INFINITI Conference 2014 Prato, June 2014 1 / 14 Convergence and - - PowerPoint PPT Presentation

Hypnosis Before Wake-up Call?!

The Revival of Sovereign Credit Risk Perception in the EMU-Crisis Ingo Bordon

German Development Institute Bonn

Kai Daniel Schmid

Macroeconomic Policy Institute Duesseldorf

Michael Schmidt

Graduate School of Economics, Finance, and Management Frankfurt

INFINITI Conference 2014

Prato, June 2014

1 / 14

Convergence and Re-widening of Public Bond Yields

(i) Long-term Interest Rate (GIIPS)

10 20 30 1980 1990 2000 2010

ESP GRC ITA IRL PRT

Data source: OECD Economic Outlook, IMF World Economic Outlook. 2 / 14

“Wake-up Call”-explosion of Spreads

What determines spectacular increase in long-term interest rates in EMU periphery countries?

important question because of severe destabilizing and self-enforcing impact in the crisis

Perspectives on this issue from the literature:

reconsideration of fundamentals (Basurto et al. 2010, Kopf 2011, Sinn

2012)

psychological component: “expectational shift”, “wake-up call”-phenomenon (Arghyrou/Kontonikas 2012, DeGrauwe 2011, IMF 2010,

von Hagen et al. 2011)

“overpricing”: disregard of fundamentals in crisis (DeGrauwe/Ji 2013)

3 / 14

“Hypnosis Before Wake-up Call”-hypothesis

What is the role of public debt for pricing sovereign risk? Has this role changed over time?

search for better understanding of re-pricing leads us to consider pricing behavior before EMU-convergence disentangle “risk bubble” from “fear bubble”

Assumptions

role of public debt changed twice as Maastricht treaty also caused shift in pricing behavior gradual (but severe) disregard of fundamentals: “hypnosis”-phenomenon pricing of public debt in crisis might be comparable to pre-convergence era

4 / 14

Data

Cross-country panel

annual data from 1980-2012 21 OECD countries: Australia (AUS), Austria (AUT), Belgium (BEL), Canada (CAN),

Switzerland (CHE), Denmark (DNK), Spain (ESP), Finland (FIN), France (FRA), Germany (GER), Great-Britain (GBR), Greece (GRC), Ireland (IRL), Italy (ITA), Japan (JPN), the Netherlands (NLD), Norway (NOR), New Zealand (NZL), Portugal (PRT), Sweden (SWE), United States (USA)

Data Sources

OECD Economic Outlook AMECO IMF World Economic Outlook

5 / 14

Regime Approach - Country- and Time-clusters

Define 9 Regimes via 3 Time Periods across 3 Country Groups

“PreEuro”-period “ConEuro”-period “Crisis”-period

1980–1993 1994–2007 2008–2012 (T = 14) (T = 14) (T = 5)

OECD-21 ex EMU-11 1 4 7

AUS, CAN, CHE, DNK, GBR, JPN, NOR, NZL, SWE, USA (N = 10)

EMU-11 ex GIIPS 2 5 8

AUT, BEL, GER, FIN, FRA, NLD (N = 6)

GIIPS 3 6 9

ESP, GRC, IRL, ITA, PRT (N = 5) 6 / 14

Regime Approach - An Illustration

(i) GIIPS, 1980-1993 (ii) GIIPS, 1994-2007

1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 19801981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993

• 30
• 20
• 10

Real Interest Rate Spread 20 40 60 80 100 120 Public Debt to GDP Ratio

ESP GRC ITA IRL PRT

1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 20052006 2007 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007

• 15
• 10
• 5

Real Interest Rate Spread 20 40 60 80 100 120 Public Debt to GDP Ratio

ESP GRC ITA IRL PRT

(iii) GIIPS, 2008-2012

2008 2009 2010 2011 2008 2009 2010 2011 2012 2008 2009 2010 2011 2012 2008 2009 2010 2011 2008 2009 2010 2011 2012

