14#July#2017# # # # # # Academic#year#2016/2017# Prof.#Dr.#R.M.W.J.#Beetsma# # # # Second#reader:#dr.#W.E.#Romp# Master#Thesis## # # # # Semester#2,#block#5## Msc#Economics## # # # # Monetary#Policy#&#Banking# ! ! ! How!Draghi!calmed!the!markets:!The!impact!of!news!on!periphery! sovereign!yields!during!the!Euro!crisis! ! ! Minke#van#der#Heijden# 11377844# # ! ! ! ! ! Abstract!! We#investigate#how#news#has#affected#the#daily#absolute#change#of#the#sovereign# yield#in#the#GIIPS#countries#during#the#recent#Euro#crisis.#News#is#collected#from# EuroIntelligence.# We# find# that# news# significantly# increases# the# daily# change# of# the# sovereign# yield# of# the# GIIPS# countries# between# July# 2009# and# July# 2016.# News#about#the#country#itself#raises#the#change#in#the#sovereign#yield,#but#there# are#also#spillover#effects#of#news#between#these#countries.#One#news#event#about# one# of# the# other# GIIPS# countries# raises# the# change# in# the# sovereign# yield# by# roughly#0.5#basis#points.#Furthermore,#news#also#has#a#significant#impact#if#it#is# weighed#by#bilateral#exports#or#the#aggregate#financial#claims#a#country#has#on# another.# Originality# of# this# paper# is# that# we# show# that# after# Mario# Draghi’s# “whatever#it#takes”#speech#the#spillover#effects#of#news#have#disappeared#and#the# impact#of#news#about#the#country#itself#has#decreased.#Thus#we#have#empirically# shown#that#this#speech#by#Mario#Draghi#has#calmed#the#markets#by#containing# the#spillover#effects#of#news.## # # # # # # # # # # *#I#am#grateful#to#Roel#Beetsma#for#all#his#helpful#suggestions.#Also#I#would#like#to# thank#Niels#Gilbert#and#Jorien#Freriks#for#the#inspiring#conversations#on#the# topic.#
Statement!of!Originality!! ! This#document#is#written#by#Student,#Minke#van#der#Heijden,#who#declares#to# take#full#responsibility#for#the#contents#of#this#document.# I#declare#that#the#text#and#the#work#presented#in#this#document#is#original#and# that#no#sources#other#than#those#mentioned#in#the#text#and#its#references#have# been#used#in#creating#it.# The#Faculty#of#Economics#and#Business#is#responsible#solely#for#the#supervision# of#completion#of#the#work,#not#for#the#contents.
# # 3# Table!of!content! 1.!Introduction!...!4# 2.!Literature!review!...!7# 3.!Data!...!10# 4.!Methodology!...!18# 5.!Empirical!results!...!20# 5.1$Individual$GIIPS$countries$...$20# 5.2$Panel$GIIPS$countries$...$27# 5.3$The$“whatever$it$takes”$split$...$29# 6.!Conclusion!...!33# References!...!35# Appendix!...!37#
# # 4# 1.!Introduction!! # After#the#new#Greek#government#announced#that#the#government#deficit# would#be#much#higher#than#expected#in#the#fall#of#2009,#yields#rose#in#many#of# the#Eurozone#countries.#In#the#following#years#Greece,#but#also#Cyprus,#Ireland,# Italy,# Spain# and# Portugal# had# issues# with# their# government# budget# and/or# banking# sector.# The# problems# of# these# countries# resulted# in# even# higher# yields# and# crisis# management# by# the# ECB,# IMF# and# Eurozone# member# governments.# During# the# first# ten# years# of# the# monetary# union# yields# of# Eurozone# countries# were# low# and# remained# close# to# each# other.# However,# after# 2009# a# divide# emerged# between# troubled# countries# and# safe# countries,# such# as# Germany,# Finland#and#The#Netherlands.#The#safe#countries#can#now#borrow#for#record#low# interest# rates# while# yields# for# mainly# southern# European# countries# are# higher# than# before.# The# Eurozone# faced# its# largest# challenge# since# its# creation.# De# Grauwe# (2012)# argues# that# the# countries# in# the# Eurozone# were# prone# to# speculative# attacks,# since# they# do# not# have# control# over# the# currency# that# they# issue#their#debt#in.#Countries#outside#of#a#monetary#union#have#a#lender#of#last# resort,#which#can#prevent#a#speculative#attack.#However,#a#country#in#a#monetary# union# such# as# the# Eurozone# without# a# lender# of# last# resort# is# vulnerable.# For# these# countries# the# situation# is# similar# to# those# in# developing# countries# that# borrow# in# a# foreign# currency.# De# Grauwe# (2012)# shows# that# good# and# bad# equilibria#can#arise#from#the#selfcfulfilling#nature#of#market#expectations.#Where# the#relative#safe#countries#such#as#Germany,#Finland#and#The#Netherlands#are#in# a#good#equilibrium,#the#GIIPS#ended#up#a#bad#equilibrium.1##
In# this# paper# we# look# at# the# effects# of# news# on# the# absolute# change# of# sovereign# yields# and# the# spillover# effects# of# news# in# Eurozone# countries,# particularly# in# the# GIIPS# countries.# Daily# absolute# changes# in# the# 10cyear# benchmark#yields#of#the#GIIPS#are#analyzed#from#July#2009#until#July#2016.#We# look# at# absolute# changes# in# the# sovereign# yield# since# news# can# both# have# a# positive# and# a# negative# effect# on# the# change# in# the# sovereign# yield.# We# do# not# make#a#difference#between#good#and#bad#news#and#therefore#look#at#the#absolute# ########################################################
#
# 5#
change#of#the#yield#to#capture#volatility#in#the#sovereign#yield#market.#Increased# market# volatility# is# undesirable# as# it# can# cause# uncertainty# and# a# decline# in# confidence#in#the#markets,#which#might#be#bad#for#investments.#We#use#the#news# database# created# by# Bahaj# (2014)# and# have# extended# this# database# until# July# 2016.# This# database# is# constructed# on# the# basis# of# EuroIntelligence.# Each# morning# EuroIntelligence# posts# a# newsflash# containing# the# most# important# economic# and# political# news# in# the# Eurozone# from# the# previous# day.# Although# this#is#not#the#only#news#source#for#investors#we#do#think#it#provides#a#consistent# overview# of# important# news.# Furthermore,# the# collected# news# is# weighed# by# bilateral# trade# relations# and# the# aggregate# financial# claims# a# country# has# on# another.#Finally,#we#look#at#whether#news#has#had#a#different#impact#after#Mario# Draghi’s#“whatever#it#takes”#speech.#On#July#26th#2012#Mario#Draghi#said:#“Within# our#mandate,#the#ECB#is#ready#to#do#whatever#it#takes#to#preserve#the#euro.#And# believe#me,#it#will#be#enough”.#By#this#statement#he#intended#to#calm#the#markets.# Even#though#the#Eurozone#officially#does#not#have#a#lender#of#last#resort,#Mario# Draghi#signaled#that#the#ECB#would#go#as#far#as#possible#within#their#mandate#to# save# the# Euro.# In# the# media# it# has# been# claimed# that# this# speech# indeed# has# calmed# the# markets.# This# paper# investigates# this# claim# empirically# by# testing# whether# Draghi’s# speech# has# reduced# the# impact# of# news# on# the# change# in# sovereign#yields#and#the#spillover#effects#of#news#between#countries.#
We# find# that# news# has# had# a# significant# impact# on# the# daily# change# of# sovereign#yields.#GIIPS#countries#respond#most#to#news#about#their#own#country.# However,# there# are# also# significant# spillover# effects# between# countries.# Especially#news#about#Greece#influences#the#daily#change#in#the#sovereign#yield# of# the# other# GIIPS# countries,# with# the# exception# of# Spain.# Moreover,# news# also# has# a# significant# impact# on# the# daily# change# of# sovereign# yields# when# it# is# weighed#by#bilateral#trade#relations#and#the#aggregate#financial#claims#a#country# has# on# another.# Most# interestingly,# Mario# Draghi’s# “whatever# it# takes”# speech# signaled# a# change.# After# this# speech# news# from# a# country# itself# became# less# important# and# news# from# other# countries# lost# all# of# its# significance.# Therefore,# with# this# speech# Draghi# controlled# the# spillover# effects# between# the# weak# Eurozone#countries.###
#
# 6#
This#paper#contributes#to#the#existing#literature#by#using#a#selective#news# database#over#an#extended#period#of#time.#Therefore,#capturing#more#time#before# and#after#the#“whatever#it#takes”#speech.#Furthermore,#news#is#selected#based#on# predetermined# selection# criteria.# Many# authors# look# at# the# largest# changes# in# yields# and# attribute# news# to# these# changes# (Mink# &# De# Haan,# 2013;# Osaka,# Fuertes# and# Kalotychou,# 2015),# or# take# the# number# of# times# a# country# is# mentioned#(Beetsma#et#al.,#2013).#Although#we#are#careful#in#selecting#news#we# do#use#a#wide#range#of#news#types,#and#do#not#focus#on#one#type#of#news#as#is# done# by# De# Santis# (2012)# and# Gödl# &# Kleinert# (2016).# Thus# capturing# a# broad# range#of#news#types#over#a#longer#period#of#time.##
The#rest#of#the#paper#is#structured#as#follows.#Section#2#gives#an#overview# of# background# literature.# Section# 3# describes# the# data,# and# section# 4# the# methodology.#In#section#5#the#results#are#presented#and#in#section#6#concludes# the#paper.##
#
# 7#
2.!Literature!review!!
