2020 ECONOMIC CONSEQUENCES OF HIGH PUBLIC DEBT: EVIDENCE FROM THREE LARGE SCALE DSGE MODELS - Banco de España

ECONOMIC CONSEQUENCES OF HIGH
PUBLIC DEBT: EVIDENCE FROM THREE
 2020
LARGE SCALE DSGE MODELS

Documentos de Trabajo
N.º 2029

Pablo Burriel, Cristina Checherita-Westphal,
Pascal Jacquinot, Matthias Schön
and Nikolai Stähler

ECONOMIC CONSEQUENCES OF HIGH PUBLIC DEBT: EVIDENCE
FROM THREE LARGE SCALE DSGE MODELS

ECONOMIC CONSEQUENCES OF HIGH PUBLIC DEBT:
EVIDENCE FROM THREE LARGE SCALE DSGE MODELS (*)

Pablo Burriel
BANCO DE ESPAÑA

Cristina Checherita-Westphal and Pascal Jacquinot
ECB

Matthias Schön and Nikolai Stähler
BUNDESBANK

(*) This paper was part of an ESCB background paper to the 2017 Public Finance Report. In that context, we
would like to thank our co-authors on that paper: Maria Campos (Banco de Portugal), Francesco Caprioli and
Pietro Rizza (Banca d’Italia) and Alexander Mahle (Bundesbank), as well as Frank Smets, Claudia Braz, Christophe
Kamps, Thomas Warmedinger, Philipp Rother, Samuel Hurtado and the members of the ESCB Working Group on
Public Finance for their useful comments and suggestions. We are also grateful to an anonymous referee for very
useful comments and suggestions on the ECB working paper. Any remaining errors are ours. The views expressed
in this paper are those of the authors and do not necessarily reflect those of the European Central Bank or the
ESCB Working Group on Public Finance and its members.

This paper has also been published as an ECB Working Paper with number 2450, July 2020.

Documentos de Trabajo. N.º 2029
2029

The Working Paper Series seeks to disseminate original research in economics and finance. All papers
have been anonymously refereed. By publishing these papers, the Banco de España aims to contribute
to economic analysis and, in particular, to knowledge of the Spanish economy and its international
environment.

The opinions and analyses in the Working Paper Series are the responsibility of the authors and, therefore,
do not necessarily coincide with those of the Banco de España or the Eurosystem.

The Banco de España disseminates its main reports and most of its publications via the Internet at the
following website: http://www.bde.es.

Reproduction for educational and non-commercial purposes is permitted provided that the source is
acknowledged.

© BANCO DE ESPAÑA, Madrid, 2020

ISSN: 1579-8666 (on line)

Abstract

The paper reviews the economic risks associated with regimes of high public debt through
DSGE model simulations. The large public debt build-up following the 2009 global
financial and economic crisis acted as a shock absorber for output, while in the recent
and more severe COVID19-crisis, an increase in public debt is even more justified given
the nature of the crisis. Yet, once the crisis is over and the recovery firmly sets in, keeping
debt at high levels over the medium term is a source of vulnerability in itself. Moreover,
in the euro area, where monetary policy focuses on the area-wide aggregate, countries
with high levels of indebtedness are poorly equipped to withstand future asymmetric
shocks. Using three large scale DSGE models, the simulation results suggest that high-
debt economies (1) can lose more output in a crisis, (2) may spend more time at the zero-
lower bound, (3) are more heavily affected by spillover effects, (4) face a crowding out of
private debt in the short and long run, (5) have less scope for counter-cyclical fiscal policy
and (6) are adversely affected in terms of potential (long-term) output, with a significant
impairment in case of large sovereign risk premia reaction and use of most distortionary
type of taxation to finance the additional debt burden in the future. Going forward, reforms
at national level, together with currently planned reforms at the EU level, need to be
timely implemented to ensure both risk reduction and risk sharing and to enable high debt
economies address their vulnerabilities.

Keywords: government debt, interest rates, economic growth, fiscal sustainability.

JEL classification: E62, H63, O40, E43.

Resumen

En este trabajo se analizan los efectos sobre la actividad económica derivados de
mantener un nivel elevado de deuda pública por medio de simulaciones realizadas con
modelos de equilibrio general dinámico estocástico (DSGE, por sus siglas en inglés).
La fuerte acumulación de deuda pública tras la crisis económica y financiera global
de 2009 contribuyó a moderar el impacto de la perturbación sobre la actividad. Esta
respuesta de política económica está más justificada, si cabe, ante la pandemia de
Covid-19, por el marcado carácter transitorio de la perturbación. No obstante, en el
medio plazo, una vez se haya superado el impacto inmediato de la pandemia y la
recuperación económica esté firmemente asentada, mantener niveles tan elevados
de deuda pública puede suponer una fuente de vulnerabilidad para la economía.
Esta cuestión es aun más relevante en el caso del área del euro, donde la política
monetaria responde al comportamiento de variables agregadas del conjunto del
área, de forma que los países con elevados niveles de deuda no cuentan con este
instrumento de política para responder a perturbaciones asimétricas. Los resultados
de las simulaciones obtenidas a partir de tres modelos DSGE, estimados o calibrados
con datos del área del euro, sugieren que las economías con elevada deuda pública:
1) experimentan caídas de la actividad más profundas durante una crisis; 2) pueden pasar
más tiempo en la cota inferior de los tipos de interés nominales; 3) muestran efectos
spillover negativos entre economías más intensos; 4) producen una mayor expulsión
de deuda privada en el corto y el largo plazo; 5) presentan una capacidad para realizar
una política contracíclica menor, y 6) sufren mayores pérdidas de producto potencial,
especialmente cuando la prima de riesgo reacciona significativamente y cuando se
financia el coste futuro de la mayor carga de deuda aumentando los impuestos más
distorsionadores. De cara al futuro, para facilitar que las economías con elevado nivel
de deuda puedan reducir sus vulnerabilidades, resulta necesario implementar reformas
estructurales a escala nacional, y que las reformas de la Unión Europea, actualmente en
discusión, permitan moderar los riesgos actuales y aumenten el grado de compartición
de riesgos entre los miembros de la Unión.

Palabras clave: deuda pública, tipos de interés, crecimiento económico, sostenibilidad
fiscal.

Códigos JEL: E62, H63, O40, E43.

Non-technical summary

 Debt – public and private – plays an important role in the normal functioning of a market economy. In
 the private sector, credit is essential to facilitate investment and growth over time. In both the public and
 private sector, debt can have beneficial effects in terms of smoothing consumption and financing lumpy
 productive investment. The 2009 global financial and economic crisis left a legacy of historically high
 levels of public debt in advanced economies, at a scale unseen during modern peace time. Given the
 extreme severity of the crisis, coined by many as the “great recession”, this debt build-up acted as a
 shock absorber for output. The coronavirus (COVID-19) pandemic is a different type of shock that has
 dramatically affected global economic activity, including the euro area, since early 2020. Fiscal positions
 are projected to be strongly hit by the crisis through both automatic stabilisers and discretionary fiscal
 measures. This substantial support from fiscal policy, together with that of monetary policy, is necessary
 and should help limit the economic scars of the crisis. Yet, once the crisis is over and the recovery firmly
 sets in, keeping public debt at high levels over the medium term is a source of vulnerability in itself. A
 high public debt burden is even more problematic in the euro area, as fiscal policies remain at national
 level while monetary policy focuses on the aggregate.

