Munich Personal RePEc Archive
The effects of a greater central bank credibility on interest rates level and volatility response to news in the U.K.
Tuysuz, Sukriye
BETA
1 September 2007
Online at https://mpra.ub.uni-muenchen.de/5263/
MPRA Paper No. 5263, posted 11 Oct 2007 UTC
interest rates mean and volatility response to news in
the U.K.
Sukriye Tuysuz
∗
Abstrat
ThispaperinvestigatestheimpatofBritishmaroeonomiandmonetary
newson English interest rateslevel and volatility. These news orrespond to
BankofEngland(BoE)targetvariablesnewsandtounexpetedmonetarypol-
iyrate hanges. It analyzes whether themarket rateresponse to these news
has hanged sine the Bank of England (BoE) was grantedoperationalinde-
pendene in May 1997. It also heksif thisredibility measurehas inreased
thepreditabilityofBoEdeisionsbythemarket. Theresultsrevealthatafter
May 1997, nanial markets appears better ableto antiipeBoE poliy dei-
sionsthan before May 1997. However,Bank of England target variable news
announementsandpoliy ratehangesdiusion inuene moreEnglishinter-
est rate volatility after May 1997. This results suggests that the redibility
and/or transpareny of BoE might have dereased after 1997. However, the
loserevolutionoftherealizedinationaroundthetargetxedbytheBoEand
the evolution of the transpareny and redibility index suggest that the BoE
transparenyandredibility degreeinreasesine 1997ompare to theperiod
priorto 1997. Onepossibleexplanationofthislastresultsrestsonunertainty
reatedbytheseveralnanialrises(theAsianrisis(July1997),theRussian
risis(August1998),theburstingofthetehnologyandinternetbubblein2002
inUSA).
JEL Classiation: E4;G1
keywords: Monetarypoliy,announements,news,redibility,transpareny,
termstruture ofinterestrates,GARCH,
∗
BETA-THEME, Université Louis Pasteur, Department of Eonomis, 61 avenue de la forêt
noire,67000Strasbourg,Frane,tuysuzournot.u-strasbg.fr
Aside from its negative eet on the ondution of monetary poliy by the en-
tral bank 1
, high interest rate volatility blurs the prevision of the monetary poliy
stanebynanial marketpartiipants. Inorder toprovideastableenvironment for
nanialmarketwhihfailitatesto reahitstarget,entral banksseekto reduein-
terestrates variability(Goodfriend,1990; Froyen andWaud, 1995;Goodhard,1996;
Woodford, 1999) 2
,
3. Indeed, it is easier for poliy makers to redue unertainty
that they reate themselves rather than unertainty due to other fators. To re-
due nanial instabilityand variabilitiesof theirvariables objetives, entralbanks
started to ommuniate more information about the ondut of their poliy and to
enfore their redibility (Faust and Svensson,2001; Dodge, 2002; Longworth, 2002).
Greater redibility of entral bank helps to redue nanial speulation and redue
theheterogeneity of market operators expetations about future monetary authori-
tiesdeisionsandfutureevolutionoftheobjetivevariables. Theseonsequenesofa
greaterredibilityshouldenhanethepreditabilityoftheentralbankratehanges.
Inaddition, a greaterredibilityshould redue the unertainty related to monetary
poliy and then redue the eets of news related to monetary poliy on nanial
marketvolatility.
Several authors argue that nanial market volatility reets unertainty about
monetarypoliystane. Mostoftheseresearhershavefousedontheroleofmaroe-
onomi news announements, related to monetary poliy, as a soure of nanial
market volatility and partiularly interest rates market (Fleming and Remolona,
1997; Jones et al.,1998; Lee, 2002). A large part of these authors supposeonstant
nanial marketresponse to news. However, market interest rate reation to these
news strongly depends on entral bank transpareny and redibility degrees (Hal-
daneandRead,1999,2000;Ellingsen andSöderström,2001;Gravelle andMoessner,
1
Interestratesinstabilityinuenestheeonomisituationandtheentralbanktargetvariables
andthenimportunesthemonetarypoliyondut.
2
Thejobof entral bankersis to ondutmonetary poliy inorder topromote prie stability,
sustainablegrowth,andastablenanialsystem.
3
Therehavebeenanumberofpapersdoumentingandanalysingso-alled"interestratesmooth-
ing"(Goodhart,1996 andWoodford,1999). SeeSakandWieland (2000)foraliteraturereviews.
Althoughtheprimaryfousofthatliteratureis theobservedtendenyforthesmoothing ofpoliy
rates,partofthemotivationforsuhbehaviorhasbeentoprovideastableenvironmentfornanial
markets.
transparenyor redibility shouldaet interest ratereation.
An inreasing numberof aademi papers analyzethe eetsof a greater trans-
pareny or redibility on nanial market. Most of these papers have foused on
thetheoretialandempirialaspetsofmonetarypoliytranspareny. Thesestudies
mainly onsidertheeets ofa greater transparenyon thereationof interestrate
levels to maroeonomi and monetary news relatedto themonetary poliyand on
thepreditabilityofentralbankdeisions(Haldane andRead,2000;Kuttner,2001;
Pooleet al.,2002; Lange et al.,2003; Poole andRashe,2003;Parent, 2003;Swans-
son, 2004). Conerning theeets onthevolatility, only Nolanand Chadha (2001),
Clare and Courtenay (2001a,b) and Tuysuz (2007) examine the eets of a greater
transparenyon the impatofthese news onnanial marketvolatility.
In ontrast with the previous studies, the number of aademi papers on the
eetsof a greater redibility ismore less than this on theeet ofa greater trans-
pareny. Several authors analyze the eets of redibility on the overall output.
Thereisgeneral agreementthatindependent, transparent,aountable,andredible
entralbanksareabletodeliverbetteroverallpoliyoutomes(Alesina,1988;Grilli,
Masiandaro,andTabellini,1991;Cukierman,1992;Cukierman,Webb,andNeyapti,
1992; Alesina and Summers, 1993; Jonsson, 1995; Loungani and Sheets, 1997; Ei-
jnger et al., 1998). As for the eets on nanial market, authors asAlesina and
Summers(1993)showthathigherentralbankindependenemaybeassoiatedwith
lower interestrate variability, suggesting that more redible regimes enjoylessvari-
able interest rates. And onerning the eets of a greater redibility on nanial
marketreationto news,itwasanalyseonlybyChadhaandNolan(2001)and Clare
andCourtenay(2001a,b). However, Clareand Courtenayonsidertheexhangerate
and the futures ontrats. And Chadha and Nolan (2001) do not take the main
maroeonominews relatedto Bank ofEngland.
Whilenanialmarketvolatilityreetsunertaintyaboutmonetarypoliystane
(Jones etal.,1998; Lee,2002)), theobjetive ofthis paperisto explore theeetof
a greaterentral bank redibility oninterest rate leveland volatility responseto its
fundamental related to the monetary poliy. Morepreisely,this paperinvestigates
whethertheBritish interestrates level andvolatilityresponseto English maroeo-
nomiandmonetarynewshashanged sinetheBankofEngland(BoE)wasgranted
operational independene in May 1997. In addition, it also analyses the impat of
kets predition about this Bank deisions. For the present analysis, two kinds of
daily interest rate series (3, 6 and 12 months rates and 3, 5, 7 and 10 years rate)
andseveralmaroeonominews relatedto BoEtarget variableswereused. Maroe-
onomi news inlude BoE target variables and the oial interest rate deisions
about English monetary poliy. All these data over the period ranging from the
rstofJanuary1994to 28February2003. Interestratedynamisareevaluatedwith
a GARCH (1,1) approah, proposedbyBollerslev (1986). To take into aount the
impatof thenewredibilitymeasure, interest rates dynamis areevaluated for the
sub-periods preeding and following May 1997. Suh an approah per sub-periods
was used by the majority of the authors analyzing the impat of monetary poliy
rate hanges on rates dynamis by takinginto aount new measurements of trans-
pareny and/or redibility (see for example Urih and Wahtel, 2001; Chadha and
Nolan, 2001;Clare andCourtenay,2001;Lee, 2002;Parent, 2003).
Thepaperproeedsasfollows. Setion2 presents howa newredibility measure
inuenesthe responseof theinterest ratelevelandvolatility toentral banktarget
variablesnewsandtomonetarypoliydeisions. Italsoputinevidenethatagreater
redibilityimprovetheabilityofnanialmarketstoantiipateentralbankdeision.
