Interest Rates Targeting of Monetary Policy: An Open-Economy Svar Study of Malaysia

Transcription

1 PROSIDING PERKEM VIII, JILID 2 (23) ISSN: X Ineres Res Trgeing of Monery Policy: An Open-Economy Svr Sudy of Mlysi Zulkefly Abdul Krim School of Economics Fculy of Economics nd Mngemen Universii Kebngsn Mlysi 436 Bngi, Selngor mz@ukm.my ABSTRACT This pper exmines he implemenion of monery policy during he ineres res rgeing in smll-open economy (i.e. Mlysi) by using n open-economy srucurl VAR (SVAR) sudy. I ess he effec of foreign shocks upon domesic mcroeconomic flucuions nd monery policy, nd exmines how effecive is monery policy in influencing mcroeconomic vribles. The resuls show h during ineres res rgeing, monery policy plys significn role in ffecing mcroeconomics vribles. This finding suggess h he imporn role of monery policy s sbilizion policy in smll-open economy. Keywords: monery policy; ineres res rgeing; open-economy; SVAR INTRODUCTION The min quesion reling o he implemenion of monery policy is eiher he cenrl bnk hs o rge monery ggrege or ineres re s n opering rge. In Mlysi s experience, he Bnk Negr Mlysi (BNM) s he nion s Cenrl Bnk hs chnged monery policy sregy from monery rgeing owrds ineres re rgeing since November 995. During ineres re rgeing, monery policy operes hrough shor-erm ineres res o in he ulime rge h is susinble long-run economic growh, ccompnied wih he price nd finncil sbiliy. The im of his sudy is o exmine empiriclly on he implemenion of monery policy during ineres res rgeing in smll nd highly open economy, wih specil reference o Mlysi. This sudy uses non-recursive srucurl vecor uoregressive (SVAR) modelling frmework in nine vribles, in which include hree foreign vribles, wo domesic monery policy vribles, nd four mcroeconomic vribles. By using SVAR sudy, we cn idenify he srucurl economic shocks by imposing se of resricions o he conemporneous inercions of he model vribles. For exmple, fer he model hs been specified, he impcs of foreign shocks on domesic mcroeconomic nd monery policy cn be nlyzed by using srucurl impulse-response funcion. The min conribuions of his pper hve wo specs. Firs, his sudy considers he role of foreign fcor in modelling n open-economy SVAR. The previous sudy of monery policy effecs in Mlysi, for exmple Azli nd Mhews (999) nd Ibrhim (25) hve used smll-scle VAR in closed-economy, which is no consider he role of foreign vribles in heir nlysis. Therefore, i is essenil o exmine he foreign shocks effecs on mcroeconomic flucuions nd monery policy becuse of he Mlysin economy is relively smll nd highly rde-dependen 2. Second, his sudy During he ineres re rgeing, monery policy in Mlysi cn be cegorized ino hree min evoluions. Firsly, from November 995 up o Sepember 998, he BNM hs inroduced new Bse Lending Re (BLR) frmework which kes ino ccoun he 3-monh iner bnk re in he BLR formul. Secondly, since Sepember 998, he BNM hs employed ineres re rgeing wih fixed exchnge re, nd modified he BLR frmework king ino ccoun he Inervenion Re in he deerminion of BLR formul. A he sme ime, due o he currency crisis h occurred in he Es Asin region, he BNM implemened cpil conrols o sbilize he economy. Thirdly, since April 24, he BNM hs inroduced new ineres res frmework, he Overnigh Policy Re (OPR) o signl he monery policy snce. During his period, he BNM hs grdully liberlized cpil conrol, nd hs elimined he pegging wih he US dollr since July In Mlysi, he degree of economic inerdependence or economic openness s mesured by he shre of expors nd impors s percenge of GDP hs incresed significnly from percen in 97 o 2.59 percen in 98 nd percen in 99. In fc, since 999 i hs been greer hn 2 percen. These sisics indice Persidngn Kebngsn Ekonomi Mlysi ke VIII (PERKEM VIII) Dsr Awm Dlm Er Trnsformsi Ekonomi: Cbrn dn Hluju Johor Bhru, 7 9 Jun 23

