ingredients and assumptions.

3.1. Limits to sustainability
As illustrated by the debt intolerance phenomenon, countries can quickly lose market confidence and
see their borrowing rates increase steeply, derailing their debt trajectories. The methodology developed in
this section provides a benchmark for debt limits taking into account market reactions.
Following Ghosh et al. (2013), primary balance reaction functions are estimated to determine debt
limits, taking into account growth, interest rate developments and the country-specific average fiscal policy
stance (Box 2). The public debt limit is defined as the maximum level of debt beyond which the
government cannot roll debt over, based on a given growth and risk-free interest rate, a given level of
uncertainty, and the previously observed capacity of governments to react to growing debt.
Box 2. Public debt limits
Public debt limits are calculated based on the framework by Ghosh et al. (2013). In this analysis, the primary
balance reaction function is estimated differently from Ghosh et al. (Fournier and Fall, 2015). It uses a different
functional form and also post-2007 data are taken into account. Using annual panel data for 31 OECD countries over
the period 1985–2013, the fiscal reaction function (primary balance) depends on the debt level and control variables:
𝑃𝑃𝑖𝑖 = 𝛽1 𝐺𝐺𝐺𝑖𝑖 + 𝛽2 𝑂𝑂𝑖𝑖 + 𝛽3 �𝐷𝑖𝑖−1 1𝐷𝑖𝑖−1<𝑑1 + 𝑑1 1𝑑1≤𝐷𝑖𝑖−1 � + 𝛽4 �𝐷𝑖𝑖−1 1𝑑1≤𝐷𝑖𝑖−1<𝑑2 + (𝑑2 − 𝑑1 )1𝑑2≤𝐷𝑖𝑡−1 � +
𝛽5 𝐷𝑖𝑖−1 1𝑑2≤𝐷𝑖𝑖−1 + 𝛽6 𝑂𝑂𝑖𝑖 + 𝑢𝑖 + 𝑣𝑖𝑖 (1)
where 𝑃𝑃𝑖𝑖 denotes the primary balance of country i at year t, 𝐺𝐺𝐺𝑖𝑖 denotes the output gap, 𝑂𝑂𝑖𝑖 denotes the
openness ratio multiplied by the terms of trade, 𝑂𝑂𝑖𝑖 denotes fiscal one-offs, 𝐷𝑖𝑖 denotes the debt level and 𝑑1 and 𝑑2
are estimated thresholds beyond which the fiscal reaction to debt changes. 𝑢𝑖 are country fixed effects and 𝑣𝑖𝑖 follows
an AR(1) process.
Estimates (Table A.1 in the appendix) confirm that governments react weakly by increasing their primary balance
when debt increases but remains below a certain level (𝑑1 in Figure 5). But, from this threshold (𝑑1 ) to a second
threshold (𝑑2 in Figure 5), governments react strongly to rising debt. Beyond this threshold, governments may abandon
fiscal discipline and reduce the primary balance. Alternative estimates also capture the effect of the business cycle on
the primary balance, and include additional control variables, such as asset prices, inflation, IMF programmes, the old
age dependency ratio, the euro area or the government size.
This fiscal reaction function (blue line in Figure 5) is included in the debt dynamics equation and interacted with
market reactions (red curve in Figure 5). The interest rate can include a risk premium reflecting the probability of
default in the next period 𝑝𝑡+1 , which is the probability that debt 𝑑𝑡+1 goes beyond its maximum level 𝑑̅ :
𝑝𝑡+1 = 𝑃(𝑑𝑡+1 > 𝑑̅ ) (2)
with the following debt accumulation dynamic:
𝑑𝑡+1 = (1 + 𝑟(𝑝𝑡+1 ) − 𝑔)𝑑𝑡 + 𝜇 + 𝑓(𝑑) + 𝜀𝑡 (3)
where 𝑔 denotes the potential growth rate, 𝜇 denotes an average stance of fiscal policy, 𝑓(𝑑) the reaction of the
primary balance to debt and 𝜀𝑡 captures macroeconomic shocks to the primary balance. The debt limit is a function of
the interest rate-growth rate differential 𝑟𝑖 − 𝑔𝑖 , the shock size 𝑉(𝜀𝑖 ), and the average past primary surplus 𝜇𝑖 .

Box 2. Public debt limits (cont)
Debt stabilises when the effect of past debt accumulation is exactly offset by the primary balance, as illustrated in
Figure 5. There is a stable equilibrium d for which the government would generate a higher surplus if a shock
increases the debt ratio. By contrast, when the debt level approaches the debt limit, then the government is facing an
interest rate spiral (green curve in Figure 5), and at the debt limit, the interest rate goes towards infinity, which means
that the government loses market access.
Figure 5. Determination of the debt limit
Note: g is the growth rate, r is the interest rate, r* is the risk-free interest rate, 𝑟(ε) is the interest rate with a risk premium, pb is the
primary balance, d is the debt-to-GDP ratio, d is the equilibrium debt-to-GDP ratio, d* is the stochastic debt limit depending on the
probability of default and d� is the deterministic debt limit corresponding to the intersection between the risk-free interest payment and
the fiscal reaction function. d1 and d2 are two estimated thresholds signalling changes in the reaction function to increasing debt.
The model is solved for each country to calculate current debt limits. The real long-term interest rate is the 10-
year government bond yield minus core inflation and the real risk free rate is set by deducting the CDS risk premium
from the real interest rate. The growth rate is the average potential growth rate between 2014 and 2017. The size of
macroeconomic shocks is derived from the standard deviation of the output gap. The past primary surplus is gauged
by country-specific fixed effects plus the median of residuals.
The gap between the interest and growth rate plays a large role in explaining differences in the debt limit across
OECD countries, especially for the countries that depart most from the OECD average (Figure 6).

Box 2. Public debt limits (cont)
Figure 6. Debt limit drivers

Deviation of the debt limit from the OECD average in per cent of GDP
125
Interest rate less growth rate Fiscal policy track record Size of output shocks
100
75
50
25
-25
-50
-75
12http://dx.doi.org/10.1787/888933224007
Note: This figure compares the debt limit of a given country to the 200% debt limit that would prevail in a virtual OECD average
country, in which each determinant is equal to the corresponding OECD average. Contributions are debt limit differences associated
with replacing a country-specific characteristic by the corresponding OECD average. For some countries no solution exists. For more
details on this decomposition, see Fournier, J-M. and F. Fall (2015), "Limits to Government Debt Sustainability", OECD Economics
Department Working Papers, No. 1229, OECD Publishing.
Debt limits are close to twice GDP in many OECD countries (Figure 7). This reflects that at medium to
high debt, the primary balance surplus reaction to debt is large. Those high debt limits are due to the few
1
countries which have high debt (Japan and Greece). Differences in debt limits between countries depend
2
strongly on the fiscal behaviour of governments. Maximum debt levels are higher in countries with a solid
fiscal policy track record (e.g. Norway), and lower in countries with low potential growth relative to the risk
free interest rate (e.g. Italy).
In some countries (Greece, Iceland, Ireland, Japan, Portugal, Spain, Slovenia and Slovak Republic),
there is no model based interest rate solution for a debt limit. It means that, given past fiscal behaviour, the
debt dynamics is not sustainable. These countries need to change fiscal behaviour by improving the
primary balance in the future to bring their debt dynamics onto a sustainable path. It also suggests high
market volatility to any news affecting these countries.
1. The high threshold beyond which fiscal fatigue kicks in is determined by the combination of the strong fiscal
policy reaction to debt by Greece, Italy and Portugal on the one hand and Japan’s weak reaction on the
other hand. This high debt limit is also due to the low current risk-free interest rate compared with the long-
term potential growth rate in many OECD countries.
2. Ballagria and Martinez-Mongay (2009) show that the fiscal reaction to debt was stronger in the euro area
countries in the run-up to monetary union than afterwards.

In 2013, Japan’s debt is higher than the debt limit based on the current interest rate while there is no
solution for the model based interest rate, if past fiscal behaviour prevails. Different factors explain why the
debt limit has not been binding: the central bank has kept the interest rate close to zero for a long period, a
strong home bias exists, underpinned by holdings of public pension reserve funds, and the net financial
assets position of the country is large.
Figure 7. Debt limits

Government financial liabilities, per cent of GDP
Per cent of GDP Gross debt in 2013 Debt limit (model based interest rate) Debt limit (current interest rate)
350
300
250
200
150
100
50
12http://dx.doi.org/10.1787/888933224011
Note: If past fiscal behaviour prevails, there is no finite interest rate solution to calculate the model based debt limit for Greece,
Ireland, Iceland, Japan, Portugal, Slovenia and Spain, and there is no stable debt level at current interest rates for Greece and
Portugal.
The debt limit based on market interest rates provides an upper bound because it does not take into
account the rise of the interest rate as debt approaches its limit. That is why it is higher than debt limits
built on a model-based interest rate, which takes into account the interest rate-debt spiral.
However, the debt limit is a moving target as it can be reached quickly, if a country loses market
confidence or if macroeconomic conditions change sharply (e.g. a change in the rate of potential growth or
in the size of shocks). Therefore, any debt target should be lower than such debt limits as uncertainties
surrounding the underlying hypothesis (e.g. on long-term growth rates) and the risk of interest rate spirals
call for a substantial buffer.
The risk that fiscal space vanishes is well illustrated by the following mechanisms:
• Estimated debt limits crucially depend on past fiscal behaviour: deteriorating fiscal balances not
only push public debt up, but also push the debt limit down.
• An increase in the gap between the risk free interest rate and the potential growth rate that is
plausible by historical standards could force governments to change fiscal behaviour to remain
on a sustainable track. An increase (decrease) in the potential growth rate improves (worsens)
the debt limit: a one percentage point higher potential growth rate increases the debt limit by
about 25% of GDP. In the case of a large permanent decrease of potential growth without a

commensurate decrease of the risk-free interest rate, the fiscal space could also vanish unless
the government changes its behaviour. Fournier and Fall (2015) show debt limits for different
combinations of the interest rate/potential growth differential.
• The fiscal behaviour of countries with high debt levels varies. In particular those under market
pressure (e.g. euro area countries) have to generate primary surpluses, while those under no
pressure (e.g. Japan) may even widen the primary deficit, although their debt is already very
high.
3.2. The debt level that maximises growth
Building on the initial work by Aschauer (2000), Checherita-Westphal et al. ( 2014) and Strasky (2015)
use a simple growth model with private and public capital to illustrate the role and size of government debt
in a long-run perspective. The main assumption is that government spending follows the “golden rule”, that
is public debt is used exclusively to finance net investment, while taxes finance current spending. In the
long run (steady-state), there is a constant public-to-private capital ratio that maximises growth. According
to the “golden rule”, the optimal stock of public capital equals the stock of public debt in the long run.
According to this model, the “golden rule” implies a non-linear relationship between the optimal ratio
of public-to-private capital stock and the “optimal” debt-to-GDP ratio:
𝛼 1−𝛼
𝑏 ∗ = 𝜑∗ (1−𝛼) = �(1−𝛼)2 �
where 𝑏 ∗ is the optimal debt-to-GDP ratio, 𝜑 ∗ the optimal public-to private capital ratio and α the elasticity
of output with respect to public capital.
Hence, the estimation of the elasticity of output with respect to public capital, and the assumption of
the “golden rule” allow comparing the actual government debt-to-GDP ratio to a simple long-run
benchmark.
New empirical evidence suggests that the elasticity of output with respect to public capital is 0.23
3
across 22 OECD countries and slightly higher, at 0.26, in 11 euro area countries (Strasky, 2015). These
values imply a somewhat higher return on public investment in the euro area countries than in the
22 OECD countries, resulting in a higher long-run public capital stock. Conversely, the results of
Checherita-Westphal et al. (2014) suggest a lower return on public capital in Europe and hence a lower
optimal public capital stock than the average across the OECD.
Optimal debt levels are considerably lower than the actual debt-to-GDP ratios in many OECD
countries (Table 2). The main conclusion of this exercise is that the growth-maximising debt level falls
between 50-80% of GDP. However, the results should be interpreted with caution as they rest on the
strong assumption about debt financing (the “golden rule”) and it is not clear whether debt financing of
public investment is the first-best option, as the “golden rule” asserts.
3. Country-specific estimations of the output elasticity are not significant for half of the countries. This is likely
due to the small sample size at the country level.

Table 2. Estimations of the growth maximising long-run debt-to-GDP ratio
OECD 22 EU 14 Euro area 11

65.5 63.5 49.6
Checherita-Westphal et al.
[50;81] [45;82] [27;72]
Strasky, including housing 48.9 49.9 55.9
capital stock [24.5;73.2] [22.2;77.5] [20.1;91.7]
65.1 72.9 83.7
Strasky
[42.0;88.2] [46.3;99.5] [47.9;119.5]
Note: Strasky, J. (2015), “Public Capital Stock and the Debt-to-GDP Ratios in 22 OECD Countries”, OECD Economics Department
Working Papers, forthcoming and Checherita-Westphal, C., A. H. Hallett and P. Rother (2014), "Fiscal Sustainability Using Growth-
maximizing Debt Targets", Applied Economics, Vol. 46(6), pp. 638-647 use a different estimation methodology. The borders of the
confidence interval are in brackets.
3.3. The effectiveness of fiscal policy in stabilising the economy
The impact of debt on the economy has been mostly analysed through the effects of changes in fiscal
policies on output. It is measured by fiscal multipliers, which show the effect of a change in a fiscal
instrument on economic activity.
The empirical evidence on the size of the multipliers is not conclusive, as multipliers depend on
whether policy focuses on spending or taxes, whether policy changes are temporary or permanent,
whether stimulus is provided in good or bad times, whether many households face liquidity constraints, on
monetary policy, the credibility of government and many other factors (Galí et al. (2007); Auerbach and
Gorodnichenko (2012); Blanchard and Leigh (2013); Ramey (2011)).
The level of debt also matters for fiscal policy effectiveness. Röhn (2010) provides an overview of
recent studies on Ricardian equivalence. The private/public saving offset of fiscal policy changes is larger
in the long term than in the short term. However, estimates vary considerably across the studies. The
estimates of the private saving offset for OECD countries ranges from 0.1 to 0.5 in the short run, and from
about 0.3 to as much as 0.9 in the long run.
The OECD's own estimations that take into account cross-country heterogeneity show that on
average across countries the savings offset is around 40%, both in the short and in the long term (Röhn,
2010). However, there is considerable heterogeneity across countries. Overall, his results provide
evidence against a strict version of the Ricardian equivalence hypothesis in the long term (full offset).
Moreover, offsets may also react in a non-linear way. Saving offsets become stronger at a tipping point of
government debt at around 75% of GDP.
More recently, Nickel and Tudyka (2014) find for a sample of 17 European countries that responses of
output to government spending shocks exhibit strong non-linear behaviour. They find that the overall
cumulative effect of a spending shock on GDP is positive at moderate debt-to-GDP ratios, but the effect
turns negative beyond debt ratios of 65%-70%.
This non-linear effect of fiscal stimulus is also confirmed by Turner and Spinelli (2013) who show that
the interest rate effect of marginal increases in external or government debt is non-linear and depends on
the level of debt. For instance, the interest rate effect has risen sharply in the post-crisis period for euro
area countries, which have a combination of both high external and government debt.
3.4. When does debt affect growth?
A high level of debt not only lessens the effectiveness of fiscal stimulus, it can also reduce growth.
Reinhart and Rogoff's (2010) conjecture that debt above 90% of GDP has a negative impact on growth,
has prompted much empirical work to investigate the relationship between debt and growth. Some studies,
such as Kumar and Woo (2010) and Cecchetti et al. (2011) find some evidence of a non-linearity, with high

levels of debt having a larger negative impact on growth. In particular, Cecchetti et al. (2011) find that
government debt has a negative impact beyond a range of 80 to 100% of GDP. Moreover, Baum et al.
(2013), focusing on 12 euro area countries, estimate that beyond debt-to-GDP ratios of 95%, additional
debt has a negative impact on economic activity. They also find that the long-term interest rate is subject to
increased pressure when the public debt-to-GDP ratio rises above 70%.
However, work by the OECD indicates that a universal non-linear relation between debt and growth is
not robust (Égert, 2013). For general government debt, the threshold beyond which negative growth effects
kick in is at about 50% and there is a large amount of cross-country heterogeneity. Also Afonso and
Alvés (2014) find different thresholds for the debt-growth relationship for some EU countries.
Finally, Chuddik et al. (2013) estimate long-run effects in dynamic heterogeneous panel data, taking
into account the possibility of reverse causality. They show that some economies have run into debt
difficulties and experienced subdued growth at relatively low debt levels, while others have been able to
sustain high levels of indebtedness for prolonged periods and grow strongly without experiencing debt
distress. They do not find a universal threshold effect in the relationship between debt and growth.
However, there is a statistically significant threshold effect in the case of countries with rising debt-to-GDP
ratios. The debt trajectory seems more important than the level of debt itself.
Overall, there is good reason to believe that causation between higher debt and lower growth runs
both ways and the empirical literature has not come to a strong conclusion on causality
(Panizza and Presbitero, 2014).
4. Defining debt targets
This section summarises the cross-country evidence of the effect of debt on the economy, and comes
up with debt thresholds, distinguishing between advanced economies with an autonomous monetary
policy, for the euro area countries and for lower income economies. It also reviews hard-to-quantify risk
factors. Finally, stochastic simulations are provided to show the amount of prudence needed to reach a
specific debt target with a certain probability.
4.1. Finding a debt threshold: Cross-country evidence