#
There#is#a#vast#increase#in#literature#on#the#impact#of#news#on#sovereign# yields# in# the# Eurozone.# Many# authors# seek# to# explain# the# rising# spreads# by# different# news# indicators.# Closest# related# to# this# paper# is# the# work# done# by# Beetsma# et# al.# (2013).# They# look# at# news# from# July# 2007# until# February# 2012,# measured#by#the#number#of#times#a#country#is#mentioned#in#the#EuroIntelligence# news# briefing.# Their# main# result# is# that# the# domestic# spread# of# the# GIIPS# countries#increased#due#to#news#during#this#time#period.#Furthermore,#they#also# find# that# the# spread# of# other# Eurozone# countries# increased# due# to# news# from# other#GIIPS#countries.#This#effect#is#especially#strong#when#the#size#of#the#crossc border# bank# holdings# is# higher# (Beetsma# et# al.,# 2013).# All# these# effects# are# particularly#strong#from#the#fall#of#2009#onwards#and#for#bad#news#(Beetsma#et# al.,#2013).#Also#Brutti#&#Sauré#(2015)#show#that#financial#linkages,#expressed#as# crosscborder# bank# exposure,# matters# for# the# transmission# of# sovereign# risk# in# the# Euro# crisis.# # Other# research# related# to# this# includes# Mohl# &# Sondermann# (2013).# They# find# that# political# communication# influenced# the# spread# viscàcvis# Germany# during# the# financial# crisis# in# the# Eurozone.# Statements# about# restructuring# raised# the# spread# of# periphery# countries,# while# statements# about# the# European# Financial# Stability# Facility# (EFSF)# lowered# the# spread.# Furthermore,#Mink#and#De#Haan#(2013)#analyze#the#impact#of#news#about#Greece# on#European#bank#stock#prices#in#2010.#They#find#that#except#for#Greek#banks,# news#about#Greece#doesn’t#lead#to#abnormal#returns#while#news#about#a#bailout# does,# even# for# banks# without# any# exposure# to# Greece,# or# other# indebted# euro# countries#(Mink#&#De#Haan,#2013).#Similar#to#this#is#research#done#by#Bhanot#et# al.#(2014).#They#also#look#at#the#spillover#effects#of#Greek#news#on#the#financial# sector#in#the#other#GIIPS#countries.#They#find#that#there#are#significant#spillover# effects# from# news,# in# excess# of# what# can# be# explained# by# other# transmission# channels.###
Also# De# Santis# (2012)# finds# that# there# are# spillover# effects# from# Greece.# He# finds# that# three# factors# explain# the# rising# spreads# in# the# Eurozone# in# the# period# 2008# until# 2011:# country# specific# risk,# increased# regional# risk# and# spillover#effects#from#Greece.#The#spillover#effects#from#Greece#mainly#function#
#
# 8#
as#a#wakecup#call,#which#is#also#found#by#Mink#and#De#Haan#(2013),#and#Bhanot# et# al.# (2014).# A# crisis# in# one# country# might# cause# investors# to# reassess# the# vulnerability#of#another#country#and#therefore#spread#the#crisis.#Also,#Aizenman# et# al.# (2016)# show# that# there# are# spillover# effects# from# Eurozone# news# to# developing#countries#during#the#global#financial#crisis.#During#the#global#financial# crisis#news#from#the#Eurozone#had#a#negative#effect#on#the#returns#of#equity#and# bond#markets#in#developing#countries.##
De# Grauwe# &# Ji# (2013)# test# the# De# Grauwe# (2012)# Eurozone# fragility# hypothesis.#This#states#that#in#the#Eurozone#individual#countries#no#longer#have# control# over# the# currency# that# they# issue# their# debt# in.# This# can# create# a# selfc fulfilling#prophecy#to#drive#a#country#into#default#when#investors#lose#confidence# in# that# country.# Therefore,# it# is# possible# that# yields# are# not# determined# by# fundamentals# but# by# market# sentiment# (De# Grauwe,# 2012).# De# Grauwe# &# Ji# (2013)# find# that# spreads# were# indeed# driven# by# timecdependent# effects# that# were#unrelated#to#the#fundamentals#after#the#crisis.#Another#observation#done#by# De#Grauwe#&#Ji#(2013)#is#that#most#of#the#increase#in#spread#for#Greece#is#due#to# deteriorating#fundamentals.#However,#in#the#case#of#Spain#(and#to#a#lesser#extent# Italy# and# Belgium)# the# opposite# is# true.# There# the# increase# in# the# spread# is# mainly# due# to# the# time# dummies,# not# fundamentals.# Osaka,# Fuertes# and# Kalotychou#(2015)# also#test#for#De#Grauwe’s#Eurozone#fragility#hypothesis#and# conclude#that#the#Outright#Monetary#Transaction#(OMT)#program#from#the#ECB# has# helped# to# overcome# part# of# the# issues# related# to# the# fragility# hypothesis.# They# look# at# sovereign# CDScspreads# before# and# after# the# OMT# announcement# (from# July# 2011# until# July# 2013),# and# how# they# respond# to# macroeconomic# fundamentals# and# Spanish# news.# The# macroeconomic# fundamentals# become# more# important# after# the# OMT# announcement.# Moreover,# before# the# announcement# Spanish# news# had# a# significant# impact# on# several# countries,# especially# Italy.# However,# the# impact# of# Spanish# news# lessened# after# the# OMT# announcement.#Also,#standcalone#countries,#such#as#the#UK,#Norway,#Sweden#and# Denmark,# respond# considerably# less# to# Spanish# news# than# Eurozone# countries# (Saka,#Fuertes,#&#Kalotychou,#2015).##
Several# authors# look# at# the# effectiveness# of# the# different# ECB# programs# that#were#started#during#the#Eurozone#crisis.#“Gödl#&#Kleinert#(2016)#find#that#
#
# 9#
the# announcement# of# the# SMP# and# OMT# decreased# the# Greek# and# Spanish# sovereign# yield# significantly.# As# these# programs# have# lowered# the# sovereign# yields#for#these#countries#they#have#helped#to#improve#their#solvency,#since#they# reduce# the# interest# payment# burden.# Furthermore,# the# programs# have# also# supported#the#national#banking#system#indirectly,#which#reduces#the#chances#of# a#government#bailout#in#the#future.##
#
# 10#
3.!Data!!
#
The# most# important# variable# for# this# paper# is# the# news# variable.# To# capture#news#we#make#use#of#the#news#database#constructed#by#Bahaj#(2014).# He#analyzed#the#EuroIntelligence#news#briefing#from#July#2009#until#March#2013.# This#news#briefing#appears#every#weekday#around#9#o’clock#in#the#morning#with# the# most# important# economic# and# political# news# from# the# Eurozone# from# the# previous# day.# He# takes# news# from# Cyprus,# Greece,# Spain,# Ireland,# Italy# and# Portugal,#as#all#these#countries#went#through#one#or#more#crises#between#2009# and#2013.#The#timing#of#the#news#is#checked#by#looking#up#the#time#at#which#the# news#appeared#for#the#first#time#on#Bloomberg.#Therefore#the#EuroIntelligence# news#briefing#is#used#as#a#measure#of#which#news#is#important,#while#the#timing# of#the#news#comes#from#Bloomberg.#News#is#only#considered#news#if#it#falls#into# one# of# five# categories# created# by# Bahaj# (2014).# These# five# categories# are:# domestic#political#events,#foreign#interventions,#technical#events,#domestic#data# and#domestic#instability.#Besides#these#categories#news#must#also#be#timeable#in# the# sense# that# it# is# possible# to# isolate# at# what# time# the# event# occurred.# Furthermore,# the# event# must# be# limited# to# a# single# crisischit# country.# Although# this# is# a# limited# fraction# of# the# amount# of# news# available# to# investors,# we# do# believe# this# captures# the# most# important# moments.# We# have# extended# this# database#until#July#2016.#For#consistency#reasons#only#one#author#has#extended# the# database.# Hence,# the# sample# period# of# the# news# database# is# from# July# 1st# 2009# until# July# 1st# 2016.# However,# EuroIntelligence# holds# a# Christmas# and# summer#break#each#year.#Therefore#there#are#no#news#events#in#the#two#weeks# between#Christmas#and#New#Year#and#two#weeks#in#August#each#year,#with#the# exception# of# August# 2015.# Due# to# the# turmoil# around# Greece# at# that# time,# EuroIntelligence# continued# to# update# when# there# were# major# developments,# although# not# on# a# daily# basis.# In# total# there# are# 732# days# where# there# is# news# from#one#of#the#countries,#out#of#1701#days#in#total.#In#total#1145#news#events# are# recorded.# Each# time# there# is# news# about# one# of# the# countries# that# fits# into# one# of# the# five# categories# it# counts# as# one# news# event.# A# news# item# on# EuroIntelligence#about#a#different#topic#in#the#same#country#on#the#same#day#is# counted#as#a#separate#news#event.#Two#news#items#on#EuroIntelligence#about#the#
# # 11# same#topic#in#the#same#country#are#still#considered#to#be#one#news#event.#Table#1# records#the#number#of#days#that#there#is#news#about#one#of#the#countries.## # Table!1!! Amount#of#news#events## #
Cyprus# Greece# Spain# Ireland# Italy# Portugal# All## Days#with#no#news# 1665# 1368# 1527# 1599# 1548# 1558# 969# 1#event# 29# 261# 158# 84# 130# 125# 456# 2#events#on#same#day# 4# 51# 13# 16# 18# 16# 189# 3#events#on#same#day# 0# 17# 3# 2# 3# 2# 52# 4#events#on#same#day# 3# 2# 0# 0# 0# 0# 23# 5#events#on#same#day# 0# 1# 0# 0# 2# 0# 9# 6#events#on#same#day# 0# 1# 0# 0# 0# 0# 3# Total#news#days# 36# 333# 174# 102# 153# 143# 732# Total#news#events# 49# 433# 193# 122# 185# 163# 1145# # Most#news#events#come#from#Greece,#which#was#to#be#expected#since#it#is# the#country#most#severely#hit#by#the#crisis.#Additionally,#Greece#has#been#under# three#different#programs#since#the#spring#of#2010,#thus#creating#news#events#for# almost#the#entire#sample#period.#The#fewest#events#are#from#Cyprus#and#Ireland.# Both#these#countries#had#only#one#major#crisis,#and#are#relatively#small.#Figure#1# shows# bar# graphs# for# each# of# the# six# news# countries.# The# bars# represent# the# amount#of#news#events#in#one#month#for#each#of#the#countries.##