 The main objective of this paper is to contribute to the stabilisation vs. sustainability debate in the euro
 area by reviewing through the lens of large scale DSGE models the economic risks associated with
 regimes of high public debt. The paper argues that a good balance between the two fiscal policy
 objectives is difficult to achieve when public debt is high. To this end, we first review the theoretical and
 empirical literature on the role and macroeconomic consequences of public debt. Thereafter, we
 evaluate the economic consequences of high public debt using simulations with three DSGE models
 that share the same new-Keynesian framework and feature an enhanced government sector: EAGLE
 – E(S)CB, GEAR – Bundesbank, and BE – Banco de España.

 To our knowledge, this paper is one of the first to systematically assess the macroeconomic implications
 of high public debt in a global DSGE setup designed for the euro area. By using three different models
 widely employed for policy simulations in our institutions, we aim at providing a robust and
 encompassing analysis for the euro area. First, we explicitly account for the heterogeneity within the
 euro area. The models are calibrated or estimated for different regions/countries, with the euro area
 being split into either core vs. periphery (for EAGLE and the BE model) or Germany vs. the rest of the
 EA (for GEAR). Second, and, more crucially, the three models enrich our analysis by bringing alternative
 perspectives. Hence, the EAGLE model features a more detailed euro area and external block, GEAR
 includes a sound labour market, while the BE model has a financial block with borrowing constrains and
 long-term debt. Third, we are able to design a broad range of simulation scenarios, which shed light
 into various risks associated with regimes of high debt. Hence, we can assess the macroeconomic
 consequences of an excessive level of debt when the economy is hit by an adverse shock, in normal
 times or at the zero lower bound (ZLB). In addition, using the BE model, we can examine the specific
 role played by private deleveraging. Finally, a set of simulations evaluate the long-term costs of a high
 debt burden.

 As indicated in the literature, first, high public debt poses significant economic challenges as it makes
 the economy less resilient to shocks; second, debt overhangs can exert adverse pressure on the
 economy through multiple channels over the long-run.

 Our DSGE simulations also suggest that high-debt economies (1) can lose more output in a crisis, (2)
 may spend more time at the zero-lower bound, (3) are more heavily affected by spillover effects, (4)
 face a crowding out of private debt in the short and long run, (5) have less scope for counter-cyclical
 fiscal policy and (6) are adversely affected in terms of potential (long-term) output, with a significant
 impairment in case of large sovereign risk premia reaction3 and use of most distortionary type of taxation
 to finance the additional public debt burden in the future. The strength of these results depends on the
 impact
BANCO DE ESPAÑA of the level
 7 DOCUMENTO ofN.ºpublic
 DE TRABAJO 2029 debt in these models. Given that the sovereign bond yield spread is the
 main transmission channel, in the short-run simulations results also depend crucially on the monetary
 policy implementation.

 Going forward, planned reforms at national and EU level to ensure both risk reduction and risk sharing

fiscal policy and (6) are adversely affected in terms of potential (long-term) output, with a significant
 impairment in case of large sovereign risk premia reaction and use of most distortionary type of taxation
 I. Introduction
 to finance the additional public debt burden in the future. The strength of these results depends on the
 impact of the level of public debt in these models. Given that the sovereign bond yield spread is the
 The 2009 global financial and economic crisis left a legacy of historically high levels of public debt in
 main transmission channel, in the short-run simulations results also depend crucially on the monetary
 advanced economies, at a scale unseen during modern peace time. Given the extreme severity of the
 policy implementation.
 crisis, coined by many as the “great recession”, this debt build-up acted as a shock absorber for output,
 through
 Going forward,the work of automatic
 planned reformsstabilisers,
 at nationalcosts and EUincurred
 level to in ensure
 the stabilisation
 both risk of the financial
 reduction sector,
 and risk and
 sharing
 fiscal to
 need stimulus
 be timelymeasures grantedtoatenable
 implemented the beginning
 high debt of the crisis. While
 economies reduce debttheir
 ratios had generallyOnce
 vulnerabilities. declined the
 in the euro area countries since that crisis, they still remained
 COVID-19 crisis is over and the economic recovery firmly re-established, further efforts to build at high and very high levels in some
 fiscal
 countries.
 buffers The coronavirus
 in good times and mitigate (COVID-19) fiscal pandemic
 risks over stroke
 the mediumthe global
 termeconomic
 are needed activity,
 at theincluding
 national in the
 level.
 euro
 At area,
 the–EU in early
 level, and
 the EU2020, as a
 recovery more severe
 fund currently and different type
 under negotiation of shock. Mainly
 is one economy. due
 important tool to the strict
 thatprivate lockdown
 may not only
 Debt private public – is integral to the functioning of a market In the sector,
 measures
 bolster implemented in most euro area countries around mid-March, euro area real GDP registered
 credit isthe foundation
 essential for sustainable
 to facilitate productive growth in the aftermath
 investment and growth of the
 over COVID-crisis,
 time. In both but the also
 public support high-
 and private
 a record
 debt decline
 countries to of 3.8% in
 address thevulnerabilities.
 their first quarter of 2020. According to Eurosystem staff’s macroeconomic
 sector, debt can have beneficial effects in terms of smoothing consumption and financing lumpy
 projections, a further decline in GDP of 13% is expected for the second quarter and what will happen
 investment. In most advanced economies, as well as in most macroeconomic models, 2 public debt has
 after that is subject to unprecedented uncertainty. 1 Fiscal positions are projected to be strongly hit by
 been perceived, at least before the 2009 crisis, to be safe (Coeuré 2016). When it carries low credit
 the crisis, through both automatic stabilisers and discretionary fiscal measures. This substantial support
 risk, by providing a relatively safe and liquid asset, also for refinancing operations, public debt plays a
 I.
 fromIntroduction
 vital
 fiscal policy, together with that of monetary policy, is necessary and should help limit the economic
 role for the functioning of the financial system and the transmission of monetary policy. Other
 scars of the crisis.
 contributions conclude that public debt can have positive effects on welfare as long as it provides a safe
 The
 asset 2009 global financial
 for insurance against and
 both economic
 idiosyncratic crisis leftaggregate
 a legacy ofrisks. historically high levels of public debt in
 aggregatedand
 3
 The general government debt ratio across euro area (EA) member countries rose from 66%
 advanced economies, at a scale unseen during modern peace time. Given the extreme severity of the
 of GDP in 2007 to 95% in 2014, slightly declining afterwards (to 86% by 2019), while total (public and
 crisis,
 Yet, when coineddebtbyismany as theit “great
 too high, recession”,
 may become riskythis
 for debt build-up acted
 the economy. as a shock
 Essentially, one absorber
 needs to for output,
 recognise
 private) debt peaked close to 240% of GDP in 2012 and declined to 223% by end-2018(Chart 1a).
 through the work debt
 that government of automatic
 even in stabilisers,
 advanced costs incurred
 economies, andin the stabilisation
 especially of the
 in those financialtosector,
 belonging monetary and
 Before the COVID-19 crisis hit, two euro area countries had public debt ratios above 120% of GDP at
 fiscal
 unions, stimulus measures
 is not risk free. A granted
 high public at thedebtbeginning
 burden is ofproblematic
 the crisis. While debt ratios
 especially had generally
 in a monetary uniondeclined
 like the
 end-2019 (Greece and Italy) and another four countries (Portugal, Belgium, France, Cyprus and Spain)
 in
 eurothearea,
 euro in area
 whichcountries
 fiscal since
 policies that crisis,atthey
 remain still remained
 national level, whileat high
 memberand very
 states highsharelevels in some
 a common
 above 90% (Chart 1b). According to the European Commission’s spring 2020 forecast, the euro area
 countries.
 currency and Thelackcoronavirus
 monetary(COVID-19)
 policy autonomy. pandemicIn thisstroke the global
 institutional set-up economic
 nationalactivity, including
 fiscal policies carryin the
 debt ratio would increase to 103% of GDP in 2020 and decline to 99% in the next year, while the above
 euro
 burden area, in early
 to adjust to2020, as a more
 asymmetric severe
 shocks. and different
 However, euro area typecountries
 of shock.with Mainly
 highdue to the
 levels of strict
 publiclockdown
 debt are
 mentioned countries will have higher debt ratios, with the largest increases by 2021 projected for Italy
 measures
 poorly equipped implemented
 to carryinout most thiseuro area countries
 stabilisation around
 task. Risks tomid-March, euro area
 debt sustainability in areal GDP registered
 member state can
 and
 a Spain
 record (countries
 decline mostinhard-hit by the pandemic).
 entail risks to theof 3.8%
 stabilisation theoffirst quarter
 the euro areaof as
 2020. According to Eurosystem staff’s macroeconomic
 a whole.
 projections, a further decline in GDP of 13% is expected for the second quarter and what will happen
 The main objective Chartis1:toPublic and private debt in the euro area
 after that is subject of to this paper
 unprecedented contribute
 uncertainty. to 1the stabilisation
 Fiscal positionsvs. aresustainability
 projected todebate be stronglyin thehit euro
 by
 area a)
 4 by reviewing through
 total, public & the lens
 private debt of
 (%large
 of scale
 GDP) DSGE
 b) The models
 most the
 the crisis, through both automatic stabilisers and discretionary fiscal measures. This substantial support
 EA aggregate: publicly macroeconomic
 indebted countries in the implications
 EA (% of GDP) of
 high public debt. The paper argues that a good balance between the
 from fiscal policy, together with that of monetary policy, is necessary and should help limit the economic two fiscal policy objectives is
 270
 difficultoftothe
 scars achieve
 crisis. when public debt is high.
 210