2 6 Zulkefly Abdul Krim employs n open economy srucurl VAR model, which permis n idenificion sregies bsed on economic heory rher hn he someimes quesionble ssumpions which underlie rdiionl recursive VAR. In Mlysin conex, lhough Tng (26) hs considered recursive VAR in openeconomy, nd Ibrhim (25) hs used recursive VAR in closed-economy, however heir idenificion of srucurl shocks is inpproprie. Thus, his sudy provides novel conribuion o he monery policy nlysis in smll-open economy (i.e. Mlysi) by idenifying he srucurl shocks ccording o he economic heory. The resuls of he sudy indice h he significn role of monery policy s sbilizion policy during ineres res rgeing in simuling mcroeconomic vribles, nd miiging he negive effecs of he exernl shock o he domesic economy. Foreign shocks ppered o ply prominen role in influencing domesic mcroeconomic nd monery policy vribles. There is lso no evidence for oupu puzzle, nd exchnge re puzzle. However, here is price nd liquidiy puzzle during he ineres res rgeing. The res of he pper is orgnized s follows. Secion 2 presens lierure review reling o monery policy idenificion scheme in he SVAR lierure. Secion 3 briefly discusses he reserch mehodology, nd Secion 4 presens he empiricl resuls by focusing on srucurl impulse-response funcion (SIRF). Secion 5 summrizes nd concludes. LITERATURE REVIEW There re limied number of sudies in smll-open economies which exmined he effec of he monery policy shocks by using n open SVAR pproch. For exmple, he mos recen SVAR sudies of smll-open economy were conduced by Cushmn nd Zh (997), Brischeo nd Voss (999), Dungey nd Pgn (2), Prrdo (2), nd Buckle e l. (27). Mos of he sudies hve used block exogeneiy resricions in modelling he inernionl economic linkges o he smll-open economy. Cushmn nd Zh (997) nd Dungey nd Pgn (2) hve imposed wo blocks in heir srucurl equion model, which is block represening he inernionl economy, nd block represening he domesic economy. In modelling SVAR for he Cndin economy, Cushmn nd Zh (997) included four inernionl vribles, nmely he US indusril producion, he US consumer prices, he US federl fund re, nd world ol commodiy expor prices. The min idenificion scheme in heir sudy hs hree folds. Firs, domesic ineres re is ssumed o rec conemporneously o foreign ineres re nd commodiy mrke, bu no on conemporneous oupu. Second, he exchnge re is ssumed o response conemporneously o ll shocks. Third, foreign vribles hve reed s sepre block, which is block (exogenous) for he domesic (smll open) economy. This mens h, domesic vribles re no llowed o ffec foreign vribles eiher conemporneously or wih lgs. The empiricl findings sed h Cndin monery policy responds significnly nd conemporneously wih home ineres res, exchnge re, foreign ineres res nd foreign price levels. In ddiion, here is no evidence of ineres res puzzles 3, exchnge res puzzles nd prices puzzles, nd he exernl shocks become he dominn source of domesic oupu flucuions. Monery policy shocks (n increse in ineres res) hs smll bu negive effec on oupu. In conrs, Dungey nd Pgn (2) included five inernionl vribles, nmely rel US GDP, rel US ineres res, he Ausrlin erm of rde, he Dow Jones Index defled wih he US consumer price index, nd rel expors. The Ausrlin monery policy (csh re) is ssumed o follow sndrd Tylor-rule, which is response conemporneously wih Ausrlin gross nionl expendiure nd inflion. The domesic vribles re ssumed cnno ffec foreign vribles eiher conemporneously or wih lgs. The min findings indiced h overses fcors re generlly subsnil conribuor o domesic civiy, nd domesic monery policy conribues o sbilize economic civiy, bu he effec is no lrge. In he Mlysin conex, here hve been few sudies reling o he effec of monery policy shocks on economic civiies in he exising lierure. For exmple, Azli nd Mhews (999) hve employed Bernnke s (986) conemporneous srucurl VAR pproch (six vribles in closedeconomy) in invesiging he relionship beween money-income nd credi-income during he prend he pos-liberlizion ers. They found h, during he pre-liberlizion period, he bnk credi h he Mlysin economy is highly dependen, nd hus vulnerble o exernl shocks, for exmple foreign income shocks from lrge counries. 3 The ineres res puzzle or oupu puzzle is monery conrcion by n increse in ineres res is ssocied wih n increse in oupu level rher hn decrese.

3 Prosiding Persidngn Kebngsn Ekonomi Mlysi Ke VIII 23 6 shock hd more impc compred wih he money shock in explining oupu vribiliy. In conrs, fer he pos-liberlizion period, money s well s credi innovions were significn in explining oupu shocks. In shor, ggrege demnd ws significnly influenced by credi innovion during preliberlizion, while money innovion plyed n imporn role during pos-liberlizion in explining oupu vribiliy. Anoher sudy by Ibrhim (25) used recursive VAR idenificion in closed economy in exmining he secorl effec of monery policy shock. His resuls suppored he rel effec of monery policy shocks. For exmple, i ws seen h rel oupu declined during monery ighening (posiive shocks of ineres res). In fc, some secors such s mnufcuring, consrucion, finnce, insurnce, rel ese, nd business services seem o decline more hn ggrege producion in responding o he ineres res shocks. In shor, hose secors h re very hevily dependen on bnk lons re more sensiive o monery policy shocks. In comprison, recen sudy by Tng (26) hs exmined he relive impornce of he monery policy rnsmission mechnism chnnel (ineres res, credi, sse price, nd exchnge re chnnel) by using 2 vribles in n open-economy VAR model. The vribles re four foreign vribles (foreign block), nd eigh domesic vribles (domesic block), which is esimed by using recursive VAR idenificion scheme. The foreign block is ssumed no compleely exogenous o he domesic block, which re he domesic vribles re llowed o ffec he foreign vribles in lgs bu no conemporneously. His finding concludes h he ineres res chnnel plys pivol role in influencing oupu nd inflion. In ddiion, he sse price chnnel is lso relevn for explining oupu vribiliy, bu for inflion, he exchnge re chnnel is more relevn hn he sse price chnnel. RESEARCH METHODOLOGY D nd Vribles Descripion All d re monhly frequency spnning from November 995 unil My 29 (during ineres res rgeing regime), nd colleced from he IMF s Inernionl Finncil Sisic (IFS), excep for sse price, where he d re colleced from Thompson Dsrem. All vribles re rnsformed ino logs excep for FFR, INF nd IBOR, which re sed in percenge poins. Specificlly, he endogenous vribles included in he VAR re; Oil Price (LOIL ) I refers o he log of world oil prices (US $ per brrel). The world oil price is n indicor of dverse supply shocks (inflionry nd conrcionry) o he Mlysin economy. Therefore, i is expeced h posiive shocks in he world oil prices will increse he cos of producion, nd subsequenly will rigger cos-push inflion nd reduced invesmen spending, subsequenly leding o conrcion in economic civiy. Foreign Income (IPIUS ) This sudy uses he US Indusril Producion Index s proxy for foreign income. The selecion of US income s proxy for foreign income is resonble since he US is he mjor rding prner of Mlysi. For insnce, on verge, from 997 o 28, expors o he US hve consiued 2 percen of he ol Mlysin expors, mos of which hve concenred on mnufcuring producs such s elecricl nd elecronic goods nd exiles, clohing nd foower 4. Foreign Monery Policy (FFR) The Federl Fund Re of he US is used s proxy for foreign monery policy. FFR is lso n pproprie indicor in nlyzing he effec of foreign monery policy on smll-open economy given h he US is lrge economy nd hs powerful impc in he inernionl ren. Any chnges in he FFR will signl he condiion of he US economy, nd subsequenly is expeced o rnsmi o he monery policy snce nd mcroeconomic flucuions in smll-open counry. 4 This figure is bsed on he uhor s clculion from he Bnk Negr Mlysi, Monhly Sisicl Bullein. Besides he US, he oher Mlysin mjor rding prners re ASEAN counries (priculrly Singpore), Jpn nd Europen Union. For insnce, in 28 he US, Singpore, Jpn nd Europen Union hve conribued percen of he ol Mlysin expors.