Figure#1a#shows#that#there#is#one#main#peak#of#news#in#the#beginning#of# 2013# for# Cyprus.# The# Cypriot# banking# sector# is# very# exposed# to# the# Greek# economy,#and#started#to#experience#problems#in#2011#due#to#the#Greek#crisis.#In# June# 2012# the# Cypriot# government# requested# an# emergency# loan# from# the# European# Union# and# the# IMF.# On# March# 25th# 2013# the# European# Commission,# IMF,#Eurogroup#and#ECB#agreed#on#a#10#billion#euro#bailout#for#Cyprus.#In#return# the# secondclargest# national# bank# had# to# be# closed,# where# only# accounts# below# 100,000#euro#were#guaranteed.##
In#Greece#we#can#identify#three#peaks#of#news.#All#three#correspond#to#the# three# bailout# programs# for# Greece.# The# turmoil# in# Greece# started# when# a# new# government# took# office# in# October# 2009# and# discovered# the# deficit# had# been# much#higher#than#stated#before.#In#the#spring#of#2010#Greece#requested#a#loan#
#
# 12#
from#the#EU#and#IMF#to#be#able#to#roll#over#its#debt.#In#response#to#this#the#EFSF# was#created.#In#2011#it#became#clear#that#the#first#loan#was#not#enough.#In#July# 2011# a# second# loan# was# agreed# upon.# In# return# the# Greek# government# had# to# implement#several#reforms.#It#took#the#Greek#parliament#until#February#2012#to# reach#an#agreement#on#the#reforms.#From#July#2011#until#February#2012#figure# 1b#also#shows#a#large#number#of#news#events,#due#to#the#uncertainty#regarding#if# the#second#bailout#program#would#be#implemented.#In#January#2015#there#were# early#elections#in#Greece,#which#were#won#by#Syriza#who#promised#to#undo#many# of# the# budget# cuts# that# were# implemented# under# the# second# bailout# program.# The#second#bailout#program#ended#in#June#2015#without#agreement#about#a#new# program,#which#was#necessary#to#fulfill#the#Greek#payment#requirements.# Alexis# Tsipras#who#was#the#new#Greek#Prime#Minister,#announced#a#referendum#about# the# proposed# new# bailout# requirements,# to# take# place# on# July# 5th# 2015.# The# majority# of# the# Greeks# voted# against# these# bailout# requirements# and# therefore# the# negations# between# the# Greek# government# and# the# European# countries# needed#to#be#continued#to#reach#an#agreement.#In#August#2015#a#final#agreement# on# a# third# bailout# package# was# agreed,# despite# the# negative# referendum# outcome.#
Ireland# was# the# second# European# country# to# experience# financial# problems.# The# Irish# government# had# nationalized# several# banks# in# 2009# and# 2010,# due# to# the# financial# crisis.# However,# at# the# end# of# 2010# the# Irish# government# itself# had# to# apply# for# IMF# and# EFSF# support,# which# is# also# where# the#largest#amount#of#news#is#concentrated.#In#total#85#billion#euro#was#lent#to# Ireland# until# the# end# of# 2013.# No# further# assistance# was# needed# in# Ireland,# although#the#country#is#still#under#surveillance#by#the#European#Commission#and# ECB,#which#report#on#the#economic#recovery#of#the#country.#
Contrary# to# Cyprus,# Greece# and# Ireland,# Italy# did# not# have# to# apply# for# European#support.#However,#the#country#experienced#much#political#turbulence# over# the# past# years.# There# were# two# specific# periods# of# heightened# market# tensions,# resulting# in# spikes# in# the# amount# of# news.# In# November# 2011# Prime# Minister# Berlusconi# resigned# after# his# government# could# not# get# the# budget# approved.#After#the#fall#of#the#Berlusconi#government#a#technocrat#government# was#formed#which#stayed#in#office#until#the#scheduled#elections#in#2013.#
! ! 13! ! Figure!1a.!Cyprus! ! ! Figure!1b.!Greece! ! Figure!1c.!Ireland! ! ! Figure!1d.!Italy! ! Figure!1e.!Portugal! ! ! Figure!1f.!Spain Figure'1.!Bar!graphs!for!each!country!showing!the!amount!of!news!events!in!one!month!
!
! 14!
The! second! concentration! of! news! is! around! these! elections! in! the! spring! of! 2013.!!
Most!news!related!to!Portugal!is!concentrated!in!the!spring!of!2011.!This! coincides!with!the!issuance!of!an!emergency!loan!by!the!Eurozone!countries!to! Portugal.!To!obtain!the!loan!Portugal!had!to!implement!several!reforms.!The!! European! Commission! and! the! IMF! evaluated! if! Portugal! fulfilled! all! requirements.!
Lastly,! two! peaks! of! news! can! be! identified! in! Spain.! The! Spanish! government!began!experiencing!problems!as!its!economy!went!into!a!downturn! and!Spanish!banks!were!experiencing!issues.!The!government!tried!to!solve!the! problems!by!itself!first.!In!2010!the!government!announced!several!budget!cuts! and! reforms! in! order! to! avoid! a! bailout.! However,! in! June! 2012! the! Spanish! government!asked!for!support!to!help!its!troubled!banks,!which!is!where!most! news! is! concentrated.! ! In! December! 2013! the! support! program! for! Spain! was! ended!and!there!was!no!prospect!that!a!new!program!would!be!needed!in!the! near!future.!!
Thus,! these! graphs! show! that! the! news! does! correspond! to! the! main! events!for!each!country!and!the!news!identification!can!be!considered!reliable.!!!
From! the! news! database! several! news! variables! are! constructed.! These! news!variables!are!dummy!variables!that!take!the!value!of!the!amount!of!news! events! on! that! day! for! each! country.! Foreign! news! is! the! summation! of! news! from!all!other!countries!except!the!home!country.!Table!2!shows!the!correlation! coefficients! between! news! from! the! different! countries! and! foreign! news.! As! most!correlations!are!relatively!small!there!is!little!worry!for!multicollinearity.! However,! there! are! a! large! number! of! days! where! there! is! no! news! simultaneously!from!country!i!and!j,$since!our!news!database!is!quite!selective.! Therefore,! the! correlation! between! the! news! from! different! countries! is! also! quite! low.! The! correlations! between! news! from! an! individual! country! and! the! summation!of!foreign!news!from!the!other!countries!are!larger,!especially!in!the! case! of! Greece.! This! is! mainly! due! to! the! fact! that! there! are! fewer! days! where! there!is!no!news!from!both!the!home!country!and!any!of!the!other!countries.!!
Furthermore,! the! 10Pyear! benchmark! government! closing! yields! for! the! GIIPS!countries!and!the!US!and!Germany!are!taken!from!Datastream.!The!dollarP
!
! 15!
euro! exchange! rate! and! ECB! main! refinancing! rate! are! also! taken! from! Datastream.!The!S&P500!index!based!VIX!is!the!daily!closing!index!taken!from! the! Chicago! Board! Options! Exchange.! There! are! some! days! missing! due! to! US! bank!holidays.!Thus!out!of!1828!weekdays!between!1!July!2009!and!1!July!2016! we!are!left!with!1701!days.!!
!
Table!2!
Correlation!of!news!variables!
! Cyprus! Spain! Greece! Ireland! Italy! Portugal! Foreign! news! Cyprus! 1! ! ! ! ! ! ! Spain! 0.00! 1! ! ! ! ! ! Greece! P0.00! 0.01! 1! ! ! ! ! Ireland! P0.01! 0.03! 0.01! 1! ! ! ! Italy! 0.05! 0.03! 0.08! 0.02! 1! ! ! Portugal! 0.01! P0.03! P0.04! 0.05! 0.01! 1! ! Foreign! news! 0.24! 0.38! 0.63! 0.35! 0.48! 0.33! 1! Note:!This!table!shows!the!correlation!coefficients!of!the!news!variables!for!each! country!pair!and!the!summation!of!foreign!news!with!each!country.! ! Table!3!shows!the!summary!statistics!for!the!change!in!sovereign!yields.! Yields!are!measured!in!percentage!points!and!thus!have!to!be!multiplied!by!100! to! obtain! basis! points.! In! our! regressions! we! take! the! absolute! value! of! this! change.!However,!as!can!be!seen!the!average!change!in!yields!is!negative!for!all! countries!except!for!Greece.!Thus!over!7!years!in!total!the!yields!went!down!for! Spain,!Ireland,!Portugal!and!Italy,!and!rose!for!Greece.!Also,!Greece!went!through! the!largest!daily!change.!! ! ! ! ! ! !
!
! 16!
Table!3!!
Summary!statistics!Δ!"#$%!
Δ!"#$%! Obs! Mean! Std.!Dev.! Min! Max!
Spain! 1701! P.00186! .07890! P0.883! 0.318!
Greece! 1701! .00019! .79787! P27.475! 7.028!
Ireland! 1701! P.00329! .09554! P1.028! .750!
Portugal! 1701! P.00236! .14542! P1.470! 1.686!
Italy! 1701! P.00183! .07719! P0.780! 0.509!
Note:! This! table! shows! the! number! of! observations! (Obs),! mean! (Mean),! standard!deviation!(Std.Dev.),!minimal!value!(Min),!and!maximal!value!(Max)!for! the!change!in!the!10Pyear!benchmark!sovereign!yield!in!percentage!points.!! ! Data!on!bilateral!trade!comes!from!the!IMF!Direction!of!Trade!Statistics.! These!are!monthly!observations!of!the!bilateral!free!on!board!(FOB)!exports!in! millions!of!US!dollars.!These!are!linearly!interpolated!to!obtain!daily!data.!These! daily! observations! are! converted! to! euros! and! divided! by! the! GDP! of! the! exporting!country.!Data!on!GDP!comes!from!Eurostat!and!is!on!a!quarterly!basis! in! euros.! This! is! also! linearly! interpolated.! This! leaves! daily! data! on! bilateral! exports!as!a!percentage!of!GDP!from!30!January!2009!until!30!November!2016! or!30!December!2016.!However!we!use!the!data!for!only!for!the!sample!period,! July! 1st! 2009! until! July! 1st! 2016.! Table! 4! shows! the! average! exports! as! a! percentage! of! GDP! from! one! country! to! another.! There! are! large! differences! in! trade!relations!between!the!countries.!For!example,!Greece!is!the!only!country! with! a! relatively! large! amount! of! exports! to! Cyprus,! while! Portugal! exports! a! large!share!of!its!GDP!to!Spain.!!
Data!for!the!aggregate!financial!claims!is!from!the!Bank!of!International! Settlements! Consolidated! Banking! Statistics! table! 4B,! which! measures! a! banks! country!risk!exposure.!The!table!reports!banks!claims!to!nonPresidents,!broken! down! by! nationality! of! borrower,! creditor,! and! four! types! of! debt:! public! debt,! private!bank!debt,!private!nonPbank!debt,!and!other!debt.!The!observations!are! end! of! quarter! in! million! US! dollars.! These! are! also! linearly! interpolated,! converted!to!euros!and!divided!by!the!GDP!of!the!creditor!country.!This!leaves! daily! data! from! 31! March! 2009! until! 30! December! 2016.! However! not! all! countries!have!data!available!for!the!entire!time!period.!For!Ireland!the!data!with!
!
