The effects of a greater central bank credibility on interest rates level and volatility response to news in the U.K.

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

^and

12

^monthsand

3

^,

5

^,

7

^and

10

^{years. Theseseries}

overthe periodranging fromthe rstofJanuary

1994

^toFebruary,

28 ^th

^,

2003

^{. This}

dataorrespondsto 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,allinterestrateseriespresenta}

unitrootandinterestrates 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 inperiod

t

^.

∆r ^∗ _τ

^and

D _k,t ^a , k = 1, . . . , K

orrespond respetivelyto the unexpeted partof themonetary poliyrate hanges

and a set of British maroeonomi news.

c

^and

d _k

^{measure these news eets}

on interest rate level. The index

τ

^{is used for the} maroeonomis announements variables instead of

t

^{. Depending on the variable,}

τ

^{will be equal to}

t

^or

t − 1

^{. All}

maroeonomiandmonetary 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 monetary}

poliydeisionsannounedthesameday(period

t

^{). Inthesameway,LIBOR'srates}

inthe period

t

^{reat to maroeonomi news announedthe same day. Inontrast,}

monetary poliy deisions diusedinperiod

t − 1

^{aet short term rateinperiod}

t

^.

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

^,for

k = 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 been}

showsbyempirialstudiessuhasCookandHahn(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

,

¹¹

methodology,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 test

α + 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,iftheexpeted}

inationis 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.