 240
 180
 To210our
 The knowledge,
 general governmentthis paper debtisratio oneaggregated
 of the first toacross systematically
 euro areaassess(EA) member the macroeconomic
 countries rose implications
 from 66%
 of high
 of180 public debt in a global DSGE setup designed for
 150 the
 GDP in 2007 to 95% in 2014, slightly declining afterwards (to 86% by 2019), while total (public and euro area. 5

 private)
 150 debt peaked close to 240% of GDP in 2012 120 and declined to 223% by end-2018(Chart 1a).
 The analysis is based on three different global DSGE models (EAGLE – E(S)CB, GEAR – Bundesbank,
 Before
 120 the COVID-19 crisis hit, two euro area countries had public debt ratios above 120% of GDP at
 and BE - Banco de España) that share the same new-Keynesian 90 framework and feature an enhanced
 end-2019
 90 (Greece and Italy) and another four countries (Portugal, Belgium, France, Cyprus and Spain)
 government sector. By using three different models widely employed for policy simulations in our
 above
 60 90% (Chart 1b). According to the European Commission’s 60 spring 2020 forecast, the euro area
 institutions, we aim at providing a robust and encompassing analysis for the euro area. First, we
 debt ratio would increase to 103% of GDP in 2020 and decline to 99% in the next year, while the above
 explicitly
 30 account for the heterogeneity within the euro30 area. The models are calibrated or estimated for
 mentioned countries will have higher debt ratios, with the largest increases by 2021 projected for Italy
 different
 0 regions/countries, with the euro area being split 0 into either core vs. periphery (for EAGLE and
 and Spain (countries most hard-hit by the pandemic). 1999 2001 2003 2005 2007 2009 2011 2013 2015 2017 2019 2021
 1999 2001 2003 2005 2007 2009 2011 2013 2015 2017
 the BE model) or Germany
 EA Public Debt
 vs. the rest of
 EA Private Debt
 the EA (for GEAR).GR Second,
 EA Total Debt IT
 and,
 PT
 more
 ES
 crucially,
 FR BE
 the three
 CY
 models enrich
 Debt – private and public –our analysis by bringing
 is integral alternative
 to the perspectives. The EAGLE model features a more
 Chart 1: Public andfunctioning
 private debtofinathe market
 euro economy.
 area In the private sector,
 detailed
 credit
 Source: is
 Eurostat, euro area
 essential
 European to(including
 Commissionfacilitate tradable
 productive
 (Ameco, Spring and
 2020 forecast) andnon-tradable
 investment and
 own calculations. sectors)over
 growth andtime.symmetric
 In both external
 the public blockand (and
 private as
 Notes: Public
 a) EAdebtaggregate:
 stands for general government
 total, public &gross debt (EDP
 private debt concept),
 (% of as defined in the
 GDP) b) Ameco
 The database.
 most Privateindebted
 publicly debt standscountries
 for the stockinofthe
 liabilities
 EA held
 (% by the
 of GDP)sectors
 such corporations
 sector,
 Non-financial is very
 debt well and suited
 canHouseholds
 have and to assess
 beneficial international
 effects
 Non-Profit institutions serving spillovers),
 inhouseholds,
 terms GEAR
 data, as includes
 of smoothing
 consolidated perconsumption a sound
 the Macroeconomic labour
 and
 Imbalance market,
 financing
 procedure while for
 (MIP)lumpy
 scoreboard
 euro area Member States (latest data available for all countries as of end-2018). The EA private debt is calculated as the GDP-weighted average across the Member States. The
 the BE
 euro investment.
 area model has
 In most
 270total debt is calculated
 a financial
 as theadvanced
 sum between theblock with
 economies, borrowing
 EA private debt and aspublic
 well as210in most macroeconomic models, 2 public debtlong-
 constrains
 debt. à la Kiyotaki and Moore (1997) and has
 term
 been debt. Third, we are able to design a broad range of simulation scenarios, which shed light into
 240 perceived, at least before the 2009 crisis, to be safe (Coeuré 2016). When it carries low credit
 various risks associated with regimes of high debt.
 180 Hence,
 risk, by providing a relatively safe and liquid asset, also for refinancing operations, public debt plays a we can assess the macroeconomic
 210
 vital
 1
 See role
 180
 for the
 June 2020 functioning
 Eurosystem of the financial
 staff macroeconomic system
 projections and
 for150the the
 euro transmission of monetary policy. Other
 area.
 contributions
 2
 Assuming a fullconclude that
 consolidation public
 between debt
 the can
 balance have
 sheet
 4
 ofpositive
 the central effects
 bank andonthat
 welfare as long
 of the fiscal as it provides
 authority, governmentadebt
 safeis
 150 120
 considered risk-free in nominal terms. For the euro area, in particular, institutional design like the prohibition of monetary
 asset for insurance against both idiosyncratic and aggregate 5 risks. 3
 financing and the no bail-out clause (articles 123 and 125 of the EU Treaty) make such models unrealistic.
 120
 90
 3
 However, the literature on the relationship between public debt and welfare remains ambiguous, with several studies indicating
 Yet,90
 a when debt
 negative is too
 impact high,
 of debt onitwelfare
 may become
 or optimalrisky for the
 debt levels economy.
 that are either Essentially, one needs
 very small or negative to recognise
 (government should
 60
 8 DOCUMENTO
BANCO DE ESPAÑA60accumulate
 that government net
 DE assets).
 TRABAJO N.º This
 debt even 2029 is especially the case when welfare (asset holding) is concentrated on a minority of households
 in advanced economies, and especially in those belonging to monetary
 or when the government seeks to smooth tax distortion over time and thus needs asset buffers to offset adverse shocks. For
 unions,
 30 is not
 a review, seerisk
 Dieppefree.
 andAGuardahigh public
 (2015). debt burden is problematic especially in a monetary union like the
 30