! 17!
respect! to! Cyprus! only! starts! from! Q4! 2013.! In! the! case! of! Italy! the! data! with! respect!to!all!other!GIIPS!countries!only!starts!from!Q4!2010.!The!data!for!Spain! with! respect! to! Greece! is! only! available! until! Q2! 2014.! Spain! with! respect! to! Ireland!continues!until!Q2!2015.!For!all!other!cases!the!data!is!available!for!the! full! sample! period.! Table! 5! shows! the! average! aggregate! financial! claims! as! a! percentage!of!GDP!of!the!creditor!from!one!country!to!another.!There!are!large! differences! in! the! exposure! the! countries! have! to! other! countries.! Greece! for! example!only!has!0.66%!of!GDP!exposure!to!Spain,!while!the!exposure!to!Spain!is! 63%! of! GDP! for! Portugal.! Furthermore,! only! Greece! has! a! large! exposure! to! Cyprus,!the!other!countries!all!have!less!than!1%.!Although,!most!countries!have! large!claims!in!Italy!and!Spain.!! ! Table!4!! Exports!as!percentage!of!GDP!of!the!exporting!country! ! ! To! ! ! ! ! ! ! !
Cyprus! Greece! Ireland! Italy! Portugal! Spain! From! Greece! 0.40324! P! 0.01135! 0.67794! 0.03722! 0.16353! ! Ireland! 0.00959! 0.08979! P! 0.78304! 0.1203! 0.97267! ! Italy! 0.02048! 0.15737! 0.03563! P! 0.11461! 0.6521! ! Portugal! 0.01091! 0.049! 0.05136! 0.48898! P! 3.48016! ! Spain! 0.01472! 0.09094! 0.04935! 0.91858! 0.91796! P! ! Table!5!! Aggregate!financial!claims!as!percentage!of!GDP!of!the!creditor! ! ! To! ! ! ! ! ! ! !
Cyprus! Greece! Ireland! Italy! Portugal! Spain! From! Greece! 32.94654! P! 1.09833! 1.23513! 0.15282! 0.66028! ! Ireland! 0.16422! 5.83971! P! 36.41664! 5.10313! 30.01812! ! Italy! 0.46964! 0.65794! 3.29912! P! 0.93991! 9.11809! ! Portugal! 0.30356! 15.72988! 35.48265! 12.9706! P! 63.44195! ! Spain! 0.0427! 0.46805! 4.42551! 18.65539! 37.63419! P! !
! ! 18! 4.!Methodology!! ! We!use!the!following!model!for!our!baseline!regressions:! ! Δ!"#$%!" = !!+! !!!!! Δ!"#$%!,!!! + !!ℎ!"#!!"#$!"+ !! !!!!!"#!×!!"#$%&'!!"#$!"+ !!!"#$%"&' ! !!! !,!!!+ !!",! ! where! Δ!"#$%!" !is!the!absolute!change!of!the!endPofPday!benchmark!10P year!government!yield.!We!use!the!absolute!yield!change!since!news!can!increase! and! decrease! the! yield.! It! is! to! be! expected! that! good! news! will! lower! the! sovereign!yield,!while!bad!news!raises!the!yield.!We!are!interested!in!the!effect!of! news!on!the!change!of!the!sovereign!yield,!therefore!we!use!the!absolute!change! of!the!daily!sovereign!yield.!Unobserved!country!fixed!effects!are!captured!by!!!.!
We!allow!for!any!number!of!lags!of!the!change!in!absolute!sovereign!yields.!As! for! the! news! variables,! home! news! includes! the! news! specific! to! the! country! i.! This!dummy!variable!measures!the!amount!of!news!events!on!a!specific!day!that! are! strictly! about! country! i.! Foreign! news! is! the! summation! of! news! from! all! countries!excluding!the!country!i.!This!term!captures!the!spillover!effects!from! one! country! to! another.! Variable!!!"#!is! a! weighing! variable! and! is! used! to! measure!the!intensity!of!the!exports!or!financial!relationship!of!country!i!and!j.! This! variable! is! interacted! with! foreign! news.! In! the! case! of! the! unweighed! summation!of!foreign!news!!!"#!is!one!since!we!do!not!weigh!the!news.!However,!
we! extend! this! model! by! weighing! foreign! news! by! bilateral! exports! and! aggregate!financial!claims,!as!is!done!by!Beetsma!et!al.!(2013).!In!these!cases!the! term!!!"#!captures! the! bilateral! exports! and! aggregate! financial! claims.! Thus,!
news!from!country!j!is!weighed!by!the!export!or!financial!relationship!of!country! i$and! j.! We! expect! that! more! exports! and! a! larger! exposure! to! banks! from! the! periphery!will!make!news!more!important!for!the!change!in!the!sovereign!yield,! thus!that!there!will!be!larger!spillover!effects!between!countries.!
The! variable!!!"#$!%&!,!!!!includes! a! set! of! control! variables! that! are! expected! to! drive! changes! in! the! sovereign! yield.! The! control! variables! include!
!
! 19!
the!change!in!the!VIX,!the!change!in!the!dollarPeuro!exchange!rate,!the!change!in! the! US! benchmark! 10Pyear! government! yield,! the! change! in! the! German! benchmark! 10Pyear! government! yield,! and! the! change! in! the! ECB! main! refinancing!rate.!Where!we!also!allow!for!any!number!of!lags.!The!VIX!is!an!index! based! on! the! market! expectation! of! near! term! volatility! by! stock! index! option! prices!of!the!S&P500!and!serves!as!a!global!risk!indicator,!as!in!De!Santis!(2012).! The!dollarPeuro!exchange!rate!serves!as!an!indicator!of!regional!risk.!The!more! risk!in!the!Eurozone!the!less!attractive!euro!assets!are.!The!German!and!US!10P year! government! yields! control! for! general! market! conditions.! The! US! 10Pyear! government!yield!controls!for!global!market!condition!and!the!German!10Pyear! government!yield!for!the!Eurozone!market!conditions.!The!ECB!main!refinancing! rate! reflects! the! ECBs! perspective! of! the! economic! outlook! for! the! Euro! area.! Lastly,!!!"!is!the!error!term.!!
This! model! is! used! to! run! regressions! for! the! GIIPS! countries.! Although! we!also!have!collected!news!for!Cyprus!we!do!not!run!separate!regressions!for! Cyprus,!as!the!number!of!news!observations!for!Cyprus!is!very!limited.!However,! we! do! use! the! news! observations! from! Cyprus! as! foreign! news! for! the! GIIPS! countries.!!
!
! 20!
5.!Empirical!results!!
!
In!this!section!we!discuss!our!results.!First!the!results!for!the!individual! GIIPS! countries! and! the! effects! of! news! with! different! weights! will! be! shown.! Then! we! will! aggregate! the! GIIPS! countries! together! in! a! fixed! effects! panel.! Lastly,! we! will! test! for! a! break! after! Draghi’s! “whatever! it! takes”! speech.! All! results!are!obtained!for!the!period!July!1st!2009!until!July!1st!2016.!!
!
5.1$Individual$GIIPS$countries$
First!we!start!with!each!of!the!GIIPS!countries!separately!and!news!split! out!for!the!different!countries.!These!results!are!reported!in!table!6.!To!control! for! hetroskedasticity! and! serial! correlation! NeweyPWest! standard! errors! are! used.!The!absolute!change!of!the!daily!10Pyear!benchmark!government!yield!is! the!dependent!variable.!The!control!variables!are!the!change!in!the!German!10P year!government!yield!Δ!"10!,!the!change!in!the!US!10Pyear!government!yield! Δ!"10!,! the! change! in! the! VIX! index! Δ!"#!!,! the! change! in! the! dollarPeuro!
exchange!rate!Δ!"!!,!and!the!change!in!the!main!refinancing!rate!from!the!ECB! Δ!"#!.! The! dollarPeuro! exchange! rate! is! defined! as! the! number! of! dollars! paid! for! one! euro.! For! all! of! the! control! variables! the! first! lag! is! also! included! to! control!for!lagged!responses.!!
Table!6!shows!that!news!from!the!country!itself!is!significant!for!all!of!the! countries.! Also,! there! are! spillover! effects! from! news! to! other! countries.! Greek! news!has!a!positive!impact!on!the!absolute!change!of!the!daily!sovereign!yield! for!Ireland,!Italy!and!Portugal.!Although!the!coefficient!estimate!for!Spain!is!not! significant,!the!point!estimate!is!positive.!Therefore!there!is!a!large!chance!that!if! Greek! news! also! influences! the! change! in! the! Spanish! yield! it! will! do! so! positively,!and!not!negatively.!Thus,!increasing!the!change!in!the!sovereign!yield.!! Furthermore,!Spanish!news!is!also!significant!for!Italy!and!Portugal,!and!Italian! news! has! an! effect! on! Portugal.! However,! the! effect! of! news! from! the! home! country!is!larger!than!from!any!foreign!country!in!all!cases.!The!lagged!absolute! change!of!the!10Pyear!government!yield!is!significant!for!all!countries!except!for! Greece.!The!change!in!the!German!10Pyear!government!yield!is!only!significant! for!Ireland,!while!the!lag!of!the!German!10Pyear!government!yield!change!is!only!
!
! 21!
significant! for! Greece.! The! lag! of! the! change! in! the! VIX! index! is! also! only! significant! for! Ireland,! and! the! lag! of! the! change! in! the! ECBs! main! refinancing! rate!is!only!significant!for!Italy.!All!other!control!variables!are!not!significantly! different!from!zero.!Thus!table!6!shows!that!each!country!responds!most!to!their! own! news,! but! there! are! significant! spillover! effects! between! the! countries.! A! single!news!event!about!Italy!on!average!moves!the!Italian!yield!by!slightly!more! than! 2! basis! points.! However,! a! news! event! from! Spain! on! average! moves! the! Italian!yield!by!1.3!basis!points,!and!a!Greek!news!event!changes!the!yield!by!0.5! basis! points! on! average.! Furthermore,! Greek! news! influences! the! Greek! yield! much!more!than!home!news!for!any!of!the!other!countries.!A!single!news!event! about! Greece! on! average! changes! the! Greek! yield! by! 25! basis! points.! For! the! other!countries!this!is!roughly!2!basis!points.!Thus,!news!about!Greece!has!the! largest!spillover!effects!on!the!other!countries!and!is!also!most!important!for!the! country!itself.!!
!
!
! 22!
Table!6!!