From the review of the cross-country evidence of the links between government debt and economic
activity, the following can be gleaned:
• At moderate levels, government debt plays a positive role for the functioning of domestic
financial markets as it provides a safe asset in a very liquid market. It is difficult to pin down the
size of this positive effect and when diminishing returns set in.
• Debt also plays a positive role when it funds public infrastructure, but only up to a point. The
analysis suggests that this point lies between 50 and 80% of GDP for OECD countries.
• Empirical estimates of non-linear effects of debt on growth, though controversial, show that
negative effects of debt on growth are likely to emerge for debt-to-GDP ratios above 80% to
100% of GDP.
• Röhn (2010) showed that at higher debt, private saving offsets of fiscal stimuli become stronger,
with a tipping point estimated around 75% of GDP. Also, Nickel and Tudyka (2014) find that the
overall (short and long-run) effect of spending shocks becomes negative beyond debt ratios of
65% to 70%.
• Sustainability limits are hard to pin down, but are high for many countries. Japan illustrates that
some countries can have a high level of debt for a long time period without adverse market
reactions, as the Japanese economy is characterised by a high household saving rate, low
external financing and a large amount of external financial assets. The fiscal limit exercise
suggests that a country may only be able to live with very high debt as long as interest rates are

low and market confidence is high. This suggests that such an equilibrium is likely to become
unstable. Because debt limits are state dependent, countries should endeavour to steer clear of
their default limit.
Advanced economies
Overall, these estimates suggest a gross debt threshold range, where negative effects of debt start to
dominate, of 70 to 90% of GDP for higher-income countries, with specific risk exposures to factors such as
4
foreign debt, bank fragilities, etc., to be taken into account (see below).
Different lessons can be drawn from the recent experience of advanced countries compared with past
developing and emerging country debt crises. First, the debt intolerance phenomenon that kicked in for
developing and emerging countries at low levels of debt may also affect advanced countries though at
higher levels of debt. Duress involves a vicious cycle of loss of market confidence, rising interest rates on
government debt, and difficulties to service debt, potentially leading to default or restructuring. This
scenario is clearly what happened in the euro area in Greece and its contagious effects. Second, as some
advanced countries are more and more dependent on capital inflows and international markets for their
debt financing, they are more vulnerable to global factors that could reverse financing flows and shake
market confidence. Finally, in advanced economies fiscal risks are linked to financial markets crises
originating either internally or externally.
Recent events suggest that debt thresholds for euro area countries are lower than for the other
advanced OECD economies, because they are constrained by the absence of a country-level monetary
policy. Federal countries can provide insights for countries in a monetary union. In federations two
countervailing forces exist. On the one hand, tight economic and monetary integration can make for high
shock absorption capacity, as shown by large federal countries such as Canada, Germany and the
United States (Hepp and von Hagen, 2009 or Carlino and Inman, 2013). On the other hand, the sharp
reaction of financial markets to crises in small sub-national jurisdictions suggests that even small events
can have severe disruptive effects, giving credence to the argument that contagion poses a considerable
threat in an integrated economy (Blöchliger, 2013). The net effect of setting risk-spreading integration
against risk-creating contagion should tilt the euro area country debt targets in the direction of prudence,
as global capital flows freely in the euro area, leading to contagion risks, while labour and goods markets
are less well integrated, making adjustment to shocks tougher and more long-lasting than in mature
federations. These considerations would suggest that the debt threshold for euro area countries is in a
range of 50 to 70% debt to GDP, though the "right" threshold is difficult to establish.
Emerging economies
Developing and emerging countries tended to default at a relatively low level of debt. Among default
episodes between 1970-2008 reported by Reinhart and Rogoff (2009), more than half of all defaults
occurred at levels below 60%. A striking feature of these defaults was that liabilities represented on
average 230% of exports and a sizable portion of government revenues. This points to two vulnerabilities:
first, borrowing was no longer in line with the exports needed to generate foreign currency with which to
service foreign currency debt; second, governments had limited capacity to collect revenues signalling
institutional weaknesses.
Vulnerabilities are also due to global factors. As noted by Reinhart and Rogoff (2009), peaks and
troughs in commodity price cycles appear to be leading indicators of peaks and troughs in the capital flow
4. The establishment of this debt threshold range is not based on a complete meta-analysis of all studies and
there is no weighting of the various links between debt and economic activity.

cycle, with troughs typically resulting in multiple defaults. One lesson is that countries experiencing sudden
5
large capital inflows are at high risk of experiencing a sovereign debt crisis.
Major financial crises are often key ingredients in sovereign defaults. The combination of bank failures
and recessions can trigger sovereign crises. Arellano and Kocherlakota (2008) find that domestic private
sector default risks are positively correlated with sovereign default risk. And sovereign defaults are often
associated with large numbers of domestic defaults, such as bank insolvencies. Reinhart and Rogoff
(2011) also confirm the link between domestic financial sector problems and sovereign debt crises.
Since the financial and debt crises of the 1990s, the fiscal performance of the emerging and lower-
income OECD economies has improved. Thanks to these improvements and stronger growth they fared
better during the recent crisis and were less affected in terms of sovereign debt problems. Indeed, most of
the emerging economies have a low level of debt, except Brazil and India, which have debt ratios above
60%. The government finances are less vulnerable to external developments thanks to low external
borrowing (on average less than 10% of GDP) and higher foreign currency reserves, matching or even
higher than their external indebtedness for most of them (Table 3).
However, some risks to the public finances appear high in some countries. For instance, Brazil is
heavily exposed to future public spending on pensions due to the indexation of pension benefits to a fast
rising minimum wage while the population is aging rapidly. India’s government appears exposed to public
enterprise risks, while the financial sector seems weak (Table 3). The budgets in several countries
(Colombia, Mexico and Russia) are heavily exposed to adverse oil price shocks.
Moreover, the financial sector and the exchange rate of most of these countries appears to be highly
sensitive to monetary policy changes in advanced countries, especially non-conventional policies of the
United States and its impact on the dollar. The volatility of capital flows remains high indicating that despite
their better performance, there are still vulnerabilities. Emerging economies remain exposed to the debt
intolerance phenomenon. Reinhart and Rogoff (2011) find that when external debt of emerging markets is
above 30-35 % of GDP, risks of a credit event start to increase significantly. This implies that fiscal space
is much more limited than in the advanced economies, although the need for infrastructure investment is
high. Taking into account the various risk factors and especially external debt, the quantity of foreign
reserves, the expected interest rate and growth differential and the maturing of health and pension
schemes, emerging economies debt threshold is probably in a range of 30 to 50% debt to GDP ratio.
5. Aguiar and Amador (2013) provide a survey of the channels that lead to sovereign debt defaults.

1
Table 3. Fiscal performance and risk indicators for emerging and lower income economies
Indicator Chile Colombia Mexico Turkey Brazil Russia India Indonesia China South Africa
Fiscal performance indicators
Financial balance
-3.4 -2.44 -0.3 -0.8 -3.3 -0.5 -7.9 -2.2 -0.7 -6.1
(% of GDP)
Primary balance
0.2 0.8 -0.5 0.8 1.9 0.3 -3.8 -1.4 -1.5 -1.8
(% of GDP)
Interest payments
2.7 9.5 9.2 21.7 1.8 24.9 10.9 3.3 9.6
(% of revenue)
Debt (% of GDP) 11.1 32.0 43.5 35.5 67.2 10.4 66.4 23.6 21.3 42.7
External debt (% of GDP) 2.1 9.4 16.4 10.8 2.8 2.9 4.0 13.1 n.a. 14.8
Fiscal risk indicators
Public ownership of enterprises
2.24 2.3 2.59 3.55 2.68 3.94 4.50 n.a. 4.15 3.39
(PMR indicator)
Terms of trade (Volatility)2 0.128 0.068 0.022 0.032 0.055 0.154 0.019 0.046 0.048 0.034
Non-performing loans
2.1 2.8 3.2 2.6 2.9 6.0 3.8 1.7 1.0 3.6
(to total gross loans (%))
Bank capital-to-assets (%) 8.1 14.8 10.4 10.9 9.3 11.5 6.9 12.5 6.7 7.9
Bank regulatory capital to risk-
13.3 17.0 15.6 15.3 16.1 13.5 12.3 19.8 12.2 15.6
weighted assets (%)
Future public spending on
pensions (change in % points n.a. n.a. 1.1 4.1 7.3 0.4 n. 1.2 0.4 0.2
of GDP 2010-50)
Future public spending on
health and long-term care
3.2 n.a. 2.8 2.7 2.4 1.7 1.6 2.0 2.1 1.5
(Increase in % points of GDP
2010-30 )
Projected interest rate–growth
0.1 1.5 -0.2 0.2 2.4 -1.2 -4.5 -5.2 -7.7 -2.1
differential (%)3
Foreign currency reserves
14.1 10.9 13.3 13.3 15.5 21.6 14.2 10.6 40.7 11.9
(% of GDP)
Exchange rate (Volatility) 0.06 0.08 0.07 0.08 0.11 0.09 0.07 0.07 0.03 0.11
1. Data refer to general government accounts for 2013 or for the latest year available.
2. Terms of trade and exchange rate volatility are calculated as the standard deviation of annual percentage changes from 2004-13.
3. IMF projection 2013-18.
4. Central government.

4.2. Debt and the vulnerabilities of countries to specific factors and risks
Fiscal risks due to explicit and implicit contingent liabilities are discussed in Bloch and Fall (2015).
Countries are also exposed to other specific risks, the prevalence and size of which is often difficult to
quantify. Yet, these specific risks as well as the explicit and implicit contingent liabilities should be taken
into account, when establishing the prudent debt target.
Total indebtedness and asset price movements
High total indebtedness can create vulnerabilities to shocks, exposing not only households and firms,
but also governments to maturity, currency and liquidity mismatches as well as potential solvency
problems (Sutherland and Hoeller, 2012). Furthermore, high indebtedness can expose the economy to
asset price movements, which can amplify shocks and macroeconomic instability. Government debt
typically rises after the onset of a recession, suggesting that there is a migration of debt across balance
sheets. This is confirmed by the latest IMF (2014) accounting of financial sector support by governments.
Since the onset of the crisis up to early 2014, the net impact of support to the financial sector on gross
public debt was 24% for Greece and 33% for Ireland. However, most of the impact of a financial crisis on
government debt is not due to bank bailout costs but due to a sharp decline in tax revenue due to the deep
recession and ensuing hysteresis effects that accompany severe financial crisis (Reinhart and
Rogoff, 2009; Ollivaud and Turner, 2014).
The vulnerabilities of budgets to financial and housing sector developments can be partly gauged by
the effects of asset price cycles on the budget. Price and Dang (2011) find that budgets of some countries
are very sensitive to house price cycles, which can account for budget balance swings of up to 3% of GDP
in either direction, notably in Ireland, the Netherlands and Spain. Equity price cycles tend to have a smaller
impact in almost all countries (apart from Korea and Australia). Turner (2006) analyses how commodity
prices affect the government balance in Australia. For countries relying strongly on some commodity
exports, there is a risk of commodity price bubbles and trend reversals that can have a considerable effect
on budget balances and therefore debt.
Taxation
A high level of taxation constrains the options to respond to a surge in debt. Among the countries with
a debt-to-GDP ratio above 100%, France, Belgium and Italy have the highest level of current receipts
amounting to close to 50% of GDP or more. In contrast, the United States and Japan, which also have a
high debt ratio, have government receipts below 35% of GDP. They have more room to raise taxation to
reduce debt than the countries with an already high tax burden and Japan has started to raise the value-
added tax with the aim of reining in spiralling government debt.
Population ageing
Population ageing is due to accelerate in the coming years with the old-age ratio (working-age
population over those older than 65 years) decreasing from 4.2 currently to 2.1 in 2050 in the OECD on
average. This development is particularly detrimental for pension schemes. It also affects health care
spending, but to a lesser degree, with health technology developments playing a prominent role, but
ageing has a strong effect on long-term care. Future public health care and pension spending pressures
could be considerable in many OECD countries (Fall et al., 2014). Pension systems need to be reformed to
make them sustainaible and efficiency gains be sought in health care systems. The remaining future
spending pressures will have to be matched by additional resources or by cuts of other government
spending items.
The composition of government debt
The composition of government debt can also have macroeconomic impacts (Lojsch et al., 2011). For
instance, in advanced economies, the government bond yield curve serves as a benchmark for pricing
private sector bonds. The maturity composition of government debt affects the yield curve and hence the

financing conditions of the private sector, with possible crowding out effects (Bloch and Fall, 2015). Also,
with a high share of short-term debt the government may be vulnerable to an increase in monetary policy
rates. As illustrated by the recent euro area crisis, some countries (Spain, Portugal) faced sharp interest
rate increases rendering the roll-over of debt very costly. Therefore, debt maturity is important for the
sustainability analysis, but its effects depends on interest rate variations and strategic debt management
and are not easy to gauge.
A high exposure to external holdings of government debt can generate vulnerabilities and result in a
crisis, especially in emerging economies. But a high exposure also plays a role in advanced economies.
The euro area countries rely more on external financing of government debt than most other countries
(Figure 8). This is due to the greater financial integration of the euro area but as seen during the recent
crisis, integration does not prevent sudden capital outflows or spikes in government borrowing rates that
can precipitate a debt crisis.
Figure 8. General government gross debt and its external component
Foreign-held government debt as a % of GDP
160
GRC
140
120
100
PRT
80 IRL
FRA
AUT
60
HUN BEL
FIN
DEU ITA
SVN
ESP
40 NLD
SVK ISL USA
POL GBR
NZL SWE
20 DNK CAN JPN
EST AUS CZE
LUX NOR
ISR
KOR CHE
0
0 50 100 150 200 250
Gross government debt as a % of GDP
12http://dx.doi.org/10.1787/888933224028
Note: External debt refers to general government debt held by external creditors (either in local currency or foreign currency). Gross
government debt refers to general government gross financial liabilities.
Source: Joint External Debt Hub Database, accessed 22 September 2014; OECD Economic Outlook Database, accessed 9 October
2014.
In the case of foreign currency denomination of debt, governments are exposed to exchange rate
risks, which could affect the cost of debt. In addition, the interest rate is more sensitive to the debt level if
more debt is issued in foreign currency (Fall and Fournier, 2015), which can be due to the absence of a
lender of last resort for foreign currency denominated debt. However, the foreign currency risk is limited for
most OECD countries. Figure 9 indicates that, for the countries for which data are available, only for
Hungary, Mexico, Poland and Sweden, debt denominated in foreign currency represents more than 20% of
total debt. However, for countries with important assets (Sweden) or currency reserve (Hungary), the net
effect of the debt composition is not straightforward.