4 62 Zulkefly Abdul Krim Domesic Income (IPIM ) Domesic income is rge vrible, which is proxied by he Mlysin Indusril Producion Index. In Mlysi, he Indusril Producion Index (IPI) is good proxy for oupu becuse i comprises he hree min indices in he economy; he mnufcuring index (weighed 7.7 %), mining index (weighed 23.4 %), nd elecriciy Index (weighed 6.%). Inflion (INF ) Anoher rge vrible is he inflion re (INF) which is clculed from he chnges in he Consumer Price Indices (CPI) 5. The inclusion of INF in he model is resonble given h he BNM hs been rgeing he inflion re s n objecive since 99. Monery ggrege (LM ) The nrrow monery ggrege M is lso considered s monery policy vrible becuse M is more liquid, nd hus demnded for rnscions purposes. The inclusion of M is lso consisen wih previous sudies, for exmple Tng (26) who sed h M is he bes cndide for monery ggrege becuse i is relively sble nd more liquid compred o he brod monery ggrege. Ineres Res (IBOR) The iner bnk overnigh re (IBOR) is lso good cndide for monery policy vrible becuse he BNM hs direcly influenced he iner bnk re hrough is inervenion in he money mrke (Domc, 999). In fc, since April 24, he BNM hs inroduced new monery policy frmework, nmely he Overnigh Policy Re (OPR) o signl he monery policy snce. Kul Lumpur Composie Index (KLCI ) Kul Lumpur Composie Indices (KLCI) is lso imporn for nlyzing he effec of he monery policy on he sse prices, nd cn exmine how he economic civiy responds o he sock mrke developmen. 6 For insnce, n esing of monery policy hrough incresing he moun of he money supply (or reducion in ineres res) leds o n increse in he sock mrke (sse price), nd subsequenly will simule he economic civiy vi wo chnnels - welh effec nd Tobin s-q effec. Nominl Effecive Exchnge Re (NEER ) Besides sse prices, he role of exchnge re is lso imporn given h Mlysin is highly rdedependen economy, nd ny chnges in exchnge re will significnly ffec he exernl secors, priculrly expor nd impor. In ddiion o he endogenous vribles, he model lso includes wo dummy vribles for breks in he inercep, which coincide wih mjor economic evens. Specificlly, he evens re: he period in which he Ringgi ws pegged o he US dollr, nd he Asin finncil crisis. I is ssumed hese wo evens only influence vribles in he domesic block rher hn he foreign block. In ddiion, he sesonl dummy hs lso considered for ll endogenous vribles. Time Series Properies Much of he SVAR lierure in monery policy hs ignored possible non-sionriy of he vribles in he VAR model 7. However, ignoring non-sionriy in he VAR sysem is inpproprie becuse only sionry vribles reurn o he long-run men fer shock. In his sudy, he Augmened Dickey-Fuller (ADF) es ws employed o invesige he presence of uni roos in he d. Afer esing for uni roos, he opimum lg in he VAR model is chosen by using he Akike informion crieri (AIC). 5 The sndrd formul o compue he inflion re is CPI CPI. CPI 6 An excellen lierure survey bou he effec of monery policy on sock prices cn be found in Sellin (2). 7 Rmswmy nd Sloek (997) hve discussed exensively wheher he VAR model should be esimed in level, difference or in vecor error correcion model (VECM).

