Pepino sovereign risk and financial crisis the international political economy of the eurozone (2015)

Tables and FiguresTables 3.1 Greece and Ireland macro-data snapshot 613.2 Comparing political economy variables in Greece and Ireland 656.1 Key economic and financial indicators 128 the s

Series Editor: Timothy M Shaw, Visiting Professor, University of Massachusetts, Boston, USA, and Emeritus

Professor, University of London, UK

The global political economy is in flux as a series of cumulative crises impacts its organization and governance The IPE series has tracked its development in both analysis and structure over the last three decades It has always had a concentration on the global South Now the South increasingly challenges the North as the centre of devel- opment, also reflected in a growing number of submissions and publications on indebted Eurozone economies in Southern Europe.

An indispensable resource for scholars and researchers, the series examines a variety of capitalisms and tions by focusing on emerging economies, companies and sectors, debates and policies It informs diverse policy communities as the established trans-Atlantic North declines and “the rest”, especially the BRICS, rise.

connec-Titles include:

Ryan David Kiggins (editor)

THE POLITICAL ECONOMY OF RARE EARTH ELEMENTS RISING POWERS

AND TECHNOLOGICAL CHANGE

Seán Ó Riain, Felix Behling, Rossella Ciccia and Eoin Flaherty (editors)

THE CHANGING WORLDS AND WORKPLACES OF CAPITALISM

Alexander Korolev and Jing Huang

INTERNATIONAL COOPERATION IN THE DEVELOPMENT OF RUSSIA’S FAR EAST AND SIBERIA

Roman Goldbach

GLOBAL GOVERNANCE AND REGULATORY FAILURE

The Political Economy of Banking

Kate Ervine and Gavin Fridell (editors)

BEYOND FREE TRADE

Alternative Approaches to Trade, Politics and Power

Ray Kiely

THE BRICS, US “DECLINE” AND GLOBAL TRANSFORMATIONS

Philip Fountain, Robin Bush and R Michael Feener (editors)

RELIGION AND THE POLITICS OF DEVELOPMENT

Markus Fraundorfer

BRAZIL’S EMERGING ROLE IN GLOBAL SECTORAL GOVERNANCE

Health, Food Security and Bioenergy

Katherine Hirschfeld

GANGSTER STATES

Organized Crime, Kleptocracy and Political Collapse

Matthew Webb and Albert Wijeweera (editors)

THE POLITICAL ECONOMY OF CONFLICT IN SOUTH ASIA

Matthias Ebenau, Ian Bruff and Christian May (editors)

STATES AND MARKETS IN HYDROCARBON SECTORS

Critical and Global Perspectives

Jeffrey Dayton-Johnson

LATIN AMERICA’S EMERGING MIDDLE CLASSES

Economic Perspectives

Andrei Belyi and Kim Talus

STATES AND MARKETS IN HYDROCARBON SECTORS

Dries Lesage and Thijs Van de Graaf

RISING POWERS AND MULTILATERAL INSTITUTIONS

Leslie Elliott Armijo and Saori N Katada (editors)

THE FINANCIAL STATECRAFT OF EMERGING POWERS

Shield and Sword in Asia and Latin America

Md Mizanur Rahman, Tan Tai Yong and Ahsan Ullah (editors)

MIGRANT REMITTANCES IN SOUTH ASIA

Social, Economic and Political Implications

Bartholomew Paudyn

CREDIT RATINGS AND SOVEREIGN DEBT

In Search of the Brazil Dream

Toni Haastrup and Yong-Soo Eun (editors)

REGIONALISING GLOBAL CRISES

The Financial Crisis and New Frontiers in Regional Governance

Kobena T Hanson, Cristina D’Alessandro and Francis Owusu (editors)

MANAGING AFRICA’S NATURAL RESOURCES

Capacities for Development

Daniel Daianu, Carlo D’Adda, Giorgio Basevi and Rajeesh Kumar (editors)

THE EUROZONE CRISIS AND THE FUTURE OF EUROPE

The Political Economy of Further Integration and Governance

Karen E Young

THE POLITICAL ECONOMY OF ENERGY, FINANCE AND SECURITY IN THE UNITED ARAB EMIRATES Between the Majilis and the Market

Monique Taylor

THE CHINESE STATE, OIL AND ENERGY SECURITY

Benedicte Bull, Fulvio Castellacci and Yuri Kasahara

BUSINESS GROUPS AND TRANSNATIONAL CAPITALISM IN CENTRAL AMERICA

Economic and Political Strategies

Leila Simona Talani

THE ARAB SPRING IN THE GLOBAL POLITICAL ECONOMY

Andreas Nölke (editor)

MULTINATIONAL CORPORATIONS FROM EMERGING MARKETS

State Capitalism 3.0

Roshen Hendrickson

PROMOTING U.S INVESTMENT IN SUB-SAHARAN AFRICA

Bhumitra Chakma

SOUTH ASIA IN TRANSITION

Democracy, Political Economy and Security

Greig Charnock, Thomas Purcell and Ramon Ribera-Fumaz

THE LIMITS TO CAPITAL IN SPAIN

Crisis and Revolt in the European South

Felipe Amin Filomeno

MONSANTO AND INTELLECTUAL PROPERTY IN SOUTH AMERICA

Eirikur Bergmann

ICELAND AND THE INTERNATIONAL FINANCIAL CRISIS

Boom, Bust and Recovery

Silvia Pepino

SOVEREIGN RISK AND FINANCIAL CRISIS

The International Political Economy of the Eurozone

International Political Economy Series

Series Standing Order ISBN 978–0–333–71708–0 hardcover

Series Standing Order ISBN 978–0–333–71110–1 paperback

(outside North America only) yy

You can receive future titles in this series as they are published by placing a standing order Please contact your bookseller or, in case of difficulty, write to us at the address below with your name and address, the title of the series and one of the ISBNs quoted above.

Customer Services Department, Macmillan Distribution Ltd, Houndmills, Basingstoke, Hampshire RG21 6XS, England

Sovereign Risk and Financial Crisis

The International Political Economy of the Eurozone

Silvia Pepino

Bank of England, UK

Softcover reprint of the hardcover 1st edition 2015 978-1-137-51163-8 The book was written while the author was at LSE Any views expressed are solely those of the author and so cannot be taken to represent those

of the Bank of England.

All rights reserved No reproduction, copy or transmission of this

publication may be made without written permission.

No portion of this publication may be reproduced, copied or transmitted save with written permission or in accordance with the provisions of the Copyright, Designs and Patents Act 1988, or under the terms of any licence permitting limited copying issued by the Copyright Licensing Agency, Saffron House, 6–10 Kirby Street, London EC1N 8TS.

Any person who does any unauthorized act in relation to this publication may be liable to criminal prosecution and civil claims for damages The author has asserted her right to be identified as the author of this work

in accordance with the Copyright, Designs and Patents Act 1988.

First published 2015 by

PALGRAVE MACMILLAN

Palgrave Macmillan in the UK is an imprint of Macmillan Publishers Limited, registered in England, company number 785998, of Houndmills, Basingstoke, Hampshire RG21 6XS.

Palgrave Macmillan in the US is a division of St Martin’s Press LLC,

175 Fifth Avenue, New York, NY 10010.

Palgrave Macmillan is the global academic imprint of the above companies and has companies and representatives throughout the world.

Palgrave®and Macmillan®are registered trademarks in the United States, the United Kingdom, Europe and other countries.

ISBN 978-1-349-57304-2 ISBN 978-1-137-51164-5 (eBook)

DOI 10.1057/9781137511645

This book is printed on paper suitable for recycling and made from fully managed and sustained forest sources Logging, pulping and manufacturing processes are expected to conform to the environmental regulations of the country of origin.

A catalogue record for this book is available from the British Library Library of Congress Cataloging-in-Publication Data

Pepino, Silvia, 1976–

Sovereign risk and financial crisis : the international political economy of the eurozone / Silvia Pepino.

pages cm — (International Political Economy Series)

1 Debts, Public—European Union countries 2 Financial crises— European Union countries 3 Eurozone 4 European Union

countries—Economic policy I Title.

HJ8615.P46 2015

4.5 Key results from the Greek case study 94

5.2 Evolving investment analysis and the Irish sovereign

v

5.3 Domestic political economy of the Irish sovereign

5.4 International political economy of the Irish sovereign

5.5 Key results from the Irish case study 119

6 The International Political Economy of the Eurozone

6.2 Lessons from the Greek and Irish crises 1226.3 Application to other Eurozone sovereigns 1276.4 Next steps in IPE research 1336.5 Implications for policy 135

Tables and Figures

Tables

3.1 Greece and Ireland macro-data snapshot 613.2 Comparing political economy variables in Greece and Ireland 656.1 Key economic and financial indicators 128

the sovereign default decision 493.1 Greece and Ireland government bond spreads to Germany 563.2 Eurozone government bond spreads to German benchmark 573.3 S-score for Greece and Ireland 643.4 Greece government bond spreads and key events 673.5 Ireland government bond spreads and key events 714.1 Greece ten-year government bond spread to Germany and