Regressions!for!individual!countries!with!a!split!in!news!countries! !! Dependent!variable:! Δ!"#$%!,! !
! Greece! Ireland! Italy! Portugal! Spain!
!"#$!!"!!! 0.0109! 0.00385! 0.00400! P0.00683! 0.000431! ! (0.0213)! (0.00687)! (0.00470)! (0.00824)! (0.00338)! !"#$!!"!!! 0.254**! 0.00772**! 0.00538**! 0.00782*! 0.00367! ! (0.123)! (0.00305)! (0.00232)! (0.00417)! (0.00287)! !"#$!!"!!! P0.0445**! 0.0228**! P0.000864! P0.00164! 0.00250! ! (0.0216)! (0.00914)! (0.00487)! (0.00834)! (0.00461)! !"#$!!"!!! P0.0168! 0.00120! 0.0228***! 0.0128*! 0.00619! ! (0.0349)! (0.00439)! (0.00751)! (0.00755)! (0.00429)! !"#$!!"!!! P0.00980! 0.00706! P0.00235! 0.0154*! P0.00128! ! (0.0252)! (0.00524)! (0.00344)! (0.00883)! (0.00367)! !"#$!!"!!! P0.00324! 0.00495! 0.0126***! 0.0117*! 0.0232***! !Δ!"#$%!,!!! !! 0.129!(0.0357)! (0.00426)! (0.00435)!0.378***! 0.229***! (0.00672)!0.325***! (0.00468)!0.246***! ! (0.0827)! (0.0580)! (0.0437)! (0.0456)! (0.0394)! Δ!"10!!! 0.0169! 0.197**! 0.0580! 0.0818! 0.0338! ! (0.649)! (0.0933)! (0.0631)! (0.112)! (0.0641)! Δ!"10!!!!! P0.485*! P0.110! P0.0646! P0.108! P0.0334! ! (0.287)! (0.0698)! (0.0473)! (0.0844)! (0.0527)! Δ!"10!!! P0.132! P0.112! P0.00921! P0.0583! 0.0155! ! (0.205)! (0.0678)! (0.0358)! (0.0723)! (0.0377)! Δ!"10!!!!! 0.416! 0.0694! P0.00109! 0.0878! 0.0226! ! (0.326)! (0.0641)! (0.0412)! (0.0762)! (0.0457)! Δ!"#!!! P0.0175! P0.00345! 0.00377! P0.000930! 0.00294! ! (0.0149)! (0.00259)! (0.00285)! (0.00311)! (0.00336)! Δ!"#!!!!! 0.00412! 0.00486**! P0.0000877! 0.00441! 0.000190! ! (0.0128)! (0.00240)! (0.00116)! (0.00326)! (0.00135)! Δ!"!!! P0.656! P0.187! P0.0530! 0.264! 0.0471! ! (1.365)! (0.346)! (0.232)! (0.371)! (0.250)! Δ!"!!!!! P4.218! 0.198! P0.0288! 0.453! 0.110! ! (3.468)! (0.239)! (0.157)! (0.292)! (0.171)! Δ!"#!!! P0.328! 0.300! P0.242! 0.157! P0.184! ! (0.315)! (0.225)! (0.183)! (0.212)! (0.118)! Δ!"#!!!!! 0.122! 0.150! P0.122*! 0.0222! 0.0407! ! (0.394)! (0.106)! (0.0734)! (0.111)! (0.0610)! !!!! 0.137***! 0.0289***! 0.0327***! 0.0513***! 0.0349***! ! (0.0186)! (0.00304)! (0.00216)! (0.00377)! (0.00235)! N! 1637! 1637! 1637! 1637! 1637! Note:!NeweyPWest!standard!errors!
!
! 23!
Next,! we! use! home! news! as! the! news! from! the! country! itself! and! aggregate!the!news!from!other!countries!into!foreign!news.!These!are!shown!in! table!7.!!
!
Table!7!
Regressions! for! individual! countries! with! unweighted! summation! of! foreign! news!
!! Dependent!variable:! Δ!"#$%!" !
! Greece! Ireland! Italy! Portugal! Spain!
!"#$!!"#$!,!!! 0.254**! 0.0228**! 0.0229***! 0.0147*! 0.0234***! ! (0.122)! (0.00910)! (0.00751)! (0.00886)! (0.00466)! !"#$%&'!!"#$!,!!! P0.0145! 0.00554***! 0.00452**! 0.00741***! 0.00317*! ! (0.0205)! (0.00184)! (0.00180)! (0.00280)! (0.00184)! Δ!"#$%!,!!! !! 0.130! 0.378***! 0.234***! 0.326***! 0.248***! ! (0.0826)! (0.0579)! (0.0441)! (0.0457)! (0.0390)! Δ!"10!!! 0.00188! 0.196**! 0.0577! 0.0810! 0.0329! ! (0.649)! (0.0930)! (0.0632)! (0.112)! (0.0637)! Δ!"10!!!!! P0.485*! P0.109! P0.0586! P0.106! P0.0323! ! (0.291)! (0.0699)! (0.0472)! (0.0846)! (0.0527)! Δ!"10!!! P0.132! P0.110! P0.0133! P0.0601! 0.0146! ! (0.207)! (0.0674)! (0.0359)! (0.0720)! (0.0376)! Δ!"10!!!!! 0.417! 0.0709! P0.00495! 0.0869! 0.0212! ! (0.332)! (0.0641)! (0.0415)! (0.0761)! (0.0456)! Δ!"#!!! P0.0174! P0.00342! 0.00376! P0.000998! 0.00289! ! (0.0148)! (0.00258)! (0.00285)! (0.00310)! (0.00336)! Δ!"#!!!!! 0.00415! 0.00492**! P0.0000983! 0.00440! 0.000168! ! (0.0127)! (0.00239)! (0.00116)! (0.00326)! (0.00135)! Δ!"!!! P0.633! P0.183! P0.0563! 0.274! 0.0393! ! (1.372)! (0.346)! (0.231)! (0.370)! (0.249)! Δ!"!!!!! P4.179! 0.184! P0.0225! 0.453! 0.109! ! (3.470)! (0.238)! (0.157)! (0.294)! (0.170)! Δ!"#!!! P0.319! 0.310! P0.242! 0.142! P0.194! ! (0.327)! (0.222)! (0.183)! (0.219)! (0.122)! Δ!"#!!!!! 0.135! 0.145! P0.118! 0.0293! 0.0430! ! (0.382)! (0.109)! (0.0744)! (0.111)! (0.0592)! !!!! 0.137***! 0.0290***! 0.0325***! 0.0513***! 0.0347***! ! (0.0185)! (0.00301)! (0.00216)! (0.00377)! (0.00234)! N! 1637! 1637! 1637! 1637! 1637! Note:!NeweyPWest!standard!errors! ! Also!here!can!be!seen!that!home!news!has!a!significant!effect!for!all!of!the! countries.!As!in!table!6,!home!news!has!a!much!larger!effect!in!Greece!than!in!any!
!
! 24!
of!the!other!countries.!Moreover,!foreign!news,!given!by!the!summation!of!news! from!all!countries!except!home,!is!significant!for!all!countries!with!the!exception! of!Greece.!The!change!in!the!German!10Pyear!government!yield!and!the!lag!of!the! change! in! the! VIX! index! is! again! only! significant! for! Ireland! and! the! lag! of! the! German!10Pyear!government!yield!is!only!significant!for!Greece.!All!other!control! variables!are!not!significantly!different!from!zero.!From!table!7!we!can!conclude! that!in!the!case!of!Greece!the!effect!of!news!is!limited!to!news!from!Greece!itself.! However,! for! the! other! countries! the! change! in! sovereign! yield! is! also! determined! by! the! amount! of! news! from! the! other! countries.! For! all! countries! except!for!Greece!one!more!news!event!from!the!country!itself!on!average!raises! the! change! in! the! sovereign! yield! by! 2! basis! points.! However,! one! more! news! event!from!one!of!the!other!countries!raises!the!change!in!the!sovereign!yield!by! roughly!0.5!basis!point.!We!also!tested!if!there!was!a!significant!effect!from!the! lag! of! home! news! and! foreign! news.! We! however! do! not! find! that! these! are! significant.!!
Furthermore,!we!weigh!the!foreign!news!by!the!bilateral!exports!and!the! aggregate! financial! claims! a! country! has! on! another.! The! stronger! the! trade! or! financial! relationship! between! two! countries! could! mean! that! news! becomes! more!important.!Both!the!bilateral!exports!and!the!aggregate!financial!claims!are! as! percentages! of! GDP,! see! the! description! in! section! 3.! These! weights! are! interacted!with!the!news!dummies!for!each!country.!This!gives,! ! !"#$%!!"#$ℎ!"#!!!"#$%&!!"#$!,!!! =! !"#$%&'!!"#$!"×!!"#$%&'$#!!"#$%&'!!"!!!!ℎ!"#!!"!!"#!"# !!! ! and,! !"#$#%&!!"#$ℎ!"#!!"#$%&'!!"#$!,!!! =! !"#$%&'!!"#$!"×!!""#$"!%$!!"#$#%"$&!!"#$%&!!"!!!!ℎ!"#!!"!!"#!"# !!! ! ! !! The!results!for!tradePweighted!foreign!news!and!finance!weighted!foreign! news!are!reported!in!tables!8!and!9.!! !
! ! 25! Table!8! Regressions!for!individual!countries!with!news!weighted!by!bilateral!trade! !! Dependent!variable:! Δ!"#$%!" ! !
