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/

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)targetvariablesnewsandtounexpetedmonetary pol-iyrate hanges. It analyzes whether themarket rateresponse to these news has hanged sine the Bank of England (BoE) was grantedoperational inde-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 moreEnglish inter-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,

∗

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 redue in-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. Mostoftheseresearhershavefousedontheroleof maroe-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

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 enjoyless vari-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.

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?

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 v ari-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;beausetheyensurethatthereisnoundueinueneorpolitial pres-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

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.

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

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

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 and mone-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

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

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

+

K

X

k

=1

d

k

D

a

k,t

+

3

X

j

=1

e

j

J S

t

+

ǫ

t

,

(1)

where

R

t

denotes interest rates dierentials inperiod

t

.

∆

r

∗

τ

and

D

a

k,t

, 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

).

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

+

K

X

k

=1

θ

k

Dum

a

k,t

+

3

X

j

=1

λ

j

J S

t

.

(2)

Asinthe meanequation(Eq. 1),theinueneofmaroeonomiandpoliy vari-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

impatsisinaordane withtheoretialexpetanies. For instane,thenegative ef-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

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'ssensitivityto un-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.

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; T uy-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. T a-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

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,thearguments pre-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

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,unertaintyabouttheBritisheonomiativityombinedwith nan-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,market par-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/or

redibil-14

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

m

0

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

m

4

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

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

m

0

-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)?.

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]

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),`Inferringmarketinterestrate expe-tationsfrommoneymarketrates',Bank ofEngland QuarterlyBulletinpp.392 402.

Caporale and Williams(2002),`Long-term nominalinterestrates and domesti fun-damentals', Review of FinanialEonomis (11),119130.

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?',DeutsheBundesbankDisussion Pa-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.

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).

Reserve Bank of Rihmond, Working Paper (90-06).

Goodhart,C.(1996),`Whydothemonetaryauthoritiessmoothinterestrates?',LSE Finanial MarketsGroup Speial Paper (81).

Gravelle,T.andMoessner,R.(2001),`Reationsofanadianinterestratesto maroe-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: Determinantsoftheir rela-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.

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),`Marketantiipationsof mone-tarypoliyations',TheFederal ReserveofSt.LouisReviewsissueJul.,6594.

Ross,K. (2002),`Marketpreditabilityofeb monetarypoliydeisions: A ompar-ative examination', IMFWorking Paper (02/233.).

iy: Areviewofreentempirialevidene.', JournalofEonomisandBusiness .

Saxton,J.(1997),`Transpareny andfederal reservemonetary poliy',A Joint Eo-nomi Committee Study .

Seo,B.(1999),`Distributiontheoryforunitroottestswithonditional heteroskedas-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 ReserveBoardFinaneandEonomis Dis-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.

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

ADF ZandA SEO

C B A C B A Model2 Model1 Model0

b

ρ

β

b

b

ρ

µ

b

ρ

b

3month -1.176 -0.000

∗

0.483 -0.003

∗

-0.912

∗

-2.576

∗

-2.083

∗

-2.572

∗

-1.650

∗

-0.521

∗

-1.433

∗

(

−

2

.

931)

(

−

0

.

65)

[03

/

97]

[07

/

97]

[09

/

96]

[0.58℄ [0.58℄ [0.58℄ 6month -1.589 -0.000

∗

0.116 -0.002

∗

-0.770

∗

-2.711

∗

-2.236

∗

-2.982

∗

-1.607

∗

-1.010

∗

-0.577

∗

(

−

2

.

89)

(

−

0

.

25)

[03

/

97]

[08

/

97]

[09

/

96]

[0.49℄ [0.50℄ [0.51℄ 12month -2.075 -0.000

∗

0.197 -0.002

∗

-0.701

∗

-2.838

∗

-2.602

∗

-3.374

∗

-1.723

∗

-0.698

∗

-1.476

∗

(

−

3

.

45)

(

−

0

.

32)

[10

/

96]

[01

/

00]

[09

/

96]

[0.62℄ [0.62℄ [0.59℄ 3year -3.867 -0.000

∗

-0.418 0.002

∗

-0.695

∗

-4.818

∗

-4.019

∗

-4.549

∗

-0.530

∗

0.197

∗

-1.717

∗

(

−

4

.

29)

(0

.

29)

[07

/

99]

[10

/

01]

[06

/

99]

[0.58℄ [0.59℄ [0.57℄ 5year -3.817 -0.000

∗

-0.439 0.002

∗

-0.741

∗

-4.908

∗

-4.139

∗

-4.709

∗

-0.496

∗

0.152

∗

-1.959

∗

(

−

4

.

14)

(0

.

30)

[06

/

99]

[10

/

98]

[06

/

99]

[0.62℄ [0.62℄ [0.62℄ 7year -3.803 -0.000

∗

-0.410 0.001

∗

-0.835

∗

-5.018

∗

-4.637

∗

-5.098

∗

-0.160

∗

-0.14

∗

-2.09

∗

(

−

4

.

04)

(0

.

24)

[05

/

99]

[10

/

98]

[06

/

99]

[0.63℄ [0.63℄ [0.64℄ 10year -3.636 -0.000

∗

-0.502 0.002

∗

-0.794

∗

-5.007

∗

-4.634

∗

-5.147

∗

-0.856

∗

-0.392

∗

-1.85

∗

(

−

3

.

81)

(0

.

33)

[11

/

97]

[10

/

98]

[04

/

97]

[0.65℄ [0.64℄ [0.65℄ *and**indiatethattheorrespondingoeientisstatistiallysigniantatthe5%and10%level,respetively.

Thevalues[./.℄intheentralpartofthetableorrespondtothemonthandtheyearofthehange. Thevalue[.℄intherighthandofthetableorrespondstothevalueof

ρ

.

3month 6month 12month 3year 5year 7year 10year Lyung-Box (LB)

testonthesquared residuals

LB

Q

2

(1)

80.00

∗

460.69

∗

18.53

∗

29.75

∗

27.71

∗

25.75

∗

33.90

∗

LB

Q

2

(5)

89.00

∗

497.68

∗

21.50

∗

78.61

∗

171.79

∗

187.98

∗

203.69

∗

LB

Q

2

(10)

92.01

∗

506.40

∗

21.73

∗

151.00

∗

297.53

∗

343.10

∗

371.06

∗

Box-Piere (BP) testonthesquared residuals

BP

_ǫ

2

(1)

79.87

∗

459.92

∗

18.50

∗

29.70

∗

27.66

∗

25.71

∗

33.85

∗

BP

_ǫ

2

(5)

88.84

∗

496.83

∗

21.46

∗

78.42

∗

171.34

∗

187.48

∗

203.17

∗

BP

_ǫ

2

(10)

91.84

∗

505.51

∗

21.69

∗

150.47

∗

296.49

∗

341.89

∗

369.79

∗

LMtest forARCH eet (Engle (1982))

LM

−

ARCH

(1)

79.87

∗

459.92

∗

18.50

∗

29.70

∗

27.67

∗

25.72

∗

33.86

∗

LM

−

ARCH

(5)

82.09

∗

475.61

∗

20.42

∗

60.62

∗

122.15

∗

133.22

∗

136.90

∗

LM

−

ARCH

(10)

84.52

∗

480.94

∗

20.63

∗

93.29

∗

159.97

∗

172.05

∗

179.59

∗

*and**indiatethattheorrespondingoeientisstatistiallysigniantatthe5%and10%level,respetively.