Setion 3 presents the data used for the analysis. In setion 4, the examination of
thedata suggeststhattheabilityofnanialmarkets toantiipateBankofEngland
deisionshangesimprovedafter1997. Setion5presentsthemodelusedtoevaluate
theresponseofinterestratelevelandvolatilitytomaroeonomiandmonetarynews
(modelGARCH).Setion 6 analyzesthe results, andnally,setion7 onludes.
2 How a greater redibility an aet interest rate re-
sponse to news?
Market operators reation to entral bank target variables news dependsmainly on
how this agents understand the eetive ondut of the monetary poliy and how
they aknowledge the apaity of the entral bank (Haldane and Read, 1999, 2000;
EllingsenandSöderström,2001;GravelleandMoessner,2001;Parent,2003;Connoly
andKohler,2004). Inanotherway,marketinterestrate leveland volatilityresponse
to entral bank target variables news and to monetary poliyrate hanges depends
redibility should aet the interestrates leveland volatilityresponseto maroeo-
nomi newsannounements andto theunexpetedpartof themonetary poliyrate
hanges. Inaddition,this measureshouldalso aetthemarketpreditabilityofthe
entral bank monetary poliy deisions. These eets of a new redibility measure
arepresentedinwhatfollows.
2.1 Impats on interest rates response to entral bank target vari-
ables news
Aording to Geraats (2000, 2002), Jensen(2001) and Clareand Courtenay (2001a,
b),anewredibilitymeasureshouldimprove theentral banktransparenydegree 4
.
More preisely, Jensen shows theoretially that a entral bank whih manages to
maintain the ination rate lose to its target an be optimally transparent. In an-
otherway,Jensen ndthatfullyredible entralbank isalsooptimally transparent.
Asfor Geraats, she argues thata greater redibility improvesentral bank politial
transpareny 5
and then the general transpareny. Indeed, politial transpareny is
enhanedbyinstitutionalarrangements, like entral bankindependeneand entral
bankontrats;beausetheyensurethatthereisnoundueinueneorpolitialpres-
sureto deviatefrom statedobjetives. Inanother world, entral bankindependene
and entral bank ontrats inrease market operators aknowledge the apaity of
4
Manytheoretialandempirialworksshowthatmonetarypoliytransparenyhasthepotential
toenhanetheredibility,reputationandexibilityofentralbanks(Saxton,1997;Geraats,2000;
Faustand Svensson,2001; Cukierman, 2001; Geraatset al.,2006). However, theinuene ofthe
redibility on the transpareny was onsidered onlyby authors, as Geraats (2000, 2002), Jensen
(2001)andClareandCourtenay(2001a,b).
5
Geraatsdistinguishveaspetsoftranspareny: politial,eonomi,proedural,poliyandop-
erationaltranspareny. Politialtransparenyreferstoopennessaboutpoliyobjetives. Eonomi
transparenyfousesontheeonomi informationthatis usedformonetary poliy. Thisinludes
theeonomidatatheentralbankuses,thepoliymodelsitemploystoonstruteonomifore-
astsor evaluatetheimpatofitsdeisions,and theinternalforeaststhe entral bankrelieson.
Proeduraltransparenyisaboutthewaymonetarypoliydeisionsaretaken.Poliytranspareny
meansapromptannounementofpoliydeisions. Inaddition,itinludesanexplanationofthede-
isionandapoliyinlinationorindiationoflikelyfuturepoliyations. Operationaltranspareny
onerns theimplementationofthe entral bank'spoliyations. Inthe sameway that Geraats,
anotherauthorsasGerbashandHahn(2000)andFaustandSvensson(2000)denedierenttypes
oftranspareny.
target. Consequently, market operators expetation about future ination will be
less heterogeneous and loser to the level xed by the entral bank. Thus, entral
bankindependeneorentralbankontratsshouldimplyaredutionofunertainty
relatedtothemonetary poliy andadereasetheheterogeneityof theinvestors'ex-
petations aboutthe futureorientation ofthemonetarypoliyandabout thefuture
evolution of the target variables. These eets expressed itself by a smaller inter-
est rates volatility reation to entral bank target variables news and by a greater
responseof these rates level tothese news.
Empirially,severalauthors,asAlesina(1988),Grilli,Masiandaro, andTabellini
(1991),Cukierman (1992),Cukierman, Webb,and Neyapti(1992), and Alesinaand
Summers (1993), Jonsson (1995), Loungani and Sheets (1997) and Eijnger et al.
(1998), among other, nd evidene of a negative orrelation of entral bank inde-
pendene with lower and more stable ination. Conerning the eets on nanial
markets,Alesina andSummers(1993), usingrosssetionevidene, showthatinter-
est rate variability is dereasing with higher entral bank independene, suggesting
thatmore redible entral banks benetfrom lessvariable interestrates. Asfor the
relationbetweenthe redibilityandthe responseofnanial marketstonews,itwas
taken into aount only by Clare and Courtenay (2001a, 2001b) and Chadha and
Nolan (2001). Clare and Courtenay investigate whether the reation of British fu-
turesontratsand exhangesrates to Englishmaroeonomi newsannounements
hashangedsinetheBankofEnglandwasgrantedoperationalindependeneinMay
1997. Their results indiatethat there maywell have been hanges intheway that
nanial marketsinorporate keyeonomidataintoseurities pries. Inpartiular,
they doument an inrease in the speed of the reation to interest rate announe-
ments. As for Chadha and Nolan (2001), they analyse the impats of numerous
hange inEnglishmonetarypoliyondut,andpartiularlytheindependeneofthe
Bankof England, on short-terminterest ratevolatilityreationto announement of
the Bank of England interest rate deisions, of publiations of the minutes of the
Monetary Poliy Committee (MPC) meetings and of the publiations of the quar-
terly Ination Report. However, Clare and Courtenay analyze only the dynamis
of the future ontrats and the exhange rate, and Chadha and Nolan onsider few
explanatoryvariables.
A greater redibility and then a better market understanding of monetary poliy
should improve the auray of market foreasts of entral bank poliy deisions.
A number of researhers have foused on the role of a new transpareny measure
on the ability of nanial markets to predit monetary poliy deisions (Tabellini,
1987; Dotsey, 1987; Rudin, 1988; Blinder, 1998; Kuttner, 2001; Haldane and Read,
2000;Pooleetal.,2002;Langeetal.,2003;PooleandRashe,2003;Swansson,2004;
Tuysuz, 2006, 2007). For the United States, Poole and Rashe (2003), Urih and
Wahtel (2001), Lange et al. (2003), Swansson (2004) and Tuysuz (2007) demon-
strated that preditability of the Fed's ations inreased after the 1994 deision to
announehangesinFedpoliyrates immediately afterFederal Open Market Com-
mittee(FOMC)meetings. However,the impatofanewredibilityonthedegreeof
foreseesabilitywasnot retainedintheempirial studies.
The lessunertainand thebetter preditabilityof entral bankdeisions should
redue interest rates response to monetary poliy deisions. Aording to Kuttner
(2001), only the unexpeted part of the entral bank deisions provide news infor-
mations about monetary poliy to markets operators' and then inuene interest
rates dynami. In addition, a greater redibility should redue investors' expeta-
tions about ination and thereby derease also interest rates response to monetary
poliy deisions. Indeed aordingto Fisher (1930)hypothesis nominalinterestrate
is expressed as the sum of expeted onstant real interest rates plus expeted rate
of ination. On the other hand, insituation of fully redibility, the derease of the
poliy rate, for example, will not generate a rise of interest rates. The inrease in
the degree of redibility thus redues the impat of monetary poliy rate hanges
on interest rates level. Empirially, several authors analyse theimpat of a greater
transpareny on nanial markets reation to monetary poliy ations (Urih and
Wahtel, 2001; Kuttner, 2001; Coppel and Connolly, 2003) 6
. However, the impat
of a greaterredibility on the nanial marketreation to entral bank ations was
retainedonly byChadha and Nolan(2001) andClare andCourtenay (2001).
Onthevolatilitylevel,anewredibilitymeasureinuenesinterestratesvolatility
6
For example, Haldane and Read (2001) found that the introdution of ination targeting in
theUnitedKingdomappearstohaveoinidedwithamarkeddampeninginyieldurveresponses,
suggesting greater transpareny and preditability as the Bankof England monetary framework
hanged.