Figure 9. General government debt by currency denomination in selected OECD countries

Fourth quarter 2013
Per cent Foreign currency Domestic currency
100
80
60
40
20
12http://dx.doi.org/10.1787/888933224031
Source: IMF Quarterly Public Debt database, accessed 23 September 2014.
4.3. Defining a prudent debt target as conveyed by stochastic simulations
Prudent debt targets should take into account the uncertainties (macroeconomic shocks) surrounding
debt developments. To minimise the risk of missing a debt target, which could undermine credibility, a
prudent target for future deficit developments needs to be set. A stochastic debt analysis is developed to
quantify the uncertainties surrounding the development of the main macroeconomic variables and
therefore debt dynamics and this uncertainty around debt dynamics is used to define a prudent debt target
(Box 3).

Box 3. Simulating the effect of shocks on public debt
A set of seven equations is used to simulate jointly six variables and public debt dynamics. The six variables are
the growth rate 𝑔𝑖𝑖 of country i at year t, the inflation rate measured by the GDP deflator 𝜋𝑖𝑖 , the overnight nominal
interest rate 𝑟𝑖𝑖𝑠 , the long-term nominal interest rate 𝑟𝑖𝑖𝑙 , the primary balance 𝑃𝑃𝑖𝑖 , and the structural primary balance
𝑃𝑃𝑖𝑖𝑠𝑠 .
The framework includes three deterministic equations and four estimated stochastic equations, which provide the
main coefficients for the simulations. The first deterministic equation is the fiscal reaction function, reflecting past
behaviour of the government, as estimated in the baseline estimate of Fall and Fournier (2015). The debt
accumulation dynamic is calculated with an equation that takes into account the share of debt that needs to be rolled
over each year. The structural balance is defined as the primary balance minus about 0.4 times the output gap,
consistent with the estimates of the impact of the business cycle on the primary balance reported in Sorbe (2012).
Four estimated stochastic equations capture short-term shocks on growth, inflation, monetary policy and on long-term
interest rates:
𝑙
(1) 𝑔𝑖𝑖 = 𝛽1,1 𝐺𝐺𝐺𝑖𝑖−1 + 𝛽1,2 �𝑟𝑖𝑖−1 −𝜋𝑖𝑖−1 � + 𝛽1,3 ∆𝑃𝑃𝑖𝑖𝑠𝑠 + 𝛽1,4 ∆𝑃𝑃𝑖𝑖𝑠𝑠 1𝑡≥2009 + 𝛽1,5 ∆𝑃𝑃𝑖𝑖𝑠𝑠 1𝑡≥2009 + 𝛽1,6 𝑒𝑒𝑒𝑖𝑖 1𝑡≥2009 +
𝛽1,7 ∆𝑃𝑃𝑖𝑖𝑠𝑠 𝐺𝐺𝐺𝑖𝑖−1 + 𝛽1,8 𝐺𝐺𝐺𝑖𝑖−1 1𝑡≥2009 + 𝑢1,𝑖 + 𝛼1,𝑡 + 𝜀1,𝑖𝑖
(2) πit = β2,1 πit−1 + β2,2 πit−2 + β2,3 πit−3 + β2,4 GAPit−1 + u2,i + α2,t + ε2,it
(3) 𝑟𝑖𝑖𝑠 = 𝛽3,1 𝐺𝐺𝐺𝑖𝑖−1 + 𝛽3,2 (𝜋𝑖𝑖−1 − 𝜋𝑡𝑡𝑡 ) + 𝑢3,𝑖 + 𝛼3,𝑡 + 𝜀3,𝑖𝑖
(4) 𝑟𝑖𝑖𝑙 = 𝛽4,1 𝑟𝑖𝑖𝑠 + 𝛽4,2 𝑔𝑖𝑖−1 + 𝛽4,3 𝜋𝑖𝑖−1

𝑐
+ 𝛽4,4 𝐷𝑖𝑖−1 + 𝛽3,5 𝑒𝑒𝑒𝑖𝑖 𝐷𝑖𝑖−1 + 𝑢4,𝑖 + 𝜀4,𝑖𝑖
where 𝑒𝑒𝑒𝑖𝑖 is a dummy equal to one for the countries that are a member of the euro area, 𝜋𝑡𝑡𝑡 is the inflation
𝑐
target of the central bank assumed to be equal to two, and 𝜋𝑖𝑖 denotes core inflation. 𝑢𝑖 and 𝛼𝑡 are country and year
fixed effects, 𝜀1,𝑖𝑖 , 𝜀3,𝑖𝑖 and 𝜀4,𝑖𝑖 follow an AR(1) process and 𝜀2,𝑖𝑖 is a white noise error term.
Country-specific residuals of each of these equations are combined with output gap revisions and the time fixed
effect of the first equation to capture four country specific short-term shocks, a country-specific potential output shock
and a common growth shock. Shocks are jointly-drawn from the estimated co-variance of the residuals of these
equations.
The simulations are run with gross debt. The mechanisms at work are the same if one replaces gross by net
debt. The level of debt is modified substantially for those governments with large financial assets and hence the
primary balance needed to keep debt at a prudent level is lower (see Fall and Fournier, 2015 for an example of net
debt dynamics). However, the simulations with net debt do not take into account the uncertainties surrounding
financial asset valuation changes.
Countries differ in terms of initial positions, the size of shocks, long-term growth potential, and heterogeneous
structural features are captured by country-specific fixed effects. Two specific features capture the specificities of the
euro area. First, the short-term interest rate is set jointly, based on the average output gap and average inflation.
Second, the long-term interest rate reacts more to public debt in the euro area. This second feature reflects the
observed market behaviour and captures a different belief on the capacity of the central bank to act as a lender of last
resort and the higher risk of debt restructuring when devaluation is not possible.
Given the high level of indebtedness, many OECD countries have to pursue fiscal consolidation policies at the
same time. The associated negative cross-country spillover effects can be important during the transition toward
higher primary balance levels (Mendoza et al., 2014), but are not included in this exercise.
The stochastic simulations can be used to produce fan charts for future debt developments. Fall and
Fournier (2015) show such fan charts for the United States. They can also be used to assess the risk of
hitting a debt threshold and calculate the cushion that is needed so that adverse macroeconomic shocks
do not push debt beyond the level, where adverse effects of debt on economic activity set in. In practice,
this can be achieved by keeping the probability that debt goes above this threshold level sufficiently low. In
the present framework, the probability of debt to go above 85% for non-euro area OECD countries and
65% for euro area countries is calculated. The prudent debt target is the median debt by 2040 such that
there is less than a 25% risk to go beyond the debt threshold (85% or 65% debt ratio) and the
corresponding fiscal deficit trajectory is calculated (see also Guillemette, 2010). Larger uncertainties in a

country are associated with a lower prudent target. As illustrated in Figure 10 (Panel A), the prudent debt
level ranges from about 35% in Greece and Ireland to about 75% in the United Kingdom and the United
States. Differences reflect the different exposure of countries to shocks and their capacity at absorbing
them as estimated on the basis of their history of adjusting the primary balance.
Reaching the prudent debt target by 2040 would require a primary surplus in nineteen OECD
countries (Figure 10, Panel B). Greece and Japan would need to make the largest effort (an average
annual primary surplus of about 5% and 6% of GDP, respectively). The high surplus required for Japan is
mainly due to the high starting point of the debt ratio. For Greece, on top of the high starting point, the
required surplus is high because large uncertainties make the prudent debt level lower.
Figure 10. Public debt ranges under a prudent scenario

Panel A. Prudent debt levels
Per cent of GDP

150
125
100
75
50
25
Panel B. Average annual primary balance between 2014 and 2040
Per cent of GDP

7
12http://dx.doi.org/10.1787/888933224043
Note: The thick horizontal lines show the median debt level, boxes show the interquartile range, and extreme values are the 5th and
the 95th percentiles. Only those countries that need to generate a primary surplus are shown.
A variant with larger shocks (standard deviation 50% higher than estimated in the past) results for all
countries in a corresponding increase in the likelihood of recessions and of uncertainties surrounding the

debt trajectories (Fall and Fournier, 2015). However, the larger shock variant does not imply any
substantial additional consolidation effort to reach the prudent debt level.
On the other hand, if the stringency of the tolerance probability is increased from 25% to 10%, for
instance, the prudent debt target diminishes for all countries and consolidation needs rise (Figure 11,
Panel A). The decrease of the prudent debt target is higher for countries with a bigger shock variance (this
is the case for Greece and Ireland, for example). Moreover, the prudent debt target is an increasing
function of the tolerance (probability) level. Figure 11, Panel B shows this relationship between the
tolerance level and the prudent debt target taking the United States as an example.
Figure 11. Sensitivity of prudent debt targets to the confidence level

Panel A. Prudent debt targets by country
25% tolerance level 10% tolerance level
Per cent of GDP
80
70
60
50
40
30
20
10
Panel B. Sensitivity of the prudent debt target to various tolerance levels (US example)
Per cent of GDP
90
85
80
75
70
65
Prudent debt target Debt threshold
60
55
50
0 5 10 15 20 25 30 35 40 45 50
Tolerance level (%)
12http://dx.doi.org/10.1787/888933224057
Note: This figure shows the sensitivity of the prudent debt level depicted in Figure 10, Panel A to a change of the tolerance level.
The stochastic simulations can only illustrate fiscal risks and their potential size. Future shocks will be
different from past ones and fiscal behaviour is time dependent. Also the shocks taken into account are
limited to some macroeconomic shocks and other country-specificities will also need to be taken into
account, when setting the prudent debt target. For instance, the OECD countries with a high share of their
debt denominated in foreign currency should take into account the accompanying exchange rate risk in
determining their prudent debt target.

5. Designing fiscal frameworks
5.1. Justifications, objectives and composition of the fiscal framework
In all countries, there is a set of rules and institutions that shape fiscal policy making. They comprise
the institutions, arrangements and procedures that govern the planning and implementation of budgetary
policies.
In most advanced countries, debt ratios have been trending up since the late 1970s, illustrating the
lack of fiscal discipline and the inability of governments to commit to sustainable public finances. Indeed,
debt breaks the link between taxation and spending, tempting governments to shift the fiscal burden to
future generations. A large literature identifies reasons, why such a deficit bias exists (see Persson and
6
Tabellini, 2000; Wyplosz, 2013; Calmfors and Wren-Lewis, 2011):
• Informational problems: The electorate may lack information on fiscal positions. This ignorance
may allow the government to increase its chances of re-election, creating a political business
cycle. As there is no similar incentive to raise taxes or cut spending after elections, this will lead
to deficit bias.
• Electoral competition: Competition between parties will drive up debt (Alesina and
Tabellini, 1990; Persson and Svensson, 1989). Governments will follow the preferences of their
electorate over public goods or the size of government and will use increasing debt as a
strategic tool to prevent future governments from implementing the programmes they prefer.
• Common-pool theory: Different groups lobby for public spending catering to their specific
interests with insufficient regard to the full budgetary costs now as well as in the future
(Krogstrup and Wyplosz, 2010).
• Intergenerational equity: Deadweight losses due to higher taxation to service the higher debt
level would fall on the future generations. Moreover, lower income people, which tend to hold
only a small amount of assets, including government bonds, would not benefit from interest
payments, but would have to pay taxes to service the higher debt.
• Pro-cyclicality: Deficit bias results in large part from a failure to exercise budget discipline in
good times (Manasse, 2006; Tornell and Lane, 1999).
The major objectives of the fiscal framework are twofold: it should underpin fiscal discipline in
reaching the prudent debt target and allow for short-run flexibility of fiscal policy to react to cyclical
developments and shocks. A rule serves as a commitment tool to increase the cost of deviations and to
anchor expectations about future fiscal developments. Some discretion remains, however, essential to
guarantee that fiscal policy can react to sizeable shocks.
The fiscal framework should make it more likely that countries with high debt will follow a path towards
a more prudent level and that countries close to the prudent debt target will seek to stay there. Five
components of the fiscal framework are considered: a debt target, a fiscal rule, fiscal councils, budgetary
processes and medium-term budgeting.
5.2. Setting a debt target
In designing a fiscal framework, the starting point is to define the targets and instruments. A debt
target can be effective in anchoring expectations about future fiscal policy. The prudent debt target serves
as the reference point to define numerical fiscal rules, in particular, for countries with high debt that have to
converge to a lower prudent debt ratio.
6. These categories are based on Calmfors and Wren-Lewis (2011).