 euro
 4
 See area,
 0 inter alia,inECB
 which fiscal
 (2016b), policies
 EC (2016), remainand
 Bańkowski at Ferdinandusse
 national0 level, while member states share a common
 (2017), IMF (2017). For more details on sustainability,
 1999Boubabdallah
 see 2001 2003 2005 2007 2009 2011 2013 2015 2017
 et al. (2017).
 currency
 5
 and lack monetary policy autonomy. In this institutional set-up national fiscal policies carry the
 1999 2001 2003 2005 2007 2009 2011 2013 2015 2017 2019 2021

debt ratio would increase to 103% of GDP in 2020 and decline to 99% in the next year, while the above
 mentioned countries will have higher debt ratios, with the largest increases by 2021 projected for Italy
 and Spain (countries most hard-hit by the pandemic).

 Chart 1: Public and private debt in the euro area
 a) EA aggregate: total, public & private debt (% of GDP) b) The most publicly indebted countries in the EA (% of GDP)
 Debt – private and public – is integral to the functioning of a market economy. In the private sector,
 credit
 270 is essential to facilitate productive investment 210 and growth over time. In both the public and private
 sector,
 Debt
 240 – debt
 privatecan
 and have beneficial
 public – is effects
 integral to in
 the terms of smoothing
 functioning of a market consumption
 economy. and
 In thefinancing lumpy
 private sector,
 180
 investment.
 credit
 210
 In most
 is essential advancedproductive
 to facilitate economies, as well asand
 investment in most
 growth macroeconomic models,
 over time. In both 2 public
 the public anddebt has
 private
 been
 sector,perceived,
 debt can athave
 leastbeneficial
 before theeffects
 2009 crisis,
 in to 150
 terms be
 of safe (Coeuré
 smoothing 2016). When
 consumption it carries
 and low lumpy
 financing credit
 180
 risk, by providing
 investment. a relatively
 In most advancedsafe and liquid
 economies, asasset,
 well asalso for refinancing
 in most operations,
 macroeconomic public
 models, debtdebt
 2 public plays a
 has
 150 120
 vital
 beenrole for the at
 perceived, functioning
 least before of the
 the financial
 2009 crisis,system
 to beand safethe transmission
 (Coeuré 2016). of monetary
 When policy.
 it carries low Other
 credit
 120
 contributions
 risk, conclude
 by providing that public
 a relatively safe debt
 and can
 liquidhave positive
 asset, also
 90 effects
 for on welfare
 refinancing as long as
 operations, it provides
 public a safe
 debt plays a
 90
 assetrole
 vital for insurance against both
 for the functioning idiosyncratic
 of the and aggregate
 financial system 60 and therisks. 3
 transmission of monetary policy. Other
 60
 contributions conclude that public debt can have positive effects on welfare as long as it provides a safe
 Yet,
 30 when debt is too high, it may become risky for 30 the economy. Essentially, one needs to recognise
 asset for insurance against both idiosyncratic and aggregate risks.3
 that
 0 government debt even in advanced economies, and especially in those belonging to monetary
 0
 1999 2001 2003 2005 2007 2009 2011 2013 2015 2017
 unions,
 Yet, when is not
 debt risk
 is free. A high
 too high, public
 it may debt burden
 become risky for is problematic
 the economy.
 1999 2001 2003 especially
 2005 2007 in
 Essentially, 2009a monetary
 one needsunion
 2011 2013 2015
 to2017 like
 2019 the
 recognise 2021
 EA Public Debt EA Private Debt EA Total Debt GR IT PT ES FR BE CY
 euro government
 that area, in which debtfiscal
 evenpoliciesin advanced remaineconomies,
 at nationaland level, while member
 especially in thosestates belonging sharetoa monetary common
 Debt – is
 currency
 unions, private
 and
 not lackand
 risk public
 monetary
 free. A –policy
 high ispublic
 integral
 autonomy.
 debt toburden
 theInfunctioning
 this
 is problematic of a market
 institutional set-up
 especially economy.
 national
 in a In the
 fiscal
 monetary private
 policies union sector,
 carry
 like the
 Source: Eurostat, European Commission (Ameco, Spring 2020 forecast) and own calculations.
 credit
 Notes:burden is
 Public debt essential
 to adjust
 stands
 eurocorporations
 area, for general
 inandwhich to facilitate
 to asymmetric
 government
 fiscal gross productive
 shocks.
 debt
 policies (EDP investment
 However,
 concept),
 remain as defined euro
 in
 at national and
 the Ameco growth
 area database.
 level, over
 countries
 Private time.
 debt In
 stands both
 with highstates
 for the the
 levels
 stock ofpublic
 of
 liabilities and
 publiccommon
 held by private
 debt
 the are
 sectors
 Non-financial Households and Non-Profit institutions serving households, consolidated data,while
 as per themember
 Macroeconomic Imbalanceshare
 procedurea(MIP) scoreboard for
 sector,
 poorly
 euro area debt
 Memberequipped
 currency and
 States can
 lack
 (latest datahave
 carrybeneficial
 tomonetary
 available out
 for
 policy
 all countries effects
 this stabilisation
 autonomy.
 as inThe
 of end-2018). terms
 task.
 In this ofdebtsmoothing
 Risks isto
 institutional
 EA private debtset-up
 calculated as consumption
 sustainability
 the
 national in and
 a across
 fiscal
 GDP-weighted average financing
 member
 policies
 the Memberstate
 carrylumpy
 can
 States.
 the
 The
 euro area total debt is calculated as the sum between the EA private debt and public debt.
 investment. toInthe most advancedof economies, as well aas inarea
 most macroeconomic with highmodels,levels of public public debt debt has
 2
 entail
 burden risks
 to adjust tostabilisation
 asymmetric the euro
 shocks. area
 However, as euro whole. countries are
 poorly equipped to carry out this stabilisation task. Risks to debt sustainability in a member statecredit
 been perceived, at least before the 2009 crisis, to be safe (Coeuré 2016). When it carries low can
 The
 risk, main objectivea of
 by providing this paper
 relatively safeis andto contribute
 liquid asset, to the also stabilisation
 for refinancing vs. sustainability
 operations, public debatedebt in theplayseuro a
 1 entail risks to the stabilisation of the euro area as a whole.
 See June
 4 by 2020
 area
 vital for Eurosystem
 role reviewing the functioning staff macroeconomic
 through theof lens
 the financial projections
 of large scale
 system forDSGE
 the euromodels
 and the area. the macroeconomic
 transmission of monetaryimplications policy. Other of
 high
 The publicobjective
 contributions
 main debt.
 concludeThe paper
 thatpaper
 of this argues
 public to that
 isdebt canahave
 contribute goodto balance
 positive between
 effects
 the stabilisation on vs. thesustainability
 welfare twoasfiscal
 long as policy objectives
 it provides
 debate in thea eurosafe is
 difficult
 asset4 by
 area fortoreviewing
 achieve when
 insurance against
 through public thedebt
 both isofhigh.
 idiosyncratic
 lens large scaleand 5aggregate
 DSGE models risks.3the macroeconomic implications of
 high public debt. The paper argues that a good balance between the two fiscal policy objectives is
 To
 Yet,our knowledge,
 when this high,
 debt is too paperitismay
 one of the first to systematically assess the macroeconomic implications
 difficult to achieve when public debtbecome
 is high. risky for the economy. Essentially, one needs to recognise
 of high
 that public debtdebt
 government in a even
 globalinDSGE setupeconomies,
 advanced designed forand
 theespecially
 euro area.in
 5
 those belonging to monetary
 unions,
 To is not risk free.
 our knowledge, A highispublic
 this paper one ofdebt burden
 the first is problematicassess
 to systematically especially in a monetary union
 the macroeconomic like the
 implications
 The analysis is based on three different global DSGE models (EAGLE –5 E(S)CB, GEAR – Bundesbank,
 euro
 of higharea,