5 Prosiding Persidngn Kebngsn Ekonomi Mlysi Ke VIII Mlysin Srucurl VAR Modelling The endogenous vribles re divided ino wo blocks; nmely foreign nd domesic. The foreign block includes he world oil price, foreign income, nd foreign monery policy mesure. The domesic block includes rge vribles (domesic oupu nd inflion), policy vribles (monery ggrege nd ineres res), nd wo oher vribles nmely sse prices nd he exchnge re. I is ssumed h smll-open economy like Mlysi is described by srucurl form represenion. The dynmic relionship of he sysem of equion in he srucurl model cn be wrien s follows; A Y D A( L) Y () Where, A is n inverible squre mrix of coefficiens reling o he srucurl conemporneous inercion beween he vribles in he sysem, Y is 9 mrix h is he vecor of sysem vribles or LOIL LIPIUS FFR LIPIM INF LM IBOR LNEER LKLCI, D is vecor of deerminisic vribles (consn, rend nd dummy vribles), A(L) mrix polynomil in he lg operor L is h k order A ( L) A L A L 2... A k 2 k L, LOIL LIPIUS FFR LIPIM INF LM IBOR LNEER LKLCI is he vecor of srucurl shocks which sisfies he condiions h E ( ), E( ' s ) I (ideniy mrix] for ll s. Equion () cnno be direcly esimed o derive he rue vlue of A, A L nd.however, equion (2.) cn be esimed by rnsforming o he reduced form represenion s follows; Y A D A A( L) Y A (2) or Y ) D ( L Y (3) Where, A, A A( L), A ' nd E( ) ' A A. The soluion o he SVAR sysem cn be genered by recovering he relionship beween he reduced-form disurbnces ( ), nd he underlying srucurl shocks ( ). This relionship cn be esimed by using he equion (3), which is or A u using he mximum likelihood esimes. In mrix form, his relionship cn be represened s follows; A LOIL LIPIUS FFR LIPIM INF IBOR LM u LNEER LKLCI LOIL LIPIUS FFR LIPIM INF IBOR LM LNEER LKLCI (4) In idenifying he srucurl VAR model, his sudy employed he SVAR-A model proposed by Amisno nd Ginnini (996). Enough resricions hve o be imposed on mrix A. According o

6 64 Zulkefly Abdul Krim he order condiion, for he sysem o be jus idenified or excly idenified, i requires K K /2 9 8 /2 36 zero resricions on he conemporneous mrix A. However, since he conemporneous mrix A in equion (4) hs 4 zero resricions, he SVAR model is over-idenified. Idenificion Scheme The vribles in foreign blocks hve been ssumed o be compleely exogenous o he domesic blocks. I is common o idenify h he foreign vribles do no respond conemporneously or wih lgs o he movemen in he domesic vribles in smll-open counry. Specificlly, he world oil price is srucurl disurbnce or exogenous vrible h is uncorreled wih oher conemporneous shocks. Menwhile, he US income nd Federl Fund Re (FFR) cn influence world oil price in lgs. The US income is ssumed o respond conemporneously wih he world oil price nd ll foreign vribles lg. The US monery policy (Jensen nd Mercer) is ssumed o respond conemporneously o he innovions in world oil price nd he US income nd ll foreign vribles lg. The FFR recs posiively o he world oil price in minimizing he inflionry pressure due o he posiive shocks of world oil price. This ssumpion is lso consisen wih prior sudies, for exmple, Hmilon (996) nd Bernnke e l. (997) who found h oil price movemens hve significn forecs power for he snce of monery policy in he US economy. In conrs, FFR responds negively wih he US income in minimizing he oupu gp. Domesic oupu growh hs been ssumed o respond conemporneously only o he world oil price shocks, nd respond in he lgs o ll endogenous vribles. Domesic oupu growh recs posiively o he curren growh of world oil price. This ssumpion is resonble given h Mlysi is ne exporer of oil. However, in he long run, he relionship beween oil price nd oupu is expeced o be negive becuse n increse in world oil price will increse he cos of producion, wheres he firms will respond by cuing he level of producion or invesmen. In ddiion, i is ssumed h Mlysin oupu does no respond conemporneously wih oher vribles in he sysem. For insnce, he monery policy vrible, h is, monery ggrege nd ineres res, cnno influence oupu conemporneously. The min plusible jusificion for his ssumpion is h firms do no chnge heir oupu nd prices insnneously in responding o he monery policy signl wihin monh, due o he ineri, djusmen coss nd plnning dely, bu hey will respond immediely o he curren oil prices following heir mrk-up rule (Kim nd Roubini, 2). This ype of resricion is lso imposed by Bernnke nd Blinder (992), nd Bernnke nd Mihov (998). Domesic inflion hs been ssumed o respond conemporneously o he innovions in oil prices nd domesic oupu. The posiive innovion in oil prices nd domesic oupu will sponneously ccelere he domesic price level. However, oher vribles in he sysem cnno influence domesic inflion sponneously becuse inflion is slow-moving vrible. However, ll endogenous vribles re ssumed o ffec he inflion re in he lgs. In mrke economies, he use of he ineres res s mjor insrumen of monery policy does no imply h i cn ignore he role of money supply. This is becuse he ineres res re deermined in finncil mrkes. For exmple, if he monery uhoriy wns o lower he ineres res bu no supporing wih required increse in he money supply, i would find h he mrke ineres res would devie from is desired level. As resul, he inended effecs on expendiures will no be chieved. Therefore, n ineres re policy mus be ccompnied by n pproprie money supply (Hnd, 29). For h reson, his sudy will consider wo ypes of monery policy shocks, nmely, money supply nd ineres res. During ineres res rgeing, i is ssumed h he monery uhoriy ses he ineres res fer observing he curren vlue of domesic vribles h re domesic oupu nd inflion, nd ll foreign vribles, bu no he curren vlue of oher vribles. The inclusion of oupu nd inflion in he monery policy recion funcion is resonble given h he cenrl bnk cn observe hese d on monhly bsis. The inclusion of foreign monery policy in he domesic monery policy recion funcion is resonble given h he Mlysin economy is highly dependen on he US economy. Therefore, i is resonble o ssume h he BNM will response posiively o he US monery policy in minimizing he cpil ouflow s well s sbilizing he domesic currency. The jusificion of he monery policy recion funcion is lso consisen wih previous sudies, for exmple, Kim nd Roubini (2), Cushmn nd Zh (997), nd Sim nd Zh (25). I is ssumed h he monery uhoriy (he BNM) ses he money supply fer observing he curren level of ll foreign vribles, domesic oupu growh, inflion, nd ineres res. This is resonble given h monery uhoriy cn observe he d on monhly bsis nd chooses he moun of money supply o offse ny negive effec resuling from foreign shocks, domesic ggrege demnd nd inflion shocks o he domesic economy.