A debt target is better than a debt limit. The experience of the EU framework is that, in the absence of
a debt target, debt drifted up towards the 60% of GDP limit or even beyond, leaving no room to absorb the
sizeable fiscal shock of the recent crisis without breaching the limit.
5.3. Fiscal rules
Justifications, objectives, types and conditions for the effectiveness of fiscal rules
The debt target needs to be complemented by fiscal rules that allow for counter-cyclical policies and
escape clauses in the case of large shocks. Many countries have adopted fiscal rules that constrain their
government’s policies and underpin fiscal discipline. However, they can also lead to sub-optimal policies.
Indeed, there is a trade-off between commitment and discretion: on the one side, rules provide valuable
commitment that can limit interest rate risk premiums by curbing policy makers spending bias and
excessive deficits; on the other hand, a strict rule leads to costs in terms of lost flexibility as rules are
necessarily incomplete, and some discretion may be optimal. Since Kydland and Prescott's (1977) seminal
contribution, it is known that time-inconsistency often prevails because short-term considerations push
policymakers to deviate from ex-ante optimal strategies. Moreover, Halac and Yared (2014) show that,
when shocks are persistent, the ex-ante optimal rule is not sequentially optimal, as it provides incentives
for governments to over-accumulate debt. Also, fiscal rules may limit the capacity of governments to
pursue macroeconomic stabilisation policies (see also Wyplosz (2013), for a discussion).
But in practice, discretion may undermine discipline. As noticed by Kopits (2011a), discretionary fiscal
policy did often not fulfil the role of macroeconomic stabilisation. There is evidence that the conduct of
fiscal policy was more often pro-cyclical than not (Taylor, 2000; Auerbach, 2002). Also Égert (2010) finds
that fiscal policy in countries with high public debt and high government deficits tends to be pro-cyclical.
Different analyses provide insights in the conditions of the effectiveness of fiscal rules. Bohn and
Inman (1996), studying the effectiveness of fiscal rules among the US States provides strong evidence that
the form of a balanced-budget rule matters. They find that more stringent balance rules with outside
enforcement have a positive impact on budget balances. Poterba (1994) finds that the no-carryover rule is
associated with a more rapid reduction of the budget deficit, mainly through lower spending, and not higher
taxes. More recently, Clemens and Miran (2013) find that during times of fiscal stress, states with strict
rules enact relatively large budget cuts in order to quickly narrow deficits.
Alesina and Bayoumi (1996) find that fiscal policy restrictions have little effect on output volatility in
US states, because fiscal policy restrictions have two opposite effects: 1) strict fiscal rules lead to lower
policy volatility (i.e. less discretion in conducting fiscal policy; 2) fiscal restrictions reduce the
responsiveness of fiscal policy to output shocks and decrease the persistence of spending fluctuations
which might amplify the business cycle. Overall, these effects cancel each other. On the other hand, Fàtas
and Mihov's (2006) results support the notion that the effects of fiscal constraints dominate and therefore
can reduce macroeconomic volatility. They show that fiscal policy is a significant source of business cycle
volatility for US States, and, as a result, constraints on politicians lead to less volatile economic
fluctuations. At the country level, Fatàs and Mihov (2003) showed that constraints on fiscal policy reduce
the volatility of the business cycle.
It is often argued that public investment that fosters long-run growth should be excluded from fiscal
targets. However, drawing the line for public investment that matters for long-run growth is not
straightforward, as not only public physical investment, but also spending on education or research and
development have a positive effect on growth. Moreover, if public investment is debt financed, it is better to
take it into account as debt is created and the positive effects on growth are reaped in the short run
(demand effect) and in the long-run (supply effect with spill-overs to the private sector).
Table 4 reviews the different rules and their likely effects on fiscal discipline and stabilisation policies.
They are also subject to different measurement uncertainties and side-effects of fiscal policy. These factors
determine their suitability as fiscal rules in a sound fiscal framework.

Table 4. Strengths and weaknesses of existing fiscal rules

Promotion of fiscal discipline Impact on stabilisation policies Risks and side effects
Numerical budget balance rules seem to Strict budget balance rules fail to They can induce governments
have a positive impact (that is higher provide enough flexibility to pursue under fiscal consolidation pressures
surpluses or lower deficits) on a counter-cyclical fiscal policy. to rely excessively on cuts to
budgetary outcomes (Debrun, X., et al. They tend to induce pro-cyclicality growth-enhancing, but politically
(2008), “Tied to the Mast? The Role of that can be reinforced by weak less sensitive, expenditure such as
Budget National Fiscal Rules in the European coordination between different public investment. To counter such
balance Union”, Economic Policy, No. 54, government levels adverse effects, it is often proposed
rules pp. 298–362). Numerical budget to exclude investment and focus on
balance rules give a concrete the current balance (Golden rule).
benchmark upon which to judge the
fiscal stance (European Commission
(2011), “Public Finances in EMU -
2011“, European Economy, No. 4/2011.
Brussels: European Commission).
Structural In principle, structural balance rules Structural or cyclically-adjusted These rules are vulnerable to
balance underpin fiscal discipline. As they set a balance rules provide some uncertainties about the
rules target for the evolution of a core flexibility as they allow the measurement of the output gap,
variable, which is purged of cyclical automatic stabilisers to play in full. which renders real-time monitoring
effects, they define the space for Indeed, they are less binding than difficult. A structural deficit rule can
discretionary policy. budget balance rules as they be misleading if the output gap and
account for the effects of the cycle. potential growth estimates are too
optimistic or too pessimistic.
Expenditure Expenditure rules enhance fiscal A well-designed spending rule Spending rules have the same
rules discipline thanks to better accountability should not prevent the automatic drawbacks as budget balance rules
and higher transparency in the budget stabilisers from operating. Also, regarding the quality of spending or
process. They allow for better control of spending rules can limit pro- cuts, in particular in difficult times.
primary spending by the government. cyclical spending in the presence They also can induce the use of tax
They played an important role in of revenue windfalls in good times. expenditure for various policy
effective fiscal frameworks in the objectives for which direct spending
Netherlands, Denmark, Sweden, and might be better suited.
Finland (European Commission, 2010,
“Public Finances in EMU – 2010”,
European Economy, No. 4/2010.
Brussels: European Commission).
Revenue Revenue rules can be useful in limiting Revenue rules fail to Revenue rules can lead to
rules the tax burden. The impact on deficit accommodate macroeconomic additional spending, as it is difficult
limitation is a by-product as limitations stabilisation. However, they can to establish the cyclical component
on revenues should constrain help to limit the spending of of revenues. For instance, in some
governments from excessive spending. revenue windfalls in good times. countries prolonged booms in
It also serves to avoid pro-cyclical property and asset prices led to
policies, in particular, when the use of additional public expenditure, and
higher-than-expected revenues is therefore increased the risk of pro-
defined ex ante in the budget law. cyclicality and unsustainable fiscal
policies.

As shown in Table 5, most of the existing rules are numerical rules that set a clear target or constrain
government policy options. Debt rules are present in many countries and are sometimes complemented by
expenditure or budget balance rules. Seven OECD countries have a debt target, while five have a debt
limit, and 21 have the supranational EU debt limit.
European Union members have supranational rules including a debt limit of 60% of GDP, a deficit
limit of 3% of GDP, and an expenditure rule: the annual growth of primary expenditure – excluding
unemployment benefits and subtracting discretionary revenue increases – should not exceed long-term
nominal GDP growth.
Table 5. Synthesis of existing fiscal rules

Budget balance Structural balance
Debt rule Expenditure rule Revenue rule
rule rule
Budget deficit target Structural deficit Debt-to-GDP <= Expenditure = X% of GDP Revenue growth =
(in % of GDP) target in % of GDP X% of GDP (where Discretionary spending caps GDP growth
X is either 40%, and automatic spending cuts
50% and 60% of (the so-called sequester)
GDP)
Primary surplus Structural budget is Debt-to-GDP = 0 Spending growth < = x% or a Multi-year revenue
target in balance target in nominal or real terms path
Structural surplus = Spending growth = growth rate Revenue windfall

X% of potential GDP of GDP rule
Structural annual Spending growth = rate of
fiscal effort >= inflation + population growth
X% of GDP
Spending growth < = potential
growth
Spending growth < = x% or a
target + a second condition or
escape clause
Non-cyclical primary spending
(real) = max x%
Source: Synthesis based on IMF (2012), “Fiscal Rules at a Glance: Country Details from a New Dataset”, IMF Working Paper,
WP/12/273.
New estimations suggest that fiscal rules have an impact on fiscal performance (Table 6). The
indicators of fiscal performance are the primary balance, government spending, government receipts and
the cyclically-adjusted primary balance. The impact of different fiscal rules on these variables is estimated
with control variables of the main determinants of fiscal performance. An increase in government gross
debt is positively correlated with improvements in the primary balance and the cyclically-adjusted primary
balance, through decreasing government spending and increasing revenues. An increase of the old age
dependency ratio has a negative effect on the primary balance, by increasing spending and reducing
receipts.
The budget balance rule appears to have a positive and significant effect on the primary balance
(Table 6, Column P2) and a negative and significant effect on spending (Table 6, Column S2). In contrast,
spending rules are significant only in restraining government spending (Table 6, Column S3).
The estimations confirm that an index of the stringency of fiscal rules (sum of rules) has a positive and
significant impact on the primary balance and a constraining and significant effect on government spending
and receipt. These results are in line with the findings by Marneffe et al. (2011) and Debrun et al. (2013).
Also, Nerlich and Reuter (2013) find that numerical fiscal rules combined with fiscal councils and medium-
term budgeting framework help to improve the primary balance for EU member states. However, the
results may also reflect that disciplined countries are likely to adopt fiscal rules.

Table 6. Estimations of the effectiveness of fiscal rules

Primary balance Spending Receipts Cyclically-adj. primary balance
(P1) (P2) (P3) (S1) (S2) (S3) (R1) (R2) (R3) (C1) (C2) (C3)
Lag of the dependent variable 0.89*** 0.88*** 0.90*** 1.02*** 1.02*** 1.03*** 0.88*** 0.89*** 0.89*** 0.86*** 0.86*** 0.87***
[0.028] [0.031] [0.028] [0.016] [0.016] [0.017] [0.026] [0.026] [0.025] [0.028] [0.028] [0.028]
GG gross debt (t-1) 0.022*** 0.021*** 0.023*** -0.0085** -0.0080** -0.0093*** 0.0138*** 0.0132*** 0.0130*** 0.0228*** 0.0225*** 0.0239***
[0.0048] [0.0049] [0.0049] [0.0034] [0.0034] [0.0034] [0.0031] [0.0033] [0.0030] [0.0041] [0.0043] [0.0041]
Output gap (t-1) 0.0022 0.0088 0.0093 0.108*** 0.101*** 0.106*** 0.060*** 0.062*** 0.062*** -0.012 -0.0097 -0.003
[0.028] [0.027] [0.027] [0.023] [0.024] [0.023] [0.021] [0.020] [0.020] [0.025] [0.025] [0.024]
Openness * Terms of trade 0.0002*** 0.0002*** 0.0002*** -0.0002*** -0.0002*** -0.0002*** -0.0001 0 -0.0001 0.0001*** 0.0001*** 0.0001***
[0.0001] [0.0001] [0.0001] [0.0000] [0.0000] [0.0000] [0.0000] [0.0000] [0.0000] [0.0001] [0.0001] [0.0001]
Dependency ratio -0.0011** -0.0012** -0.0014*** 0.0002 0.0002 0.0004 -0.0006* -0.0005 -0.0005 -0.0013*** -0.0013*** -0.0015***
[0.0005] [0.0005] [0.0005] [0.0003] [0.0003] [0.0003] [0.0003] [0.0003] [0.0003] [0.0004] [0.0004] [0.0004]
Fiscal rule stringency index 0.24*** -0.23*** -0.12* 0.11
[0.079] [0.062] [0.071] [0.076]
Fiscal council dummy 0.61 0.76 0.24 -0.77** -0.23 -0.42 -0.14 0.14 0.14 0.032
[0.38] [0.47] [0.29] [0.34] [0.25] [0.41] [0.25] [0.35] [0.44] [0.26]
Fiscal rule index * Fiscal council -0.17 -0.025
[0.106] [0.116]
Fiscal council safeguard on budget -0.54 -0.89**
[0.38] [0.42]
Fiscal rule index * Fiscal council
0.24** 0.33***
safeguard on budget
[0.106] [0.1188]
Fiscal budget balance rule 0.55** -0.40** -0.21 0.24
[0.24] [0.19] [0.22] [0.23]
Fiscal budget balance rule * Fiscal
-0.66 0.86*** 0.46 -0.033
council
[0.48] [0.32] [0.41] [0.48]
Fiscal expenditure rule 0.36 -0.73*** -0.33 -0.045
[0.23] [0.20] [0.22] [0.19]
Fiscal expenditure rule * Fiscal
-0.070 0.47* 0.28 0.26
council
[0.30] [0.24] [0.28] [0.33]
Time fixed-effects Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes
Country fixed-effects Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes
N 619 619 619 622 622 622 623 623 623 611 611 611
Countries 30 30 30 30 30 30 30 30 30 30 30 30
Note: Standard errors are in parentheses. * significant at 10%, ** significant at 5%, *** significant at 1%. Dependent variables are adjusted for one-off expenditure.
Source: OECD Economic Outlook 95 and IMF fiscal council and fiscal rules dataset.

Designing fiscal rules consistent with debt targets and stabilisation objectives
The debt trajectory is the core concern of the fiscal framework. Given the uncertainties surrounding
the macroeconomic determinants of debt developments and also that these macroeconomic variables are
beyond the control of the government, though they are endogenously affected by government policies,
fiscal rules will target the main determinants of debt directly under government control.
Fiscal rules should have two objectives: (1) anchor fiscal policy expectations by targeting a prudent
debt level and (2) allow for macroeconomic stabilisation that enhances economic growth. Fiscal stimulus to
mitigate recessions increases uncertainties surrounding the debt path: there is a trade-off between these
two objectives (Figure 12). Real time stimulus may stabilise the cycle, but also affects debt dynamics. The
macroeconomic model, set out in Box 3, is used to illustrate this trade-off.
Figure 12. The trade-off between counter-cyclicality and hitting the debt target
Panel A. Long-term recession risks
Primary balance is kept constant Automatic stabilisers Automatic stabilisers and stimulus
Per cent
40
35
30
25
20
15
10
Panel B. Debt ratio uncertainties

Per cent of GDP
60
50
40
30
20
10
12http://dx.doi.org/10.1787/888933224069
Note: The long-term recession risk is the probability of GDP per capita growth to become negative. The uncertainty surrounding the
debt trajectory is assessed by the interquartile range of the debt level in 2040. The “Primary balance is kept constant” simulation is a
stylised scenario in which the actual primary balance is kept constant such that the prudent debt target is reached, with no automatic
stabilisers. In the scenario labelled “Automatic stabilisers”, a one percentage point negative surprise in the output gap is associated
with a 0.4% of GDP stimulus. In the “Automatic stabilisers and stimulus” scenario, the government is taking discretionary measures
on top of automatic stabilisers in reacting to changes in the output gap.