Courtenay, 2001a,b;Lee, 2002). Indeed, unertaintyrelated tomonetary poliyand
heterogeneityofinvestorsantiipationsdependnegativelyonentral bankredibility
degree.
3 Data Desription and Preliminary Tests
This setion presents the dataset and its statistial properties. The empirial part
uses data series on interest rates, maroeonomi announements and unexpeted
variations of keyinterest rates.
3.1 Interest rates series
Two kinds of daily interest rate series are onsidered: a short term rate (London
Interbank Oered Rates; LIBOR) and a Government bond rate orresponding to
maturities ofrespetively
3
,6
and12
monthsand3
,5
,7
and10
years. Theseseriesoverthe periodranging fromthe rstofJanuary
1994
toFebruary,28 th
,2003
. Thisdataorrespondsto the quotesat loaltime market losure: 17:30 AMGMT.
In order to determine the order of integration of these series we arry out a
series of unit-root tests. Three dierent kinds of unit-root tests are performed: the
standard ADF test, the Zivot and Andrews (1992) test and nally the Seo (1999)
test. Aording tothe results oftheADF test,displayedintable6,weannotrejet
thenullhypothesisofunitrootforanyofthefourseries. Theseresultsareonrmed
for the Zivot and Andrews test as well as the Seo test. The Seo statisti allows to
takeintoaountforstruturalhangesintheserieswhiletheformeraountsforthe
preseneofonditional heteroskedastiity. Indeed,using Box-Piere,Ljung-Boxand
LMstatistis(seetable7), thenullhypothesisofhomoskedastiityisrejetedat the
5%
levelfor allassetsonsideredinour study. Thus,allinterestrateseriespresentaunitrootandinterestrates dierentialswillbeusedintheempirialanalysis. These
interestrate seriesarealso onditionally heterosedasti.
3.2 Announements and surprises
Aording to Balduzzi et al. (1997), it is not the announement per se that is im-
portant, but rather the information it onveys to market partiipants. Indeed, if
behavioral hanges. Sine the aim of this paperis to study theeet of announe-
ments on the dynamis of interest rates, series that reet unantiipated variations
for the relevant series are needed. These "surprises" are dened as the dierene
between the observed values for thevariables and the values that were antiipated.
As antiipations annot be observed diretly some approximation are needed. The
surveyspublishedbyReutersforUKmaroeonomiannounementsareusedinthis
paper. Thisorganization ollets every Fridayforeasts froma panelofmarket par-
tiipantsforthefollowingweekannounements. Medianvaluesforeahvariablewere
omputed. Thosevalueswereretainedasproxiesofmarketpartiipantexpetations.
In more detail, these variables orrespond to possible targets for entral banks.
That is, primarily, news onerning the ination rate and the global health of the
eonomiesonsidered. Theonsideredannounementsonern unemployment (UE),
ConsumerPrieIndex(CPI),ProdutionPrieIndex(PPI),Produtionindex(PROD),
retailsales(RET) andthe aggregate M4(M4). Allthese maroeonomi andmone-
tarynews areannouned around9:30a.m..
Conerning the unexpetedpartofmonetary poliydeisions, two methodshave
been used in the literature for their omputation. The rst method uses surveys
as previously disussed for maroeonomi announements. The alternative is to
approximateentralbanksdeisionsthroughsomearefullyhosenassetsquotations.
Thissolution waspreferred to theuseof surveys sine, aspointed byEhrmann and
Fratzher (2003),(2005),theweeklyfrequeny of thosesurveys prevent from taking
into aount themost reent expetations. Ontheother side,theassets priesused
are those from the day preeding entral bankers deisions. The nanial assets
allowing this deomposition must show some harateristis (Brooke et al., 2000),
namely(i)itsmaturityislose tothatofthekeyinterestrate,(ii)itisaliquidasset
and (iii) its maturity is shorterthat the time interval between two monetary poliy
meetings. In the ase of United Kingdom, assets that an be used to extrat the
unexpeted part of English monetary authorities deisions orrespond to the short
termforward rates deduedfromthe twoalternativeforward urves estimatebythe
BoE 7
(Ross, 2002). However, the both short term forward rates are available only
7
BoEestimates two alternative forward urvesfrom two alternativessets ofinstruments. One
urveistted usingmostlyGCreposandgiltyields(knowngeneriallyasthe VRPurve),while
January1994-May1997 May1997-February2003
AtualChange
-ExpetedNoChange 77.78% 44.00%
-ExpetedChange 22.22% 56.00%
AtualNoChange
-ExpetedChange 3.03% 1.89%
-ExpetedNoChange 96.97% 98.11%
Total
-Inorretexpetation 19.05% 15.38%
-Corretexpetation 80.95% 84.62%
from1997. OurperiodofstudystartsinJanuary1994. WethusreliedontheReuters
pollforthisountry,althoughthismeansthattheagentsexpetationsareonlyknown
onaweekly frequeny. Asshownin,e.g.,GravelleandMoessner(2001)orEhrmann
andFratzher(2005), surveyexpetations prove to be unbiased and eient.
Theexpetedandunexpetedpartofthepoliymonetaryrateanservetoassess
anddisuss the preditabilityof BoEdeisions.
4 Eets of a greater redibility on the preditability of
the Bank of England poliy deisions
In order to hek if the Bank of England independene improves thepreditability
of its deisions, the perentages of the expeted and unexpeted part of the BoE
poliydeisionsforthesub-periodspreedingandfollowingMay1997aredetermined.
Aording to the table1, aftertheadoption ofinstrument independeneof theBoE
in 1997, the nanial markets appear better able to antiipe BoE poliy deisions
thanbeforemay1997. Indeed,only80.95%monetarypoliydeisionswereantiiped
before mai1997 and 84.62% of deisions areantiiped afterthis date. Theseresults
revealthatmarketoperatorsbetterunderstandtheeetiveondutofthemonetary
poliyand/ortheyaknowledgetheapaityoftheentralbankaftermay1997than
theseond(knownasthebankliability urve)isttedusingsynthetibondpriesfrominterbank
oerrates,short-sterlingfutures,FRAs,andswaps. Afteradjustingforbiases,theseurvesanbe
seenasthebestmeasureofthemarket'sexpetationoftwo-weekforwardrates.
Inaddition, the adoption ofinstrument independene oftheBank of Englandin
May 1997 should aet interest rates meanand volatilityresponseto eonomiand
monetarynews. Thisimpatisanalyse inthe following setions.
5 Empirial model
Giving the unit-root test in Setion 2, interest rates rst-dierened response to
maroeonomiand poliy news hasbeen modelised asfollows:
∆R t = a + b∆R t − 1 + c∆r ∗ τ + X K k =1
d k D a k,t + X 3 j =1
e j J S t + ǫ t ,
(1)where
R t
denotes interest rates dierentials inperiodt
.∆r ∗ τ
andD k,t a , k = 1, . . . , K
orrespond respetivelyto the unexpeted partof themonetary poliyrate hanges
and a set of British maroeonomi news.
c
andd k
measure these news eetson interest rate level. The index
τ
is used for the maroeonomis announements variables instead oft
. Depending on the variable,τ
will be equal tot
ort − 1
. Allmaroeonomiandmonetary variablesretainedinthis paper areannounedaround
9:30 a.m. (loal time) and BoE deisions are diused around 12:00 a.m. Thus,
Government bond rates inperiod
t
respond to maroeonomi news and monetarypoliydeisionsannounedthesameday(period
t
). Inthesameway,LIBOR'sratesinthe period
t
reat to maroeonomi news announedthe same day. Inontrast,monetary poliy deisions diusedinperiod
t − 1
aet short term rateinperiodt
.Inadditionto maroeonomiand poliynews,three daysof theweek aretake into
aount;namely Monday (
M o
),Wednesday(W e
)and Friday(F r
).Theterm
ǫ t
orrespondsto theinnovation series. Several authorsestimateequa-tion(1 ) supposing thatthe innovations area Gaussian whitenoise (Balduzzi et al.,
1999; Bernhardsen, 2000; Ellingsen and Söderström, 2001; Favero, 2001; Kearney,
2001; Caporale and Williams, 2002; Parent, 2003). In the same line, equation (1 )
was estimated, rst by supposing that the innovations are a Gaussian white noise
and Engle Arh LM statistiswas thenapplied to hek whether theinnovations
ǫ t
areonditionally homosedasti. Table8 ,intheAppendix,enablesto rejetthenull
hypothesis and then aept the hypothesis that the interest rates volatility is on-
ditionally heterosedasti. Sine Bollerslev proposed the GARCH models in 1986,
tional varianes of nanial market. As these authors, we also apply the GARCH
approahofBollerslevtoestimatetheonditionalvarianesoftheinterestrates. This
modelan be expressedas:
h t = w + αε 2 t − 1 + βh t − 1 + γDum r ∗ τ +
X K k =1
θ k Dum a k,t + X 3 j =1
λ j J S t .