7 Prosiding Persidngn Kebngsn Ekonomi Mlysi Ke VIII The exchnge re is ssumed o be respond conemporneously wih ll foreign nd ll domesic vribles (excep sock prices). In conrs, he sse price is ssumed o respond conemporneously o ll foreign nd ll domesic vribles. This ssumpion is resonble given h exchnge re nd sock mrke re he fs-moving vribles in he sysem. The sse price is ssumed o respond conemporneously o he exchnge re becuse ny chnges in he exchnge re will influence inernionl cpil mobiliy, which in urn ffecs he sock mrke. For insnce, n ppreciion in domesic currency mkes domesic sses more expensive o he inernionl invesors, nd herefore decreses he demnd for domesic sse, subsequenly leding o decline in he sse price. EMPIRICAL RESULTS Preliminry Anlysis Tble repors he resul of he uni roo es by using he Augmened Dickey-Fuller es. As cn be seen, only hree vribles, which re he FFR, inflion (INF), nd iner bnk overnigh re (IBOR) re sionry level form les percen significnce level; menwhile oher vribles re sionry in firs difference form 8. The opimum lg in he VAR sysem is four ccording o he Akike informion crierion (AIC). Since he bseline SVAR model is over-idenified, we need o es he vlidiy of he over-idenificion resricions. The vlue of 2 (wih four degrees of freedom) is 7.97 nd he probbiliy is., which indices h he over-idenificion resricions re vlid 9. Srucurl Impulse Response Funcion (SIRF) Figures -Figure 3 illusres he srucurl impulse response funcions of he endogenous vribles in his sudy. The min focused is o nlyze he foreign shock effecs on domesic vribles (monery policy nd mcroeconomic vribles), he effec of domesic monery policy shocks (monery ggrege nd ineres res) on domesic mcroeconomics vribles, nd how effecive is monery policy in miiging he negive effecs of exernl shocks on domesic mcroeconomics. The solid line represens he esimed responses; menwhile he wo dshed lines represen he confiden bnds or error bnds. The error bnds of he SIRF re derived from Hll s boosrpping mehodology, which hs 68 percen confidence inervl wih 3 being he number of boosrp replicions. Accumuled impulse-response funcions will be discussed for he firs difference vribles responses, while he usul IRF will be used for he level form vrible responses. By ccumuling he responses of he firs-difference vrible on is srucurl shocks, we cn inerpre he impc of srucurl shocks on he level form of endogenous ( Y ) vribles. Foreign Shocks Effecs on Domesic Vribles Response of domesic monery policy Pnel A in Figure shows he response of ineres res o he innovion in foreign shocks, nd Pnel B describes he ccumuled response of money supply o he innovion in foreign shocks. As cn be seen in Figure (Pnel A), he response of he ineres res o he innovion in world oil price growh is negive, decying fer 3 monhs. The ccumuled response of money supply hs posiive effec o he innovion in world oil price growh. This findings indice h he monery uhoriy concern he negive effecs of world oil price shock on domesic oupu, nd herefore implemening n esing monery policy hrough decrese in ineres res, nd n increse in money supply. This sregy is crucil o offse he dverse supply shocks on economic civiy hrough simuling he ggrege demnd. 8 According o Blnchrd nd Quh (989), Gli (992), nd Bjørnlnd nd Leiemo (29) he level nd difference form vribles cn be combined in he VAR sysem s long s he sysem is sionry. An excellen survey bou he issue of sionry sysems, non-sionry I() sysems, nd mixure of I() nd I() vribles in he SVAR model cn be found in Levchenkov e l. (998). 9 The full resul of SVAR conemporneous coefficien esimion is vilble upon reques. The SVAR model hs been esimed by using J-Muli sisicl sofwre developed by Lukepohl nd Krzig (24).