A good fiscal rule should have two main properties:
• The fiscal rules should define a confidence interval or a band for its main variables to
accommodate macroeconomic fluctuations. If the economy has a budget or structural deficit
close to a fixed limit, the limit constitutes an important restriction for the stabilisation effects of
both the automatic stabilisers and counter-cyclical discretionary fiscal policy in a downturn.
• A combination of rules should ensure compliance with the chosen debt trajectory. A budgetary
constraint cannot only refer to the budget balance, if it is to avoid potential adverse effects on
the underlying components of the budget balance (e.g. undesirable tax increases to
compensate for structural problems on the expenditure side).
The rules differ in their ability to fulfil objectives of reaching a prudent debt target and of stabilisation.
The structural budget balance rule combines, in principle, the capacity of satisfying the two objectives, but
it has important drawbacks in terms of observability and real time assessment. Structural balance
measures, despite some progress on measurement, are highly dependent on volatile and often biased
estimates of the output gap and subject to frequent revisions (Hers and Suyker, 2014). For instance, for
Slovakia, Klein et al. (2013) report that the structural deficit in 2010 using pre-crisis estimates of potential
growth would have been nearer to 4% than 8% of GDP.
Also, structural deficits can be revised sharply in the case of a crisis as potential growth is not well
measured in real time (Figure 13). In particular, the downward revision of the structural primary balance
was about 7% of GDP for Greece and about 3% of GDP in Ireland. These revisions follow growth
surprises: on average across countries, a 1% real output shock is associated with a 0.2% revision of the
potential output level (Fall and Fournier, 2015).
Figure 13. The pre-crisis overestimation of structural balances

2007 structural primary balance as a per cent of potential GDP
6
Published in May 2008 Published in November 2014
4
-2
-4
-6
-8
12http://dx.doi.org/10.1787/888933224071
Note: The structural primary balance published in May 2008 was calculated with the available information at that time, such as the
general government accounts, the provisional GDP estimate and the potential output calculated in 2008. In practice, the difference
between this publication and the latest one is mainly due to revisions in potential output, though there was also a sharp downward
revision of the actual primary balance for Greece.
Source: Economic Outlook No. 83 and No. 96 databases.
Therefore, the adoption of a budget balance rule complemented by an expenditure rule could suit
most countries well. As shown in Table 7, the combination of the two rules responds to the two objectives.
A budget balance rule encourages hitting the debt target. And, well-designed expenditure rules appear
decisive in ensuring the effectiveness of a budget balance rule (Guichard et al., 2007). Carnot (2014)
shows also that a binding spending rule can promote fiscal discipline while allowing for stabilisation

policies. The marginal benefit of adding a revenue rule is likely outweighed by its costs in terms of
complexity and reduction in fiscal flexibility.
Table 7. Synthesis of the effects of rules with respect to fiscal discipline and stabilisation
Budget balance Structural balance Expenditure rule Revenue rule

Fiscal stabilisation - + + -
Fiscal discipline ++ + + -+
Side-effects and risks - -- - -
Source: OECD calculations, based on Table 4.
The macroeconomic model, set out in Box 3, is used to illustrate the impact of a budget balance rule
combined with a spending rule. In the central scenario, the primary balance follows the estimated primary
balance reaction function (Figure 14, Panel A). The budget balance and spending rules lead to a higher
primary balance surplus and, thus, to a lower public debt path (Figure 14, Panel B).
Figure 14. Illustration of a budget balance rule combined with a spending rule: The US example
Panel A. Past behaviour primary balance
Panel B. Budget balance rule + spending rule
Tail events happen, but they need not undermine credibility. Clear escape clauses should be set
allowing the temporary suspension of fiscal rules. A temporary suspension should be conditional on
exceptional events such as natural catastrophes or a sharp output contraction. However, the definition of

these escape clauses must be clear to make sure they cannot be used in normal times. Determining the
existence of exceptional circumstances can be delegated to a body outside the government or submitted to
a validation by qualified majority in the parliament. To cope with tail events, a “rainy day” fund can underpin
the respect of the rule over the cycle and would allow greater room for fiscal stabilisation. Unexpected
surpluses would be saved and used later to finance unexpected deficits and/or short-term stabilisation
policies.
In the design of fiscal rules two cases should be considered: countries pursuing fiscal consolidation
towards a lower prudent debt target and countries for which the starting debt level is already below or
around the medium to long-run prudent debt target.
Fiscal rules for countries in transition towards a lower prudent debt level
For countries in transition towards a prudent debt level, a path either in terms of budget balance or
debt needs to be defined. The consolidation path links the current debt level to the prudent debt target.
There are many OECD studies on consolidation strategies that provide useful lessons.
Rawdanowicz (2012) shows that it is possible to choose an optimal consolidation path to bring the deficit
down and stabilise debt at a long-run target in a finite horizon. A previous OECD project on fiscal
consolidation (Sutherland et al. (2012), Barrell et al. (2012) and Merola and Sutherland (2012) stressed the
need to structure a consolidation strategy such that instruments with low multipliers are used initially and to
enhance the institutional framework for fiscal policy in order to minimise the trade-offs with growth in the
short run. Hageman (2012) shows that in most countries there is scope to target spending programmes
more effectively and eliminate distortions in taxation. Moreover, Cournède et al. (2013) find that there is
room in half of the OECD countries to reduce debt mainly through adjustments in instruments (such as
subsidies, pensions or property taxes) that have at most limited side-effects on other policy objectives,
such as growth or equity.
Finally, the consolidation strategy should involve all levels of government. During past consolidation
episodes, sub-central governments helped stabilise general government debt. In general, consolidation at
the state and – albeit less – at the local level increased the success of debt stabilisation at the general
government level (Blöchliger et al., 2012).
Therefore, choosing the consolidation path is the cornerstone of the calibration of fiscal rules (budget
7
balance rule complemented by an expenditure rule). Once the consolidation path is defined in terms of the
debt trajectory toward the prudent debt level, the corresponding budget balance and spending rules can be
deducted. The stringency of the consolidation path and of the fiscal rules depends on the time horizon to
reach the prudent debt level, which may depend on the distance to the prudent debt target. Also, the
stringency of the fiscal rules can be kept constant or increased to minimise the risk of missing the timing
and the target. When fiscal consolidation is initiated under pressure, then it may be important to build
credibility by imposing more stringency from the beginning.
The fiscal rule should include a mechanism to correct for past slippages in upcoming budgets. This
could be achieved by means of rules similar to the German “debt brake”, where fiscal slippages are
recorded in a special account that has to be balanced over time (Baumann and Kastrop, 2007 and
Kastrop et al., 2010).
7. See Rawdanowicz (2012) and Carnot (2014) who provide a rule for determining the structural budget
balance effort with the two objectives of debt sustainability (with debt target) and macroeconomic
stabilisation and illustrate the calibration.

Fiscal rules for preserving prudent debt levels
For countries that are already close to the debt target, the objectives of the rules are to guarantee
debt stability while allowing for stabilisation policies. If the objective is only to stabilise the debt level, then
the primary balance needs to be equal to the real cost (difference between real interest rate and real
growth times the debt level) of debt. But, as the objective is also to include room for macroeconomic
stabilisation, the rules need to allow enough flexibility to dampen fluctuations in economic activity.
There are three dimensions that need to be taken into account when designing the rule:
• First, annual budgeting for year t is realised in year t-1 based on the projections of
macroeconomic variables. There is a need for an independent fiscal institution to ensure
unbiased projections (see below). However, even with a fiscal institution, realised variables will
likely differ from the projections. Thus the rules should allow for contemporaneous involuntary
(unexpected shocks, automatic stabilisers) or voluntary (fiscal stabilisation) deviations.
• Second, because of these contemporaneous deviations, the rules should also include a
backward dimension requiring that past deviations get offset over a limited amount of time. That
is, the stringency of the budget balance and the spending rules is increased over time up to full
correction, taking the form of a binding constraint of ex-post corrections of deviations.
• Third, there is a need to anchor the spending path. A medium-term budget framework, in which
the government sets a spending trajectory over the medium term complements the budget rule
and enhances its effectiveness (see below).
Country-specific macroeconomic properties and fiscal rules
The macroeconomic properties of countries differ in terms of the business cycle, trade openness, and
exposition to financial developments that can affect the suitability of fiscal rules. The small macroeconomic
model developed to assess the macroeconomic uncertainties surrounding debt developments is used to
assess how different rules perform for the different countries in terms of debt uncertainty and recession
risk.
First, the central scenario presented in Box 3 is used to simulate the dynamics of the six main
variables (the growth rate 𝑔𝑖𝑖 , the inflation rate 𝜋𝑖𝑖 , the overnight nominal interest rate 𝑟𝑖𝑖𝑠 , the long-term
nominal interest rate 𝑟𝑖𝑖𝑙 , the primary balance 𝑃𝑃𝑖𝑖 , and the structural primary balance 𝑃𝑃𝑖𝑖𝑠𝑠 ). But, the
primary balance equation (fiscal reaction estimations) is replaced by stylised fiscal rules to determine the
primary balance and therefore debt developments. Short-term shocks are jointly-drawn as defined in Box 3
and Fall and Fournier (2015).
Box 4 presents the different fiscal rules that are simulated. In the simulations, the rules are
mechanically binding. However, governments have room to take discretionary fiscal measures in practice.
Adding random discretionary spending to this framework would increase the uncertainties of the debt path.

Box 4. Simulated fiscal rules
Six rules are simulated:
• In the baseline simulation, the annual budget plan is set so that the primary balance is equal to the target if
the output gap is in line with expectations. During the year, the government lets the automatic stabilizers
play around this plan (equation 1). The primary balance target is set to zero for countries that do not need to
generate a surplus to reach a prudent debt level by 2040. For the other countries, the primary balance target
is adjusted so that the prudent debt level is reached in 2040.
(1) 𝐵𝑡 = 𝑏𝑏𝑏𝑏𝑏𝑏𝑏 𝑡𝑡𝑡𝑔𝑔𝑔 + 𝛼. (𝑔𝑔𝑔𝑡 − 𝐸𝑡−1 (𝑔𝑔𝑔𝑡 )) − 𝑟. 𝐷𝑡−1
• Two simulations are considered to investigate the counter-cyclical role of fiscal policy. In the first one, there
are no automatic stabilisers, that is, the primary balance is kept constant such that the prudent debt target is
reached (equation 2).
(2) 𝐵𝑡 = 𝑏𝑏𝑏𝑏𝑏𝑏𝑏 𝑡𝑡𝑡𝑡𝑡𝑡 − 𝑟. 𝐷𝑡−1
• In the second one, the government is assumed to take discretionary measures (𝛼 = 0.4) on top of automatic
stabilisers to react to the output gap (equation 3).
(3) 𝐵𝑡 = 𝑏𝑏𝑏𝑏𝑏𝑏𝑏 𝑡𝑡𝑡𝑡𝑡𝑡 + 𝛼. (𝑔𝑔𝑔𝑡 − 𝐸𝑡−1 (𝑔𝑔𝑔𝑡 )) + 𝛼. 𝑔𝑔𝑔𝑡 − 𝑟. 𝐷𝑡−1
• The effect of a spending rule is investigated in a simulation in which structural spending grows less fast than
GDP so that the structural spending to GDP ratio decreases by 0.5 GDP point each year, for countries for
which the structural spending level is above the pre-crisis OECD average (37%). In this simulation, the
primary balance follows the same path as in the baseline (equation 4).
𝐵𝑡 = 𝑏𝑏𝑏𝑏𝑏𝑏𝑏 𝑡𝑡𝑡𝑡𝑡𝑡 + 𝛼. (𝑔𝑔𝑔𝑡 − 𝐸𝑡−1 (𝑔𝑔𝑔𝑡 )) − 𝑟. 𝐷𝑡−1

(4) �
𝑆𝑡 = min(𝑆𝑡−1 − 0.5, 𝑆 𝑇 ) 𝑖𝑖 𝑆𝑡−1 > 𝑆 𝑇
• The role of frontloading is investigated. If lagged debt is higher than the debt ceiling, then the government
th
generates an additional surplus equal to one 20 of the difference between lagged debt and this ceiling.
This comes on top of the effort made otherwise, which is set with the aim to reach a prudent debt level by
2040 (equation 5).
(5) 𝐵𝑡 = 𝑏𝑏𝑏𝑏𝑏𝑏𝑏 𝑡𝑡𝑡𝑡𝑡𝑡 + 0.05 ∗ max(𝐷𝑡−1 − 𝐷𝑇 , 0) + 𝛼. (𝑔𝑔𝑔𝑡 − 𝐸𝑡−1 (𝑔𝑔𝑔𝑡 )) − 𝑟. 𝐷𝑡−1
• The government’s target is set in terms of the actual balance including interest payments, instead of the
primary balance (equation 6).
(6) 𝐵𝑡 = 𝑏𝑏𝑏𝑏𝑏𝑏𝑏 𝑡𝑡𝑡𝑡𝑡𝑡 + 𝛼. (𝑔𝑔𝑔𝑡 − 𝐸𝑡−1 (𝑔𝑔𝑔𝑡 ))
𝐵𝑡 denotes the government balance, 𝛼 is the elasticity of the primary balance to the output gap, 𝑔𝑔𝑔𝑡 denotes the
output gap, 𝐷𝑡−1 denotes the debt ratio in the previous year, 𝑆𝑡 denotes the structural level of government spending in
per cent of GDP, 𝑆 𝑇 the targeted level of structural spending in per cent of GDP and 𝐷𝑇 the debt ceiling.
Second, two criteria are used to benchmark the performance of the different rules in the model: the
long-term recession risk, measured by the probability of GDP per capita growth to become negative and
the uncertainty surrounding the debt trajectory, assessed by the interquartile range of the debt level in
2040. The uncertainties surrounding the debt trajectory can be reduced by debt management. For
instance, longer debt maturities reduce roll-over risks, but at the cost of higher interest payments.

The simulations suggest the following:
• The capacity of fiscal policy to mitigate shocks and its adverse consequence on debt trajectory
uncertainty varies substantially across countries. Rules with stimulus compared to the baseline
scenario reveal that highly indebted countries have less potential to counteract large adverse
shocks, due to the binding budget balance component and the higher debt and GDP growth
uncertainties.
• Spending rules entail no trade-off between minimising recession risks and minimising debt
uncertainties. They can boost potential growth and hence reduce the recession risk without any
adverse effect on debt. Indeed, estimations show that public spending restraint is associated
with higher potential growth (Fall and Fournier, 2015).
• For all countries, sticking to initial annual budget plans and restoring the primary balance in the
following years reduces debt uncertainty without a substantial increase of recession risks. In
terms of cyclical corrections, the budget balance rule compares well with the structural balance
rule because when setting each year a numerical target, the deviations due to the cycle can be
taken into account. That is, the correction of the budget balance target due to the cycle can be
incorporated ex-post.
• Rules based on a measure of the balance that includes interest payments are useful to reduce
debt trajectory uncertainty in countries in transition towards a prudent debt level.
Countries are grouped into six groups depending on a combination of their characteristics and their
responses to the different rules. Four criteria are used for the grouping of countries: the effectiveness of
8
counter-cyclical fiscal policy, having room to restrain spending, belonging to the euro area and
consolidation needs. In the first group are countries where fiscal policy has a strong counter-cyclical effect
and where this strong effect does not have a large effect on the uncertainty surrounding the debt path (see
Fall and Fournier, 2015 for details). In the second group are countries with a low level of public spending
and a moderate effectiveness of fiscal policy to damp short-term shocks. The third group includes
countries that have room for spending restraint, but excluding those euro area countries that need to
generate large primary surpluses. The fourth group includes the euro area countries that face a moderate
need to consolidate and have room to restrain spending. The fifth group gathers euro area countries that
need to raise the primary balance to 2% of GDP or even more to hit the prudent debt ratio in 2040. Last,
Japan has such a high debt level, that hitting a prudent debt target by 2040 would imply an unrealistically
large fiscal tightening in the short run.
The fiscal rules have a different performance with regard to the two criteria (long term recession risks
and debt uncertainties) for six groups of countries (Figure 15):
• Group 1 (Australia, Israel, Korea, New Zealand, Poland, Switzerland and the United States):
Countries that can rely more on fiscal policy to mitigate short-term shocks. In these countries, a
stronger reaction of fiscal policy to the cycle is efficient in reducing recession risks with a
moderate cost in terms of debt uncertainty. Among these countries, the United States also
needs to generate surpluses to bring debt back to a prudent level.
• Group 2 (Canada, Czech and Slovak Republic): Countries with a low level of public spending
and a moderate effectiveness of fiscal policy to damp short-term shocks. The low level of public
spending reveals there is little to gain from focusing on spending restraint, but a spending rule
that makes sure this low level remains in the long run is appropriate. In these countries,
counter-cyclical policy entails a considerable rise in debt uncertainty for a moderate reduction of
recession risks.
8. Belonging to the euro area is a criterion, because of the absence of monetary policy at the country level.