 publicin debt
 which in a fiscal policies
 global DSGEremain at national
 setup designed for level, while
 the euro member states share a common
 area.
 and BE - Banco de España) that share the same new-Keynesian framework and feature an enhanced
 currency and lack monetary policy autonomy. In this institutional set-up national fiscal policies carry the
 government
 The analysis sector.
 is based Byon using three different
 three different modelsmodels
 widely(EAGLE
 employed for policy simulations in our
 burden to adjust to asymmetric shocks. global
 However, DSGE euro area countries – E(S)CB,
 with GEAR
 high levels of–public
 Bundesbank,
 debt are
 institutions,
 and BE - we aim
 Banco de at providing
 España) that a robust
 share the and encompassing
 same new-Keynesian analysis
 framework for and
 the euro
 featurearea.
 an First, we
 enhanced
 poorly equipped to carry out this stabilisation task. Risks to debt sustainability in a member state can
 explicitly account
 government for Bythe using
 heterogeneity within the euro area. The models arefor calibrated or estimated for
 entail risks tosector.
 the stabilisation ofthree different
 the euro area models widely
 as a whole. employed policy simulations in our
 different regions/countries,
 institutions, we aim at providing with thea euro robustareaandbeing split into either
 encompassing core vs.
 analysis forperiphery
 the euro (for EAGLE
 area. First, andwe
 the
 TheBE main
 explicitlymodel)
 account or for
 objective Germany
 ofthethis vs. the
 paper
 heterogeneity rest
 is to of thethe
 contribute
 within EA (for
 toeuro GEAR).
 the area. TheSecond,
 stabilisation and,calibrated
 more crucially,
 vs. sustainability
 models are debate inthe
 thethree
 or estimated euro
 for
 models
 area4 byenrich
 different our analysis
 reviewing throughwith
 regions/countries, by
 thebringing
 lenseuro
 the alternative
 of large beingperspectives.
 area scale DSGE eitherThe
 models
 split into the EAGLE
 core model features
 macroeconomic
 vs. periphery a more
 implications
 (for EAGLE andof
 detailed
 highBE
 the eurodebt.
 public
 model) area (including
 or The
 Germany papervs. tradable
 argues
 the rest and
 thatofanon-tradable
 good
 the EA balance sectors)
 (for GEAR). betweenand the
 symmetric
 Second, external
 two fiscal
 and, more policy block (and
 objectives
 crucially, the as
 threeis
 such
 models is very
 difficult to wellour
 achieve
 enrich suited
 when topublic
 analysis assess
 by debt international
 is high.
 bringing spillovers),
 alternative GEAR includes
 perspectives. The EAGLE a sound labour
 model market,
 features while
 a more
 the BE model
 detailed has a(including
 euro area financial tradable
 block withand borrowing constrains
 non-tradable à laand
 sectors) Kiyotaki and Moore
 symmetric (1997)
 external block and(andlong-
 as
 To ouris
 term
 such knowledge,
 debt.
 veryThird, wethisare
 well suited paper
 to able isto
 assess one of theafirst
 design
 international to
 broad systematically
 range ofGEAR
 spillovers), assess
 simulation theamacroeconomic
 scenarios,
 includes soundwhich
 labourshed implications
 market,lightwhile
 into
 of high
 various
 the public
 risks debt
 BE model has ainfinancial
 associated a global
 with DSGE
 blockregimes setup
 with of designed
 high debt.
 borrowing for Hence,
 the euro
 constrains area.
 à lawe
 5
 can assess
 Kiyotaki and Moorethe (1997)
 macroeconomic
 and long-
 term debt. Third, we are able to design a broad range of simulation scenarios, which shed light into
 The analysis is based on three different global DSGE models (EAGLE – E(S)CB, GEAR – Bundesbank,
 various
 Assuming risks associatedbetween with regimes of high debt. Hence,
 bank andwe can
 of theassess the macroeconomic
 BE - aBancofull consolidation
 de España) thatthe balance
 thesheet of the central that fiscal authority, government debt is
 2
 and share same new-Keynesian framework and feature an enhanced
 considered risk-free in nominal terms. For the euro area, in particular, institutional design like the prohibition of monetary
 government
 financing andsector. By using
 the no bail-out clausethree different
 (articles 123 and 125models widely
 of the EU Treaty)employed for policy
 make such models simulations in our
 unrealistic.
 institutions, we aimonattheproviding relationshipabetween
 robustpublic
 anddebtencompassing analysis for thewith euro
 3 Assuming a full consolidation between the balance sheet of the central bank and that of the fiscal authority, government debt is
 2
 However, the literature and welfare remains ambiguous, severalarea.
 studiesFirst, we
 indicating
 considered
 a risk-freeofindebt
 negative impact nominal terms. For
 on welfare the euro
 or optimal area,
 debt in particular,
 levels institutional
 that are either designorlike
 very small the prohibition
 negative of monetary
 (government should
 explicitly
 financing
 accumulate account
 and the
 net forbail-out
 no
 assets). theThis heterogeneity
 clause (articles
 is especially within
 the123
 caseand the euro
 of the area.
 125welfare
 when EU The
 Treaty)
 (asset models
 make
 holding) issuch are
 modelscalibrated
 concentrated or estimated
 unrealistic.
 on a minority of householdsfor
 3different
 or when regions/countries,
 the government seeks with
 to the
 smooth euro
 tax area
 distortionbeing
 over split
 time andinto
 thus either
 needs core
 asset vs. periphery
 buffers to offset (for
 adverseEAGLE
 shocks. and
 For
 However, the literature on the relationship between public debt and welfare remains ambiguous, with several studies indicating
 a
 aBEreview,
 negative see Dieppe
 impact of and
 debtGuarda (2015). optimal debt levels that are either very small or negative (government should
 on welfare
 the model) or Germany vs. theorrest of the EA (for GEAR). Second, and, more crucially, the three
 4
 Seeaccumulate
 inter alia, ECBnet assets).
 (2016b), ThisECis(2016),
 especially the case
 Bańkowski andwhen welfare (asset
 Ferdinandusse holding)
 (2017), IMFis(2017).
 concentrated
 For moreon details
 a minority of households
 on sustainability,
 models
 or when enrich
 the our
 governmentanalysis
 see Boubabdallah et al. (2017). seeks by
 to bringing
 smooth tax alternative
 distortion perspectives.
 over time and thus needs The EAGLE
 asset buffers tomodel features
 offset adverse a more
 shocks. For
 a review, see Dieppe and Guarda (2015).
 5
 detailed
 Andrés eteuro area
 al (2020) (including
 assess the role tradable
 of high debtand
 in anon-tradable
 monetary union sectors) and symmetric
 under an endogenous external
 fiscal limit block However,
 a la Bi (2012). (and as
 4
 Seetheinter alia, ECBof(2016b),
 introduction this novel ECfeature
 (2016), Bańkowski
 restricts and Ferdinandusse
 significantly the number of(2017),
 nominal IMFand(2017). For more
 real rigidities of details on sustainability,
 the model in comparison
 suchsee is
 to the very well
 Boubabdallah suited
 models usedetinal. to
 the(2017). assess
 current paper.international spillovers), GEAR includes a sound labour market, while
 5the BE model has a financial block with borrowing constrains à la Kiyotaki and Moore (1997) and long-
 Andrés et al (2020) assess the role of high debt in a monetary union under an endogenous fiscal limit a la Bi (2012). However,
 termthedebt.
 introduction
 Third, of this
 we novel
 are able featuretorestricts
 design significantly
 a broadthe number
 range ofofsimulation
 nominal and scenarios,
 real rigidities of the model
 which in comparison
 shed light into
BANCO DE ESPAÑA to 9theDOCUMENTO
 models used in the
 DE TRABAJO current paper.
 N.º 2029 6
 various risks associated with regimes of high debt. Hence, we can assess the macroeconomic