(2)Asinthe meanequation(Eq. 1),theinueneofmaroeonomiandpoliyvari-
ables are taking into aount. Contrary to the mean equation, dummies insteadof
atual news inorderto avoid multiollinearitywith theonditional mean regressors
are used. More preisely,
Dum r ∗
τ
is equal to 1 during entral bank deisions an-nounement days. Inthe same way, eah dummy variable
Dum a k,t
,fork = 1, ..., K
,isequal to 1during the announement daysof thevariable k.
6 Empirial results
Inordertotake intoaount theimpatofthe adoptionofinstrumentindependene
of the BoE, interest rates dynamis have been estimatedas desribed by equations
(1)and (2 ) for the sub-periods preedingand following May 1997 8
. The results are
presented anddisussed inwhat follows.
6.1 Interest rate reation to maroeonomi and monetary news
Over the period former to the BoE independene, interest rates level reat parti-
ularly to the unemployment, produer prie index and retail sales news (see Table
1). Onthe seondsub-period, itis interesting to notethat unemployment and pro-
duer prie index news have lost all the impat they had before 1997. During this
period, marketinterest rates reat mainly to prodution and retail sales news. Un-
employmentnewsinuenenegativelyinterestratesdynamiwhereasretailsalesand
8
Suh an approah per under-period was used by the majority of the authors analyzing the
impatofthemonetarypoliyratehangesonthedynamisofinterestratesbytakingaountof
newmeasurementsoftransparenyand/orredibilityoftheentralbank(UrihandWahtel,2001;
Lee,2002;Tuysuz,2007). ItwasalsousedbyParent(2003),whostudiestheimpatoftheshoks
ofthevariablesrelatingtotheCanadianmonetarypoliyontheleveloftheratesofthemarket.
impatsisinaordane withtheoretialexpetanies. For instane,thenegativeef-
fetofunemployment news anbeexplained ifmarket operatorstrustthemonetary
poliiesabouttheirapaitytoontrolinationaryshoks. Inotherwords,theyhave
enough ondene inentral bank to ahieve its employment target byreduing in-
terestrateswithoutimperilingtheir ination objetive. Asforproduerprie index,
itan serve asa proxy for theination level. Thus, a positive surprise orresponds
to an underestimation of the ination level and market investors will revise their
expetations about BoE monetary poliy rate. Lastly, the retail sales an be used
asa omponent of the eonomiativity. Mosttheories predit thatan unexpeted
inreases in real ativity and ination should inrease bond rates (Hess,2001; An-
dersen, Bollerslev, Diebold and Vega, 2004). More preisely, if inreasing eonomi
ativity is oupled with inreasing investments, and thus with a higher demand for
apital, interest rates shouldrise given a nite elastiity of apitalsupply. Informa-
tion about higher eonomi ativity might also alter agents' expetations of future
inationrates,sineinationouldbespurred byanoverheatingeonomy. Thus,an
unexpeted inrease inretail salesthenin real ativityould drive interestrates up
throughhigher realrates and/or higherination expetations.
Asfor the BoEpoliy rate, ineah sub-period theunexpeted partof monetary
poliy deisions inuene positively interest rates and the amplitude of this eet
is dereasing with maturity (see Table 2 )(
c
). Thispositive eet hasalready beenshowsbyempirialstudiessuhasCookandHahn(1989), Kuttner(2001), Kimand
Sheen (2000) or Lee (2002). This observation supports the expetations theory of
theterm struture 9
. Inaddition, table2 point outan important inrease ininterest
ratereationtounexpetedpoliydeisionsafter1997. Indeed,theunexpetedpoliy
deisionsinueneonlythe3-yearsand5-yearsinterestratesbefore1997. Inontrast,
after 1997, short term and medium term interest rates reat to poliy deisions. In
addition,theoverallsize ofinterest rateresponseto unexpetedhangesintheBoE
ratetendsto inreaseafter 1997. Toillustrate this eet,the12-month interestrate
reationto unexpeted poliy deisions was 0.084before 1997 whereas thisreation
inreases down to 0.4478after 1997.
9
Theexpetationstheorysaysthata longterminterestrateshouldbe equalto theaverageof
theshortterminterestratesoverthesame periodoftimeplusatermpremium;thus,aninrease
intherstoupleofshortrateshoulddriveupthelongrateinalesserextent.
hoks during theboth sub-periods. During the rst sub-period, there was greater
unertainty onerning unemployment and eonomi growth than there was about
ination. The periodprior to 1997 wasmarked by a relatively lowEnglish ination
rate (see g. 1 in appendix). This rate osillated around 2%. Contrary, the un-
employment rate was important (see g. 2 in appendix). During this period, the
BoEinreasedmore theirmaininterestratethatdereasedthem. Thesedeisionsof
theBritish monetary authorities allowed to maintain the ination rate lose to her
target but they aetednegatively theeonomi growth and unemployment. These
observations an explain the greater unertainty onerning theunemployment and
theeonomisituationpriorto1997. After1997,marketoperator'ssensitivitytoun-
employmentdisappeared. Duringthisseondperiod,theseagentsseemedtobemore
sensitive to eonomi growth hoks. As for therst sub-period, our results an be
explainedbythegreaterunertaintyonerningtheeonomigrowththan therewas
about unemployment rate and ination rate. These both rates were relatively low
after1997. However,this period wasmarked byan important dereaseandinrease
of the gross domesti produt rate (see Fig. 3 in appendix). These observations
an explainthe sensitivityof nanial agentsto eonomi growth hoks during the
seondperiod.
Finally,thelastobservationonernsthedynamiofinterestratevolatility. Table
5 shows that Wald test enables to rejet the null hypothesis of integrated GARCH
model(IGARCH).However,thesumoftheGARCHoeients(
α
+β
) isquitehigh.In another world, the importane of the sum of the GARCH oeients suggests
quiteimportant autoorrelatedinterestratevolatility. However, thiseventouldnot
be explained by unertainty related to the BoE monetary poliy. Indeed, during
theboth sub-periods,maroeonomiandmonetaryvariablesannounementsdonot
have important impat on interest rates volatility. These results are in aordane
with results obtained by Jones et al. (1998). These authors nd that the impat
of maroeonomi announements on interest rate volatility does not persist at all,
onsistent withthe immediate inorporation ofinformation's into pries. Aording
to these authors, maroeonomi announement days do not explain Government
bond ratepersistene.
sion
The result that the eets of maroeonomi news announements on interest rate
volatility inrease in the seond sub-period suggests that BoE redibility and/or
transpareny derease sine May 1997. The greater impat of BoE deisions an-
nounementsonmarketratelevelandvolatilityafter1997suggestsalsothederease
of BoE redibility and/or transpareny degree. Indeed, aording to setion 2, in
the period following the implementation of a new redibility measure, interest rate
volatility should be less inuened by the announements on maroeonomi and
monetary variables (Chadha and Nolan, 2001; Clare and Courtenay, 2001a,b; Tuy-
suz, 2007). Similarly, a greater redibility should imply a derease of interest rate
levelreationto unexpeted monetary poliy ratehanges.
A smaller redibility means that BoE does not manage to respet ination sta-
bility around its target. However, British ination rate evolution after May 1997,
represented in gure 1 in the appendix, does not emphasize any inationary ten-
denyduringtheseondsub-period. Thisobservationsuggests,then,thatwhihan
all into question the BoE redibility. In addition, using Cukierman and Meltzer
(1986) 10
,
11methodology,weonstrutBoEredibilitydegreefrom1994to2003. Ta-
ble2 showsthatBoEredibilitydegree inreaseovertime. Theseboth observations
suggestthattheBoEredibilitydegree hasnot dereasesine 1997.