8 66 Zulkefly Abdul Krim Ineres re responds significnly posiively o foreign income shocks wih he highes response being.3 percenge poins wihin 25 monhs. However, fer 25 monhs he posiive response of he ineres re begins o reduce nd decys fer 6 monhs. Money supply responds negively o foreign income shocks. For insnce, n ccumuled response of money supply in response o one percen increse in foreign income growh is ppropriely of -.6 percen fer 25 monh period. One possible explnion is h he monery uhoriy hs been concerned wih sbilizing he domesic price level, since he expnsion of foreign income hs riggered n increse in domesic inflion due o n increse in exernl demnd. The domesic ineres re hs responded posiively o he FFR shocks, wih he highes response being.5 percenge poins chnge in 5 monhs. Money supply hs negively responded o he posiive innovion in FFR. This finding indices h monery uhoriy is concerned bou he cpil inflow from domesic counry o he US due o he monery ighening in he US. Therefore, o ensure h porfolio invesmen in Mlysi is compeiive, he BNM hs o respond by incresing he domesic ineres res. Response of domesic mcroeconomic vribles Figure 2 in Pnel A-Pnel C describes he response of domesic mcroeconomic vribles o he innovion in foreign shocks. As depiced in Figure 2 in Pnel A, posiive innovion in world oil prices growh led o n increse in domesic oupu in wihin wo monh. However, fer four monhs he posiive innovion of he world oil price growh leds o decline in domesic oupu. There is no significn effec of world oil price o domesic oupu fer 2 monhs. The posiive response of oupu wihin wo monh period is resonble given h Mlysi is ne oil exporer counry. An increse in world oil price hs genered higher income for he peroleum indusry, nd subsequenly leds o n increse in domesic oupu. However, fer four monhs, oupu responds negively o he world oil price becuse of he dverse supply shocks. For exmple, n increse in he world oil price will led o n increse in firms producion coss, nd subsequenly he firms will respond by conrcing heir invesmen spending. Domesic inflion lso posiively responds o he posiive shocks in he growh of world oil price. For insnce, he highes effec is recorded in 5 monh period, which poin every percen increse in world oil prices growh leds o n increse in he inflion re of.25 percen. However, fer 5 monhs he response of inflion hs grdully declined, nd vnishes wihin 4 monhs. The exchnge re responds posiively nd significnly o he posiive shock in he world oil price growh, nd he ccumuled effec is pproximely.6 percen in 6 monhs. In oher words, n increse in he world oil price riggers n ppreciion in he domesic currency relive o oher currencies becuse Mlysin is ne exporer of oil. The sock mrke responds posiively o he posiive shock in he world oil price growh, nd decying fer 4 monhs. Pnel B in Figure 2 plos he response of domesic mcroeconomic vribles o innovion in foreign income (US oupu). As cn be seen, domesic oupu growh posiively responds o he foreign demnd shock. The posiive response of domesic income is resonble since mos of he Mlysin expors re concenred in he US mrke. The effec of he US income shock o domesic inflion is negive wihin 2 monhs, wheres fer 2 monhs domesic inflion hs responded posiively. This indiced h fer 2 monhs, foreign income shocks cn influence domesic inflion hrough n increse exernl demnd, which is subsequenly simuling he ggrege demnd. A posiive shock in he US income hs cused he exchnge re o respond negively wihin 25 monhs; however, fer 25 monhs he response is posiive. The posiive response of exchnge re (ppreciion in domesic currency) occurs becuse n increse in foreign income hs simuled domesic expors, nd ferwrds hs incresed he demnd for domesic currency. Asse price hs responded posiively o he posiive innovion in he foreign income, which is incresed by.8 percen wihin 6 monhs period. Pnel C in Figure 2 shows he effec of foreign monery shocks on domesic mcroeconomic vribles. As cn be seen, he ccumuled response of domesic oupu is posiive wihin 5 monh period; however, he effec is relively smll. Afer 5 monhs, he ccumuled response of domesic oupu is being negive. For exmple, wihin 2 monhs period; he ccumuled response of domesic oupu hs decresed by.4 percen in responding o one percenge poin increse in FFR. A possible explnion is h n increse in he foreign ineres re hs conrced foreign economic civiy, which ferwrds decresed foreign demnd, nd subsequenly conrced he domesic economy. Domesic inflion hs responded negively o posiive innovion in FFR fer 2 monhs, nd decying fer 25 monhs. For exmple, in monhs, domesic inflion hs decresed by. percen in responding o one percenge poin increse in foreign monery policy.

9 Prosiding Persidngn Kebngsn Ekonomi Mlysi Ke VIII The ccumuled response of exchnge re o he innovion in FFR is posiive. A possible reson is h he BNM hs incresed he domesic ineres res o offse he cpil ouflow o he foreign counry. The ccumuled response of sock mrke o he FFR shocks is posiive nd relively smll in 5 monhs, which is indiced h foreign monery policy is no imporn in influencing he domesic sock mrke. However, fer 7 monhs, he sock price hs responded negively wih he posiive innovion in FFR. Monery Policy Effecs on Domesic Vribles Ineres Re Shocks Figure 3 in Pnel A indices he effec of he ineres re shock on domesic mcroeconomic vribles. As cn be seen, domesic oupu growh hs responded negively o posiive shock in ineres res. For insnce, he ccumuled response of domesic oupu is -.3 fer 2 monhs, which is indiced h one percenge poin increse in ineres res (IBOR) leds o conrcion in domesic oupu of.3 percen. The negive response of domesic oupu o monery policy ighening cn be explined by sndrd economic heory, by which n increse in ineres res increses he firms cpil coss, while in urn he firms will respond by reducing heir cpciy of fixed invesmen. The response of inflion o monery policy ighening is posiive, which indices he exisence of price puzzle. Brh nd Rmey (2) hve provided n lernive explnion of he price puzzle, where hey rgued h monery policy ighening operes on ggrege supply s well s ggrege demnd. Specificlly, n increse in ineres res rises he cos of holding invenories (cos chnnel of monery policy). This negive supply effec rises domesic inflion nd lowers oupu. Money supply hs responded negively o posiive shock in ineres res, where he ccumuled effec is -.2 percen fer 2 monhs. The exchnge re responds posiively o he posiive innovion in ineres re. For insnce, fer monhs, he ccumuled response of nominl exchnge re is.2, which is indiced h one percenge poin increse in ineres res leds o pprecie he nominl exchnge re by.2 percen. A possible reson is h n increse in ineres res hs simuled he cpil inflows o he domesic counry, nd subsequenly incresed he demnd for domesic currency, nd fuel exchnge re o pprecie. The sock mrke lso responds posiively wihin 5 monhs, however fer 5 monhs, sock price responses negively o he innovion in ineres res. One possible reson is h n increse in ineres res hs conrced economic civiy nd dividend pymens, which in urn decresed he sse price. Money Supply Shocks Figure 3 in Pnel B plos he money supply (nrrow money M) impulses on domesic mcroeconomic vribles. As cn be seen, in one monh, he ccumuled response of domesic oupu is negive in response o posiive innovion in money supply growh. However, fer one monh, here is posiive cumulive effec of he shocks in money supply growh on domesic oupu. For insnce, he highes cumulive effec is shown in 3 monhs, where one percen increse in money supply growh leds o n increse in domesic oupu by pproximely.3 percen. However, fer five monhs, he posiive ccumuled response of domesic oupu hs decresed. Specificlly, fer 5 monhs, he ccumuled response of domesic oupu is. percen. This finding indices h money supply plys n imporn role in simuling domesic oupu growh. The response of inflion on money supply growh shows hump-shpe pern, nd reurns o he equilibrium ph fer 5 monhs. Ineres res respond posiively o he innovion in money supply growh, which indices he exisence of liquidiy puzzle. Some empiricl sudies, for exmple Reichensein (987), nd Leeper nd Gordon (99) hve suppored he exisence of he liquidiy puzzle. In ddiion, he posiive effec of ineres res grdully decreses nd vnishes fer 3 monhs. The effec of money on he exchnge re is no sisiclly significn. In comprison, he sock mrke hs posiive response o money supply shocks, which indices h n expnsionry monery policy hs simuled he sse price o increse.