• Group 3 (Denmark, Luxembourg, Sweden and the United Kingdom): These countries have
room for spending restraint. These countries do not need to generate primary surpluses to
target a prudent debt level by 2040, except the UK that needs to generate a moderate one. In
Luxembourg and Sweden, a stronger reliance on fiscal policy to mitigate short-term shocks
would increase debt uncertainty with moderate gains in terms of lowering recession risks.
• Group 4 (Austria, Belgium, Finland, France, Germany, the Netherlands and Slovenia): These
euro area countries need to generate primary surpluses and Germany has to keep a primary
surplus to bring debt back to a prudent level and they should restrain spending. These countries
need both a spending and a debt rule to stick to the debt trajectory.
• Group 5 (Greece, Ireland, Italy, Portugal and Spain): These euro area countries need to
generate large primary surpluses. These countries also face a large debt dynamics uncertainty.
This uncertainty can be reduced with a debt rule, and by making sure that any widening of the
primary balance triggered by a weaker economy is temporary. Because of the debt dynamics
uncertainty, they have little room to mitigate recession risks with fiscal policy.
• Group 6 (Japan): The most indebted country is in a class of its own. Japan needs to reduce
debt, but this process should be protracted as there is no strong adverse effect of the debt level
on interest rates. Setting a prudent debt target by 2040 implies an unrealistic and unnecessarily
large fiscal tightening in the short run. Counter-cyclical fiscal policy should not be used
intensively as it leads to large debt trajectory uncertainty for quite a small reduction of recession
risks. The low level of public spending suggests that most of the consolidation effort will have to
come from the revenue side.

Figure 15. Comparison of fiscal rules by country group

Panel A. Long-term recession risks
Constant primary balance Automatic stabilisers Additional stimulus

Per cent
Spending rule Include interest payments Debt rule
35
30
25
20
15
10
0
Group 1 Group 2 Group 3 Group 4 Group 5 Group 6
Panel B. Debt level uncertainties

Per cent GDP
45
40
35
30
25
20
15
10
5
0
Group 1 Group 2 Group 3 Group 4 Group 5 Group 6
12http://dx.doi.org/10.1787/888933224084
Note: The bars are averages for the countries in the group. The “Constant primary balance” simulation is a stylised scenario in which
the actual primary balance is kept constant such that the prudent debt target is reached, with no automatic stabilisers. In the scenario
labelled “Automatic stabilisers”, a one percentage point negative surprise in the output gap is associated with a 0.4% of GDP
temporary stimulus. In the “Additional stimulus” scenario, the government is taking discretionary measures on top of automatic
stabilisers to react to the output gap. In the “Spending rule” scenario, the government lets the automatic stabilisers play during the
current year only as in the baseline (rule 1); and structural spending grows by 0.5 percentage points less than potential GDP, for
countries for which the structural spending level is above the pre-crisis OECD average (37%) until it reaches this average. In the
“Include interest payments” scenario, the government’s target is set in terms of the actual balance including interest payments,
instead of the primary balance. In the “Debt rule” or frontloading scenario, the primary balance is increased by one twentieth of the
difference between the debt level and its ceiling when debt is above this ceiling. This comes on top of the effort made otherwise. The
long-term recession risk is the probability that GDP per capita growth becomes negative. The uncertainty surrounding the debt
trajectory is assessed by the interquartile range of the debt level in 2040.

The simulation results are summarised in Table 8.
Table 8. Simulation results: Grouping of countries and fiscal rules
Groups of countries Rules

Group 1: Australia, Israel, Korea, New Zealand, Poland, Budget balance rule and considerable room for counter-
Switzerland and the United States cyclical fiscal policy
Group 2: Canada, Czech and Slovak Republic Budget balance rule
Group 3: Denmark, Luxembourg, Sweden and the United Budget balance rule and spending rule with some bite
1
Kingdom
Group 4: Austria, Belgium, Finland, France, Germany, the Budget balance rule and spending rule with strong bite
Netherlands and Slovenia
Group 5: Greece, Ireland, Italy, Portugal and Spain Budget balance rule that aims at a considerable primary
surplus with some frontloading
Group 6: Japan Considerable primary surplus with some frontloading
1. The United Kingdom has a spending-to-GDP ratio somewhat above the OECD average, while the other countries in this group
have high spending ratios.
5.4. Fiscal councils
There is a growing literature on the design of fiscal councils and their effectiveness (see for surveys
Debrun et al. 2009, Hagemann, 2011, Kopits, 2011b, and Debrun and Kinda, 2014). This section reviews
the arguments for setting up fiscal councils, their role and the conditions for their effectiveness.
Why do we need a fiscal council?
Fiscal councils foster fiscal discipline and thereby complement fiscal rules. As the rules are designed
to hit fiscal targets and allow for stabilisation policies, deficit bias behaviour should be disentangled from
accommodation of short-run shocks to make sure that governments respect their commitments. To that
end, an external body can verify whether the government complies with the fiscal rules.
A fiscal council can underpin transparency and thus credibility. The adoption of fiscal rules, in
particular of complex rules, increases the need for transparency. Governments may be biased toward
overestimating expected revenues, underestimating spending, or unduly exercise escape clauses (Debrun
et al. 2013). A fiscal council can help discipline government behaviour.
Finally, a fiscal council can help address the common-pool problem, in particular in federal states, by
warning on externalities of fiscal policies generated by sub-central governments that may benefit them at
the expense of others.
Design and remit of fiscal institutions
The role of fiscal councils is to monitor ex ante that fiscal policy is likely to meet the short and long-run
targets while allowing fiscal flexibility. Rating agencies have shown clear limitations in terms of monitoring
public finances.
There is “no one size fits all” in terms of designing a fiscal council. The design should depend on the
country’s political culture, legal traditions and economic characteristics. In some countries, a fiscal council
with a monitoring role is not deemed necessary due the high standards of accountability and transparency
already in place (New Zealand); while in others, an independent authority is in charge of fiscal surveillance
(United Kingdom).

Fiscal councils often have very different roles and remits. In addition to analysis of budget proposals
and current fiscal developments, common functions include producing independent projections and
forecasts or endorsing or assessing government projections and forecasts; monitoring compliance with
fiscal rules and targets; analysis of long-term fiscal sustainability; costing of policy proposals; and analytical
studies on selected issues. A minority of fiscal councils also provide normative assessments.
The exact combination of these features in the remit of the fiscal council should be guided by the
weaknesses of and risks surrounding fiscal policy. Table 9 shows the different mandates of existing fiscal
councils. If the government tends to miss its targets due to over-optimistic fiscal projections, then these
forecasts should be delegated to a fiscal council, or alternatively a council could audit government
forecasts. If the main issue is insufficient assessment of the long-run consequences of decisions, providing
fiscal sustainability calculations highlighting these consequences might alleviate the bias.
Another important role for fiscal councils is the assessment of the need for counter-cyclical fiscal
policy. Governments often miss the opportunity to build fiscal buffers in good times and a council could call
for adjustments in that case. Moreover, when the government calls for the application of escape clauses to
deviate from the fiscal rules, then the fiscal council could be in charge of assessing the correctness of this
call.
Conditions for the effectiveness of a fiscal institution
Fiscal councils are relatively new, but their number within the OECD has tripled since the crisis started
to 20 in 2013. They differ in terms of mandate and features, and the qualitative nature of an important part
of their work renders the assessment of their effectiveness difficult. In addition, the analysis is subject to
reverse causation as there is the possibility that disciplined countries reveal their preferences by adopting
such institutions.
The estimations in Table 6 show that it is difficult to capture the effectiveness of fiscal councils. The
impact of fiscal councils on the primary balance is not significant in most of the specifications. However,
fiscal councils appear to limit spending when associated with a budget balance rule (Table 6, Column S2).
These results are in line with the findings by Debrun et al. (2013) who conducted a thorough analysis
of the performance of fiscal councils. On average the mere existence of a fiscal council appears only
loosely related to stronger fiscal outcomes. But, higher primary balances are associated with fiscal councils
featuring certain characteristics. For instance, the task of monitoring compliance with fiscal rules is unlikely
to be sufficient to affect fiscal performance if it is not paired with strict independence and a presence in the
public debate. They also find that, in the European Union, countries with fiscal councils have stronger fiscal
positions, more accurate and less optimistic forecasts and policy is less pro-cyclical.

Table 9. The remit of fiscal institutions in OECD countries
Mandate
Normative Forecast Forecast Long-term Costing of Monitoring of Ex-post
Optimality2 beyond
analysis preparation assessment sustainability1 measures3 fiscal fules analysis
fiscal policy
Australia Parliamentary Budget Office X X
Austria Fiscal Advisory Council X X X X X
Belgium High Council of Finance X X X X
Belgium Federal Planning Bureau X X X X X X
Canada Parliamentary Budget Office X X X X X X
Denmark Danish Economic Council X X X X X X X X
Finland National Audit Office of Finland X X X X X X
France High Council of Public Finance X X X X
Germany Stability Council X X X X X X
Hungary4 Fiscal Council X X X X X X
Ireland Irish Fiscal Advisory Council X X X X X X
Italy Parliamentary Budget Office X X X X X X
Fiscal System Council and Council on Economic
Japan X X
and Fiscal Policy
Korea National Assembly Budget Office X X X X X X
Mexico Center for Public Finance Studies X X X X
Netherlands CPB Bureau for Economic Policy
Netherlands X X X X X X X
Analysis
Portugal Portuguese Public Finance Council X X X X X
Slovak Republic Council for Budget Responsibility X X X X X
Institute of Macroeconomic Analysis and
Slovenia X X X X X X X
Development
Slovenia Fiscal Council X X X X
Sweden Swedish Fiscal Policy Council X X X X X
United Kingdom Office for Budget Responsibility X X X X X X X
United States Congressional Budget Office X X X X X
1. Long-term sustainability is defined as the long-term forecast of government balance and debt level.
2. "Optimality" is defined as the assessment of government budgetary and fiscal performance in relation to fiscal objectives and strategic priorities.
3. "Costing of Measures" is defined as the quantification of either short-term or long-term effects, or both, of measures and reforms.
4. The council has been significantly weakened in 2010.
Source: IMF Fiscal Council Dataset (February 2014) and OECD (2015), Principles for Independent Fiscal Institutions and Country Notes, forthcoming.

These empirical findings are in line with Kopits (2011a; 2011b), who identifies the following features
as critical for the effectiveness of a fiscal institution: a) home-grown and home-owned design and
operations; b) independence, non-partisanship, technical competence, and accountability to the legislature;
c) sufficient technical support staff, with unlimited access to timely information from the government; d) a
remit consisting of the assessment of the fiscal stance and debt sustainability – including monitoring of the
compliance with rules or targets – through real-time estimation of the budgetary effects of legislative
proposals; and e) effective means of communication to the public, ensuring a high level of transparency.
These features correspond to the principles for independent fiscal institutions agreed by OECD
member countries set out in OECD (2014). Among the nine headline principles emerge the key roles of the
relationship with the legislature and the involvement in budget procedure, independence and non-
partisanship and access to information for an effective fiscal institution.
5.5. Medium-term budgetary frameworks
A medium-term budgetary framework (MTBF) is useful for the effectiveness of the fiscal framework.
The role of the MTBF is to guarantee time-consistency between policies and targets. The MTBF is crucial
to convert fiscal targets into detailed revenue and expenditure plans. Successful medium-term budget
frameworks provide binding restrictions on multi-year expenditure and a clear and consistent statement of
the government’s medium-term priorities within an overall expenditure ceiling (Gupta and Yläoutinen,
2014).
The MTBF needs to be anchored in medium-term numerical objectives (Blondal, 2005). These
objectives are intermediary objectives in line with the long-term fiscal targets. Fixed medium-term
objectives imply that fiscal rules (budget and spending targets) are not adjusted over time unless
unexpected exceptional events arise during the period covered by the framework. For instance, in the
Netherlands, Sweden, Finland and the new framework in the United Kingdom, the MTBFs are based on a
multi-annual spending rule providing binding expenditure limits. In the case of the Netherlands, each newly
elected government announces its medium-term budgetary objectives in accordance with the long- run
fiscal target. Then, over the legislature, policies are assessed with regard to their impact on the target.
The key factor in the effectiveness of a medium-term budgetary framework is its influence on annual
budgeting. The medium-term path derived from the MTBF should be binding for the budget balance and
spending rules. The ex-ante setting of numerical targets for the budget balance and spending rules from
the MTBF implies an adjustment of spending and revenue plans to ensure the respect of the different
targets. In particular, a prioritisation of spending plans is necessary to make sure that whenever cuts in
spending are necessary to respect the fiscal framework, it will not hit policy priorities.
The MTBF is subject to the same risks as fiscal rules, which include non-binding fiscal targets, limited
political commitment and rosy projections. Therefore, as for fiscal rules, a fiscal council is needed to
guarantee transparency and credibility of the framework.
5.6. Budgetary procedures
The budgetary process is also important in ensuring that fiscal rules, in particular spending targets,
are met. OECD (2014) sets out the principles of budgetary governance agreed by OECD member
countries. Along the ten headline principles emerge the key roles of transparency, sincerity and
coordination. Countries deemed successful in the implementation of their fiscal framework (Sweden and
the Netherland, for instance) have a centralised budget process (Ministry of finance, Central Budget
Administration) ensuring the consistency between ex-ante spending plans and actual spending. Also the
centralisation of budgeting procedures is deemed necessary for the effectiveness of top-down budgeting
(EC, 2010). In particular, the central budget office should have the possibility to veto any over-spending
until a new law is voted to authorise it.
Successful budgetary procedures are likely to be country-specific depending on the size of the
country, its central or federal nature and the features of the political system. Von Hagen and Harden (1994)

find that budget processes of all governments of large states that successfully limited spending and deficits
in the 1970s and 1980s (France, Britain, and Germany) are based on a procedure-oriented approach, that
is when there is a process of negotiations between spending ministers and a central minister (budget or
finance) which has some power. In contrast, the budget processes of smaller countries (Denmark, the
Netherlands and Luxembourg) that successfully limited spending and deficits are target-oriented. This
suggests that country size matters for the complexity of administrations, which makes it more difficult to
monitor compliance with numerical budget targets.
OECD (2014) and Gupta and Yläoutinen (2014) find that top-down budgeting is one of the key
elements that underpin a credible fiscal strategy. In particular, during fiscal consolidation episodes, a top-
down budgeting approach imposes limits to spending both at the aggregate and sectoral level, which
increases the likelihood that the budget execution is consistent with the ex-ante fiscal plan. The spending
limit can be reinforced by requiring that a particular expenditure item cannot be raised without cutting
spending within the same area.
6. Political economy issues
6.1. The credibility of the consolidation path matters for avoiding fiscal fatigue
The choice of the consolidation path is crucial to limit the occurrence of fiscal fatigue, which could
lead to higher interest rates with negative effects on demand. It takes time to bring debt back to a prudent
level. In recent years, successful fiscal consolidation policies took on average ten years. Canada’s
consolidation started in 1993 and was interrupted only by the 2008-09 crises. For Belgium, the Netherlands
and Sweden, debt consolidation followed the same dynamics, starting in the mid-1990s and lasted more
than ten years. Given these consolidation durations, a strong frontloading in the initial years, though
deemed efficient, should be avoided and a smoother path privileged. For instance, it took seven years for
Canada’s debt-to-GDP ratio to decrease from 102% in 1993 to 84% in 2000. During the four years to 1997,
the improvement in the general government financial balance was significant, averaging 2% of GDP per
annum.
6.2. Political economy of fiscal consolidation
Some political economy lessons can be drawn from successful fiscal consolidation episodes. Posner
and Sommerfeld (2013) reviewed fiscal consolidation episodes lasting 6 to 9 years in Australia, Canada,
New Zealand, Sweden and the United States. Many OECD countries experienced consolidation episodes
that lasted over long periods and achieved much in terms of consolidation (Guichard et al., 2007). For
instance, Blöchliger et al. (2012) found 13 cases of fiscal consolidation between 1980 and 2000 where
deficits were reduced from 8% to 1% of GDP on average within five years. The lessons from Posner and
Sommerfeld’s (2013) review and case studies can be summarised as follows:
• Triggers for consolidation: As spending cuts or tax increases are contentious, they are seldom
undertaken in normal times. As a general rule, fiscal consolidation is undertaken only when
public finances are weak, as measured by rising deficits and debt levels (Molnar, 2012). Some
studies suggest that a crisis is needed to get countries to undertake deeper and more
prolonged consolidations (Larch and Turrini, 2008). For instance, Sweden undertook fiscal
consolidation policies after unemployment jumped from 2% in 1990 to 8% in 1993, a budget
deficit of 12% of GDP in 1993 and negative growth.
• Economic context: There is a tension between fiscal austerity and growth. However, many
countries undertake consolidation during times when economies have emerged from
recessions. In these times, deficit reduction can boost growth, partly by convincing central
banks and markets to lower interest rates (Von Hagen and Strauch, 2001). For instance,
Sweden undertook significant deficit reduction in 1994 just as its economy was embarking on a
strong recovery.