Table 2: BoEredibility andtranspareny degreeevolution (1994-2003)
1994 1995 1996 1997 1998 1999 2000 2001 2002 2003
Credibility
degree
1 0,99 1 1 1 1 1 1 1 1
Transpareny
degree
0,73 0,80 0,80 0,80 0,80 0,80
Theindexofredibility(resp.transpareny)takesavaluebetween0and1.
Theindexofredibility(transpareny)takesavalueof1ifentralbankisfullyredible(transparent).
Theindexofredibility(transpareny)takesavalueof0ifentralbankisnotredible(transparent).
10
Inthe literature, the most frequentlyused methodology to onstrut redibility indexis the
methodology proposed by Cukierman and Meltzer (1986) (Faust and Svensson, 1998; Huthison
andWalsh,1998;CehettiandKrause,2002).
11
CukiermanandMeltzer(1986)methodologyispresentedinappendix.
on the interest rates volatility ould also be explained by a derease of the BoE
transparenydegree. However, afterMay1997,BankofEnglishisamongstthemost
transparent entral bank (Chadha and Nolan, 2001; Clare and Courtenay, 2001;
Diner and Eihengreen, 2007). Indeed, sine 1997 the nal objetive of monetary
poliy has been madeexpliit and passed to an independent entral bank, the date
of the Monetary Poliy Committee meetings are known around a year in advane,
the deision of the BoE is announed at a set time, often with an explanation for
the deision, minutes detailing voting patterns are published and regular quarterly
foreasts of the intermediate variable undera varietyof assumptionsare published.
Inpartiular,itmakesknownthevotingreordoftheninemembersoftheMonetary
PoliyCommittee(MPC),alongwithadetailedsummary/ommentaryoftheMPC's
deliberations. In addition, aording to Dinerand Eihengreen (2007) 12
,theBank
ofEngland transpareny degreewasabout 73%in1998 and 88%between 1999 and
2005 (table 2). In sum, these both observations suggest that the greater eets of
maroeonomi and monetarynews on marketvolatilityannotbe explainedbythe
degreeof transpareny.
Inaordane withChadhaand Nolan(2001) 13
observations,theargumentspre-
sented in the both previous paragraph suggest that the stronger eets of British
maroeonomi andmonetary newsannounementsoninterest ratesvolatilityould
not be explained neither by a derease of BoE redibility nor by a derease of BoE
transpareny. A possible explanation of the ampliationof the poliy deision im-
12
Several authors onstrut entral bank transpareny index (Fry et al., 2000; Mahadeva and
Sterne,2000;Bini-SmaghiandGros,2001;Siklos,2002;DeHaan,AmtembrinkandWaller(2004),
Eijnger and Geraats, 2006; Diner and Eihengreen, 2007). Amongthese authors, Diner and
Eihengreen onstrut a transpareny index for a large numberof ountry (124) and for a long
period(from1998to2005). Forthisreason,weusetheirtransparenyindex.
13
ChadhaandNolan(2001)ndthattheBritishmarketinterestratevolatilityhigherafterMay
1997. These authors argue that the highest level of interest rate volatility has been assoiated
with the period of ination targeting following the adoption of entral bank independene (May
1997). Theseauthorsexaminewhetherthisvolatilityisattributedtotheinformationowsrelated
toBritishmonetarypoliy. Theirresultssuggeststhatinformationowsintheformofminutesof
poliymeetings,publishedinationforeastsand announementsof theBankofEngland interest
rates deisions, show little sign of aeting, jointly or individually, the volatility of short-term
nominalinterestrates. ChadhaandNolanargue thatthe higherinterestratevolatility ould not
beexplainedbythedereaseofBoEredibility.
ial situation (Banerjee, 1992; Bikhandani et al., 1992; MQueen and Roley,1993;
Fleming and Remolona, 1997; Veronesi, 1999). Aording to these authors, the
main maroeonomi and monetary news an strongly inuene market operators
behaviournot only during of monetarypoliyunertainty but also duringeonomi
or/and nanial instability. Thus, without questioning the redibility and trans-
pareny oftheBoE, various nanial rises ourringafter 1997 14
andtheeonomi
situationmayhavereatedunertaintyonnanialmarketwhihexplainsthegreater
impatofthe BoE'sdeisionson interestrates volatility. Speially,byaeting neg-
atively the English eonomy and of other industrialized ountries (Lahrèhe-Révil,
2002; Heitz et al., 2004) theAsian risis and the Russianrisis reated unertainty.
Theunertainty relatedto thenanial situation results fromthe dierent nanial
risis and more partiularly the Russian risis and the bursting of the tehnology
andinternet bubble in2002 inUSA.Theseboth riseshad a generalized eet ona
world. Insum,unertaintyabouttheBritisheonomiativityombinedwithnan-
ial unertainty ouldexplain thegreater eet ofnews on interestrate volatility.
7 Conlusion
ThispaperinvestigatestheimpatoftheBankofEnglandindependeneonnanial
marketreationto news related to themonetarypoliy. Speially, itanalyzes the
eets of this new redibility measure on the reation of British Treasury rate and
Government bond rate level and volatility to news related to theBank of England
poliy. These news orrespond to theBoE target variables news and to unexpeted
partof the poliy ratehanges. It also analyzes how a greater redibility inuenes
thepreditability of the BoE rate hanges. The results obtained suggestthat sine
May1997,periodwheretheBoEwasgrantedoperational independene,marketpar-
tiipants have been able to antiipate better the deisions of the British monetary
authorities. Contrary to the theoretial waiting of a greater redibility eets, our
resultsshow thatthe eetsof the announements of theBoEtarget variables news
and poliy rate deisions diusion on interest rate volatility inrease after 1997. A
priori,theseresults ansuggesta dereaseof theBoEtransparenyand/orredibil-
14
ForexampletheAsianrisis(July1997),theRussianrisis(August1998),theBrazilianrisis
(January1999)andtheArgentinarisis(November2001).
the target xed by the BoE and the evolution of the transpareny and redibility
index suggest that the BoEtranspareny and redibility degree inrease sine 1997
ompare to theperiod prior to 1997. Giving these observations and the results ob-
tained by Banerjee (1992), Bikhandani et al. (1992), MQueen and Roley (1993),
Flemingand Remolona(1997)andVeronesi(1999), thegreater impatof maroeo-
nomiand monetary newsannounements oninterestrates volatilityanbeexplain
by the unertainty related to the nanial risis (the Asian risis (July 1997), the
Russian risis (August 1998) and the bursting of the tehnology and internet bub-
blein2002inUSA).Insum, theeet ofthemaroeonomiand monetarynewson
marketvolatilitydependsonmonetarypoliyunertaintyaswellaseonomiand/or
nanial unertainty.