When the theme of Sovereign Risk and Financial Crisis was being defined, it

was far from obvious that the issue of Euro-area sovereign risk would become

as prominent as it has in the last few years Meanwhile, a severe financial sis hit the US and Europe, and governments emerged as the only remaininglifeline for failing banks and a collapsing real economy Since then, a number

cri-of Euro-area governments have come into the firing line, burdened by bankrescues, automatic stabilisers and countercyclical fiscal policy, and one of themain events in the financial history of the last few decades has unfolded.For decades, scholars, investors and policymakers treated sovereign defaultrisk as a defining feature of emerging market economies Recently, sovereignrisk has re-emerged as an empirical issue for advanced economies, raisingnew questions for academic research, policymaking and popular debate Thekey tenet of this book is that domestic and international political factorsinfluence sovereign credibility in financial markets in developed democra-cies as well as in emerging markets, and the analysis reveals which politicaleconomy factors matter most for investment decisions The unfolding ofthe Eurozone debt crisis at the time of writing provided a valuable real-lifetest of the theoretical arguments Indeed, the book combines an innovativeinternational political economy framework with the results of in-depth anal-ysis of government bond market fluctuations during the crisis A strong linkemerges between political economy factors and financial market perceptions

of sovereign risk

The result is an original theoretical and empirical approach to open thedebate on the political determinants of sovereign risk premia in devel-oped democracies The constantly evolving nature of the debt crisis added

a number of challenges to the project, but the resulting efforts added to itstimeliness and interest for the academic and policy community Crucially,its conclusions sound like a cautionary tale for the current historical phase,alerting policymakers as to the importance of carefully managing the politi-cal economy aspects of the painful macroeconomic adjustments that manyEuropean countries are now going through

viii

of developed economies, including the US, the UK and Japan, as well asmost Eurozone economies However, financial markets punished Eurozonesovereigns, and particularly the weaker economies in the Eurozone,1moreseverely than stand-alone developed democracies.2The particular nature ofEurozone sovereign debt, as argued by De Grauwe (2012), likely contributed

to accelerating market concerns about sovereign creditworthiness At thetime of writing, it is too early to tell whether the remaining developed-worldsovereigns will remain immune to market reversals in the coming years.Against this uncertain backdrop, the re-emergence of sovereign risk as anempirical issue for a new group of countries is bound also to ignite interest

in academia in studying its features

This book places itself in the historical and empirical context describedabove It aims to contribute to understanding how international financial

1

markets price sovereign risk in developed democracies, and specifically inthe developed democracies of the Eurozone It alerts readers to the role thatthe political economy backdrop can play in influencing sovereign credi-bility, particularly during a sovereign debt crisis Indeed, the overarchinghypothesis of this study is that politics matters for investors’ assessment ofdeveloped democracies as well as for emerging-market sovereigns In particu-lar, as a result of theoretical and empirical analysis, this book argues that thesocio-political landscape influences government bond pricing in developeddemocracies, particularly during sovereign debt crisis In so doing, it adds anew perspective to the existing academic tradition, which argues that poli-tics had little impact on bond spreads in advanced economies prior to thecrisis (mainly Mosley, 2003) Moreover, it extends the coverage of political-economy factors from the domestic to the international sphere, integratingthe domestic and international perspectives that have so far mostly beenanalysed separately The empirical focus on the Eurozone sovereign debtcrisis provides a timely and policy-relevant analysis of an issue of highimportance from both positive and normative perspectives For Eurozonecountries, issues related to sovereign creditworthiness go well beyond theboundaries of national policymaking, reaching the heart of the monetaryunion’s governance infrastructure The core of this study is concerned withthe interaction of politics, economics and financial markets In particular,

it focuses on an area where interactions between an unusually neous set of factors blend into observable variables in financial markets

heteroge-To address such an interdisciplinary issue, it gathers insights from a ety of disciplines and aims to avoid artificial boundaries between politicaland economic sciences, in line with the international political economytradition

vari-1.1.1 Key arguments and existing views

This book looks at how the political economy context influences sovereigncredibility in developed democracies In particular, it highlights the domesticand international political economy factors that impacted government bondmarkets during the Eurozone sovereign debt crisis We present the study intwo steps First, we develop an international political economy frameworkfor the analysis of sovereign risk perceptions in developed democracies thatcan be applied to Eurozone countries;3second, we test the framework empir-ically through the investigation of two events: the sovereign debt crises thathit Greece and Ireland in 2010 We advance three key arguments in theinterpretation of the Eurozone sovereign debt crisis, and the events observed

in Greece and Ireland in particular First, investment analysis evolves overtime, so static categorisations of countries such as the traditional divisionbetween developed democracies and emerging markets may not hold in thelong term Second, the domestic socio-political system affects sovereign riskperceptions in developed democracies, particularly during a sovereign debt

crisis Third, the financial market credibility of a sovereign under fiscal stress

is influenced also by the role of external de facto veto players and the degree

of proximity between debtor and international creditors

The most complete study of this matter published so far in the tional political economy sphere is Mosley (2003): in her analysis of gov-ernment bond markets between 1981 and 1997, she finds that investors indeveloped democracies focus on a limited number of macroeconomic short-cuts to inform their country choices, with little interest in politics Bernhardand Leblang (2006b) identify an impact of political processes on interestrate volatility, but do not analyse the broader role of institutional and soci-etal features Similarly, Alesina and Roubini (1997) uncover some evidencethat US bond markets are impacted by fluctuations in opinion polls ahead

interna-of presidential elections, consistent with the predictions interna-of partisan ical cycle theories Meanwhile, empirical studies of Eurozone governmentbond spreads (for example those of Codogno et al., 2003; Manganelli andWolswijk, 2009; Attinasi et al., 2009) do not engage with the role of politicalfactors However, the fundamentally political nature of sovereign debt itself(as highlighted by Eaton et al., 1986) and a preliminary observation of eventsduring the Eurozone sovereign debt crisis suggest that political institutionsand the international political economy context should be found to assumeconsiderable relevance for financial markets In order to explain the nar-rower focus of investors in developed democracies, Mosley (2003) argues thatinvestors distinguish between developed democracies, which are assumed to

polit-be “good credits”, and emerging markets, which carry default risk Whilefinancial markets impose broad constraints on government policy auton-omy in emerging markets, they impose only narrow constraints in developeddemocracies To take a step further in the analysis, we propose a dynamicmodel of investor behaviour, where markets update their pricing strate-gies over time, and where the distinction of sovereign borrowers betweendeveloped democracies and emerging markets may not hold all the time

In so doing, we place financial market behaviour in an intermediate positionbetween full efficiency and complete irrationality, as in the work of Lo (2004)and Willett (2000) Specifically, we argue that whenever default risk becomessalient, and typically in a sovereign debt crisis, investors will broaden thescope of their analysis to include political factors, in developed democracies

as well as in emerging markets When a sovereign borrower approaches asituation in which a choice concerning default is potentially to be made,political trade-offs emerge that may not have been as strong, or even rele-vant, in good times As a result, politics becomes increasingly important inassessing sovereign creditworthiness, and financial markets will take account

of this, adapting valuation models to changing circumstances

North and Weingast (1989) provide the seminal paper for analysing thedomestic political sources of sovereign credibility Comparing the experi-ences of England and France in the early modern era, they argue that a

higher number of institutional checks and balances in the political tem boost sovereign credibility in financial markets The findings of Northand Weingast and their disciples appear, however, to conflict with the pre-dictions of the “consolidation” literature – represented, for example, byRoubini and Sachs (1989) and by Alesina and Roubini (1997) – whichpostulates that greater political fragmentation makes fiscal consolidationmore difficult MacIntyre (2001) proposes a compromise between the twoapproaches to explain differing degrees of capital outflows across coun-tries during the Asian financial crisis of the 1990s: financial markets dislikeexcesses in both policy volatility and rigidity, and thus prefer intermedi-ate veto-player configurations MacIntyre’s focus is strictly on institutionalveto players and on emerging markets.4 Developed democracies, however,present considerably less variation in the distribution of formal veto author-ity and do not occupy the extremes found in emerging markets In thecontext of this debate, we argue that differences in institutional veto-playerconstellations are not sufficient for understanding how markets distinguishbetween sovereign borrowers in developed democracies Instead, we positthat investors take into account the broader socio-political system In par-ticular, we identify the degree of socio-political contestation, as well as theinteraction between the number of formal veto players and socio-politicalcontestation, as relevant for sovereign credibility Moreover, strong exter-nal creditors may act as external de facto veto players, particularly whenthe possibility of external bail-out – from the International Monetary Fund(IMF), European institutions or bilateral sources – emerges as an additionaloption available to a government under fiscal stress.5In these circumstances,

sys-we argue that the preferences of those players will also influence sovereignrisk perceptions in the debtor country Broadly, we argue that financial mar-kets will assess sovereign borrowers more favourably when there is greatereconomic, financial and ideological proximity between debtor and creditorcountries Indeed, the higher the direct and indirect costs are perceived to

be, the less likely is a sovereign borrower to default on its debt to nal creditors The issue-linkages approach to sovereign debt highlights, forexample, that trade sanctions can act as an incentive to sovereign debtrepayment (Bulow and Rogoff, 1989a) Indirectly, reputational theories ofsovereign debt (Eaton and Gersovitz, 1981; Tomz, 2007) also underscore theimportance of external considerations for the decisions of ailing sovereignborrowers On the other hand, the strong creditor country is more likely to

exter-be willing to provide assistance if it faces a high level of exposure (and thuspotential losses) towards the debtor, either directly or indirectly through itsbanks or companies In a similar vein, the literature on the political economy

of IMF lending highlights the role of US interests in IMF lending decisions(Woods, 2003; Oatley and Yackee, 2004) Crucially, the focus of this book onthe analysis of political economy factors is not intended to downplay theimportance of economic and financial variables as sources of sovereign debt

crisis and indicators of sovereign stress Sovereign debt crises are very plex events, and a huge set of factors can interact to determine the overallcrisis outcomes Political factors should be seen as contributory factors to

com-a sovereign debt crisis, rcom-ather thcom-an com-as exclusive drivers With regcom-ard to theEurozone sovereign debt crisis, in particular, a number of scholars have high-lighted the specific fragilities of the Economic and Monetary Union (EMU)governance system that increase the vulnerability of member countries tosovereign debt crisis (for example, Featherstone, 2011; De Grauwe, 2012; DeGrauwe and Ji, 2012) We take a different approach, adopting concepts fromthe international political economy sphere in order to analyse the crisis as itunfolded, rather than focusing on the economic and institutional conditionsthat led to it