 6
 2
 Assuming a full consolidation between the balance sheet of the central bank and that of the fiscal authority, government debt is

of high public debt in a global DSGE setup designed for the euro area.5

 The analysis is based on three different global DSGE models (EAGLE – E(S)CB, GEAR – Bundesbank,
 and BE - Banco de España) that share the same new-Keynesian framework and feature an enhanced
 government sector. By using three different models widely employed for policy simulations in our
 institutions, we aim at providing a robust and encompassing analysis for the euro area. First, we
 consequences
 explicitly account of for
 an the
 excessive level ofwithin
 heterogeneity debt when
 the eurothearea.
 economy is hit byare
 The models an calibrated
 adverse shock, in normal
 or estimated for
 times or at the zero lower bound (ZLB). In addition, using the BE model, we
 different regions/countries, with the euro area being split into either core vs. periphery (for EAGLE can examine the specific
 and
 role BE
 the played by private
 model) deleveraging.
 or Germany Finally,
 vs. the rest a set
 of the EAof(forsimulations
 GEAR). Second, evaluateand, the long-term costs the
 more crucially, of athree
 high
 debt burden.
 models enrich our analysis by bringing alternative perspectives. The EAGLE model features a more
 detailed euro area (including tradable and non-tradable sectors) and symmetric external block (and as
 The main results of our simulations are the following: (i) A high level of public debt makes the economy
 such is very well suited to assess international spillovers), GEAR includes a sound labour market, while
 more vulnerable to shocks (crises); (ii) High public debt prolongs the time spent at the ZLB; (iii)
 the BE model has a financial block with borrowing constrains à la Kiyotaki and Moore (1997) and long-
 International spillovers increase the time spent at the ZLB for the high debt economy; (iv) A higher level
 term debt. Third, we are able to design a broad range of simulation scenarios, which shed light into
 of public debt crowds-out private debt in the short and long run; (v) High public debt restricts the scope
 various risks associated with regimes of high debt. Hence, we can assess the macroeconomic
 for counter-cyclical fiscal policy; (vi) A high level of public debt affects adversely potential (long-term)
 consequences of an excessive level of debt when the economy is hit by an adverse shock, in normal
 output, with a significant impairment in case of large sovereign risk premia reaction and use of most
 times or at
 Assuming theconsolidation
 a full zero lowerbetween
 boundthe (ZLB).
 balance In addition, usingbank
 the and
 BE that
 model, we can examine the specific
 thesheet of the central of the fiscal authority, government debt is
 2
 distortionary type of taxation to finance additional debt burden in the future.
 roleconsidered
 played by private
 risk-free deleveraging.
 in nominal Finally,
 terms. For the euroa area,
 set ofin simulations evaluate
 particular, institutional the like
 design long-term costsofofmonetary
 the prohibition a high
 financing and the no bail-out clause (articles 123 and 125 of the EU Treaty) make such models unrealistic.
 debt burden.
 These results show that in the presence of high debt, both the stabilisation and the sustainability
 3
 However, the literature on the relationship between public debt and welfare remains ambiguous, with several studies indicating
 objectives
 a negative ofimpact
 national fiscal
 of debt policies
 on welfare or in the euro
 optimal area that
 debt levels arearemore
 eitherdifficult to attain
 very small in case
 or negative of a severe
 (government should
 Theaccumulate
 main results of ourThis
 net assets). simulations
 is especiallyare
 the the
 casefollowing:
 when welfare(i)(asset
 A high level isofconcentrated
 holding) public debt onmakes the
 a minority of economy
 households
 idiosyncratic
 or when the government seeks to smooth tax distortion over time and thus needs asset buffers to offset adverse shocks. the
 shock. The main explanation relates to the constraint of monetary policy against For
 more vulnerable to shocks (crises); (ii) High public debt prolongs the time spent at the ZLB; (iii)
 backdrop
 a review,ofsee
 theDieppe
 euro andarea institutional
 Guarda (2015). design, with the main channel being that of heightened sovereign
 International
 See inter and spillovers
 alia, ECB (2016b),increase
 EC (2016), the time spent
 Bańkowski andtheat the ZLB for
 Ferdinandusse the high
 (2017), IMF debt economy; (iv) on
 A higher level
 of(2017). For more details sustainability,
 4
 spreads uncertainty. These results fit into broader literature sovereign vulnerability, according
 of public debt crowds-out
 see Boubabdallah private debt in the short and long run; (v) High public debt restricts the scope
 et al. (2017).
 toAndrés
 whichethigher risk premia
 al (2020) assess
 and borrowing
 the policy;
 role of high
 costs can union
 spill over to other sectors or jurisdictions, especially
 (vi)debt in a monetary under an endogenous fiscal limitpotential
 a la Bi (2012). However,
 5
 for counter-cyclical fiscal A high level of public debt affects adversely (long-term)
 in integrated
 the introductioneconomic
 of this novelandfeaturemonetary unions. the
 restricts significantly Investors
 number ofmaynominalthus more
 and real easily
 rigidities question
 of the both the
 model in comparison
 output,
 to thewith
 models a significant impairment
 used in the current paper. in case of large sovereign risk premia reaction and use of most
 sustainability of fiscal policies of a sovereign with high debt burden, particularly when its fiscal track-
 distortionary type of taxation to finance the additional debt burden in the future.
 record and growth prospects are poor, as well as its capacity to effectively implement counter-cyclical
 6
 fiscal
 Thesepolicies
 resultsand show stabilise
 that inthe theeconomy.
 presence of high debt, both the stabilisation and the sustainability
 objectives of national fiscal policies in the euro area are more difficult to attain in case of a severe
 The paper is structured as follows. Section 2 reviews the theoretical and empirical literature on the
 idiosyncratic shock. The main explanation relates to the constraint of monetary policy against the
 macroeconomic impact of public debt, with a special focus on the channels through which high debt
 backdrop of the euro area institutional design, with the main channel being that of heightened sovereign
 can ultimately affect growth. Section 3 reports the simulation results and Section 4 concludes.
 spreads and uncertainty. These results fit into the broader literature of sovereign vulnerability, according
 to which higher risk premia and borrowing costs can spill over to other sectors or jurisdictions, especially
 in integrated economic and monetary unions. Investors may thus more easily question both the
 II. Related literature
 sustainability of fiscal policies of a sovereign with high debt burden, particularly when its fiscal track-
 record
 The and growth
 literature prospects
 on the are poor, as
 macroeconomic well as
 effects of itspublic
 capacity to in
 debt, effectively implement
 particular counter-cyclical
 the empirical literature,
 fiscal policies and stabilise the economy.
 suggests that debt overhangs can exert adverse pressure on the economy through multiple channels. 6