94-97 97-03
3-month 6-month 12-month 3-year 5-year 7-year 10-year 3-month 6-month 12-month 3-year 5-year 7-year 10-year
a -0.0043
∗
-0.0014 0.0027 -0.0037 -0.0024 -0.0054
∗
0.0008 -0.003
∗∗
-0.0042 0.0004 0.0005 0.0013 0.0014 0.0011
( − 2.07) ( − 1.35) (1.41) ( − 1.27) ( − 0.82) ( − 2.00) (0.23) ( − 1.88) ( − 1.48) (0.25) (0.28) (0.67) (0.71) (0.55)
b -0.0613 -0.0226 0.0578 0.1075
∗
0.0627
∗∗
0.0330 -0.0123 -0.1570
∗
-0.0033 0.1039
∗
0.0905
∗
0.0561
∗
0.0369 0.0238
(−0.91) (−0.45) (1.10) (3.04) (1.86) (1.02) (−0.31) (−3.62) (−0.05) (3.73) (3.25) (2.17) (1.47) (0.97)
c
0.0918 0.1462 0.0842 0.2060∗
0.1240
∗
0.0451 -0.0186 0.5319
∗
0.4942
∗
0.4478
∗
0.2125
∗
0.1042 0.0298 -0.0336
(0.73) (1.33) (0.68) (2.50) (2.07) (0.81) (−0.29) (7.65) (6.22) (8.72) (2.66) (1.48) (0.43) (−0.50)
d cho
0.0000∗
0.0000∗
0.0000∗
0.0000∗
0.0000∗
0.0000∗
0.0000∗
0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000(−2.93) (−3.70) (−3.19) (−3.14) (−2.91) (−2.70) (−2.54) (1.59) (1.34) (0.78) (0.61) (0.09) (0.01) (0.12)
d ipc
-0.0192 0.076∗∗
-0.0116 0.118
∗∗
0.0229 0.0262 0.0997
∗
0.0313
∗
0.0713
∗
0.0907
∗
0.0875 0.0612 0.0506 0.0291
(−0.44) (1.78) (−0.52) (1.86) (0.53) (0.75) (2.22) (2.00) (3.50) (8.81) (1.38) (1.11) (1.01) (0.73)
d ppii
0.0016 0.0042 0.0142 0.0276∗
0.0308
∗
0.0250
∗
0.0229
∗
0.0009 -0.0004 0.007
∗∗
0.0051 0.0055 0.0050 0.0020
(0.19) (0.61) (0.84) (2.93) (3.68) (2.97) (2.29) (0.27) (−0.07) (1.89) (0.76) (0.83) (0.77) (0.30)
d ppio
0.0151 0.0736∗
0.0395 0.0153 -0.0065 0.0008 0.0097 -0.0090 -0.032
∗∗
-0.0097 -0.0048 -0.0007 -0.0005 -0.0079
(0.48) (3.47) (1.07) (0.41) (−0.18) (0.02) (0.26) (−0.32) (−1.96) (−0.81) (−0.20) (−0.03) (−0.03) (−0.47)
d vd
0.0190∗
0.0218∗
0.0354∗
0.0629∗
0.0488∗
0.0313∗∗
0.0226 0.0139∗
0.0131 0.0286∗
0.0250∗
0.014∗∗
0.0095 0.0036(2.25) (2.10) (2.11) (3.80) (2.97) (1.80) (1.02) (2.25) (1.07) (4.11) (3.01) (1.69) (1.13) (0.39)
d prod
0.0008 0.0146 0.0131 0.023∗∗
0.0125 0.0037 0.0061 0.0032 0.016∗∗
0.0180∗
0.0262∗
0.0238∗
0.0220∗
0.0187∗
(0.22) (1.13) (0.56) (1.67) (0.91) (0.28) (0.33) (1.44) (1.74) (2.26) (3.60) (3.71) (3.41) (2.94)
d m0
0.0039 0.0061 -0.0044 0.0083 0.0178 0.0307∗
0.037∗∗
0.0124∗
0.0053 -0.0062 -0.0171∗
-0.016∗∗
-0.0104 -0.0078(0.71) (1.53) (−0.61) (0.54) (1.28) (2.16) (1.70) (4.13) (0.37) (−0.78) (−2.23) (−1.88) (−1.08) (−0.78)
d m4
0.0276∗
0.0221∗
0.022∗∗
0.0244 0.0240 0.0114 0.0088 -0.0050 -0.0297∗
-0.0009 0.012∗∗
0.0098 0.0065 0.0050(2.39) (2.31) (1.95) (1.61) (1.51) (0.83) (0.45) (−0.57) (−2.83) (−0.08) (1.84) (1.08) (0.81) (0.72)
e mon
0.0086∗
0.0075
∗
0.0011 0.0104
∗
0.0078 0.0118
∗
0.0032 0.0004 0.006
∗∗
-0.0023 -0.0028 -0.0042 -0.0044 -0.0037
(2.25) (2.26) (0.36) (1.97) (1.50) (2.28) (0.51) (0.32) (1.65) (−1.19) (−0.94) (−1.39) (−1.48) (−1.21)
e wen
0.006∗∗
-0.0018 -0.0062 -0.0011 -0.0054 -0.0004 -0.0022 0.0013 -0.0001 -0.0018 -0.0041 -0.005
∗∗
-0.006
∗∗
-0.0050
(1.99) (−0.56) (−1.59) (−0.18) (−0.93) (−0.07) (−0.33) (0.49) (−0.04) (−0.72) (−1.20) (−1.68) (−1.87) (−1.49)
e f ri
0.0028 0.005∗∗
-0.0060 0.0078 0.0045 0.0056 0.0044 -0.0008 0.0044 -0.0053 -0.006
∗∗
-0.0077
∗
-0.0077
∗
-0.007
∗∗
(1.04) (1.71) ( − 1.42) (1.37) (0.76) (0.94) (0.65) ( − 0.37) (1.00) ( − 1.38) ( − 1.88) ( − 2.22) ( − 2.15) ( − 1.85)
Notes: Thevaluesin
(.)
arethet-statistisproposedbyBollerslevandWooldridge(1992)?.*and**indiatethattheorrespondingoeientisstatistiallysigniantatthe5%and10%level,respetively.
P P
19
94-97 97-03
3-month 6-month 12-month 3-year 5-year 7-year 10-year 3-month 6-month 12-month 3-year 5-year 7-year 10-year
w 0.0010
∗
0.0000 0.0004
∗
0.0003 0.0001 0.0001 0.0044
∗
0.0012
∗
0.0032
∗
0.0003
∗
0.0006
∗
0.0006
∗
0.0007
∗
0.0007
∗
(5.09) (0.60) (2.56) (0.68) (0.37) ( − 0.25) (5.42) (11.58) (13.88) (25.25) (2.76) (3.37) (3.96) (3.44)
α
0.2053∗
0.3588
∗
0.2079
∗
0.0506
∗
0.0440
∗
0.0393
∗
0.1467
∗
0.1485
∗
0.3420
∗
0.0719
∗
0.0492
∗
0.0291
∗
0.0301
∗
0.0305
∗
(2.19) (2.84) (3.23) (3.43) (2.99) (3.42) (3.06) (3.49) (10.51) (9.49) (3.22) (3.05) (3.48) (3.21)
β
0.5667∗
0.5388
∗
0.6464
∗
0.9313
∗
0.9440
∗
0.9512
∗
0.5929
∗
0.5941
∗
0.5581
∗
0.6436
∗
0.8906
∗
0.9494
∗
0.9446
∗
0.9372
∗
(4.81) (6.25) (10.07) (40.04) (46.73) (57.56) (4.91) (8.09) (48.47) (23.70) (26.56) (63.39) (73.62) (69.57)
γ
0.0011 0.0040 0.00256 0.0012 0.0018∗
0.0017∗
0.00010 0.0001 0.0136∗
0.0033∗
0.0009∗
0.0008∗
0.0007∗
0.0007∗
(0.29) (0.67) (0.36) (1.19) (2.11) (1.97) (0.04) (0.12) (9.57) (4.21) (2.10) (2.21) (2.01) (1.94)
θ cho
-0.001∗∗
-0.0005 0.00030 0.0001 0.0001 -0.0004 -0.0020 0.0000 0.0002∗∗
0.0008∗
-0.001∗
-0.0003 -0.0002 -0.0004(−1.73) (−1.41) (0.60) (0.16) (0.08) (−0.68) (−1.42) (−0.03) (1.72) (3.75) (−1.74) (−0.72) (−0.66) (−1.01)
θ ipc
0.0001 0.0006 -0.0010∗
-0.0001 -0.0012
∗
-0.001
∗∗
-0.0015 -0.0014
∗
-0.0016
∗
0.0006
∗
0.0009
∗
0.0003 0.0001 0.0001
(0.27) (1.16) (−2.61) (−0.25) (−2.27) (−1.80) (−1.36) (−4.39) (−7.26) (4.86) (2.51) (0.73) (0.42) (0.38)
θ ppii
0.0004 -0.0076∗
0.0002 -0.0022 -0.0002 0.0000 -0.0013 0.0000 0.0001 0.0026 0.0009 0.0007 0.001
∗∗
0.0009
∗
(0.19) (−4.55) (0.12) (−1.62) (−0.27) (0.00) (−0.55) (0.00) (0.01) (0.77) (0.99) (1.26) (1.95) (2.35)
θ ppio
0.0007 0.0074∗
0.002551 0.0017 0.0002 0.0002 -0.0013 0.0000 -0.0012 -0.0029 -0.0010 -0.0007 -0.0010∗
-0.0010∗
(0.59) (4.00) (1.46) (1.37) (0.35) (0.29) (−0.57) (0.00) (−0.08) (−0.87) (−1.03) (−1.18) (−2.04) (−2.41)
θ vd
0.0000 -0.0001 0.0011 -0.0019∗
-0.0010 -0.0009 -0.0020∗
-0.0014∗
-0.0002∗∗
-0.0001 -0.0009∗
-0.0003 -0.0003 -0.0002(0.08) (−0.45) (1.58) (−2.47) (−1.43) (−1.36) (−2.38) (−8.14) (−1.68) (−0.77) (−2.34) (−1.10) (−1.01) (−0.71)
θ prod
-0.0005∗
0.0005 0.001 -0.0012∗∗
-0.0015∗
-0.0016∗
-0.0021∗
-0.0001 0.0005 0.0010∗
-0.0008∗