1.2 Sovereign risk, developed democracies and financial

crisis

1.2.1 A major test for state–market relations

The global financial crisis which started in 2007 shook many of the victions prevailing among economists, financial market practitioners andpolicymakers In particular, the relationship between financial market play-ers and national governments assumed new, unexpected contours Thefinancial crisis followed a period when the balance of power had appeared to

con-be shifting from national governments towards the increasingly globalisedmarketplace (Strange, 1996) But, as crisis hit, national governments, centralbanks and international institutions had to intervene forcefully to shore upmarkets, bail out financial institutions and support the real economy Publicmoney flowed in billions from governments to banks in the US, the UK andmany other countries In the process, numerous banks were nationalised,

de jure or de facto Public authorities made it a priority to strengthen their

regulatory and supervisory grip on financial institutions and markets

In these circumstances, the contradictory position of financial marketswith respect to the desired role of the state emerged starkly In spite oftheir advocacy in good times of a retrenchment in the role of the state,markets more than welcomed state protection during the crisis As Walterand Sen pointed out, “The financial turmoil brought home once again

the lesson that financial sector actors prefer rapid and deep state tion during crisis” (2009, p 168) Meanwhile, the rescue programmes of the

interven-US administration received mixed reviews outside the financial institutionsthat benefited from these Stiglitz (2009a) highlighted the risks of “the priva-tizing of gains and the socializing of losses”, while suggesting that the toxicasset purchase plan outlined in late March 2009 amounted to a “robbery

of the American people” (Stiglitz, 2009b) By unveiling striking weaknesses

in financial markets and institutions, the crisis appeared at first to haveput some power back into the hands of nation-states Willingly or under

compulsion, states found themselves the determinant forces in the future offinancial markets, choosing which institutions and sectors to support, own-ing a large part of the banking sector, and dictating new and increasinglymore pervasive rules However, this also raises the question of whether thenew state of affairs was one additional symptom of “regulatory capture” orrepresented a real re-balancing of power away from the globalised privatesector and towards the nation-state Indeed, the additional financing needsfaced by sovereigns soon started to push in the opposite direction: sovereignswere more than ever in need of raising abundant and reasonably pricedfinancing in international financial markets, and this strengthened a keychannel for the capacity of the financial market to sanction and influencegovernment policies.

Indeed, the flip side of the increase in the real or perceived role of thestate was a burgeoning financial burden In many cases, the financial risksassumed by financial institutions were transferred wholesale to national gov-ernments The destiny of banking systems and their respective sovereignsbecame inextricably linked While smaller countries with proportionallyhuge banking sectors (as in the case of Iceland, where bank assets amounted

to more than 1,000% of GDP prior to the collapse) tipped over relativelyquickly, larger or more diversified countries, such as the UK and the US,faced a massive increase in public debt, set to haunt the nations for years tocome The increase in actual and contingent liabilities assumed by the pub-lic sector in financial sector rescues was compounded by a sharp underlyingdeterioration in public finances as a consequence of real-estate crisis, reces-sion and surging unemployment, as well as by the cost of expansive fiscalpolicy measures put in place to try to cushion the fallout of the financial cri-sis for the real economy and society in general The situation was made worse

by the fact that a number of advanced economies had failed to adjust theirpublic finance situation during good times, and aging populations added tothe longer-term sustainability risks

The fulcrum of the global financial crisis and its real economy and publicfinance repercussions was in developed democracies, while emerging mar-ket economies fared much better overall In the advanced economies as awhole, public debt rose sharply from 74% of GDP in 2007 to 104% of GDP

in 2011, having already been on an upward trajectory in the preceding threedecades Meanwhile, over the same period, public debt was fairly stable inemerging markets, hovering in a range between 33% and 39% (IMF, 2012).General government debts are now above 100% of GDP in the US, Japan and

a number of Eurozone countries;6only a few advanced economies, such asAustralia, Switzerland, Sweden and Norway, have maintained healthy publicfinances True, the extent and nature of the deterioration in the last few yearsand the magnitude of overall problems differ across countries; nevertheless,this was a widespread trend in the advanced economies In turn, the increase

in debt did not go unnoticed in financial markets, which started to require

higher rewards in order to provide funding for the governments of a number

of advanced economies and to insure against government default As the

“risk-free” status of developed democracies, as a group, was put in doubt,investors started to differentiate more markedly among sovereign borrowerswithin the category On the one hand, borrowing conditions deterioratedsharply for troubled advanced economy sovereigns: government bond yieldsrocketed for countries such as Greece, Portugal and Ireland, hit by out-right sovereign debt crisis, and increased significantly in other developeddemocracies, such as Italy and Spain By the end of June 2012, ten-year gov-ernment bond yields were 5.8% in Italy, 6.3% in Spain, 10.2% in Portugaland 25.8% in Greece On the other hand, sovereigns considered relativelystronger have seen falling funding costs in the period since the beginning

of the global financial crisis: government bonds in these countries benefitedfrom a mixture of safe-haven flows, monetary easing and, in some cases,outright central bank purchases.7By the end of June 2012, ten-year govern-ment bond yields had fallen to 1.6% in the US, 1.7% in the UK and 1.5% inGermany (Figure 1.1)

1.2.2 Sovereign risk and developed democracies

As a result of the developments described above, the issue of sovereigndefault risk in developed democracies re-emerged in financial markets, with

Figure 1.1 Ten-year government bond yields in selected developed democracies

investors discriminating more carefully within the group In this context,the concept of “risk-free” government debt in developed democracies was acrucial casualty of the financial crisis Doubts about the absolute credibility

of developed democracies’ capacity and willingness to repay their debts inthe near and distant future emerged first and most dramatically in developedeconomies considered to be in the most vulnerable positions, but they werenot limited to these and reached all the way to US Treasury debt Some evenstarted to question whether academics and market practitioners could stilluse Treasury yields as the reference “risk-free” interest rate (De Keuleneer,2008) The marked deterioration in the ratings attributed to the governmentbonds of developed democracies by specialised agencies (as shown for exam-ple by BIS,82012) is another symptom of this trend The sharp downgrades

in the debt of the most troubled countries, such as Greece and Portugal, arenot surprising and indeed came late relative to underlying and market devel-opments, but it is remarkable how even the US and French governments losttheir AAA credit ratings The US’s loss of its Standard and Poor’s AAA rating

in August 2011 is most striking from this perspective, and is clearly tomatic of the progressive deterioration in the perceived creditworthiness ofadvanced economy sovereigns

symp-For decades before these developments, sovereign default risk had beenidentified with developing and emerging economies Anderson et al (1996)point out that “[i]nvestors don’t perceive default risk for developed coun-try bonds” (p 2) Similarly, Mosley (2003) finds that investors distinguishbetween developing and developed economies on the basis of presence orlack of outright sovereign default risk.9The increase in perceived sovereigndefault risk across the globe since the 2008 financial crisis suggests thatscholars and practitioners need to allow for the possibility that doubts aboutthe solvency of developed economies may emerge in extreme situations,such as in the case of a financial crisis The absence of perceptions of defaultrisk for developed democracies in recent decades may indicate that investorsand scholars were unduly influenced by the lack of outright sovereigndefault experiences in advanced economies since the 1950s Another featurethat may have encouraged investors and academics to consider advancedeconomy debt as risk-free is the fact that it is normally issued in domes-tic currency While emerging market economies often suffer from so-calledoriginal sin (Eichengreen et al., 2003, p 1) and thus need to raise finance inforeign currency, developed democracies typically issue debt in domestic cur-rency Domestic currency denomination opens the additional option of debtmonetisation in response to fiscal stress, as an alternative to outright default

or fiscal tightening Debt monetisation can be seen as an indirect form ofdebt default, albeit with different distributional implications, and histori-cally inflation has often been seen as a less controversial way of eroding thereal value of domestic debt (Reinhart and Rogoff, 2009, p 174) However,inflationary default may not always be the route of choice, particularly when

the costs in terms of output loss and rampant inflation become extremelyelevated Indeed, Reinhart and Rogoff (2008b) show that outright default ondomestic debt has occurred a number of times in history, although “undersituations of greater duress than for pure external default” (p 3) More-over, in the current institutional setting, a significant subset of developeddemocracies, the Eurozone countries, have lost the option of unilaterallymonetising their debt Accordingly, recent experience shows that investorscan at times face the eventuality of one or more developed democraciesmoving into the “bad credit risk” category previously considered to includeonly emerging markets A longer historical perspective (that is, looking also

at the period before World War II) confirms that a specific set of countriescannot forever be considered completely immune from default risk, partic-ularly in the event of severe financial crisis In a series of papers reviewinghistorical episodes of financial crisis, Reinhart and Rogoff highlight somepoints in this regard First, sovereign defaults are much more frequent

in emerging markets, and no major sovereign credit event had occurred

in advanced economies since 1952 However, a number of governmentsnow considered highly creditworthy did experience credit events (includ-ing default, debt restructuring, cuts or delays in payments) in the first part

of the twentieth century Only a small set of countries had never defaulted(Reinhart and Rogoff, 2008a, p 14) Second, sovereign debt defaults aremost frequent for foreign-denominated and foreign-held debt, but they dooccur also in the case of domestic-denominated debt (Reinhart and Rogoff,2008b, p 10) Third, the incidence of sovereign defaults increases in theevent of severe financial crises, both in emerging and in developed markets

In particular, during the Great Depression sovereign debt defaults picked

up for both developed and developing economies (Reinhart and Rogoff,

2009, p 73) Fourth, severe banking crises dramatically weaken fiscal tions in both emerging and developed economies (Reinhart and Rogoff,2008c, p 1)

posi-1.2.3 Defining sovereign risk

The term “sovereign risk” is widely used and rarely specifically defined, andcan include a broader or narrower range of specific risk factors, depending

on the user and the context Business practitioners often attribute to theterm “sovereign risk” a broad meaning, akin to “country risk”, representingthe mix of all the risks involved in investing in or doing business with aparticular country Meanwhile, credit rating agencies use sovereign ratings

to quantify a narrower concept of sovereign risk, and specifically the “creditrisk of national governments”10 (Standard and Poor’s, 2008, p 19) This isalso the prevailing (albeit not unique) interpretation among financial marketpractitioners and policymakers Thus, before progressing further, it is impor-tant to specify the definition of sovereign risk that will be used throughoutthis book By “sovereign risk” we mean the risk that a national government

(“sovereign borrower”) will default (i.e will not repay its debts), a conceptthat can also be better specified as “sovereign default risk” or “sovereigncredit risk”, in line with the credit rating agencies’ use of the term.