10 68 Zulkefly Abdul Krim Robusness Checking For robusness checking, severl lernive procedures hve been considered in his sudy; (i) esiming he recursive SVAR model; (ii) lernive conemporneous srucurl idenificion schemes; (iii) model wih money demnd; nd (iv) model wihou money. Overll, he srucurl IRF re robus wih he bseline resricion. SUMMARY AND CONCLUSIONS The focl poin of his sudy is o exmine he role of monery policy during ineres res rgeing in smll-open economy (i.e. Mlysi) by using n open-economy SVAR sudy. Specificlly, his sudy invesiges he effec of n exernl shock, nmely he world oil price nd he US inernionl rnsmission (US income nd US monery policy) o domesic monery policy nd mcroeconomic flucuions. In ddiion, his sudy exmines he effeciveness of domesic monery policy in simuling he mcroeconomic vribles (oupu nd inflion). Severl mjor conclusions cn be drwn from he empiricl findings. Firs, he world oil price shock hs genered significn effec in influencing domesic monery policy implemenion nd mcroeconomic vribles in he Mlysin economy. Second, he US rnsmission in erms of he US income nd he US monery policy shocks hve lso plyed prominen role in influencing domesic monery policy nd mcroeconomic flucuions. Third, domesic monery policy, eiher money supply or ineres res hve plyed significn role in sbilizing economic civiy (oupu nd inflion). These findings hve severl implicions for domesic monery policy implemenion. Firs, since he foreign shocks cn influence domesic mcroeconomic flucuions, he monery uhoriy (in priculr, he BNM) hs o mke n ccure ssessmen bou he exernl evens by monioring inernionl condiions in implemening monery policy. This is very imporn for minimizing he negive effec of he foreign shocks on he mcroeconomic flucuions hrough procive monery policy mesures. Second, here is relevn role for monery policy insrumens (money supply nd ineres res) in simuling mcroeconomic vribles, in which suggess h monery policy cn be used s sbilizion policy. Third, during he ineres re rgeing regime, domesic monery policy hs posiively responded o he foreign monery policy shocks. I seems h he BNM hs o follow he US monery policy. This sregy is crucil under perfec cpil mobiliy o minin he compeiiveness of he domesic finncil sses, nd i cn sbilize he domesic currency. As noed by Umizki (27), he BNM no only hs o consider he domesic fcors such s inflion res nd oupu bu lso foreign vribles, priculrly foreign monery policy nd exchnge re, in designing n pproprie monery policy rule. Fourh, since domesic oupu hs responded negively o ineres res, he BNM hs greer opporuniy o simule economic civiy by conrolling he iner bnk overnigh ineres res. Finlly, he monery policy lso plys n imporn o sbilize economic civiy vi he exchnge re nd sock mrke effecs. For exmple, he cenrl bnk cn simule he expor secor by implemening n esing monery policy (eiher by n increse in money supply or decrese in ineres res). This sregy will deprecie domesic currency, nd subsequenly simule exernl demnd nd economic civiy. REFERENCES misno, G. & Ginnini, C Topics In Srucurl Vr Economerics, Berlin, Springer. Azli, M. & Mhews, K Money-Income And Credi-Income Relionships During The Pre- And The Pos-Liberlizion Periods: Evidence From Mlysi. Applied Economics, 3, 6-7. Brh, M. J. & Rmey, V. A. 2. The Cos Chnnel Of Monery Trnsmission. Nber Working Pper Series. Cmbridge: Nionl Bureu Of Economic Reserch. Bernnke, B. S Alernive Explnions Of The Money-Income Correlion. Crnegie- Rocheser Conference Series On Public Policy, 25, 49-. I do no repor he full resul of robusness checking o sve spce. However, he full resuls re vilble upon reques.