• 5

5 10 15 20 Real Interest Rate Spread 50 100 150 200 Public Debt to GDP Ratio

ESP GRC ITA IRL PRT

7 / 14

Empirical Model

Regress Yield Spreads Against Germany on Public Debt to GDP-ratio and Standard Set of Macroeconomic Fundamentals1 Spread = (Debt ⊗ R)β1 + (Debt2 ⊗ R)β2 +(X ⊗ R)β3 + TDβ4 + CDβ5 + U Average Partial Effect of Public Debt to GDP Ratio λ = ∂Spread ∂Debt = β1 + 2β2 × Debt

1Fixed-effects model as baseline (Beirne/Fratzscher 2013, DeGrauwe/Ji 2013) 8 / 14

Hypotheses

Hypotheses for GIIPS Country Cluster H1: Effect of Debt changes across time periods λ3 > λ6 & λ6 < λ9 H2: Effect of Debt in “Crisis”-period is not different from “PreEuro”-period λ3 = λ9 H3: Effect of Debt is not different from core-EMU countries in “ConEuro”-period λ5 = λ6 H4: Effect of Debt is different from core-EMU countries in “PreEuro”-period and “Crisis”-period λ2 < λ3 & λ8 < λ9 H5: Contribution of Debt to spreads falls from the “PreEuro”-period to the “ConEuro”-period and rises from the “ConEuro”-period to the “Crisis”-period

9 / 14

Baseline Estimation Results

Total Effects of Public Debt to GDP Ratio Across Regimes2

.12 .08 .04

• .04

1 2 3 4 5 6 7 8 9 2Marginsplot: 95%-CI; regression with robust standard errors, year dummies and country fixed effects 10 / 14

Selected Robustness Dimensions

Alternative Panel Estimators

exploit large T asymptotics and impose some more structure on covariance matrix (Driscoll/Kraay (1998) Standard Errors and Panel

Corrected Standard Errors)

Dynamic Model (include lagged dependent variable)

POLS on differenced model (ordinary FE) GMM estimators (Arellano/Bond (1991) and Blundell/Bond (1998))

Pairwise Variation of Regime Break Dates (1993/94, 1994/95, 1995/96

versus 2007/08, 2008/09)

Consider Additional Fiscal Information (Primary Balance to GDP-ratio)

11 / 14

Explanatory Power Analysis - Levels

Contributions to Levels of Yield Spreads (GIIPS)3

• 10

10 20 30

ESP GRC IRL ITA PRT

3 6 9 3 6 9 3 6 9 3 6 9 3 6 9

• Publ. Debt

Time Component Other Fundament. Resid. Country-FE & Const.

3Regimes - 3: “PreEuro”-period; 6: “ConEuro”-period; 9: “Crisis”-period 12 / 14

Explanatory Power Analysis - Changes

Contributions to Changes in Yield Spreads (GIIPS)4

• 10

10 20

ESP GRC IRL ITA PRT

3/6 6/9 3/6 6/9 3/6 6/9 3/6 6/9 3/6 6/9

Public Debt Time Component Other Fundamentals Residual

4Regimes Changes - 3/6: “PreEuro”-period to “ConEuro”-period; 6/9: “ConEuro”-period to “Crisis”-period 13 / 14

Conclusion and Implications

Institutional Changes of EMU Heavily Affected Financial Markets’ Weighting Schemes

• verwhelmingly strong impact of expectational shifts due to

institutional changes (EMU) and macroeconomic shocks (crisis)

Pricing Mode has Changed Twice

Maastricht treaty initiates “hypnosis”-effect “wake-up call” causes second rebalancing effect bursting “risk bubble”

Qualification of “Overpricing”-hypothesis

no clear indication of “fear bubble” as default risk assessment in “Crisis”-period comparable to “PreEuro”-period

14 / 14