The next step is to define more specifically what we mean by the term

“default” itself: this is particularly important given the specificities ofsovereign borrowing in developed democracies When a sovereign borrows

in domestic currency, it can reduce the net present value of its outstandingobligations in two key ways: either by ceasing to make interest and/or prin-cipal payments on its debt (“outright default”), or by eroding the value ofthe debt by creating inflation and depreciating the currency (“inflationarydefault”) As highlighted earlier, the option of monetising debt reduces theneed to resort to “outright default”, but does not eliminate it completely.Domestic currency denomination is typical for developed democracies, withthe exception of Eurozone countries In their analysis of domestic debt,Reinhart and Rogoff (2008b) make a clear distinction between “de jure” or

“overt” default, on the one hand, and inflation, hyperinflation or currencydebasement, on the other hand In their survey and classification of financialcrisis in the past eight centuries, they define sovereign default as “the failure

of a government to meet a principal or interest payment on the due date (orwithin the specified grace period)” (Reinhart and Rogoff, 2009, p 11) On thebasis of this definition, they identify episodes of outright default (“debt cri-sis”) as separate from episodes of inflation and currency crisis, while referring

to default through debasement as “an old favourite” (p 174) The concept of

“sovereign default” adopted by Reinhart and Rogoff corresponds to the nition employed by rating agencies Indeed, it matches Standard and Poor’sgeneral definition of default: “the failure to meet a principal or interest pay-ment on the due date (or within the specified grace period) contained in theoriginal terms of a debt issue” (2008, p 22) Standard and Poor’s further clar-ifies that sovereign default includes “a sovereign’s failure to service its debt

defi-as payments come due” defi-as well defi-as “distressed debt exchanges (even when nopayment is missed)” (p 21) In this book, we follow the stricter definition ofdefault used both by Reinhart and Rogoff and by credit rating agencies Theanalysis of default risk as a separate issue from that of inflation risk puts us inthe tradition of Alesina et al (1992) and of Lemmen and Goodhart (1999);both these papers make a distinction between sovereign credit risk and infla-tion (or exchange rate) risk components in their analysis of Organisation ofEconomic Co-operation and Development (OECD) government bond yields.For Eurozone countries, where exchange rate risk differentials have disap-peared, bond spreads to Germany are generally considered to reflect outrightdefault risk perceptions (see, for example, De Grauwe and Ji, 2012).11 Thefocus on the stricter definition of sovereign default risk in this study allows

us to concentrate on the analysis of the sovereign default decision, dently of the monetisation option and therefore of the particular monetarysetting of the country in question

indepen-1.2.4 Defining sovereign debt crisis

Sovereign debt crises are rarely defined per se Meanwhile, analyses

of sovereign debt crises tend to be focused on episodes involving outrightsovereign defaults This book employs a broader definition of sovereign debtcrisis, encompassing a broader set of episodes characterised by a sovereign’sactual or perceived difficulty in servicing its debt and reflected in bond mar-ket turbulence Pescatori and Amadou (2007) propose a similar approach foremerging markets, defining sovereign debt crises as “events occurring wheneither a country defaults or its bond spreads are above a critical thresh-old” In this broader definition, outright sovereign default as defined inSection 1.2.3 is just one of the possible outcomes of a sovereign debt cri-sis Alternative outcomes include various forms of rescheduling or externalbail-out – including, for example, IMF loans – as well as domestic measuressuch as freezing of bank deposits or public finances consolidation As shown

in Reinhart and Rogoff (2009), sovereign debt crises can be considered atype of financial crisis Other key types of financial crisis include bankingcrises and exchange rate crises (Reinhart and Rogoff, 2009, pp 3–14) Impor-tantly, Reinhart and Rogoff (2009) find that a financial crisis can often bebroken down into a sequence of sectoral crises, with banking and exchangerate crises typically preceding sovereign debt crises (p 271), and sovereigndebt crises in turn exacerbating banking crises The 2008 global financial cri-sis also reflects this pattern, with banking crisis followed by sovereign debtcrisis in a number of countries

The empirical part of this study investigates the sovereign debt crises inGreece and Ireland, identified on the basis of government bond market tur-bulence, rather than on the basis of the fundamental source of the crisis.Indeed, while both Greece and Ireland suffered a government bond marketrun and required external assistance as market funding dried up, the sources

of Greece’s troubles can more clearly be linked to public finances agement, while the Irish sovereign suffered from its intervention in whatwas originally a banking crisis Still, both sovereigns faced government bondmarket turbulence, as seen in both a sharp acceleration in bond yields insecondary markets (both in absolute terms and relative to Germany) anddifficulty in raising market financing in primary markets Regarding morespecifically the critical bond spread threshold for the identification of asovereign debt crisis, this is likely to vary depending on the type of debt(emerging markets, Eurozone, developed democracies) and to be influenced

misman-by the particular circumstances of the event Pescatori and Amadou (2007)estimate the spreads threshold for emerging market external debt at 1,000basis points over comparable US Treasury yields This numerical threshold isnot directly applicable to the much less volatile government bond markets ofdeveloped democracies and Eurozone countries in particular The behaviour

of bond markets during the Eurozone sovereign debt crisis suggests that twospread thresholds acted as discriminating levels for bond investors: ten-year

bond spreads from Germany of about 300 basis points generated cant worries and accelerated sell-offs in government bond markets, whilethe 500 basis points mark represented a “point of no return” for sovereigndebt crises.

signifi-1.3 Sovereign risk and the debt crisis in the Eurozone

In 2009–2012, the Eurozone was in the eye of the storm when it came tosovereign risk re-discovery after the global financial crisis The Eurozonesovereign debt crisis represents the most dramatic expression of sovereignrisk re-pricing in developed democracies in recent years A higher degree ofdifferentiation of the perceived intra-Eurozone sovereign risk profiles ini-tially emerged between late 2008 and early 2009, bringing intra-regionalspreads to levels not seen since the creation of the EMU The relatively mod-erate first phase of spread widening turned into outright sovereign debt crisis

in a number of countries in the following three years Having started in a gle member country, Greece, the sovereign debt crisis spread to a number ofother sovereigns; it had such pervasive repercussions on the regional eco-nomic, financial and policy landscape that it effectively came to be known

sin-as the “Eurozone sovereign debt crisis” or, even more broadly, sin-as the “Eurocrisis”

The increase in intra-EMU government bond spreads was all the morestriking because it came after a long period of very low intra-regional dif-ferentiation, which had in turn challenged many earlier predictions.12Bondmarkets overlooked intra-regional economic and structural differences forthe first nine and a half years of EMU With the crisis, investors were re-alerted to the differences remaining across national boundaries The movelooked like a belated recognition that not all EMU bonds can be considered

“equal” when it comes to default risk – particularly not when crisis strikes

1.3.1 The Eurozone sovereign debt crisis

Intra-EMU government bond spreads started widening moderately during

2008, reaching a local peak early in 2009 While this represented a firstmarket attempt to differentiate among Eurozone sovereigns, the move wasmodest compared with what was to come The real crisis started at theend of 2009, when the Greek sovereign, with a history of elevated govern-ment debt and deficits, started to lose credibility in financial markets Theconsequence was the start of an uptrend in Greek borrowing costs whichextended into the first part of 2010: in late April 2010, Greek ten-year bondyields moved above 8%, and sovereign credit default swap (CDS)13 pricesimplied a higher probability of default in Greece than anywhere else inthe world, including Argentina and Venezuela Right from the start of theGreek debacle, the crisis raised some crucial policy dilemmas for Europeanpolicymakers, which would permeate and contribute to defining all the

following stages of the Eurozone crisis This dilemma went to the core ofthe EMU institutional structure and the nature of the monetary union itself.The original EMU conception of monetary union without fiscal union, asexpressed in the Maastricht Treaty, became increasingly untenable and grad-ually evolved towards a higher degree of burden-sharing In May 2010,the 110 billion euro bilateral bail-out of the Greek sovereign (including

an IMF contribution) and the start of secondary market purchases of ernment bonds by the European Central Bank (ECB) were the first stepstowards a progressive re-definition of the no-bail-out element of the mone-tary union On that occasion, European policymakers also agreed to set up

gov-a common rescue fgov-acility, the Europegov-an Fingov-ancigov-al Stgov-ability Fgov-acility (EFSF),backed by a system of guarantees by Eurozone member states and aimed

at ensuring financial assistance to EMU members facing difficulties TheEFSF was initially endowed with 440 billion euros of guarantees, althoughthe actual lending capacity was to be around 250 billion euros, due to thecomplex guarantee structure put in place in order to obtain the AAA rat-ing The European Financial Stability Mechanism (EFSM), managed by theEuropean Commission and available to all 27 EU member states, would also

be available to contribute to future rescues, with an allocation of 60 billioneuros