• Political institutions and timing: Successful consolidations require strategic timing and strong
support from political parties. With regard to timing, governments are most successful when
consolidations are introduced in the immediate aftermath of an election (Guichard et al., 2007).
A mandate to consolidate is also important: for instance, the newly elected governments of
Canada and Sweden had both campaigned on the need for consolidation, providing a mandate
for strong action. The research is ambiguous with regard to the impact of strong ruling parties
on the prospects for fiscal consolidation. However, strong majorities (one party ruling majority,
unified presidential majority, bi-partisan coalition) are associated with more decisive action,
longer lasting fiscal consolidations and fundamental reforms of major spending areas (Alesina
et al., 2006 and Larch and Turrini, 2008).
• The composition of consolidation initiatives: Sharing sacrifices and balancing spending and
revenue actions can help to cushion the view that consolidation is unfair or at the advantage of
specific groups. For instance, Sweden imposed an 11% across-the-board cut on nearly all
programmes and agencies at the outset of its consolidation in 1994. Compensating losers with
packages that cement coalitions by providing gains to offset a portion of the losses of major
groups is also a strategy.
• Political competition and consolidation: Conventional wisdom suggests that political leaders are
short-sighted, and the political rewards of consolidation are likely to be overshadowed by near-
term political considerations. Brender and Drazen (2006) found that governments achieving
lower deficits through policy actions actually increased the probability of their re-election. Using
data on 164 elections from 23 OECD countries from 1960 through 2003, they found that
controlling for changes in the economy, a reduction of 1 percentage point in the deficit/GDP
ratio increased the probability of re-election for existing governments by 5.7 percentage points.
• Alesina et al. (1998) find that during sharp adjustments that rely primarily on spending cuts and,
in particular, on the major components of government wages and transfers, the probability of
government survival increased. Also Posner and Sommerfeld (2013) find that the nine OECD
countries running persistent surpluses were re-elected in 63% of the 24 elections in the eight
years up to the Great Recession. By contrast, countries with the highest deficits during this
period were re-elected 40% of the time. More importantly, the governments in surplus countries
had far greater success in elections following the Great Recession. Incumbent governments
won in six of eight elections in the surplus group, but only won in one of the five elections in the
deficit group.
• In the European countries hard hit by the crisis the ruling party that undertook important
consolidation policies lost the following elections. In many of these countries, the legislature did
not last until its term and elections were held in advance:
− Greece: The ruling party at the outset of the crisis (New Democracy) lost the election in
October 2009. The winner Pasok, which undertook the consolidation policies lost the election
in May-June 2012 by a wide margin. The New Democracy party, which succeeded and
continued the consolidation progamme, lost also by a wide margin in January 2015.
− Ireland: The winner of the May 2007 election (Fianna Fáil) lost the February 2011 election
after starting the consolidation programme.
− Portugal: The ruling party at the outset of the crisis (Socialist Party) won the September 2009
election by a small margin, but after undertaking the consolidation programme it lost the 2011
election.
− Spain: The Spanish socialist party won the 2008 election and in the aftermath of the crisis
started the consolidation programme. It lost the 2011 election.

BIBLIOGRAPHY
Afonso, A. and J. Alves ( 2014). “The Role of Government Debt in Economic Growth,” Working Papers
Department of Economics 2014/16, ISEG - School of Economics and Management, Department of
Economics, University of Lisbon.
Aguiar, M. and M. Amador (2013), “Sovereign Debt: A Review”, NBER Working Paper, No. 19388.
Alesina, A. and G. Tabellini (1990), “A Positive Theory of Fiscal Deficits and Dovernment Debt”, The
Review of Economic Studies, Vol. 7, Issue 3, pp. 403-414.
Alesina, A., S. Ardagna and F. Trebbi (2006), “Who Adjusts and When? On the Political Economy of
Reforms”, NBER Working Paper, No. 12049.
Alesina, A., R. Perotti and J. Tavares (1998), “The Political Economy of Fiscal Adjustments”, Brookings
Papers on Economic Activity, 1:197-266.
Alesina, A. and T. Bayoumi (1996), “The Costs and Benefits of Fiscal Rules: Evidence from U.S. States”,
NBER Working Paper, No. 5614.
Arellano, C. and N. R. Kocherlakota (2008), “Internal Debt Crises and Sovereign Defaults”, NBER Working
Paper, No. 13794.
Aschauer , D. A. (2000), “Do States Optimise? Public Capital and Economic Growth”, Annals of Regional
Science, Vol. 34, pp. 343-63.
Auerbach, A. and Y. Gorodnichenko (2012), “Measuring the Output Responses to Fiscal Policy”, American
Economic Journal: Economic Policy, Vol. 4(2), pp. 1-27.
Auerbach, A.J. (2002), “Is There a Role for Discretionary Fiscal Policy?”, in Rethinking Stabilization Policy,
a symposium sponsored by the Federal Reserve Bank of Kansas City, Jackson Hole (WY), August
29-31, pp. 109-50.
Ballagria, F. and C. Martinez-Mongay (2009), “A Further Inquire about the Sustainability of Fiscal Policy in
the EU”, in: Larch, M. and J. Nogueira-Martins (eds), Fiscal Policy Making in the European Union,
Brussels.
Barnes, S. and D. Smyth (2013), The Government's Balance Sheet after the Crisis: A Comprehensive
Perspective, Irish Fiscal Advisory Council.
Barrell, R., D. Holland and I. Hurst (2012), “Fiscal Consolidation: Part 2. Fiscal Multipliers and Fiscal
Consolidations”, OECD Economics Department Working Papers, No. 933, OECD Publishing, Paris.
Baum, A., Checherita - Westphal, C., and Rother, P. (2013), “Debt and Growth: New Evidence for the Euro
Area”, Journal of International Money and Finance, 32, 809-821.
Baumann, E. and C. Kastrop (2007), “A New Budget Rule for Germany”, in Banca d’Italia (ed.), Fiscal
Policy: Current Issues and Challenges, acts of the 9th Public Finance Workshop held in Perugia,
29-31 March.
Blanchard, O. J. and D. Leigh (2013), “Growth Forecast Errors and Fiscal Multipliers”, IMF Working Paper,
13/01.

Bloch, D. and F. Fall (2015), “Government Debt Indicators: Understanding the Data”, OECD Economics
Department Working Papers, No. 1228, OECD Publishing, Paris.
Blöchliger, H. (2013), “Fiscal Consolidation across Government Levels – Part 1. How Much, What
Policies?”, OECD Economic Department Working Paper, No. 1070. OECD Publishing, Paris.
Blöchliger, H., D. Song and D. Sutherland (2012), “Fiscal Consolidation: Part 4. Case Studies of Large
Fiscal Consolidation Episodes”, OECD Economics Department Working Papers, No. 935, OECD
Publishing, Paris.
Blondal, J. (2005), “The Reform of Public Expenditure Management Systems in OECD Countries”, in
Banca d’Italia (ed.), Fiscal Policy: Public Expenditures, acts of the 7th Public Finance Workshop held
in Perugia, 31 March – 2 April.
Bohn, H. and R. Inman (1996), “Balanced-budget Rules and Public Deficits: Evidence from the U.S.
States”, Carnegie-Rochester Conference Series on Public Policy.
Brender, A. and A. Drazen (2006), “Political Implications of Fiscal Performance in OECD Countries”, in
Banca d’Italia (ed.), Fiscal Policy: Fiscal Indicators, acts of the 8th Public Finance Workshop held in
Perugia, 30 March – 1 Avril.
Calmfors, L. and S. Wren-Lewis (2011), “What Should Fiscal Councils Do?”, Economic Policy,
pp. 649-695.
Carlino G. and R. P. Inman (2013), "Local Deficits and Local Jobs: Can U.S. States Stabilize their own
Economies?", NBER Working Paper, No. 18930.
Carnot, N. (2014), “Evaluating Fiscal Policy: A Rule of Thumb”, European Economy, Economic Papers,
No. 526.
Cecchetti, S., M. Mohanty and F. Zampolli ( 2011), “The Real Effects of Debt”, BIS Working Papers,
No. 352, Bank for International Settlements.
Checherita-Westphal, C., A. H. Hallett and P. Rother (2014), “Fiscal Sustainability Using Growth-
maximizing Debt Targets”, Applied Economics, Vol. 46(6), pp. 638-647.
Chudik, A., K. Mohaddes, H. Pesaran and M. Raissi (2013), “Debt, Inflation and Growth: Robust Estimation
of Long-Run Effects in Dynamic Panel Data Models”, Federal Reserve Bank of Dallas Globalization
and Monetary Policy Institute Working Paper, No. 162.
Clemens, J. P. and S. I. Miran (2012), “Fiscal Policy Multipliers on Sub-National Government Spending”,
American Economic Journal: Economic Policy, 4, pp. 46-68.
Cournède, B., A. Goujard and Á. Pina (2013), “How to Achieve Growth- and Equity-friendly Fiscal
Consolidation? A Proposed Methodology for Instrument Choice with an Illustrative Application to
OECD Countries”, OECD Economics Department Working Papers, No. 1088, OECD Publishing,
Paris.
Debrun, X. and T. Kinda (2014), “Strengthening Post-Crisis Fiscal Credibility—Fiscal Councils on the Rise.
A New Dataset”, IMF Working Paper, WP/14/58.
Debrun X., T. Kinda, T. Curristine, L. Eyraud, J. Harris, and J. Seiwald, (2013), “The Functions and Impact
of Fiscal Councils”, IMF Policy Paper, July.
Debrun, X. et al. (2008), “Tied to the Mast? The Role of National Fiscal Rules in the European Union”,
Economic Policy, No. 54, pp. 298-362.

Debrun, X., D. Hauner and M. Kumar (2009), “Independent Fiscal Agencies”, Journal of Economic
Surveys, No. 23, pp. 44-81.
Égert, B. (2013), “The 90% Public Debt Threshold: The Rise and Fall of a Stylised Fact”, OECD
Economics Department Working Papers, No. 1055, OECD Publishing, Paris.
Égert, B. (2010), “Fiscal Policy Reaction to the Cycle in the OECD: Pro- or Counter-cyclical?”, OECD
European Commission (EC) (2011), “Public Finances in EMU – 2011”, European Economy, No. 4/2011,
European Commission, Brussels.
European Commission (EC), 2010, “Public Finances in EMU – 2010“, European Economy, No. 4/2010
European Commission, Brussels.
Eurostat (2015), “First Time Release of Data on Contingent Liabilities and Non-performing Loans in EU
Member States”, news release, http://ec.europa.eu/eurostat/documents/2995521/6616449/2-
10022015-AP-EN.pdf/d75df6fe-100b-4ae7-a09e-00400edb183a.
Fall, F., D. Bloch, P. Hoeller, J.K. Pareliussen and M. Pisu (2014), “Vulnerability of Social Institutions”,
OECD Economic Policy Papers, No. 11, OECD Publishing, Paris.
Fall, F. and J-M. Fournier (2015), “Macroeconomic Uncertainties, Prudent Debt Targets and Fiscal Rules”,
OECD Economics Department Working Papers, No. 1230, OECD Publishing, Paris.
Fatás, A. and I. Mihov (2003), “The Case for Restricting Fiscal Policy Discretion”, The Quarterly Journal of
Economics, Vol. 118(4), pp. 1419-1447.
Fatas, A. and I. Mihov (2006), “The Macroeconomic Effects of Fiscal Rules in the US States”, Journal of
Public Economics, Vol. 90(1-2), pp. 101-117.
Fournier, J-M. and F. Fall (2015), “Limits to Government Debt Sustainability”, OECD Economics
Galí, J., J. D. López-Salido, and J. Vallés (2007), “Understanding the Effects of Government Spending on
Consumption”, Journal of the European Economic Association, Vol. 5(1), pp. 227-70.
Ghosh, A. R., J. D. Ostry and M. S. Qureshi (2013), “Fiscal Space and Sovereign Risk Pricing in a
Currency Union”, Journal of International Money and Finance, Vol. 34, pp. 131-163.
Giavazzi, F. and M. Pagano (1990), “Can Severe Fiscal Contractions Be Expansionary? Tales of Two
Small European Countries”, in: NBER Macroeconomics Annual 1990, ed. by O. J. Blanchard and S.
Fischer.
Guichard, S. et al. (2007), “What Promotes Fiscal Consolidation: OECD Country Experiences”, OECD
Guillemette, Y. (2010), “A Simulation Model of Federal, Provincial and Territorial Government Accounts for
the Analysis of Fiscal Consolidation Strategies in Canada”, OECD Economics Department Working
Papers, No. 818, OECD Publishing, Paris.
Gupta, S. and S. Yläoutinen (2014), “Budget Institutions in Low-Income Countries: Lessons from G-20”,
IMF Working Paper, No. 14/164.