-0.0003 -0.0002 -0.0003(−6.49) (0.93) (1.40) (−1.95) (−2.28) (−3.13) (−2.29) (−0.42) (1.63) (6.27) (−2.02) (−1.29) (−0.97) (−1.23)
θ m0
-0.0011∗
-0.0002 0.0012∗
-0.0003 -0.001∗∗
-0.001∗∗
-0.0048∗
-0.0014∗
-0.0010∗
0.0003∗
0.0006 -0.0002 -0.0003 -0.0003(−6.80) (−0.74) (2.13) (−0.39) (−1.65) (−1.71) (−5.93) (−7.10) (−8.88) (4.29) (0.74) (−0.56) (−0.89) (−0.67)
λ lu
0.0001 0.0005 -0.0011∗
0.0000 -0.001∗∗
-0.001∗∗
-0.0026∗
-0.0009 -0.0029∗
-0.0015∗
-0.0009∗
-0.0013∗
-0.0016∗
-0.0016∗
(0.40) (1.03) (−3.76) (−0.07) (−1.73) (−1.69) (−2.68) (−1.07) (−9.83) (−17.28) (−2.20) (−3.48) (−4.18) (−3.99)
λ me
-0.0010∗
0.0011
∗
-0.00001 0.0022
∗
0.002
∗∗
0.0022
∗
-0.0015 -0.0002 -0.0042
∗
0.0001
∗∗
-0.0002 -0.0006 -0.001
∗∗
-0.0005
(−3.56) (3.05) (−0.24) (2.49) (1.86) (2.85) (−1.10) (−0.77) (−10.97) (1.81) (−0.35) (−1.53) (−1.92) (−1.19)
λ ve
-0.0014∗
0.0001 0.0002 0.0004 0.0003 0.0006 -0.0032
∗
0.0000 -0.0036
∗
0.0014
∗
-0.0011
∗
-0.0008
∗
-0.0008
∗
-0.001
∗∗
( − 5.19) (0.53) (0.39) (0.50) (0.35) (0.81) ( − 3.17) ( − 0.02) ( − 13.42) (9.17) ( − 2.53) ( − 2.09) ( − 2.04) ( − 1.80)
Notes: Thevaluesin
(.)
arethet-statistisproposedbyBollerslevandWooldridge(1992)?.20
94-97 97-03
3-month 6-month 12-month 3-year 5-year 7-year 10-year 3-month 6-month 12-month 3-year 5-year 7-year 10-year
α + β
0.7720 0.8976 0.8543 0.9819 0.9880 0.9905 0.7396 0.7427 0.9001 0.7155 0.9398 0.9785 0.9747 0.9677 Wald testα + β = 1
[0.000] [0.000] [0.000] [0.000] [0.000] [0.000] [0.000] [0.000] [0.000] [0.000] [0.000] [0.000] [0.000] [0.000]
21
Alesina,A.andSummers,L.(1993), `Centralbankindependeneandmaroeonomi
performane',Journal of Money, Credit, and Banking25,157162.
Balduzzi, P., Bertola,G. and Foresi, S. (1997), `A modelof target hanges and the
termstrutureofinterestrates',JournalofMonetaryEonomis39(2),223249.
Balduzzi, P., Elton, E. and Green, T. (1999), `Eonomi news and the yield urve:
Evidenefromtheu.s.treasurymarket',NewYorkUniversity,Leonard N.Stern
Shool Working Paper .
Banerjee, A. (1992), `A simple model of herd behavior', Quarterly Journal of Eo-
nomis107,797818.
Bernhardsen, T. (2000), `The relationship between interest rate dierentials and
maroeonomi variables: a panel data study for european ountries', Journal
of International Money andFinane (19), 289308.
Bikhandani,S., Hirshleifer,D.andWelh,I.(1992), `Atheoryoffads,fashion,us-
tom and ultural hange asinformational asades',Journal of Politial Eon-
omy100,9921026.
Bini-Smaghiand Gros(2001),`Istheeb suiently aountable and transparent?',
ENEPRI Working Paper (N°7).
Blinder,A.S.(1998),`Central banking intheoryand pratie',MITPress .
Bollerslev,T.(1986),`Generalizedautoregressiveondionalheteroskedastiity',Jour-
nalof Eonometris 31,307327.
Brooke,M.,Cooper,N.andSholtes,C.(2000),`Inferringmarketinterestrateexpe-
tationsfrommoneymarketrates',Bank ofEngland QuarterlyBulletinpp.392
402.
Caporale and Williams(2002),`Long-term nominalinterestrates and domesti fun-
damentals', Review of FinanialEonomis (11),119130.
Cehetti, S. and Krause, S. (2002), `Central bank struture, poliy eieny and
maroeonomi performane: Exploring the empirial relationships', The Fed-
eral Reserve Bank of St. Louis(July/August).
volatility: Dithing'monetarymystique' intheuk',Journalof Maroeonomis
23(3), 349366.
Clare,A.andCourtenay,R.(2001a),`Assessingtheimpatofmaroeonomis news
annonements on seurities pries under dierent monetary poliy regimes',
Bank of England Working Paper N° 125 .
Clare, A. and Courtenay, R. (2001b), `What an we learn about monetary poliy
transparenyfromnanialmarketdata?',DeutsheBundesbankDisussionPa-
per(06/01).
Connolly,R.A.andKohler (2004), `News andinterest rateexpetations: a studyof
sixentral banks', Researh Disussion Paper (2004-10).
Cook, T. and Hahn, T. (1989), `The eet of hanges in the federal funds target
rate on market interest rates in the 1970s', Journal of Monetary Eonomis
24, 331351.
Cukierman, A. (2001), `Aountability, redibility, transpareny and stabilization
poliyintheeurosystem', In Charles Wyplosaed., The impat of EMU on Eu-
rope and the developing Countries,Oxford UniversityPress hap. 3,4075.
Cukierman, A. and Meltzer, A. (1986), `The theory of ambiguity, redibility, and
inationunderdisretionandasymmetriinformation', Eonometria54,1099
1128.
Diner,N. N.and Eihengreen,B. (2007), `Central banktranspareny: Where, why
and withwhat eets?',NBER WORKINGPAPERSERIES (13003).
Dodge, D. (2002), `La poursuite de ibles d'ination au anada: expériene et en-
seignements'.
Dotsey,M. (1987),`Monetary poliy,serey,andfederal fundsratebehavior',Jour-
nalof MonetaryEonomis20, 463474.
Ehrmann,M.andFratzher,M.(2003),`Monetarypoliyannounementsandmoney
markets: A transatlanti perspetive', International Finane 6(3),309328.
terdependenebetween theeuroareaand theunitedstates', Eonomi Journal
115(506), 930948.
Eijnger,S.andGeraats,P.(2006), `Howtransparentareentralbanks?',European
Journal of Politial Eonomy 22(1).
Ellingsen,T.andSöderström,U.(2001),`Monetarypoliyandmarketinterestrates',
Amerian Eonomi Review 91(5),15941607.
Engle, R.F. (1982),`Autoregressive onditional heterosedastiity withestimatesof
the variane ofunited kingdomination', Eonometria 50, 9871006.
Faust, J. and Svensson, L. (1998), `Transpareny and redibility: Monetary poliy
withunobservablegoals', NBER Working Paper .
Faust, J.and Svensson,L.E. (2001), `Transpareny andredibilitymonetary poliy
withunobservablegoals', InternationalEonomiReview 42(2),369397.
Favero (2001), `Does maroeonomis help us to understand the term struture of
interestrates?', CEPRdisussion paper series (2749).
Fleming,M.andRemolona,E. (1997),`Prieformationandliquidityintheu.s.trea-
surymarket: Evidenefromintradaypatternsaroundannounements',Reserve
Bank of New York(27).