In the second half of 2010, the sovereign debt crisis expanded fromGreece to Ireland The latter country had been battling a huge banking andreal-estate crisis since 2008, and this was increasingly weighing on publicfinances In November 2011, Ireland accepted a 67.5 billion euro rescuepackage financed by funds from the EFSF, the EFSM, the IMF and bilateralcontributions from the UK, Sweden and Denmark An additional 17.5 billioneuros came from Ireland’s own National Pension Fund and Treasury reserves.Portugal was the third country to be affected by the sovereign debt crisisand to need external help For a number of years, the country had sufferedfrom falling competiveness, troubled public finances and very low growth.Portugal tapped into the EFSF, the EFSM and the IMF for 78 billion euros

in May 2011 Each sovereign rescue operation was accompanied by strictconditionality: the disbursement of money required ongoing compliancewith tough adjustment programmes While Ireland and Portugal remainedmore or less on track with their programmes, Greece struggled to comply,and this resulted in ongoing uncertainty about the chances of success ofthe rescue measures The EFSF had initially been calibrated so as to be able

to bail out the smaller peripheral economies, but would be insufficient torescue the much larger Spanish economy, let alone Italy As a result, theEFSF’s firewall capacity was later ramped up In July 2011, European leadersagreed to an increase in EFSF guaranteed capital to 780 billion euros Thisimplied an actual lending capacity of 440 billion euros; with IMF and EFSMcontributions, the bail-out potential reached 750 billion euros Then, inOctober 2011, the EFSF was allocated new instruments of action, including

the possibility of intervening in primary and secondary markets, acting onthe basis of precautionary programmes, and financing the re-capitalisation

of banks through loans to governments

However, all this was insufficient to deter market fears of further sis contagion to larger economies Italian and Spanish government bondspreads widened sharply in the following few weeks, and the two countriesfaced intense market pressures for much of the second half of 2011 and thefirst half of 2012 Until then, the sovereign debt crisis had affected onlysmall, peripheral economies in the Eurozone, but by threatening to topplethe sovereigns of two large economies it assumed a much broader regionaldimension Italy was in the eye of the storm in the second half of 2011.Italian ten-year government bond yields moved above 7% in November

cri-2011 The country suffered from chronically high public debt, althoughthe public sector deficit, and particularly the primary public sector balance,was at that time in much better shape than that of many other developeddemocracies The turmoil in the sovereign debt market triggered the fall ofSilvio Berlusconi’s government and the creation of a technocratic govern-ment led by Mario Monti, who set the country on a path of strict fiscaltightening and structural reform, contributing to calming market fears, atleast for a while In the first half of 2012, doubts about the sustainability ofthe Spanish situation prevailed in bond markets Spanish ten-year govern-ment bond yields increased throughout much of the first half of 2012 andcrossed the 7% line in June The burden of bailing out the troubled bankingsector, hit by a severe real-estate crisis, was the main concern with regard

to Spain The country also suffered from prolonged recession and surgingunemployment Eventually, Spain was allocated a rescue package of approx-imately 100 billion euros, funded by the EFSF and specifically earmarked forbank re-capitalisation Meanwhile, Greece’s difficulties in complying withthe consolidation requirements of the first bail-out package, generating anever-growing funding gap, had come to a head in the autumn of 2011

In October 2011, European partners and the IMF organised a second rescuepackage, which was subsequently ratified in February 2012 The new pack-age was worth 130 billion euros, and, crucially, required a restructuring ofGreek sovereign liabilities as a condition for disbursement The Greek debtrestructuring took place in early March

As Eurozone policymakers adapted to face the evolving nature of the crisis,the EMU governance model saw additional institutional and “philosoph-ical” changes At the end of 2010, the European Council agreed to thecreation of a permanent rescue mechanism, the European Stability Mech-anism (ESM), to replace the EFSF by the time of its expiry in 2013 As acounterpart to this, it was also decided to introduce a “fiscal compact”, inorder to impose tougher controls on spending and borrowing, as well moresevere sanctions Amendments to the Lisbon Treaty were judged necessary

to introduce a permanent rescue mechanism and more severe sanctions,

and a new intergovernmental treaty was planned Then, at the end of June

2012, the European Council moved in the direction of a banking union,with agreement on the creation of a single bank supervisor at the ECB, andthe attribution to the ESM of the power to lend to banks directly once thenew supervisor was in place The modus operandi and philosophy of theECB also evolved significantly Its bank liquidity provision increased signif-icantly from the start of the global financial crisis, and in December 2011and February 2012, it carried out two unprecedented three-year long-termre-financing operations (LTROs) that injected about a trillion euros into theEuropean banking system But the biggest step, relative to its previous stance,was the overture to buying government bonds in secondary markets, ini-tially on a limited scale with the Securities Markets Programme (SMP) andthen on a possibly much larger scale with outright monetary transactions(OMTs).14At the time of writing it is unclear how much the crisis will finallyaffect the EMU, the EU, and the economies and societies of member coun-tries What is clear is that, while strictly defined sovereign debt crises wereconcentrated in a restricted number of EMU member countries, all membercountries were more or less directly affected, through their sovereigns, banks

or real economies, as were the institutional structure and vision of the EMUand the EU themselves

1.3.2 Sovereign risk and monetary union

The debt issued by Eurozone governments shares numerous features withthe debt of other developed democracies, including the back-up of sta-ble and generally well-developed institutional frameworks and of relativelyadvanced economic systems However, the nature of the debt issued byEurozone member states is also influenced by their membership of the mon-etary union and its governance structure When Eurozone countries joinedthe monetary union, member governments lost the option, available tosovereign borrowers in stand-alone developed democracies, of unilaterallymonetising their debt So, while stand-alone developed democracies issuedebt in domestic currency – where they are in full control of central bankpolicy – Eurozone governments issue debt in a currency, the euro, over whichthey lack direct control This is a feature that Eurozone sovereign debt has incommon with foreign-currency-denominated sovereign debt, issued mostly

by emerging markets, and with the debt of subnational governments infederal states

The loss of the monetisation option increases the risk that a sovereign willhave to resort to outright default in the face of fiscal and economic hardship(Lemmen and Goodhart, 1999; Reinhart and Rogoff, 2009).15 Interestingly,the higher outright default risk embedded in EMU countries’ debt wasreflected in their credit-rating structure when the monetary union was cre-ated Up to 1998, all 11 of the prospective EMU member countries enjoyed

an AAA (or equivalent) rating When EMU entry was finally confirmed,

Standard and Poor’s and the other major rating agencies merged the tic and foreign currency ratings of member countries.16As a result, a broaderdispersion emerged in the ratings of the 11 EMU entrants.17On Standard andPoor’s scale, the new ratings went from AA– for Portugal to AAA for Germany,France, the Netherlands and Luxembourg The recognition of increased out-right default risk should lead rational investors to require higher default riskpremia for monetary union bonds than for domestic currency-denominatedbonds issued by stand-alone sovereigns, other things being equal More-over, it should lead to increased differentiation in the sovereign risk premiarequired from sovereign borrowers within the monetary union itself, reflect-ing different domestic fundamentals This is generally found to be the case,for example, among subnational government borrowers in the US (Bayoumi

domes-et al., 1995) The other side of the coin is that the “club membership” ture of belonging to a monetary union could increase the likelihood of acountry in difficulty receiving some form of rescue from partner countries.The incentives for cross-country bail-outs would likely be higher in a mone-tary union than for stand-alone countries, due to greater default externalities

fea-as well fea-as possible solidarity among partners However, the actual bility of a bail-out would depend on the institutional structure in place

proba-as well proba-as on the incentives to comply with such a structure Thus, in arational market, the default risk premium on monetary union governmentbonds relative to stand-alone economies, as well as the degree of differentia-tion among member states, would decline to the extent that the probability

of a bail-out increased, and would therefore depend also on the presence

of burden-sharing institutions Consistently with this, the evidence fromnational monetary unions suggests that default risk premia and differenti-ation according to fiscal performance are lower in subnational entities thatreceive fiscal transfers in the context of solidarity schemes or are eligible forbail-out by the federal government: examples of this are found in Germanyand Canada (Schuknecht et al., 2008)

The EMU was originally designed as a monetary union with no fiscal fers and no bail-outs As a result, the dominant prediction was that defaultrisk premia would become more differentiated within the EMU (Alesina

trans-et al., 1992; Lemmen and Goodhart, 1999); this was, however, cruciallydependent on the credibility of the no-bail-out clause Meanwhile, fiscalrules (particularly the Stability and Growth Pact) were designed to ensurethat member states would not live beyond their means In the first few years

of EMU, intra-regional bond spreads fell to very low levels and there was noclear evidence of significantly higher differentiation in default risk premia(Buiter and Sibert, 2005) In spite of considerable research efforts, it remainedunclear whether this was mainly due to low credibility of the no-bail-outclause, high credibility of the fiscal rules and real convergence efforts, tech-nical factors such as the ECB collateral policy, deeper and more integrated

financial markets, or exogenous factors such as a generalised global increase

in risk appetite.18 Eventually, the Eurozone entered the sovereign debt crisiswith no burden-sharing or lender of last resort arrangement in place, and at

a time when the credibility of the Stability and Growth Pact had been tered by repeated violations that went unpunished As the crisis unfoldedover the following three years, institutional arrangements and policymakerattitudes evolved, as described in Section 1.3.1 Innovations included thecreation of joint rescue facilities (the EFSF/ESM) and the ECB becomingincreasingly relaxed, first in terms of bank liquidity provision and then interms of government bond purchases in secondary markets Additionally,the fiscal rules were toughened, and fiscal and macroeconomic coordinationand surveillance systems were reinforced

shat-The theoretical counterpart of an increase in credit risk premia in a etary union is that differentials in currency risk disappear as the monetaryframework is unified (Alesina et al., 1992; Lemmen and Goodhart, 1999).This prediction in turn is crucially dependent on the credibility of the mon-etary union itself If investors start to fear that the union may break up orthat one member country may exit, they may factor some currency riskdifferentials back into government bond yields While, in the first decade

mon-of EMU, differences in exchange rate risk premia were generally thought

to have disappeared from regional bond markets (Codogno et al., 2003),

in summer 2012 ECB President Draghi hinted at the return of a premiumrelated to “fears of the reversibility of the euro” in some government bondmarkets (ECB, 2012) A final point worth making in this context concernsthe prevailing analytical treatment of the euro as a “foreign” currency fromthe perspective of Eurozone sovereign bond issuers, as, for example, men-tioned by Mosley (2003) We would argue, instead, that the euro should beconsidered a “negotiated” currency True, individual member states do nothave full and direct control of monetary policy as in stand-alone economies.However, they are not as completely excluded from the decision process ascountries issuing in outright foreign currency would be The economic andfinancial conditions of EMU member states do have an influence on thedecisions of the ECB, although this influence tends to be related to the size

of the country, given that the central bank is required to act in the ests of the region as a whole and that a larger country tends, by definition,

inter-to have a higher influence on the aggregate Moreover, the independence

of the ECB from political influences may not be as absolute in periods ofextreme distress, particularly when the survival of the EMU itself is put intoquestion, as in normal times Recognising that the euro may be a “negoti-ated” currency also implies that the relative size and political influence ofeach member state may affect the way in which area-wide authorities, and

in particular the ECB, will react to episodes of stress in different membercountries