Hagemann, R.P. (2012), “Fiscal Consolidation: Part 6. What Are the Best Policy Instruments for Fiscal
Consolidation?”, OECD Economics Department Working Papers, No. 937, OECD Publishing, Paris.
Hagemann, R.P. (2011), “How Can Fiscal Councils Strengthen Fiscal Performance?”, OECD Journal:
Economic Studies, Vol. 2011/1.
Halac, M. and P. Yared (2014), “Fiscal Rules and Discretion under Persistent Shocks”, Econometrica,
Vol. 82, No. 5, pp. 1557–1614.
Hepp, R. and J. von Hagen (2009), “Fiscal Federalism in Germany: Stabilization and Redistribution Before
and After the Unification”, CEPR Discussion Paper, No. 7246.
Hers, J. and W. Suyker (2014), “Structural Budget Balance: A Love at First Sight Turned Sour”, CPB Policy
Brief, 2014/07
IMF (2014), “Public Expenditure Reform: Making Difficult Choices”, Fiscal Monitor, Washington D.C., April.
Kastrop, C., G. Meister-Scheufelen and M. Sudhof (2010) (eds.), Die neuen Schuldenregeln im
Grundgesetz, Berliner Wissenschaft Verlag, pp. 121-23, Berlin.
Klein, C., R. W. Price and A. Wörgötter (2013), “Improving the Fiscal Framework to Enhance Growth in an
Era of Fiscal Consolidation in Slovakia”, OECD Economics Department Working Papers, No. 1018,
OECD Publishing, Paris.
Kopits, G. (2014), “Coping with fiscal risk: Analysis and practice”, OECD Journal on Budgeting, Vol. 14/1.
http://dx.doi.org/10.1787/budget-14-5jxrgssdqnlt.
Kopits, G. (2011a), “Can Fiscal Discipline be Reconciled with Fiscal Sovereignty?”, in Banca d’Italia (ed.),
Fiscal Policy: Rules and Institutions for Sound Fiscal Policy After the Crisis, acts of the 13th Public
Finance Workshop held in Perugia, 31 March – 2 Avril.
Kopits, G. (2011b), “Independent Fiscal Institutions: Developing Good Practices”, OECD Journal on
Budgeting, Vol. 11/3, http://dx.doi.org/10.1787/budget-11-5kg3pdgcpn42.
Krogstrup, S. and C. Wyplosz (2010), “A Common Pool Theory of Supranational Debt Ceilings”, European
Economic Review 54, 269-78.
Kumar, M. S. and J. Woo (2010), “Public Debt and Growth”, IMF Working Paper, No. 10/174, International
Monetary Fund.
Kydland, F. and E. Prescott (1977), “Rules Rather Than Discretion: The Inconsistency of Optimal Plans”,
Journal of Political Economy, Vol. 85, pp. 473-91.
Larch, M. and A. Turrini (2008), “Received Wisdom and Beyond: Lessons from Fiscal Consolidation in the
EU”, European Economy, Economic Papers, No. 320, European Commission, Brussels.
Lojsch, D.H., M. Rodríguez-Vives and M. Slavík (2011), “The Size and Composition of Government Debt in
the Euro Area”, ECB Occasional Paper Series, No. 32.
Manasse, P. (2006), “Pro-cyclical Fiscal Policy: Shocks, Rules, and Institutions: A View from MARS”, IMF
Working Paper, No. 06/27.
Marneffe, W. (2011), Marneffe, W., B. van Aarle, W. van der Wielen and L. Vereeck (2011), “The Impact of
Fiscal Rules on Public Finances in the Euro Area”, CESifo DICE Report, Vol. 9, Issue 3, pp. 18-26.

Mendoza, E., L. Tesar and J. Zhang (2014) “Saving Europe?: The Unpleasant Arithmetic of Fiscal
Austerity in Integrated Economies,” NBER Working Papers 20200, National Bureau of Economic
Research, Inc.
Merola, R. and D. Sutherland (2012), “Fiscal Consolidation: Part 3. Long-Run Projections and Fiscal Gap
Calculations”, OECD Economics Department Working Papers, No. 934, OECD Publishing, Paris.
Molnar, M. (2012), « Fiscal Consolidation: Part 5. What Factors Determine the Success of Consolidation
Efforts?”, OECD Economics Department Working Papers, No. 936, OECD Publishing, Paris.
Nerlich, C. and W. H. Reuter (2013), “The Design of National Fiscal Frameworks and their Budgetary
Impact”, European Central Bank Working Paper Series, No. 1588.
Nickel, C. and A. Tudyka (2014), “Fiscal Stimulus in Times of High Debt: Reconsidering Multipliers and
Twin Deficits”, Journal of Money, Credit and Banking, Vol. 46(7), pp. 1313-1344.
OECD (2014), “Draft Recommendation of the OECD Council on the Principles of Budgetary Governance”,
OECD Publishing, Paris,
www.oecd.org/gov/budgeting/OECD-Recommendation-on-Principles-for-Independent-Fiscal-
Institutions.pdf.
Ollivaud, P. and D. Turner (2014), “The Effect of the Global Financial Crisis on OECD Potential Output”,
OECD Economics Department Working Papers, No. 1166, OECD Publishing, Paris.
Panizza, U. and A. F. Presbitero (2014), “Public Debt and Economic Growth: Is there a Causal Link?”,
Journal of Macroeconomics, Vol. 41, pp. 21-41.
Persson, T.and G. Tabellini, (2000), “Political Economics: Explaining Economic Policy”, Cambridge, MA:
MIT Press.
Persson, T. and L.E. Svensson (1989), “Why a Stubborn Conservative Would Run a Deficit: Policy with
Time-Inconsistent Preferences”, Quarterly Journal of Economics, Vol. 104, 325-45.
Posner, P. L. and M. Sommerfeld (2013), “The Politics of Fiscal Austerity: Democracies and Hard
Choices”, OECD Journal on Budgeting, Vol. 13/1,
http://dx.doi.org/10.1787/budget-13-5k3w6lk42l33.
Poterba, J. (1994), “State Responses to Fiscal Crises: The Effects of Budgetary Institutions and Politics”,
Journal of Political Economy, Vol. 102, No. 4.
Price, R. W. and T. Dang (2011), “Adjusting Fiscal Balances for Asset Price Cycles”, OECD Economics
Ramey, V. (2011), “Can Government Purchases Stimulate the Economy?”, Journal of Economic Literature,
Vol. 49(3), pp. 673-685.
Rawdanowicz, Ł. (2012), “Choosing the Pace of Fiscal Consolidation”, OECD Economics Department
Working Papers, No. 992, OECD Publishing, Paris.
Reinhart, C. and K. Rogoff (2011), “From Financial Crash to Debt Crisis”, American Economic Review,
Vol. 101, No. 5, pp. 1676-706.
Reinhart, C.M. and K.S. Rogoff (2010), “Growth in a Time of Debt”, American Economic Review, Vol. 100,
No. 2, pp. 573-78.

Reinhart, C. and K. Rogoff, (2009), This Time Is Different: Eight Centuries of Financial Folly, Princeton,
New Jersey: Princeton University Press.
Röhn, O. (2010), “New Evidence on the Private Saving Offset and Ricardian Equivalence”, OECD
Sorbe, S. (2012), “Portugal - Assessing the Risks around the Speed of Fiscal Consolidation in an
Uncertain Environment”, OECD Economics Department Working Papers, No. 984, OECD
Publishing, Paris.
Strasky, J. (2015), “Public Capital Stock and the Debt-to-GDP Ratios in 22 OECD Countries”, OECD
Economics Department Working Papers, forthcoming.
Sutherland, D. and P. Hoeller (2012), “Debt and Macroeconomic Stability”, OECD Economics Department
Working Papers, No. 1003, OECD Publishing, Paris.
Sutherland, D., P. Hoeller and R. Merola (2012), “Fiscal Consolidation: Part 1. How Much is Needed and
How to Reduce Debt to a Prudent Level?”, OECD Economics Department Working Papers, No. 932,
OECD Publishing, Paris.
Taylor, J.B. (2000), “Reassessing Discretionary Fiscal Policy”, Journal of Economic Perspectives,
American Economic Association, Vol. 14(3), pp. 21-36.
Tornell, A. and P. Lane (1999), “The Voracity Effect”, American Economic Review, Vol. 89, pp. 22–46.
Turner, D. (2006), “Should Measures of Fiscal Stance be Adjusted for Terms of Trade Effects?”, OECD
Turner, D. and F. Spinelli (2013), “The Effect of Government Debt, External Debt and their Interaction on
OECD Interest Rates”, OECD Economics Department Working Papers, No. 1103, OECD Publishing,
Paris.
Von Hagen, J. and R.R. Strauch (2001), “Fiscal Consolidations: Quality, Economic Conditions, and
Success”, Public Choice, Vol. 109(3-4), pp. 327-346.
Von Hagen, J. and I. Harden (1994), “National Budget Processes and Fiscal Performance”, European
Economy. Reports and Studies, Vol. 3, pp. 315–418.
Wyplosz, C. (2013), “Fiscal Rules, Theoretical Issues and Historical Experiences”, in A. Alesina and F.
Giavazzi (eds.), Fiscal Policy after the Financial Crisis, NBER, The University of Chicago Press,
Chicago.

APPENDIX
Table A.1. Fiscal reaction function
(1) (2) (3) (4) (5) (6) (7)

Primary Primary Primary Primary Primary Primary Primary
Dependent variable
balance balance balance balance balance balance balance
Gap 0.42*** 0.38*** 0.43*** 0.49*** 0.42*** 0.43*** 0.39***
(0.029) (0.043) (0.029) (0.034) (0.029) (0.029) (0.031)
Lag (debt) below d1 0.017* -0.008 0.025 0.059** 0.018* 0.011 0.018*
(0.009) (0.007) (0.016) (0.024) (0.009) (0.010) (0.010)
Lag (debt) between 0.064** -0.074** 2.60** 0.003 0.065** 0.046* 0.067**
d1 and d2 (0.027) (0.037) (1.09) (0.011) (0.027) (0.024) (0.028)
Lag (debt) above d2 -0.11*** -0.14** -0.83* -0.11*** -0.11*** -0.098** -0.11***
(0.041) (0.064) (0.43) (0.040) (0.040) (0.040) (0.041)
One-off -1.00*** -1.02*** -1.01*** -1.03*** -1.04*** -0.97*** -1.01***
(0.036) (0.046) (0.038) (0.041) (0.039) (0.038) (0.037)
Openness x 0.034*** 0.011*** 0.029*** 0.035*** 0.033*** 0.033*** 0.034***
terms of trade (0.008) (0.004) (0.008) (0.009) (0.008) (0.0083) (0.009)
Lag (assets (a)) 0.022
(0.018)
Lag (debt below d1 x a) -0.017
(0.026)
Lag (debt between -3.01**
d1 and d2 x a) (1.39)
Lag (debt above d2 x a) 0.86*
(0.52)
Lag (real interest rate (r)) -0.027
(0.090)
Lag (debt below d1 x r) 0.24
(0.25)
Lag (debt between 0.32**
d1 and d2 x r) (0.14)
Lag (debt above d2 x r) 0.91
(0.67)
Lag (One-off) -0.12***
(0.039)
IMF programme 0.011 0.017**
(0.007) (0.007)
Inflation 0.14*** 0.097***
(0.055) (0.035)
Old age dependency ratio 0.001**
(5e-04)
Lag (debt x r) 0.26***
(0.051)
Lag (Government size) -0.099***
(0.028)
Euro area -0.005
(0.004)
House price change 0.073***
(0.016)
Stock price change 0.012***
(0.003)
d1 123% 137% 156% 51% 123% 117% 123%
d2 168% 180% 164% 157% 168% 168% 168%
Sample 1985-2013 1985-2013 1985-2013 1985-2013 1985-2013 1985-2013 1985-2013
Country fixed effects yes no yes Yes yes yes yes
N 709 512 699 604 706 709 658
Rho 0.75 0.71 0.73 0.67 0.75 0.75 0.76
R2 0.633 0.653 0.635 0.661 0.637 0.636 0.637
Note: Panel regression with AR(1) Prais-Winsten correction and panel heteroskedasticity-robust standard errors. Standard errors are in
parenthesis. Norway is excluded from the third specification because it is not comparable to the other OECD countries: its high net
financial worth mainly reflects past oil-related government receipts.

Govt Debt and Fiscal Frameworks

Uploaded by

Copyright:

Available Formats

Govt Debt and Fiscal Frameworks

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Govt Debt and Fiscal Frameworks

Uploaded by

Copyright:

Available Formats

PRUDENT DEBT

TARGETS AND FISCAL

This paper has been prepared by:

Authorised for publication by Catherine Mann

More detailed information is contained in the following papers:

Series: OECD Economic Policy Papers

Executive summary ...................................................................................................................... 6

3. The effects of debt on the economy .................................................................................. 13

4. Defining debt targets ......................................................................................................... 20

5. Designing fiscal frameworks ............................................................................................. 30

6. Political economy issues ................................................................................................... 48

OECD ECONOMIC POLICY PAPERS, NO. 15 © OECD 2015 3

Box 1. Debt data: Understanding the terminology ....................................................................................... 9

4 OECD ECONOMIC POLICY PAPERS, NO. 15 © OECD 2015

Prudent debt targets and fiscal frameworks

JEL classification codes: E27; E61; E62; H62; H68

Keywords: Macroeconomic uncertainties, debt, fiscal policy, fiscal rules

Cibles de dette prudentes et cadres budgétaires

Classification JEL : E27 ; E61 ; E62 ; H62 ; H68

Mots-clés : incertitudes macroéconomiques, dette, politique budgétaire, règles budgétaires

OECD ECONOMIC POLICY PAPERS, NO. 15 © OECD 2015 5

Prudent Debt Targets and Fiscal Frameworks

• For higher-income countries, a debt threshold range of 70 to 90% of GDP.

6 OECD ECONOMIC POLICY PAPERS, NO. 15 © OECD 2015

OECD ECONOMIC POLICY PAPERS, NO. 15 © OECD 2015 7

8 OECD ECONOMIC POLICY PAPERS, NO. 15 © OECD 2015

Box 1. Debt data: Understanding the terminology

Non-debt instruments Non-de

The following definitions can be distinguished:

OECD ECONOMIC POLICY PAPERS, NO. 15 © OECD 2015 9

Box 1. Debt data: Understanding the terminology (cont.)

Figure 2. Calculating general government net worth

General General General

The terminology can therefore be expanded to include:

10 OECD ECONOMIC POLICY PAPERS, NO. 15 © OECD 2015

Box 1. Debt data: Understanding the terminology (cont.)

General government gross

General government Explicit contingent Implicit liabilities

1. Includes explicit unfunded or underfunded civil servant pension liabilities.

The key terms related to these future liabilities follow:

OECD ECONOMIC POLICY PAPERS, NO. 15 © OECD 2015 11

Figure 4. Gross and net debt concepts

Per cent of GDP

12 OECD ECONOMIC POLICY PAPERS, NO. 15 © OECD 2015

Table 1. Long-term interest rate determinants

(1) (2) (3) (4) (5) (6)

3. The effects of debt on the economy

OECD ECONOMIC POLICY PAPERS, NO. 15 © OECD 2015 13

3.1. Limits to sustainability

Box 2. Public debt limits

14 OECD ECONOMIC POLICY PAPERS, NO. 15 © OECD 2015

Box 2. Public debt limits (cont)

Figure 5. Determination of the debt limit

OECD ECONOMIC POLICY PAPERS, NO. 15 © OECD 2015 15

Box 2. Public debt limits (cont)

Figure 6. Debt limit drivers

16 OECD ECONOMIC POLICY PAPERS, NO. 15 © OECD 2015

Figure 7. Debt limits

OECD ECONOMIC POLICY PAPERS, NO. 15 © OECD 2015 17

3.2. The debt level that maximises growth