Froyen,R.andWaud,R.(1995),`Optimalseigniorageversusinterestratesmoothinh',
Journal of Maroeonomis 17,111129.
Fry,M.,Julius,D.,Mahadeva,L.,Roger, S.andSterne,G.(2000),`Keyissuesinthe
hoie of a monetary poliy framework', in L. Mahadeva and G. Sterne (eds),
MonetaryPoliyFrameworksinaGlobalContext,London: Routledgepp.1216.
Geraats, P. M. (2000), `Why adopt transpareny? the publiation of entral bank
foreast',CEPR DisussionPaper (2582).
Geraats,P.M.,Eijnger,S.C.W.andvanderCruijsen,C.A.(2006),`Doesentral
banktranspareny redueinterest rates?',mimeo .
Reserve Bank of Rihmond, Working Paper (90-06).
Goodhart,C.(1996),`Whydothemonetaryauthoritiessmoothinterestrates?',LSE
Finanial MarketsGroup Speial Paper (81).
Gravelle,T.andMoessner,R.(2001),`Reationsofanadianinterestratestomaroe-
onomisannounements: Impliationsformonetarypoliytranspareny',Bank
of Canada Working Paper2001-5.
Haan,J.D.,Amtenbrink,F.andWaller,S.(2004),`Thetransparenyandredibility
of the european entral bank',Journalof CommonMarket Studies42(4).
Haldane, A. and Read, V. (1999), `Monetary poliy and the yield urve', Bank of
England Quarterly Bulletin pp.171176.
Haldane, A. and Read, V. (2000), `Monetary poliy surprises and the yield urve',
Bank of England Working Paper (106), 141.
Heitz, B., Hild, F. and Monfort, B. (2004), `Sysnhronisation desyles au sein du
g7 et intégration ommeriale et nanière',INSEE, Note de onjonture (juin
2004), 3040.
Hess,D.(2001),`Surprisesinu.s.maroeonomireleases: Determinantsoftheirrela-
tiveimpatont-bondfutures', Centerof FinaneandEonometris, University
of Konstanz, Disussion Paper (No. 2001/01).
Huthison,M.andWalsh,C.(1998),`Theoutput-inationtrade-oandentralbank
reform: evidene fromnew zealand', Eonomi Journal108,703725.
Jones, C., Lamont, O. and Lumsdaine, R. (1998), `Maroeonomi news and bond
markets volatility',Journal of Finanial Eonomis47, 315337.
Kearney (2001), `A note on modeling the impat of eonomi announements on
interestrates', Eonomi letters (71), 8389.
Kim, S.-J. and Sheen, J. (2000), `International linkages and maroeonomi news
eets in interest rate volatility - autralia and the us', Pai-Basin Finane
Journal 8(1),85113.
fed fundsfutures market',Journalof MonetaryEonomis47(3), 523544.
Lahrèhe-Révil, A. (2002), `Intégration internationale et interdépendanes mondi-
ales', L'éonomie mondiale 2003; Ed. La Déouverte; Col. Repères, p. 54-63
.
Lange, J., Sak, B. and Whitesell, W. (2003), `Antiipations of monetary poliy in
nanial markets', Journal of Money35(6), 889908.
Lee,J. (2002), `Federal fundstarget hangesand interest ratevolatility',Journal of
Eonomis andBusiness54(2), 159191.
Longworth, D.(2002),`Ination etmaroéonomie: hangements survenus entreles
années 1980 à1990', Revue de la Banque du Canada(printemps), 319.
Mahadeva, L. and (Eds.), G. S. (2000), `Monetary poliy frameworks in a global
ontext', Routledge, London.
MQueen, G. and Roley, V. (1993), `Stok pries, news, and business onditions',
Review of Finanial Studies6(3),683707.
Parent, N. (2003),`Transparene et réatindes taux d'intérêt à lapubliation péri-
odique desdonnéesmaroéonomiques', Revue dela Banque du CanadaHiver
2002-2003,3137.
Poole, W. and Rashe, R. H. (2003), `The impat of hanges in fom dislosure
praties on the transpareny of monetary poliy: Are markets nd the fom
better 'synhed, The Federal Reserve of St.LouisReviewspp.110.
Poole,W.,Rashe,R.H.andThornton,D.L.(2002),`Marketantiipationsofmone-
tarypoliyations',TheFederal ReserveofSt.LouisReviewsissueJul.,6594.
Ross,K. (2002),`Marketpreditabilityofeb monetarypoliydeisions: A ompar-
ative examination', IMFWorking Paper (02/233.).
Rudin, J.R. (1988),`Central bankserey,'fed wathing', and the preditability of
interestrates', Journal of Monetary Eonomis22, 317334.
iy: Areviewofreentempirialevidene.', JournalofEonomisandBusiness
.
Saxton,J.(1997),`Transpareny andfederal reservemonetary poliy',A Joint Eo-
nomi Committee Study .
Seo,B.(1999),`Distributiontheoryforunitroottestswithonditionalheteroskedas-
tiity', Journalof Eonometris91(1), 113144.
Siklos, P. (2002), `The hanging fae of entral banking: Evolutionary trends sine
world war ii', New York: CambridgeUniversity Press.
Swansson,E. T.(2004),`Federal reserve transparenyandnanial marketforeasts
ofshort-terminterestrates',Federal ReserveBoardFinaneandEonomisDis-
ussion Paper 2004-6.
T. G. Andersen, T. Bollerslev, F. X. D. and Vegad, C. (2004), `Real-time prie
disoveryinstok,bondandforeignexhangemarkets',The JournalofFinane
.
Tabellini,G.(1987),`Sereyofmonetarypoliyandthevariabilityofinterestrates',
Journal of Money, Credits andBanking19, 425436.
Tuysuz,S.(2006),`Eets delatransparene etde larédibilité surleméanismede
transmissiondelapolitiquemonétairesurlestauxd'intérêt: leasdesetats-unis
etdu royaume-uni',mimeo .
Tuysuz, S. (2007), `Central bank transpareny and the u.s. interest rates level and
volatility response tou.s. news', mimeo .
Urih, T. and Wahtel, P. (2001), `Finanial market responses to monetary poliy
hange inthe 1990s',Contemporary Eonomi Poliy19(3), 254267.
Veronesi,P.(1999),`Stokmarketoverreationtobadnewsingoodtimes: Arational
expetationsequilibriummodel',Review of Finanial Studies12(5),9751007.
Woodford, M. (1999), `Optimal monetary poliy inertia', NBER Working Paper
(7261).
shok, and unit-root hypothesis', Journal of Business and Eonomi Statistis
10(3), 251270.
Cukierman and Meltzer (1986) authors dene monetary poliy redibility as "the
absolute value of the dierene between the poliymaker's plans and the publi's
beliefs about those plans". In this approah, theredibility index an be expressed
as:
Cre = 1 if E(π) < π t ,
Cre = 1 −
E(π) − π t
0.2 − π t if π t < E (π) < 20%,
Cre = 0 if E(π) > 20%.
Themoretheexpetedination(
E(π)
)divergesfromthelevelofthetargetination(
π t
),thelessredibletheentralbankis(Cre → 0
). Inthesamevein,iftheexpetedinationis smaller than or lose to thetarget levelof ination, thenthe redibility
ofthe entral bankattains itsmaximumvalue(
Cre → 1
).Some authors, as Cehetti and Krause (2002), while using this approah, sup-
posed the same level for the ination target for all the ountries they retained in
their empirial analysis. In addition, they also assume that the expeted ination
used inorder to onstrut the redibilityindex is based on therealized ination of
the previous period. Contrary to these authors, we x the same ination target
for the industrialized ountriesand the same target for theemerging ountries. For
the industrialized ountries, we suppose that the ination target is 2.125 15
, whih
orresponds to the average of the target x by some entral bank of industrialized
ountriespratiing inationtarget. Asfor the emergingountries, we supposethat
theinationtargetisequalto3.25 16
. Furthermore,theexpetedinationisobtained
usingdatafrom Datastream.
15
2.125orrespondstotheaveragevalueoftheinationtargetlevelxedbyindustrialountries,
asUnitedKingdomandAustralia,during90s.
16
3.25orrespondtotheaveragevalueoftheinationtargetlevelxedbyemergingountries,as
BrazilandMexio,during90s.