1.4 The focus of the book

1.4.1 Towards a political economy approach to sovereign risk

perceptions

In the academic literature, sovereign risk perceptions have been analysedfrom numerous different perspectives and within the context of differentdisciplines, including economics, finance, international political economyand international relations Equivalent concepts are sometimes given differ-ent names according to the branch of literature concerned, but in realitythey represent essentially the same issue approached from different angles.Thus, in the international political economy literature, we find an importantgroup of authors concerned with the analysis of the constraints imposed

on sovereign borrowers and their policies by internationally mobile globalcapital In another branch of the international political economy and eco-nomic history literature, we find authors looking at the “credibility” ofsovereign borrowers in financial markets In the economics and Europeanpolitical economy literature, we find authors studying financial markets’

“disciplining” role towards government borrowers Finally, the finance erature refers more directly to the market pricing of bond spreads and thevarious components driving these This book is explicitly designed to reflectthe breadth of these approaches and denominations, drawing from each

lit-of them as necessary to add value to the analysis, rather than being strained by the “silos” created by the separation of academic disciplines.The unfolding of the Eurozone sovereign debt crisis, and, broadly, there-emergence of sovereign risk as an issue for the region’s economies, pro-vided the empirical inspiration for this study Moreover, a review of theliterature on sovereign risk perceptions reinforced the motivation for choos-ing to focus on developed democracies rather than on emerging markets.The existing studies were highly concentrated on emerging markets, whilesovereign risk perceptions in developed democracies, and Eurozone coun-tries in particular, had been much less investigated The finance literatureproduced quantitative and technical studies of the determinants of bondyields in advanced economies, but these were generally concerned with fac-tors other than credit risk (interest rate risk, liquidity risk, internationalrisk aversion) A few studies looking at the evolution of Eurozone gov-ernment bond yields before and after the creation of the monetary unionhad engaged with issues of credit risk, but these left many questions open,partly because of the young and evolving nature of the monetary union.Meanwhile, the economics and political economy literature had generallyanalysed sovereign risk perceptions within the context of emerging markets.Overall, sovereign risk perceptions in developed democracies in general, and

con-in the Eurozone con-in particular, emerged both as an con-insufficiently studied nomenon and as a highly relevant theme for both positive and normativepurposes

phe-Sovereign risk and sovereign debt crisis are very complex phenomena,spanning economic, financial, political and behavioural domains In thiscontext, we need to consider the essential nature of sovereign risk itself andask ourselves: What is it that specifically identifies sovereign risk and differ-entiates it from other forms of credit risk? We can find the answer in theseminal paper on the study of sovereign debt itself, in the international eco-nomics field, by Eaton et al (1986) Eminent economists here highlight itspolitical nature as the defining feature of sovereign debt In contrast to pri-vate debtors, sovereign borrowers cannot be coerced to make good on theircommitments, due to the lack of enforcement mechanisms So, they argue,for a sovereign borrower, “willingness to pay” can determine default deci-sions long before its “ability to pay” becomes binding Reinhart and Rogoffalso find that “most country defaults happen long before a nation literallyruns out of resources” (2009, p 51) Thus, sovereign default is essentially

a political decision rather than a purely economic determination, implyingthat government creditworthiness, or sovereign risk, needs to be assessed onpolitical at least as much as on economic grounds However, there is a dis-connect between the theoretically recognised importance of political factors

in the determination of sovereign creditworthiness and the relatively limitedroom afforded to political factors by the empirical literature on sovereignrisk Caouette, Altman and Narayanan (2001) argue that the lack of inclu-sion of political and political economy variables in traditional approaches

to sovereign risk analysis is due to the greater difficulty of measuring these.Moreover, we found that the vast majority of studies looking at the role ofpolitical factors in driving sovereign risk perceptions are focused on emerg-ing markets, a finding consistent with the greater focus on emerging markets

in sovereign risk analysis overall A branch of the literature focuses on therole of political institutions in early modern Europe However, only limitedanalysis has been applied to contemporary developed democracies, eitherwithin or outside the Eurozone The existing literature relating political fac-tors to financial market performance focuses on political processes, such aselections, referenda and cabinet formation, rather than on institutional andsocietal factors (see, for example, Bernhard and Leblang, 2006b) Mosley(2003), the key reference in this area in international political economy,provides a comprehensive empirical analysis of interest rates on governmentbonds across the world, relying on a strong distinction between developeddemocracies and emerging markets In her study of bond markets between

1981 and 1997, she finds that, while investors in emerging markets consider

a broad set of variables, including political factors, when pricing sovereigndebt, investors in developed democracies focus on a limited number ofmacro-shortcuts to inform their country choices, with minor interest in thedirect observation of political factors.19 Mosley’s arguments are based ontime-dependent empirical evidence gathered in a specific historical period.Thus, her findings do not rule out the possibility that political and political

economy factors may indeed be found to matter in investor choices in oped democracies in different institutional and historical circumstances, andparticularly in episodes of fiscal stress.

devel-1.4.2 Domestic and international political aspects of

sovereign risk

In order to identify the nature of the political and political economy tors that may be expected to influence sovereign risk perceptions, we dissectthe components of the sovereign default decision itself, that is, the politi-cal decision faced by a sovereign when it encounters difficulties in servicingand repaying its debts Indeed, when facing a fiscal sustainability problem,

fac-a sovereign borrower hfac-as essentifac-ally three policy options: deffac-ault, consolidfac-a-tion or external bail-out.20Each of these options has different distributionalimplications, hitting government constituents and bondholders in differentways We can identify in a stylised way the main distributional implica-tions for each of the three options First, debt default will hurt bondholders,which can be domestic constituents or external lenders Meanwhile, it willrelieve domestic taxpayers and public spending beneficiaries of at least part

consolida-of the debt burden, depending on the magnitude and features consolida-of the actualdebt restructuring Second, fiscal consolidation will benefit domestic andforeign bondholders, which will enjoy the ongoing creditworthiness of thesovereign Meanwhile, it will hurt domestic taxpayers and public spendingbeneficiaries, as the government will need to increase taxes and cut spending

in order to restore debt sustainability through fiscal adjustments The ments required to achieve sustainability are often large, long and painful.Third, external bail-out will benefit bondholders, domestic and foreign Evi-dently, it will also add a burden to the external rescuer, be this an individualsovereign or an international institution re-grouping the funds of multiplesovereign lenders The impact on domestic taxpayers and public spendingbeneficiaries will depend on the toughness of the conditionality attached

adjust-to the external rescue Normally, the domestic taxpayers and users of publicservices should be better off than in the case of pure domestic consolidation,but worse off than in the case of outright debt default

Of course, the political trade-offs described above are summary tions, while the reality of winners and losers can be much more complex Forexample, debt default may, in fact, end by hurting domestic taxpayers andpublic spending beneficiaries, in a second instance, if the sovereign is not

stylisa-in a financially autonomous position (that is, if it does not have a primarysurplus) and it loses access to external financing as a consequence of thedefault That said, it is evident that major political trade-offs emerge when

a sovereign needs to restore fiscal sustainability In “good times”, the cal element of sovereign debt may fall into the background; meanwhile, in

politi-“bad times”, the political dynamic takes centre stage, as decisions need to

be made concerning the group that bears the inevitable losses These points

find broader resonance in Gourevitch’s Politics in Hard Times, in which the

author remarks: “Prosperity blurs a truth that hard times make clearer: thechoice made among conflicting policy proposals emerges out of politics”(1986, p 17) The default of a government has important distributional con-sequences, and, as such, its likelihood cannot appropriately be assessed inabstraction from the political sphere In the case of the EMU, the politicaldimension potentially assumes additional importance and displays signifi-cant ramifications across domestic and international layers of government.Indeed, given the high risk of spillovers and contagion across EMU members,

as well as fears that a sovereign default may eventually lead to EMU exit,issues related to sovereign creditworthiness go well beyond the boundaries

of national policymaking, reaching to the heart of the monetary union’sgovernance infrastructure In the Eurozone, the political decision really hap-pens on two levels: first at the national level, and second at the EMU level.Indeed, in the face of a material risk of default by a member country, EMUpartners face the decision of whether to bear the cost of a bail-out of thecountry in question, or to suffer the possible spillovers that the default of

a member country would imply for other members or for EMU (or evenfor the EU project) as a whole As a result, the government default decisionbecomes a “two-level” political decision.21 Before the Eurozone sovereigndebt crisis, the economics literature generally assumed that the rules set bythe Maastricht Treaty would be entirely credible and would not be the object

of revision or re-interpretation In so doing, they clearly underestimatedthe role of political negotiations in the default decision, and particularlythe fact that new political decisions can always overturn earlier ones In theevent, governments did not appear to have “tied their hands” (Giavazzi andPagano, 1988) sufficiently tightly, as the various forms of rescues and res-cue mechanisms were eventually put in place to avoid even more seriousconsequences from the debt crisis

Overall, from the described conception of the sovereign default decision,two separate sets of political and political economy factors emerge as likely

to have an impact on sovereign risk perceptions First, there are domesticnational political and political economy factors: the potential relevance ofthese factors derives from the trade-offs generated by the sovereign defaultdecision at the domestic level Second, there are international political andpolitical economy factors: the potential relevance of these factors derivesfrom the trade-offs originating in the relationship with external creditorsand rescuers The described mix of empirical observations and theoreticaldeductions leads to the overarching hypothesis of this book: politics mat-ters for sovereign risk pricing in developed democracies It also leads to theidentification of a crucial subject requiring further investigation: the domes-tic and international political determinants of sovereign risk perceptions indeveloped democracies, and during the Eurozone sovereign debt crisis inparticular

1.5 The approach in the book

1.5.1 The international political-economy framework

This study is composed of two key parts: first, the theoretical part aimed

at defining an international political economy framework for the analysis

of sovereign risk perceptions in developed democracies that can be applied

to Eurozone countries;22 second, the empirical part, in which the ical framework is operationalised and validated through the investigation

theoret-of the Greek and Irish sovereign debt crises The theoretical part developshypotheses concerning the domestic and international political economyfactors influencing sovereign risk perceptions in developed democracies