 The paper
 First, a highisdebt
 structured as follows.
 burden makes Section more
 the economy 2 reviews the theoretical
 vulnerable and empirical
 to macroeconomic shocks.literature on the
 By restraining
 macroeconomic
 the impact of public
 room for counter-cyclical debt,
 fiscal withand
 policy a special
 throughfocus on the
 spillover channels
 effects to thethrough
 private which
 sector,high debt
 a public
 can ultimately affect growth. Section 3 reports the simulation results and Section 4 concludes.
 debt overhang can deepen volatility, restrain economic recovery or hurt the economy even in the short-
 run if aggressive consolidation needs to be implemented in recessions. High government borrowing
 requirements can make a country more prone to liquidity shocks and defaults. Perceived sovereign
 II. Related
 vulnerability, literature
 reflected in higher risk premia and borrowing costs, can spill over to other sectors or
 jurisdictions, especially in integrated economic and monetary unions. In particular, the “diabolic loop"
 The literature on the macroeconomic effects of public debt, in particular the empirical literature,
 between sovereign and bank credit risk through the so-called “balance sheet channel” was considered6
 suggests that debt overhangs can exert adverse pressure on the economy through multiple channels.
 by many to be the hallmark of the sovereign debt crisis in the periphery of the euro area (Brunnermeier
 et al.,a2016).
 First, At the
 high debt same
 burden time,the
 makes through the more
 economy country-spillovers
 vulnerable tochannel, the materialisation
 macroeconomic of risks to
 shocks. By restraining
 sovereign
 the room for debt sustainability can
 counter-cyclical have
 fiscal adverse
 policy and implications not only
 through spillover for the
 effects to country concerned,
 the private sector, abut also
 public
 for other members of the monetary union.
 debt overhang can deepen volatility, restrain economic recovery or hurt the economy even in the short-
 run if aggressive consolidation needs to be implemented in recessions. High government borrowing
 requirements can make a country more prone to liquidity shocks and defaults. Perceived sovereign
 vulnerability, reflected in higher risk premia and borrowing costs, can spill over to other sectors or
 jurisdictions, especially in integrated economic and monetary unions. In particular, the “diabolic loop"
 6 See also ECB (2016a) for a discussion of risks and vulnerabilities associated with regimes of high debt.
 between sovereign and bank credit risk through the so-called “balance sheet channel” was considered
 by many to be the hallmark of the sovereign debt crisis in the periphery of the euro area (Brunnermeier
 et al.,
BANCO DE ESPAÑA 10 2016).
 DOCUMENTOAt the same
 DE TRABAJO N.º 2029 7
 time, through the country-spillovers channel, the materialisation of risks to
 sovereign debt sustainability can have adverse implications not only for the country concerned, but also
 for other members of the monetary union.