Greece! Ireland! Italy! Portugal! Spain! !!"#!!"#$!,!!! 0.254**! 0.0227**! 0.0231***! 0.0149*! 0.0236***! ! (0.123)! (0.00902)! (0.00752)! (0.00881)! (0.00467)! !"#$%!!"#$ℎ!"# !"#$%&'!!"#$!,!!! P0.0268! 0.00806**! 0.0188***! 0.00430**! 0.00344! !Δ!"#$%!"!! !! (0.0427)! (0.00334)!0.129! 0.378***! (0.00579)!0.233***! (0.00194)! (0.00295)!0.328***! 0.248***! ! (0.0828)! (0.0581)! (0.0439)! (0.0458)! (0.0390)! Δ!"10!!! P0.0133! 0.192**! 0.0558! 0.0784! 0.0329! ! (0.661)! (0.0934)! (0.0631)! (0.112)! (0.0636)! Δ!"10!!!!! P0.484*! P0.109! P0.0612! P0.109! P0.0314! ! (0.291)! (0.0698)! (0.0472)! (0.0846)! (0.0527)! Δ!"10!!! P0.128! P0.106! P0.00997! P0.0573! 0.0149! ! (0.207)! (0.0677)! (0.0358)! (0.0722)! (0.0376)! Δ!"10!!!!! 0.416! 0.0749! P0.00206! 0.0911! 0.0220! ! (0.331)! (0.0640)! (0.0412)! (0.0760)! (0.0456)! Δ!"#!!! P0.0175! P0.00339! 0.00378! P0.00104! 0.00290! ! (0.0149)! (0.00258)! (0.00285)! (0.00309)! (0.00336)! Δ!"#!!!!! 0.00416! 0.00502**! P0.0000796! 0.00443! 0.000193! ! (0.0127)! (0.00237)! (0.00116)! (0.00325)! (0.00134)! Δ!"!!! P0.642! P0.182! P0.0608! 0.251! 0.0316! ! (1.366)! (0.346)! (0.231)! (0.371)! (0.249)! Δ!"!!!!! P4.152! 0.161! P0.0369! 0.409! 0.0916! ! (3.438)! (0.239)! (0.158)! (0.294)! (0.170)! Δ!"#!!! P0.335! 0.314! P0.248! 0.123! P0.194! ! (0.327)! (0.223)! (0.183)! (0.207)! (0.124)! Δ!"#!!!!! 0.140! 0.135! P0.126*! 0.0209! 0.0385! ! (0.388)! (0.109)! (0.0754)! (0.121)! (0.0582)! !!!! 0.134***! 0.0304***! 0.0327***! 0.0534***! 0.0357***! ! (0.0189)! (0.00294)! (0.00226)! (0.00381)! (0.00222)! N! 1637! 1637! 1637! 1637! 1637! !Note:!NeweyPWest!standard!error! ! When!foreign!news!is!weighted!by!the!bilateral!exports!the!results!do!not! differ!much!from!the!unweighted!foreign!news.!However,!trade!weighted!foreign! news!is!only!significant!for!Ireland,!Italy!and!Portugal,!and!no!longer!for!Spain.! The! mean! of! trade! weighted! foreign! news! for! Italy! is! 0.130! and! its! standard! deviation! 0.259.! Hence,! an! increase! of! trade! weighted! foreign! news! by! one!
! ! 26! standard!deviation!raises!the!change!in!the!Italian!sovereign!yield!by!almost!0.52! basis!points.!Roughly!the!same!is!found!for!Ireland!and!Portugal3.!! ! Table!9! Regressions!for!individual!countries!with!news!weighted!by!aggregate!financial! claims! !! Dependent!variable:! Δ!"#$%!" !
! Greece! Ireland! Italy! Portugal! Spain!
!"#$!!"#$!"!! 0.253**! 0.00563! 0.0255***! 0.0144! 0.0241***! ! (0.122)! (0.00589)! (0.00791)! (0.00890)! (0.00506)! !"#$#%&!!"#$ℎ!"# !"#$%&'!!"#$!,!!!!! 0.000159! P0.000551! 0.00110*! 0.000215***! 0.00000664! !Δ!"#$%!"!! !! (0.000579)! (0.000708)! (0.000563)! (0.0000750)! (0.0000887)!0.129! 0.0785! 0.229***! 0.326***! 0.264***! ! (0.0828)! (0.0511)! (0.0492)! (0.0456)! (0.0427)! Δ!"10!!! P0.0112! 0.0652! 0.0204! 0.0710! 0.00578! ! (0.661)! (0.0749)! (0.0736)! (0.112)! (0.0828)! Δ!"10!!!!! P0.479*! 0.0820*! P0.0367! P0.107! P0.0180! ! (0.291)! (0.0476)! (0.0577)! (0.0844)! (0.0645)! Δ!"10!!! P0.124! P0.0304! 0.0208! P0.0549! 0.0243! ! (0.208)! (0.0415)! (0.0462)! (0.0721)! (0.0463)! Δ!"10!!!!! 0.417! 0.00600! P0.0229! 0.0901! 0.0478! ! (0.332)! (0.0386)! (0.0555)! (0.0761)! (0.0560)! Δ!"#!!! P0.0175! P0.000265! 0.00585*! P0.000968! 0.00353! ! (0.0148)! (0.00105)! (0.00327)! (0.00310)! (0.00489)! Δ!"#!!!!! 0.00426! 0.000305! P0.000805! 0.00440! 0.000745! ! (0.0127)! (0.00110)! (0.00127)! (0.00326)! (0.00203)! Δ!"!!! P0.626! 0.0239! P0.105! 0.282! 0.238! ! (1.373)! (0.276)! (0.299)! (0.369)! (0.380)! Δ!"!!!!! P4.171! P0.0940! P0.0967! 0.430! 0.0788! ! (3.457)! (0.235)! (0.209)! (0.294)! (0.210)! Δ!"#!!! P0.294! 0.164! P0.215! 0.119! P0.197! ! (0.330)! (0.115)! (0.172)! (0.215)! (0.129)! Δ!"#!!!!! 0.124! P0.139! P0.128! 0.0213! 0.0523! ! (0.372)! (0.224)! (0.0797)! (0.117)! (0.0576)! !!!! 0.131***! 0.0286***! 0.0370***! 0.0521***! 0.0375***! ! (0.0208)! (0.00219)! (0.00259)! (0.00373)! (0.00278)! N! 1637! 585! 1285! 1637! 1227! Note:!NeweyPWest!standard!errors! !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! 2!The!standard!deviation!is!0.259!and!the!estimated!coefficient!is!0.0188.!Calculation!shows! 0.259x0.0188=0.00487!in!percentage!points,!hence!an!increase!of!0.487!basis!points.! 3!For!Ireland!the!standard!deviation!is!0.521,!thus!0.521x0.00806=0.00420.!The!standard! deviation!for!Portugal!is!1.255,!thus!1.255x0.00430=0.00540!
!
! 27!
!
When!foreign!news!is!weighed!by!the!aggregate!financial!claims!it!is!only! significant! for! Italy! and! Portugal.! The! mean! for! Italy! is! 1.556! and! its! standard! deviation! 3.43.! Hence,! an! increase! of! finance! weighted! foreign! news! by! one! standard! deviation! will! increase! the! change! in! the! Italian! sovereign! yield! by! roughly!0.44!basis!points.!When!the!same!is!done!for!Portugal!it!would!increase! the!change!of!the!Portuguese!sovereign!yield!by!roughly!0.75!basis!points.!Note! however,! that! due! to! availability! of! data! on! aggregate! financial! claims! we! only! have!the!full!sample!period!for!Greece!and!Portugal.!Italy!only!starts!from!2011,! Spain!only!goes!until!mid!2014!and!Ireland!only!starts!from!2014!onwards.!The! lack!of!significance!for!Ireland!in!this!case!might!thus!also!be!due!to!the!lack!of! data.!!
As! for! the! control! variables! in! tables! 8! and! 9,! for! Ireland! the! lag! of! the! change!in!the!VIX!and!the!lag!of!the!change!in!the!sovereign!yield!are!no!longer! significant!when!foreign!news!is!weighed!by!aggregate!financial!claims.!For!Italy! the!lag!of!the!change!in!the!ECB!main!refinancing!rate!becomes!significant!when! foreign!news!is!weighed!by!bilateral!exports,!and!the!change!in!the!VIX!becomes! significant!when!foreign!news!is!weighed!by!the!aggregate!financial!claims.! ! 5.2$Panel$GIIPS$countries$ We!aggregate!the!GIIPS!countries!into!a!fixed!effects!panel.!However,!we! exclude!Greece!from!this!panel.!From!tables!7,!8,!and!9!it!can!be!seen!that!the! Greek!point!estimates!are!not!in!line!with!the!ones!from!the!other!countries.!The! other!countries!all!have!relatively!similar!point!estimates.!While!all!other!GIIPS! countries! respond! to! news! from! the! other! countries,! this! is! not! the! case! for! Greece.!Therefore,!it!can!be!concluded!that!Greece!responds!differently!to!news! than! the! other! countries,! and! thus! does! not! belong! in! the! panel! regression.! A! panel! including! Greece! can! be! found! in! the! appendix.! To! control! for! hetroskedasticity! and! serial! correlation! we! use! clustered! standard! errors.! Furthermore,!we!performed!a!Hausman!test!to!test!the!model!specification.!This!
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
4!3.43x0.00110=0.00377!
! ! 28! test!shows!that!a!fixed!effect!model!is!appropriate!for!the!data,!and!not!a!random! effect!model.!! ! Table!10! Panel!regressions!without!Greece! !! Dependent!variable:! Δ!"#$%!" ! ! (1)! (2)! (3)! (4)! !"#$!!"#$!"!! 0.0216***! 0.0212***! 0.0216***! 0.0215***! !Δ!"#$%!"!! !! (0.00225)! (0.00207)!0.316***! 0.313***! (0.00205)!0.314***! (0.00309)!0.295***! ! (0.0239)! (0.0234)! (0.0239)! (0.0264)! Δ!"10!!! 0.0911*! 0.0896*! 0.0882*! 0.0328! ! (0.0360)! (0.0362)! (0.0355)! (0.0167)! Δ!"10!!!!! P0.0734**! P0.0737**! P0.0745**! P0.0438! ! (0.0174)! (0.0180)! (0.0189)! (0.0307)! Δ!"10!!! P0.0409! P0.0401! P0.0381! P0.00820! ! (0.0281)! (0.0274)! (0.0269)! (0.0233)! Δ!"10!!!!! 0.0443! 0.0422! 0.0452! 0.0403! ! (0.0207)! (0.0208)! (0.0216)! (0.0278)! Δ!"#!!! 0.000678! 0.000678! 0.000685! 0.00213! ! (0.00170)! (0.00169)! (0.00169)! (0.00182)! Δ!"#!!!!! 0.00225! 0.00221! 0.00226! 0.00141! ! (0.00136)! (0.00136)! (0.00136)! (0.00144)! Δ!"!!! 0.0146! 0.0222! 0.0139! 0.114! ! (0.0877)! (0.0908)! (0.0864)! (0.102)! Δ!"!!!!! 0.152! 0.182! 0.154! 0.144! ! (0.0969)! (0.101)! (0.0951)! (0.146)! Δ!"#!!! 0.000951! 0.00886! 0.00506! P0.0928! ! (0.130)! (0.133)! (0.131)! (0.0995)! Δ!"#!!!!! 0.0390! 0.0412! 0.0344! P0.00921! ! (0.0543)! (0.0534)! (0.0536)! (0.0485)! !"#$%&'!!"#$!"!! ! 0.00518**! ! ! ! ! (0.000936)! ! ! !"#$%!!"#$ℎ!"# !"#$%&'!!"#$!,!!!!! ! ! 0.00516**! ! ! ! ! (0.000921)! ! !"#$#%&!!"#$ℎ!"# !"#$%&'!!"#$!,!!!!! ! ! ! 0.000198**! ! ! ! ! (0.0000420)! !!!! 0.0390***! 0.0362***! 0.0377***! 0.0399***! ! (0.00123)! (0.00160)! (0.00125)! (0.00145)! N! 6548! 6548! 6548! 4734! Note:!(i)!clustered!standard!errors,!(ii)!fixed!effect!panel!regression,!(iii)! countries!included:!Ireland,!Italy,!Portugal!and!Spain!