In addition, it proposes a dynamic approach to financial market behaviour,

in order to explain the transition of investors’ focus from the few shortcuts identified by Mosley (2003) to a wider set of variables duringthe Eurozone sovereign debt crisis As a result, our international political-economy framework is based on three pillars: evolving investment analysis;

macro-a link between the veto-plmacro-ayer constellmacro-ation in the politicmacro-al system macro-andsovereign risk perceptions; and a role for a country’s international politi-cal economy position in influencing the sovereign’s credibility in financialmarkets The empirical part operationalises the identified theoretical con-cepts into empirical objects and tests the theoretical hypotheses against theempirical experience First, we identify ten-year government bond spreads

to Germany as the main metric for the dependent variable in our study,sovereign risk perceptions in financial markets Then, through the investiga-tion of two case studies drawn from the broader Eurozone crisis experience,the Greek and Irish sovereign debt crises, we look for empirical validation ofour theoretical hypothesis

An important clarification concerns the choice of theoretical approachand how this relates to the empirical sample of the analysis Eurozonecountries represent the largest subset of the broader group of developeddemocracies, while at the same time presenting some features that differenti-ate them from stand-alone advanced economies In particular, as highlightedearlier, the government debt issued by Eurozone member states is influenced

by the particular EMU governance structure, which removes the option ofunilateral debt monetisation and magnifies cross-country spillovers It isthus useful to further clarify how the two categories of country relate toeach other in the approach followed in this book In the light of thehybrid nature of Eurozone sovereign debt, when considering the set ofpolitical-economy factors that may be influential in determining sovereigncredibility in the region, two types of approach are possible First, one couldfocus on the peculiarities of the Eurozone governance framework and howthis affects risk perceptions vis-à-vis stand-alone developed democracies.This is the route chosen, for example, by De Grauwe (2012), highlight-ing the fragilities created by the EMU institutional framework A different

approach, which we follow in this book, is to engage with the academicliterature and develop a theory with sufficient breadth to be potentiallyapplied to the broader set of developed democracies, not only to EMU mem-bers Indeed, our theory uses broader political and international politicaleconomy concepts rather than adopting a stricter European political econ-omy approach, which would likely have resulted in stronger emphasis onspecific Eurozone governance features This approach has two advantages.First, it allows us to abstract from the specific EMU institutional setting,which has been in flux in the last few years Indeed, the ongoing pro-cess of reform of EMU governance makes any consideration specificallyrelated to this crucially time-dependent Second, it generates a frameworkthat could potentially be applicable to analysing a broader set of developeddemocracies hit by sovereign debt crisis in the future Meanwhile, thereare some distinct empirical advantages of focusing on Eurozone countries.The common macro-policy framework and similar economic institutionsgreatly increase comparability and the likelihood of obtaining robust empir-ical results on the variables under analysis More specifically, differences inmonetary policy setting do not come into play in determining intra-regionalyield differentials In stand-alone developed democracies, domestic currencydebt denomination means that monetary factors tend to prevail in the deter-mination of bond yields (as shown, for example, by De Grauwe and Ji, 2012).

As a result, comparative analysis of non-EMU developed democracies oftenboils down to issues related to central banking, for example central bankindependence Looking, for example, at US or UK bond yields in the 2008–

2012 period, it is hard to distinguish23 how much of the decline in yields

to record low levels was due to central banks buying large quantities of ernment bonds, rather than being a reflection of fundamental market views

gov-on the evolutigov-on of each government’s fiscal soundness and lgov-ong-term debtsustainability Looking at Eurozone countries during the sovereign debt cri-sis allows us to abstract from the debt monetisation option in the empiricalanalysis, as well as in the theoretical discussion, without the analysis beingblurred by differences in the degree of fiscal dominance of monetary pol-icy Empirically, differences in sovereign risk perceptions across Eurozonecountries can be gauged directly by comparing government bond yields

In particular, government bond spreads to the German benchmark are erally considered mostly to reflect credit risk premia for Eurozone sovereigns(De Grauwe and Ji, 2012)

gen-1.5.2 The case studies

Traditional approaches to the analysis of sovereign risk in developed racies, and the Eurozone in particular, privileged a macro, quantitativeapproach However, the quantitative studies conducted so far have providedonly partial answers In order to obviate the shortfalls of necessarily syn-thetic large-n quantitative studies, this study adopts a small-n approach,

democ-looking in detail at two sovereign debt crisis episodes This small-n approachfinds validation in the experience of emerging-markets scholars: compara-tive case studies have been used with success in the study of sovereign debtcrisis in emerging markets, for example by MacIntyre (2001) As a result,this book differs from the existing literature on sovereign risk in developeddemocracies not only in terms of content, but also in its innovative empiricalapproach, focused on case studies rather than large-n quantitative estimates.Two case studies drawn from the broader experience of the Eurozonesovereign debt crisis are analysed: those of the sovereign debt crises in Greeceand in Ireland The case study selection aims to maximise unit homogeneity,variability in the dependent variable under investigation, and variability inthe independent variables.24Both Greece and Ireland were hit by sovereigndebt crisis in 2010 and needed to access external support However, therewere remarkable differences in their overall experience: while the sovereigndebt crisis was clearly a dramatic event in both countries, its severity andlength appeared significantly greater in Greece than in Ireland These dif-ferences provide the required variation in our dependent variable Crucially,Greece and Ireland also differ in terms of our key explanatory variables (oftenbeing at opposite extremes on the Eurozone spectrum), while at the sametime displaying similar characteristics in residual areas These features make

a comparison between the two episodes both highly relevant in the light

of our theory and appropriate from the methodological perspective On theone hand, Greece has a low number of formal veto players, a high level ofsocio-political contestation, and a low level of economic and financial inte-gration with the rest of the EMU and the global economy On the otherhand, Ireland has a higher number of formal veto players and a low level

of socio-political contestation, and is highly integrated with the rest of theEMU and the global economy.25

Meanwhile, the two countries have residual features that allow an priate comparison and should minimise distortions in the analysis andconsequent misattribution of causal relationships First of all, they are bothEMU countries: EMU countries share the central bank, a common macro-policy framework and similar economic institutions This markedly increasestheir comparability and therefore the likelihood of obtaining robust empir-ical results Moreover, Greece and Ireland are both small economies withinthe Eurozone, reducing the risk that differing outcomes may be related todifferences in size Greece accounts for 2.6% of Eurozone GDP and 3.4% ofEurozone population, while Ireland has 1.8% of Eurozone GDP and 1.4%

appro-of Eurozone population.26 Size matters primarily because this could be amajor driver of the countries’ influence on area-wide decision-making pro-cesses, blurring the role of other factors The similarly small size of Greeceand Ireland means that they can be treated as having similarly low formalinfluence on overall decision-making at the Eurozone and EU level or ondomestic policymaking in partner countries.27 This means that they both

tend to be “policy takers” in the European (and global) context, includingwith regard to monetary policy Additionally, foreign ownership was preva-lent at the onset of the crisis in the government bond markets of bothcountries: 77% of Greek government bonds and 83% of Irish governmentbonds were owned by foreigners in late 2009 This eliminates possible dis-tortions that could derive from behavioural differences of different investorgroups, particularly since domestic investors are typically considered to bemore stable than foreign investors.28Finally, they did not present major dif-ferences in terms of bond market liquidity in the period under consideration,both being relatively small bond markets in the Eurozone context In 2010,government debt securities outstanding were 253 billion euros in Greece and

96 billion euros in Ireland, versus 1.5 trillion for Italy and more 1.4 trillionfor France and Germany.29 This means that we do not need to account forliquidity risk possibly impacting bond prices in different ways in the twocountries

Clearly, a multitude of economic, financial and political events and tors can influence bond spreads during a sovereign debt crisis, and at times

fac-it may be hard to disentangle and quantify the effect of each Indeed, tiple events or factors can have a simultaneous impact, and, conversely, acertain event or fact may take some time to fully filter through to marketprices Our case studies are designed to minimise the risk of misinterpretingcasual links.30 As mentioned earlier, Eurozone government bond spreads toGermany are identified as the operational metric for the concept of sovereignrisk perceptions in financial markets Moreover, the collection of empiricalevidence on which the discussion in each case study is based reflects a com-bination of three elements The first of these is direct evidence of the causalmechanism of government bond spreads reacting to news concerning thementioned political economy factors The second is indirect evidence onthe causal chain whereby government bond spreads are found to respond tofacts or features that are typical of a particular political economy constella-tion The third is a comparative analysis of the political economy backdrop

mul-of the Greek and Irish sovereigns

1.6 Outline of the book

The structure of the book reflects the described approach of analysis.Chapter 2 proposes an international political economy framework foranalysing sovereign risk The political economy framework is composed ofthree pillars: the first deals with financial market behaviour, the secondfocuses on the relationship between the domestic socio-political system andsovereign credibility in financial markets, and the third covers the connec-tion between international proximity and sovereign risk premia Chapter 3sets the stage for the case studies First, it introduces government bondspreads as an empirical metric of financial market perceptions of sovereign

risk Second, it reviews developments in Eurozone bond spreads in the last 20years and highlights their role in the design of the monetary union Third, itzooms into a comparative analysis of the macroeconomic and political econ-omy backdrops of Greece and Ireland in the run-up to the crisis as well asconsidering the impact of politics on bond spreads pre-2008 Chapters 4 and

5 analyse the Greek and Irish sovereign debt crises in light of our tional political economy framework Mirroring the structure of the theory,each case study is broken down into three separate sections First, we focus

interna-on market behavioural traits, and in particular interna-on understanding the tion into “crisis mode” Second, we focus on the role of the domestic politicalsystem in driving the sovereign risk premia And third, we focus on the role

transi-of international political economy factors Chapter 6 concludes by marising the key lessons from this study, highlighting the implications ofour key arguments for the analysis of sovereign debt crises in internationalpolitical economy and for government policy

of an international bail-out is introduced into the picture Indeed, stream economists highlight the fact that the quality of political institutions

main-is a fundamental determinant of a country’s debt tolerance level (Reinhart

et al., 2003) Accordingly, economists and political scientists have made afew attempts at investigating the political features more or less conducive todefault in emerging markets (for example Kohlscheen, 2007) However, therehas been limited investigation of the role of politics in affecting risk pre-mia in what are normally denominated “developed democracies”, includingEuro area members, be these recent or long-standing “graduates” to the club