debt overhang can deepen volatility, restrain economic recovery or hurt the economy even in the short-
 run if aggressive consolidation needs to be implemented in recessions. High government borrowing
 requirements can make a country more prone to liquidity shocks and defaults. Perceived sovereign
 vulnerability, reflected in higher risk premia and borrowing costs, can spill over to other sectors or
 jurisdictions, especially in integrated economic and monetary unions. In particular, the “diabolic loop"
 between sovereign and bank credit risk through the so-called “balance sheet channel” was considered
 Second,
 by many to andberelated to theofabove,
 the hallmark the theoretical
 the sovereign debt crisis and empirical
 in the peripheryliterature
 of the suggests
 euro area that high debt
 (Brunnermeier
 burdens can ultimately impede long-term growth. 7 This is particularly the case when debt is contracted
 et al., 2016). At the same time, through the country-spillovers channel, the materialisation of risks to
 to finance unproductive
 sovereign debt sustainability expenses
 can haveor beyond
 adverse optimal (growth-maximising)
 implications not only for thelevels country of concerned,
 public capital butstock
 also
 (Checherita-Westphal et al.,
 for other members of the monetary union. 2014). Moreover, the quality of a country’s institutional framework is also
 likely to affect the relationship between debt and growth, with particularly low growth performance in
 situations
 Second, and of high debttocoupled
 related the above,with poor institutions and
 the theoretical and empirical
 conversely, a cushioned
 literature impact
 suggests thatof high
 high debt
 in situations
 burdens canof very strong
 ultimately institutions
 impede long-term(Masuch
 growth. et al.,
 7 This2016). A long string
 is particularly the ofcaseresearch
 when debttendsistocontracted
 conclude
 that high public
 to finance debt can expenses
 unproductive adversely or affect growth
 beyond through
 optimal the channels of sovereign
 (growth-maximising) levels ofspreads (confidence
 public capital stock
 effects) and sovereignet yields,
 (Checherita-Westphal
 6 See also ECB (2016a) for a
 8
 al., 2014). financial
 Moreover, intermediation
 the quality of(bank credit),institutional
 a country’s 9 higher future distortionary
 framework is also
 discussion of risks and vulnerabilities associated with regimes of high debt.
 taxation, 10 future crowding-out of private investment11 (debt is a deadweight burden on the economy),
 likely to affect the relationship between debt and growth, with particularly low growth performance in
 lower
 situationsscope for counter-cyclical
 of high debt coupled with fiscal
 poorpolicy, including
 institutions and aconversely,
 reduced capacitya cushioned to finance
 impactfuture
 of highpublic
 debt
 7
 in situations12ofincreased
 investment, very strong uncertainty
 institutions and catalyser
 (Masuch for 2016).
 et al., banking crisis.string
 A long 13
 of research tends to conclude
 that high public debt can adversely affect growth through the channels of sovereign spreads (confidence
 The literature also points out 8 that the level is not the only relevant9 dimension in debt-related
 effects) and sovereign yields, financial intermediation (bank credit), higher future distortionary
 vulnerabilities.
 taxation,10 future Other characteristics
 crowding-out of privatedefining
 investmentits composition,
 11 (debt is a like the maturity
 deadweight burden structure, financing
 on the economy),
 method and the resulting composition of public debt can also have sizable economic effects. 14 A greater
 lower scope for counter-cyclical fiscal policy, including a reduced capacity to finance future public
 share of short-term
 investment, 12 increased debt uncertainty
 may make aand government
 catalyser more vulnerable
 for banking crisis.during
 13 a crisis, because of the need
 to roll over increased amounts of debt. Moreover, in case a debt crisis mixes elements of illiquidity and
 insolvency,
 The literature like also
 the euro
 pointsareaout
 sovereign
 that the crisis,
 level theisgovernment
 not the only would be vulnerable
 relevant dimension to badin news, whose
 debt-related
 real impact would
 vulnerabilities. be amplified
 Other by creditors’
 characteristics defining unwillingness
 its composition, to roll over
 like thetheirmaturity
 claims. structure, financing
 method and the resulting composition of public debt can also have sizable economic effects.14 A greater
 In our ofpaper,
 share short-termwe use debtDSGE
 may makemodel simulations more
 a government to investigate
 vulnerable the two main
 during a crisis,aspects
 because related
 of thetoneedthe
 macroeconomic
 to roll over increased consequences
 amounts ofofdebt. high Moreover,
 public debtinemphasized
 case a debtby the mixes
 crisis literature: a reduced
 elements capacity
 of illiquidity of
 and
 high debt economies to withstand shocks and adverse effects on potential
 insolvency, like the euro area sovereign crisis, the government would be vulnerable to bad news, whose (long-term) output. In doing
 so,
 realwe coverwould
 impact topicsbe that belong to
 amplified byvarious
 creditors’strings of the literature:
 unwillingness to roll on
 oversovereign
 their claims.default and nonlinearities
 between debt and output (Corsetti et al., 2013; Bi and Leeper, 2010), on fiscal multipliers (Coenen et
 al.,
 In our 2012),
 paper, fiscal
 we usespillovers
 DSGE (Alloza et al., 2019a,
 model simulations 2019b; Attinasi
 to investigate the twoetmain 2017; Auerbach
 al., aspects related to and the
 Gorodnichenko,
 macroeconomic consequences2012), or on the interaction
 of high public debt between fiscal and
 emphasized bymonetary
 the literature: policy, in particular
 a reduced capacity when
 of
 nominal
 high debtinterest
 economies ratestoare constrained
 withstand shocks by an andeffective
 adverselower effects bound (Bouakeza,
 on potential et al., 2017;
 (long-term) output. Bletzinger
 In doing
 and
 so, we Lalik, 2017;
 cover Arce
 topics et belong
 that al., 2016; Hills and
 to various Nakata,
 strings 2014).
 of the Compared
 literature: with these
 on sovereign papers,
 default andours assesses
 nonlinearities
 more
 between systematically
 debt and output the macroeconomic
 (Corsetti et al., implications
 2013; Bi andofLeeper, high public
 2010), debtoninfiscal
 a global DSGE(Coenen
 multipliers setup with et
 three different
 al., 2012), models
 fiscal designed
 spillovers for theeteuro
 (Alloza al.,area.
 2019a, 2019b; Attinasi et al., 2017; Auerbach and
 Gorodnichenko, 2012), or on the interaction between fiscal and monetary policy, in particular when
 nominal interest rates are constrained by an effective lower bound (Bouakeza, et al., 2017; Bletzinger
 and
 7
 Lalik,
 Several 2017; studies
 empirical Arce et(spanning
 al., 2016; Hills and
 all major Nakata,
 institutions) find2014). Compared
 a negative, with these
 mostly nonlinear, papers,
 impact of publicours
 debtassesses
 on growth.
 Woo and Kumar (2015) find evidence of nonlinearity with higher levels of initial debt (above 90% of GDP) having more
 more systematically
 significantly the macroeconomic
 negative effects on subsequent growth. implications
 For the euroof areahigh public
 sample beforedebt in a global
 the crisis, DSGE
 Checherita setup(2012)
 and Rother with
 find an inverted U-shape relationship between
 three different models designed for the euro area. public debt and growth with a median debt threshold endogenously determined
 at 90-100% of GDP. Subsequent research for the euro area (Baum et al., 2012) investigates the short-run impact of public
 debt on GDP growth for a more recent period (1990-2010).The authors find a positive and highly statistically significant effect,
 which decreases however to around zero and loses significance beyond public debt-to-GDP ratios of around 67%. For high
 debt-to-GDP ratios (above 95%), additional debt is found to have a negative impact on economic activity. Similar debt
 7
 Several empirical
 thresholds studiesin(spanning
 are found the literatureall major institutions)
 on early signals offind a negative,
 sovereign mostlyFor
 distress. nonlinear,
 instance, impact of public
 the current debt
 debt on growth.
 sustainability
 Woo andframework
 analysis Kumar (2015) of the find
 IMF evidence
 (in placeof since
 nonlinearity with higher
 2013) adopts levels
 a debt ratioofofinitial
 85% debt
 of GDP(above 90%
 to flag of GDP)
 fiscal risks having more
 in advanced
 significantlyalong
 economies negativethis effects on subsequent growth. For the euro area sample before the crisis, Checherita and Rother (2012)
 dimension.
 find an inverted U-shape relationship between public debt and growth with a median debt threshold endogenously determined
 8
 Seeat Codogno
 90-100% et of al.
 GDP.(2003), Attinasi et
 Subsequent al (2010)
 research forand
 theCorsetti
 euro area et al. (2013)
 (Baum et for
 al., the effect
 2012) on spreads
 investigates theand Laubach
 short-run (2009),
 impact of Baum
 public
 et
 debtal.on
 (2012)
 GDPfor the effect
 growth on long-term
 for a more sovereign
 recent period interest rates.authors find a positive and highly statistically significant effect,
 (1990-2010).The
 9
 See De Bonis
 which and Stacchini
 decreases however(2013)to aroundfor the
 zeroimpact on bank
 and loses credit andbeyond
 significance Jorda etpublic
 al. (2016) for the effects
 debt-to-GDP ratiosinofthe aftermath
 around 67%.ofForcrises.
 high
 The latter paper
 debt-to-GDP explores
 ratios (abovea 95%),
 historical database
 additional on is
 debt public-
 foundand private-sector
 to have a negative debt build-ups
 impact for advanced
 on economic countries
 activity. Similarfordebt
 the
 period
 thresholds1870–2011
 are found andinfinds
 the that although
 literature high public
 on early signalsdebt build-ups are
 of sovereign not correlated
 distress. with athe
 For instance, greater
 currentlikelihood of financial
 debt sustainability
 crisis,
 analysis a high level of of
 framework public debt (in
 the IMF does tendsince
 place to exacerbate
 2013) adoptsthe negative effects
 a debt ratio of of post-crisis
 85% of GDPfinancial sectorrisks
 to flag fiscal deleveraging.
 in advanced
 10
 Ineconomies along
 line with Barro this dimension.
 (1979). This can, in turn, affect precautionary savings and consumption today.
 8
 11See
 WooCodogno
 and Kumar et al. (2003),
 (2015), Attinasi
 Ostry et al.et(2015).
 al (2010) and Corsetti et al. (2013) for the effect on spreads and Laubach (2009), Baum
 et al. (2012) for the effect on long-term sovereign interest rates.
 12
 Chalk and Tanzi (2004), Checherita and Rother (2012).
 9
 See De Bonis and Stacchini (2013) for the impact on bank credit and Jorda et al. (2016) for the effects in the aftermath of crises.
 13
 See
 TheHemming et al.explores
 latter paper (2013); a Brunnermeier et al. (2016).
 historical database on public- and private-sector debt build-ups for advanced countries for the
 14
 See, for example,
 period 1870–2011 Bacchiocchi
 and finds that and although
 Missale (2005), Faraglia
 high public debtetbuild-ups
 al. (2008),
 areornotHatchondo
 correlated and Martínez
 with a greater(2013).
 likelihood of financial
 crisis, a high level of public debt does tend to exacerbate the negative effects of post-crisis financial sector deleveraging.
 10
 In line with Barro (1979). This can, in turn, affect precautionary savings and consumption today.
BANCO DE ESPAÑA 11 DOCUMENTO DE TRABAJO N.º 2029 8
 11
 Woo and Kumar (2015), Ostry et al. (2015).
 12
 Chalk and Tanzi (2004), Checherita and Rother (2012).
 13
 See Hemming et al. (2013); Brunnermeier et al. (2016).
 14
 See, for example, Bacchiocchi and Missale (2005), Faraglia et al. (2008), or Hatchondo and Martínez (2013).