!
! 29!
!
Regression! (1)! in! table! 10! shows! the! panel! regression! with! only! home! news!included.!In!(2),!(3)!and!(4)!foreign!news!is!added!with!different!weights.! In!all!four!regressions!a!single!home!news!event!roughly!increases!the!change!in! the! sovereign! yield! by! 2! basis! points.! Also,! for! all! regressions! the! lag! of! the! change!in!the!yield!is!significant.!Furthermore,!the!change!in!the!German!10Pyear! yield! and! the! lag! of! the! change! in! the! German! 10Pyear! yield! are! significant! in! regressions! (1),! (2),! and! (3).! However,! together! their! effect! is! reduced! to! zero! again.! Regression! (2)! shows! that! unweighted! foreign! news! significantly! raises! the! change! in! the! sovereign! yield.! One! more! foreign! news! event! on! average! raises!the!change!in!the!sovereign!yield!by!0.5!basis!points,!consistent!with!the! results!in!table!7.!There!are!thus!spillover!effects!from!one!country!to!another.! The! trade! and! finance! weighted! foreign! news! is! also! significant! as! can! be! seen! from! regressions! (3)! and! (4).! The! mean! of! the! trade! weighted! foreign! news! is! 0.265!and!its!standard!deviation!is!0.784.!Hence,!if!trade!weighted!foreign!news! is! raised! by! one! standard! deviation! the! absolute! change! in! the! sovereign! yield! will!increase!by!0.46!basis!points.!For!finance!weighted!foreign!news!the!mean!is! 8.132! and! the! standard! deviation! 22.844.! Thus! if! the! finance! weighted! foreign! news! is! increased! by! one! standard! deviation! the! absolute! change! of! the! sovereign! yield! will! increase! by! roughly! 0.457!basis! points.! Thus! there! are! spillover! effects! of! news! between! Ireland,! Italy,! Portugal! and! Spain,! also! when! news!is!weighed!by!bilateral!trade!and!aggregate!financial!claims.!
!
5.3$The$“whatever$it$takes”$split$
Lastly,!we!investigate!if!Mario!Draghi’s!“whatever!it!takes”!speech!has!had! an!impact!on!how!news!effects!the!absolute!change!in!daily!sovereign!yields.!In! this! speech! at! the! Global! Investment! Conference! in! London! Mario! Draghi! signaled!that!the!ECB!would!do!anything!within!their!mandate!to!make!sure!the! Euro!survived.!By!this!message!he!intended!to!calm!the!sovereign!yield!markets.! To! empirically! test! this! we! interact! all! variables! with! a! dummy! variable,!!!",! !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
6!0.784x0.00516=0.00405! 7!22.844x0.000198=0.00452!
! ! 30! which!becomes!1!on!the!day!of!this!speech,!July!26th!2012.!Also!here!we!look!at!a! fixed!effects!panel!without!Greece!with!clustered!standard!errors.!! ! Table!11! Panel!regression!with!whatever!it!takes!split! !! Dependent!variable:! Δ!"#$%!" ! ! (1)! PPvalue! break!test! (1)! (2)! PPvalue! break!test! (2)! Δ!"#$%!"!! !! 0.314***! ! 0.306***! ! ! (0.0269)! ! (0.0235)! ! !"#$!!"#$!"!! 0.0295***! 0.1126! 0.0265***! 0.0577! ! (0.00258)! ! (0.00242)! ! Δ!"10!!! 0.182! 0.5784! 0.170! 0.7029! ! (0.0829)! ! (0.0808)! ! Δ!"10!!!!! P0.141**! 0.7356! P0.124**! 0.8344! ! (0.0400)! ! (0.0358)! ! Δ!"10!!! P0.0855! 0.7197! P0.0797! 0.5276! ! (0.0463)! ! (0.0439)! ! Δ!"10!!!!! 0.0606*! 0.5718! 0.0570*! 0.4624! ! (0.0239)! ! (0.0242)! ! Δ!"#!!! 0.00106! 0.1064! 0.00137! 0.0660! ! (0.00260)! ! (0.00250)! ! Δ!"#!!!!! 0.00424! 0.2013! 0.00450! 0.2380! ! (0.00218)! ! (0.00228)! ! Δ!"!!! 0.0855! 0.1265! 0.149! 0.1449! ! (0.194)! ! (0.189)! ! Δ!"!!!!! 0.238! 0.1917! 0.304! 0.2058! ! (0.186)! ! (0.199)! ! Δ!"#!!! P0.0818! 0.0252! P0.0553! 0.0311! ! (0.165)! ! (0.179)! ! Δ!"#!!!!! 0.0332! 0.3266! 0.0338! 0.4024! ! (0.0755)! ! (0.0772)! ! !!"!×!!"#$!!"#$!"!! P0.0188**! ! P0.0136**! ! ! (0.00417)! ! (0.00347)! ! !!"!×!Δ!"10!!! P0.162! ! P0.157! ! ! (0.0940)! ! (0.0932)! ! !!"!×!Δ!"10!!!!! 0.150*! ! 0.130*! ! ! (0.0588)! ! (0.0552)! ! !!"!×!Δ!"10!!! 0.0906! ! 0.0885! ! ! (0.0438)! ! (0.0415)! ! !!"!×!Δ!"10!!!!! P0.0494! ! P0.0425! ! ! (0.0286)! ! (0.0272)! ! ! ! ! ! !
! ! 31! !!"!×!Δ!"#!!! 0.000259! P0.0000620! ! (0.00270)! ! (0.00259)! ! !!"!×! Δ!"#!!!!! P0.00510! ! P0.00531! ! ! (0.00236)! ! (0.00243)! ! !!"!×!Δ!"!!! P0.189! ! P0.243! ! ! (0.196)! ! (0.190)! ! !!"!×!Δ!"!!!!! 0.117! ! 0.0470! ! ! (0.306)! ! (0.320)! ! !!"!×!Δ!"#!!! 0.232! ! 0.191! ! ! (0.149)! ! (0.163)! ! D !!"!×!Δ!"#!!!! 0.0336! ! 0.0200! ! ! (0.0983)! ! (0.0987)! ! !"#$%&'!!"#$!"!! ! ! 0.00928**! 0.4787! ! ! ! (0.00197)! ! !!"!×!!"#$%&'!!"#$!"!! ! ! P0.0106**! ! ! ! ! (0.00318)! ! !!!! 0.0392***! ! 0.0371***! ! ! (0.00132)! ! (0.00148)! ! N! 6548! ! 6548! ! Note:!(i)!clustered!standard!errors,!(ii)!fixed!effect!panel!regression,!(iii)! countries!included:!Ireland,!Italy,!Portugal!&!Spain,!(iv)!the!break!test!column! refers!to!the!pPvalue!of!the!test!if!the!variable!before!+!variable!after!the! “whatever!it!takes”!speech!is!zero.! !
As! can! be! seen! in! table! 11! both! the! reaction! to! home! news! and! foreign! news! becomes! significantly! smaller! after! the! speech.! In! regression! (1)! which! includes! only! home! news,! we! test! for!!"#$!!"#$!+ !!"×!!"#$!!"#$! = 0.! This!cannot!be!rejected!at!the!10!percent!level.!Thus!in!this!case!news!from!the! home!country!does!not!affect!the!change!in!the!sovereign!yield!anymore.!We!test! the!same!for!regression!(2),!where!foreign!news!is!also!included.!In!this!case!the! null!hypothesis!that!home!news!is!zero!can!be!rejected!at!the!10!percent!level.! However,! the! reaction! became! smaller,! it! almost! halved,! after! the! “whatever! it! takes”! speech.! Regression! (2)! shows! that! before! the! speech! one! more! foreign! news!event!raised!the!change!in!the!sovereign!yield!by!almost!1!basis!point.!This! is! almost! double! the! effect! of! what! we! found! in! table! 10! for! the! entire! sample! period.! We! do! the! same! test! as! we! did! for! home! news! for! foreign! news! in! regression! (2).! !"#$%&'!!"#$! !and! !!"×!!"#$!!"#$! !combined! is! not! significant!at!the!10!percent!level.!Thus,!foreign!news!does!not!have!an!effect!on! the!change!in!sovereign!yield!after!the!speech.!Therefore!it!can!be!concluded!that! Mario!Draghi’s!“whatever!it!takes!speech”!did!calm!the!markets.!Spillover!effects!
!
! 32!
of! news! between! the! GIIPS! countries! were! present! before! the! speech! but! disappeared! in! the! period! afterwards.! Also,! news! about! the! home! country! became!less!important!for!the!absolute!change!in!the!sovereign!yield.!!
Furthermore,! the! change! in! the! ECB! main! refinancing! rate! becomes! significant! after! the! “whatever! it! takes”! speech! in! both! regressions.! Also,! the! change! in! the! VIX! becomes! significant! in! regression! (2)! after! the! “whatever! it! takes”! speech.! Both! have! positive! point! estimates! after! the! speech! and! thus! a! larger! change! in! either! of! these! also! leads! to! a! larger! change! in! the! sovereign! yield.! World! and! Eurozone! economic! conditions! thus! have! become! more! important!after!the!“whatever!it!takes”!speech.!!
We!also!tested!for!a!break!at!other!dates.!The!“whatever!it!takes”!speech! was!an!indication!for!the!start!of!the!OMT!program.!Therefore!we!also!tested!the! official!announcement!of!the!OMT8!and!the!starting!date!of!the!OMT9.!However! these!do!not!give!very!different!results.!The!main!point!that!the!effect!of!news! has!become!smaller!or!has!disappeared!still!stands!in!these!cases.! ! !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! 8!August!2nd!2012! 9!September!6th!2012!