In order to fill this gap, in this chapter we develop a theoretical frameworkfor the analysis of the relationship between the domestic and internationalpolitical-economy context of sovereign borrowing, on the one hand, andfinancial markets, on the other hand, with specific reference to moderndeveloped democracies facing crisis We progress in three steps First, weidentify a mechanism of adaptive investor beliefs which underpins andmotivates the subsequent investigation of the role of politics in the pricing

27

of sovereign risk in modern developed democracies Second, we build theappropriate theoretical foundations to anticipate the impact of both domes-tic and international political factors on sovereign risk perceptions acrossdifferent democratic systems Third, we distil our main theoretical argu-ments from the broader analysis in order to formulate specific hypotheses

to be tested empirically The theoretical framework built in this chapter will

be the foundation for the subsequent empirical analysis: to this end, in thenext chapter we will develop an empirical framework that operationalisesour theory and creates the conditions for empirically testing our hypotheses

in Chapters 4 and 5

2.2 Adaptive markets and sovereign risk models

2.2.1 Beyond the developed/emerging distinction in financial

markets

Historically, political factors are mostly found to have modest tory power in the market pricing of government bonds in developeddemocracies, both in and outside the Euro area Mosley (2000, 2003) pro-vides an explanation for this She suggests a distinction between the waysinvestors assess sovereigns in developed economies and emerging markets.She highlights how institutional investors adopt shortcuts in pricing gov-ernment bonds of developed democracies, focusing on a very limited set ofmacroeconomic variables to make their decisions She contrasts this withthe case of emerging market economies, where market participants con-sider a broader set of variables, including political and micro-level factors

explana-as well explana-as macroeconomic fundamentals In turn, to explain this behaviour,she adopts the concept of “economization”: since investors have a limitedamount of resources for their research, they concentrate their investigativeefforts on those areas where the costs of error are higher Government bondinvestors in developed democracies employ the working assumption thatdeveloped democracies are “good credits” on the basis of pre-establishedbeliefs about types of government In developed democracies, the over-all probability of outright default and the default risk differences withinthe group are considered to be too limited to justify the time and effortnecessary to collect additional information Meanwhile, emerging marketinvesting requires a more refined understanding of the features of the coun-tries involved, due to the higher saliency of default risk To be sure, Mosley’sstudy does not rule out the possibility that a sovereign may at some pointshift from the “good credit” to the “bad credit” category, and that investorassessment methods may change, but it leaves scope for additional exam-ination of how countries may shift from one group to the other Indeed,her analysis refers to observation of the situation in a delimited period oftime (1981–1997), during which economies do not move from one group

to the other and the set of metrics used to assess one group of economies

or the other does not change over time In the period covered by Mosley’sanalysis, neoliberal preferences based on monetarism became widespread

at the global level, influencing policymakers (Chwieroth, 2007), tators and observers “Macro-stability”, measured in terms of inflation andbudget deficit reduction, was considered a goal in itself, particularly in theindustrialised world The emphasis on “stability” as a policy goal was notpresent before the 1980s and has faded somewhat in recent years, sincelow inflation was achieved on a large scale; the applications of neoliberalprecepts brought little success in a number of emerging economies; and,more recently, Keynesianism acquired a new lease of life during the 2007financial crisis Moreover, 11 of the 19 developed democracies considered

commen-in Mosley’s paper were engaged commen-in macro-convergence ahead of EMU for

a meaningful part of the period under consideration, with the MaastrichtTreaty putting particular emphasis on inflation and deficit variables This isnot entirely unrelated to the previous point, as some claim that the emergingneoliberal consensus was one of the key reasons for European governmentssigning the treaty itself (McNamara, 1998) And it may have mattered evenmore since all Mosley’s practitioner interviews were carried out in Europe(London and Frankfurt) at the end of the period (1997) It is quite likelythat the intellectual “framing” generated by the common beliefs about thedesirable features of advanced economies expressed along a few stability-oriented rules of thumb also influenced investors and, thus, the way thatmarket participants assessed sovereigns in the period under consideration.This reinforces the impression that Mosley’s findings may be time- and/orstate-dependent

Indeed, an observation of market behaviour over a longer period of timeshows that the distinction between emerging markets and developed democ-racies is not always as straightforward as it seemed in the last part ofthe twentieth century and the first few years of the twenty-first Accord-ingly, the financial markets’ assessment of what makes default “salient”cannot be reduced to a predefined categorisation into “developed democ-racies” and “emerging markets”, but may depend on a more complex set

of changing factors Default risk can become salient also in countries mally classified as “developed democracies” and considered as “risk-free”for long periods of time The historical evidence documented by Reinhartand Rogoff (2009) provides a strong case for an open-minded approach towhere default risk may next become salient Indeed, a number of govern-ments considered excellent credits for several decades had experienced creditevents (including default, debt restructuring, cuts or delays in payments)earlier, while only a small set of countries had never defaulted The Euro-area sovereign debt crisis was a stark reminder of how things can changeover time: countries previously seen as totally or almost risk-free, such asGreece, Ireland, Portugal, Spain and Italy, suffered from sharp increases intheir risk premia, with three of them needing external bail-outs.1 Clearly,

nor-just as well-behaved countries can over time “graduate” from being seen asbad credit risks (Reinhart and Rogoff, 2009), they can also lose their status

if circumstances change, with recent “graduates” or borderline economiesbeing particularly at risk of sudden swings

These observations raise the broader issue that sovereign default riskshould not be treated as a binary variable, but as a continuous one His-tory shows that it is present even for countries normally considered highlycreditworthy, such as the US or the UK As in other markets (for exam-ple housing markets), the perceived tail risk of a catastrophic deterioration

in asset quality tends to fall during long periods of stability, then riserapidly and unexpectedly Thus, the context and path dependence of mar-ket perceptions cannot be ignored In this perspective, “risk-free” has alwaysbeen a misleading concept, since it was founded on the mistaken assump-tion that tail risk had permanently approached zero in advanced economies.Overall, both recent and longer-term experience argues against a distinc-tion between developed democracies and emerging markets as a permanentdeterminant of where saliency of default lies and of investor valuation mod-els, requiring a better understanding of how investors’ attitude may changeover time

2.2.2 Adaptive markets and valuation models

Our framework proposes a dynamic model of investor behaviour, wherebymarkets update their pricing strategies over time In particular, investors

“adapt” the set of variables under scrutiny to the extent that default risk

is perceived as more or less salient, rather than relying on the static tion between “developed” and “emerging” markets Our dynamic approach

distinc-to invesdistinc-tor behaviour finds support in recent advances in the finance ture, particularly in the Adaptive Market Hypothesis (AMH) First presented

litera-by Andrew Lo of MIT in a 2004 issue of the Journal of Portfolio

Manage-ment, the AMH aims at reconciling efficient markets with the findings of t

behavioural finance It argues that market behaviour is “consistent with anevolutionary model of individuals adapting to a changing environment viasimple heuristics” (Lo, 2004, p 1) Our treatment of investors as dynamic,adaptive actors differs from the approach of the existing empirical studies,which are crucially founded on the assumptions of efficient markets andrational investors The AMH takes its inspiration from biology, particularlyevolutionary biology, moving away from the long-standing “physics envy”

of rational economics It finds additional support in cognitive neuroscienceresearch Specifically, this new paradigm highlights how environmentalforces – particularly competition and natural selection – determine the evo-lution of markets and institutions, determining efficient market outcomesand the eventual death or survival of investment products and businesses

as well as institutional and individual fortunes (Lo, 2004) Lo refers tothis phenomenon as “survival of the richest”, as opposed to “survival of

the fittest” in evolutionary biology (2004, p 20) In the fight for survival,innovation becomes crucial The AMH evolutionary perspective “impliesthat behaviour is not necessarily intrinsic and exogenous, but evolves bynatural selection and depends on the particular environment though whichselection occurs” (Lo, 2005, p 16) In this framework, efficiency is even-tually achieved as a result of the evolutionary process: “Prices reflect asmuch information as dictated by the combination of environmental con-ditions and the ecology”, making market efficiency “context-dependent”

and “dynamic” (Lo, 2004, p 18).2

Starting from Simon’s “bounded rationality” framework (Simon, 1955), Lodescribes the process of adaptation as follows:

Individuals develop heuristics to solve various economic challenges, and

as long as those challenges remain stable, the heuristics will eventuallyadapt to yield approximately optimal solutions to them If on the otherhand, the environment changes, then, it should come as no surprisethat the heuristics of the old environment are not necessarily suited tothe new

(Lo, 2004, p 17)

As a consequence, “markets are not always efficient, but they are usuallyhighly competitive and adaptive, varying in their degree of efficiency as theenvironment and investor population change over time” (Lo, 2011, p 1)

Lo also highlights how the adaptive nature of markets and particularly of

“human risk preferences” can consequently contribute to the formation ofasset bubbles and subsequent market crashes, since risk perceptions may dif-fer for a while from the actual reality of risk and a “flight to safety” mayprevail for a period What in normal times appears to reflect the “wisdom ofcrowds” turns into something more akin to the “madness of mobs” whenexcessive fear or greed prevails (Lo, 2011, p 11).3 Applying the AMH toour specific interest in portfolio managers’ choice of valuation models, itemerges that dramatic changes, or shocks, in the investment climate maycause changes in the way institutional investors assess investment alterna-tives As in the case of investment strategies, valuation models may comeinto and go out of fashion depending on the surrounding environment, andnew models may be developed and adopted over time For example, the needfor innovation in order to ensure survival (which in the asset managementindustry equates to beating the competition in terms of asset returns) maylead one or more players to explore new ways of looking at investment alter-natives, thus triggering price movements likely to induce other players to

do the same Indeed, the nature of investors’ interest in fundamental ables may vary depending on the specific episode under consideration Forexample, as a consequence of a banking crisis, the size of the banking sec-tor relative to GDP or the results of banking sector stress tests may suddenly