!
! 33!
6.!Conclusion!!
!
In! this! paper! we! have! explored! how! news! affects! the! daily! absolute! change! in! the! sovereign! yields! of! the! GIIPS! countries! from! July! 2009! until! July! 2016.!We!used!a!news!database!with!news!about!Cyprus,!Greece,!Ireland,!Italy,! Portugal! and! Spain,! where! news! was! collected! on! a! daily! basis! from! EuroIntelligence.! We! find! that! news! from! the! country! itself! increases! the! absolute!change!of!the!daily!sovereign!yield!in!all!five!GIIPS!countries.!This!effect! is! almost! ten! times! larger! in! the! case! of! Greece! than! for! the! other! countries.! Furthermore,! there! are! spillover! effects! of! news! between! the! GIIPS! countries.! News! from! the! other! GIIPS! countries! also! raises! the! absolute! change! in! the! sovereign! yield! of! the! home! country,! for! all! countries! expect! for! Greece.! One! more!domestic!news!event!raises!the!average!daily!change!of!the!sovereign!yield! by!approximately!2!basis!points!for!Italy,!Ireland,!Portugal!and!Spain.!Likewise,! one! more! foreign! news! event! raises! the! average! daily! change! of! the! sovereign! yield! by! roughly! 0.5! basis! points! for! these! four! countries.! Greece! responds! differently!to!news!than!the!other!GIIPS!countries.!Home!news!is!more!important! for! Greece! and! the! absolute! change! in! the! Greek! sovereign! yield! does! not! respond! significantly! to! news! from! the! other! countries.! Furthermore,! when! foreign! news! is! weighted! by! bilateral! exports! and! aggregate! financial! claims! it! significantly! influences! the! absolute! change! of! the! daily! sovereign! yield! for! the! GIIPS! countries! without! Greece.! Lastly,! we! find! that! news! has! had! a! different! effect!after!Mario!Draghi’s!“whatever!it!takes”!speech.!Foreign!news!lost!all!of!its! significance!after!this!speech.!Thus!by!stating!that!the!ECB!would!do!everything! within!its!mandate!to!save!the!Euro,!the!spillover!effects!of!news!are!contained.! After!this!speech!the!effect!of!home!news!is!also!reduced.!The!effect!of!one!more! news!event!about!the!home!country!affects!the!change!in!the!sovereign!yield!by! about!half!compared!to!before!the!“whatever!it!takes”!speech.!!
Popular! media! has! often! already! said! that! Mario! Draghi! calmed! the! markets! by! his! “whatever! it! takes”! speech.! However,! we! now! also! have! shown! this!empirically.!Before!there!were!significant!spillover!effects!of!news!between! the!GIIPS!countries,!making!the!sovereign!bond!market!more!volatile.!As!we!have! shown! the! effects! of! news! have! changed! after! the! speech.! The! spillover! effects!
!
! 34!
between! countries! have! completely! vanished! and! the! effect! of! home! news! has! been!significantly!reduced,!thus!making!the!sovereign!bond!market!less!volatile.! These!results!are!in!line!with!those!of!Osaka,!Fuertes!&!Kalotychou!(2015).!They! find!that!the!announcement!of!the!OMT!has!reduced!the!impact!of!Spanish!news! on!the!CDS!spreads.!We!extend!this!to!news!from!all!the!other!GIIPS!countries,! and! that! the! effect! of! foreign! news! on! the! sovereign! yield! not! only! has! been! reduced!but!has!completely!disappeared.!!
The! OMT! program,! which! followed! Mario! Draghi’s! speech,! is! a! step! forward! in! the! survival! of! the! Eurozone.! A! monetary! union! without! an! official! lender!of!last!resort!is!vulnerable!to!speculative!attacks.!By!the!creation!of!the! OMT! program! this! risk! has! been! reduced,! although! the! program! has! not! been! used! so! far.! Only! time! will! tell! if! the! OMT! program! is! enough! to! withstand! a! future!crisis.!However,!as!we!have!shown!in!this!paper!it!has!already!reduced!the! impact!of!news!on!sovereign!yields!in!the!GIIPS!countries.!Hence,!the!(implicit)! announcement! of! the! OMT! program! has! already! been! effective.! Therefore! it! might!not!be!necessary!to!implement!a!program!in!the!future!as!long!as!there!is!a! credible! announcement.! Based! on! the! credible! announcement! investors! will! change!their!behavior!before!it!is!necessary!to!take!action.!!
! ! 35! References! Aizenman,!J.,!Jinjarak,!Y.,!Lee,!M.,!&!Park,!D.,!2016.!Developing!countries!financial! vulnerability!to!the!eurozone!crisis:!an!event!study!of!equity!and!bond!markets.! Journal!of!Economic!Policy!Reform,!19(1),!1P19.! Bahaj,!S.!A.,!2014.!Systemic!sovereign!risk:!macroeconomic!implications!in!the! euro!area.!CFM!discussion!paper!series.! ! Bank!for!International!Settlements,!2017.! http://www.bis.org/statistics/consstats.htm.! ! Beetsma,!R.,!Giuliodori,!M.,!de!Jong,!F.,!&!Widijanto,!D.,!2013.!Spread!the!news:! The!impact!of!news!on!the!EUropean!sovereign!bond!markets!during!the!crisis.! Journal!of!International!Money!and!Finance!,!34,!83P101.! ! Bhanot,!K.,!Burns,!N.,!Hunter,!D.,!&!Williams,!M.,!2014.!News!spillovers!from!the! Greek!debt!crisis:!Impact!on!the!Eurozone!financial!sector.!Journal!of!Banking!&! Finance,!38,!51P63.! ! Brutti,!F.,!&!Saure,!P.,!2015.!Transmission!of!sovereign!risk!in!the!Euro!crisis.! Journal!of!International!Economics,!97(2),!231P248.! ! De!Grauwe,!P.,!2012.!The!governance!of!a!fragile!Euozone.!Australian!Economic! Review,!45(3),!255P268.! ! De!Grauwe,!P.,!&!Ji,!Y.,!2013.!SelfPFulfilling!crises!in!the!Eurozone:!An!emperical! test.!Journal!of!International!Money!and!Finance,!34,!15P36.! ! De!Santis,!R.A.,!2012.!The!Euro!Area!Sovereign!Debt!Crisis:!Safe!Haven,!Credit! Rating!Agencies!and!the!Spread!of!the!Fever!from!Greece,!Ireland!and!Portugal.! ECB!Working!Paper,!No.!1419.! !
! ! 36! Direction!of!Trade!Statistics,!2017.!IMF.!http://www.imf.org/external/data.htm.! ! EuroIntelligence,!2017.!http://www.eurointelligence.com/archive.html.! ! Eurostat,!2017.!http://ec.europa.eu/eurostat.! ! Gödl,!M.,!&!Kleinert,!J.,!2016.!Interest!rate!spreads!in!the!Eurozone:!Fundaments! or!sentiments?!Review!of!World!Economics!,!152!(3),!449P475.! ! Mink,!M.,!&!de!Haan,!J.,!2013.!Contagion!during!the!Greek!sovereign!debt!crisis.! Journal!of!International!Money!and!Finance!,!34,!102P113.! ! Mohl,!P.,!&!Sondermann,!D.,!2013.!Has!political!communication!during!the!crisis! impacted!soverign!bond!spreads!in!the!euro!area?!Applied!Economics!Letters,!20! (1),!48P61.! ! Saka,!O.,!Fuertes,!A.PM.,!&!Kalotychou,!E.,!2015.!ECB!policy!and!Eurozone!fragility:! Was!De!Grauwe!right?!Journal!of!International!Money!and!Finance!,!54,!168P185
! ! 37! Appendix! ! Table!12! Panel!with!Greece! !! Dependent!variable:! Δ!"#$%!" ! ! (1)! (2)! (3)! (4)! !"#$!!"#$!!! 0.121! 0.121! 0.121! 0.136! ! (0.0708)! (0.0710)! (0.0707)! (0.0744)! Δ!"#$%!"!! !! 0.153***! 0.153***! 0.153***! 0.149***! ! (0.0170)! (0.0169)! (0.0169)! (0.0151)! Δ!"10!!! 0.0440! 0.0433! 0.0416! 0.00241! ! (0.0427)! (0.0413)! (0.0414)! (0.0300)! Δ!"10!!!!! P0.128! P0.128! P0.129! P0.135! ! (0.0771)! (0.0768)! (0.0764)! (0.102)! Δ!"10!!! P0.0467! P0.0463! P0.0441! P0.0307! ! (0.0287)! (0.0285)! (0.0277)! (0.0367)! Δ!"10!!!!! 0.119! 0.119! 0.120! 0.143! ! (0.0805)! (0.0818)! (0.0802)! (0.0980)! Δ!"#!!! P0.00305! P0.00305! P0.00304! P0.00324! ! (0.00384)! (0.00384)! (0.00384)! (0.00495)! Δ!"#!!!!! 0.00316**! 0.00316**! 0.00318**! 0.00294*! ! (0.00103)! (0.00104)! (0.00103)! (0.00116)! Δ!"!!! P0.163! P0.161! P0.163! P0.158! ! (0.138)! (0.144)! (0.138)! (0.192)! Δ!"!!!!! P0.839! P0.830! P0.837! P1.136! ! (0.920)! (0.941)! (0.921)! (1.120)! Δ!"#!!! P0.0173! P0.0146! P0.0123! P0.0938! ! (0.104)! (0.108)! (0.105)! (0.0785)! Δ!"#!!!!! 0.0240! 0.0243! 0.0195! P0.0393! ! (0.0798)! (0.0796)! (0.0788)! (0.0619)! !"#$%&'!!"#$!!! ! 0.00176! ! ! ! ! (0.00423)! ! ! !"#$%!!"#$ℎ!"# !"#$!!! ! ! 0.00543**! ! ! ! ! (0.00120)! ! !"#$#%&!!"#$ℎ!"# !"#$!!! ! ! ! 0.000218***! ! ! ! ! (0.0000444)! !!!! 0.0633***! 0.0624***! 0.0621***! 0.0679***! ! (0.00800)! (0.00606)! (0.00776)! (0.00950)! N! 8185! 8185! 8185! 6371! Note:!(i)!clustered!standard!errors,!(ii)!fixed!effect!panel!regression,!(iii)! countries!included:!Greece,!Ireland,!Italy,!Portugal,!and!Spain! !