11 Prosiding Persidngn Kebngsn Ekonomi Mlysi Ke VIII Bernnke, B. S. & Blinder, A. S The Federl Funds Re And The Chnnel Of Monery Trnsmission. The Americn Economic Review, 82, Bernnke, B. S., Gerler, M. & Wson, M Sysemic Monery Policy And The Effecs Of Oil Shocks. Brookings Ppers On Economic Aciviy,, Bernnke, B. S. & Mihov, I Mesuring Monery Policy. The Qurerly Journl Of Economics, 3, Bjørnlnd, H. C. & Leiemo, K. 29. Idenifying The Inerdependence Beween Us Monery Policy And The Sock Mrke. Journl Of Monery Economics, 56, Blnchrd, O. & Quh, D The Dynmic Effecs Of Aggrege Demnd And Supply Disurbnces. Americn Economic Review, 79, Brischeo, A. & Voss, G A Srucurl Vecor Auo Regression Model Of Monery Policy In Ausrli. Reserch Discussion Pper, Reserve Bnk Of Ausrli. Buckle, R. A., Kim, K., Kirkhm, H., H., M. & Shrm, J. 27. A Srucurl Vr Business Cycle Model For A Volile Smll Open Economy. Economic Modeling, 24, Cushmn, D. O. & Zh, T Idenifying Monery Policy In A Smll Open Economy Under Flexible Exchnge Res. Journl Of Monery Economics, 39, Domc, I The Disribuionl Consequences Of Monery Policy : Evidence From Mlysi. Policy Reserch Working Pper The World Bnk, Es Asi And Pcific Region. Dungey, M. & Pgn, A. 2. A Srucurl Vr Model Of The Ausrlin Economy. The Economic Record, 76, Fuller, W. A Inroducion To Sisicl Time Series, New York, John Wiley. Gli, J How Well Does The Is-Lm Model Fi Poswr U.S D?. The Qurerly Journl Of Economics, Hmilon, J This Is Wh Hppened To The Oil Price Mcro Economy Relionship. Journl Of Monery Economics, 38, Hnd, J. 29. Monery Economics, London And New York, Rouledge, Tylor & Frncis Group. Ibrhim, M. 25. Secorl Effec Of Monery Policy: Evidence From Mlysi. Asin Economic Journl, 9, Jensen, G. R. & Mercer, J. M. 22. Monery Policy And The Cross-Secion Of Expeced Sock Reurn. The Journl Of Finncil Reserch, Xxv, Kim, S. & Roubini, N. 2. Exchnge Re Anomlies In The Indusril Counries: A Soluion Wih A Srucurl Vr Approch. Journl Of Monery Economics, 45, Leeper, E. M. & Gordon, D. B. 99. In Serch Of The Liquidiy Effec. Journl Of Monery Economics, 29, Levchenkov, S., Pgn, A. R. & Roberson, J. C Shocking Sories. Journl Of Economic Survey, 2, Levy, J. & Hlikis, I Aspecs Of Monery Trnsmission Mechnism Under Exchnge Re Trgeing: The Cse Of Frnce. Imf Working Pper Series. Ludvigson, S., Seindel, C. & Leu, M. 22. Monery Policy Trnsmission Through The Consumpion-Welh Chnnel. Frbny Economic Policy Review, Lukepohl, H. & Krzig, M. (Eds.) 24. Applied Time Series Economerics, New York: Cmbridge Universiy Press. Prrdo, E. 2. Effecs Of Foreign And Domesic Monery Policy In A Smll Open Economy : The Cse Of Chile. Working Ppers, No 8.: Cenrl Bnk Of Chile. Rmswmy, R. & Sloek, T The Rel Effecs Of Monery Policy In The Europen Union: Wh Are The Differences?. Imf Working Pper, 97/6. Rmey, V. A How Imporn Is The Credi Chnnel In The Trnsmission Of Monery Policy? Nber Working Pper Series. Msschuses Avenue, Cmbridge. Reichensein, W The Impc Of Money On Shor-Term Ineres Res. Economic Inquiry, 25, Sellin, P. 2. Monery Policy And The Sock Mrke: Theory And Empiricl Evidence. Journl Of Economic Surveys, 5, Sim, C. A. & Zh, T. 25. Does Monery Policy Genere Recession?. Mcroeconomic Dynmics, 4, -42. Tng, H. C. 26. The Relive Imporn Of Monery Policy Trnsmission Chnnels In Mlysi. Cm Working Pper Series. The Ausrlin Nionl Universiy. Umizki, S. 27. Monery Policy In A Smll Open Economy: The Cse Of Mlysi. The Developing Economics, Xlv(4),

12 7 Zulkefly Abdul Krim FIGURE : Srucurl Impulse-Response Funcion: The Effec of Foreign Shocks on Monery Policy Vribles

13 Prosiding Persidngn Kebngsn Ekonomi Mlysi Ke VIII 23 7 FIGURE 2 : Srucurl Impulse-Response Funcion: The Effec of Foreign Shocks on Mcroeconomic Vribles

14 72 Zulkefly Abdul Krim FIGURE 3 : Srucurl Impulse-Response Funcion: The Effec of Monery Policy Shocks on Domesic Vribles

15 Prosiding Persidngn Kebngsn Ekonomi Mlysi Ke VIII TABLE : Uni roo es : Augmened Dickey Fuller (ADF) Level Form Firs Difference Consn nd no Consn nd no Vribles rend Consn nd rend rend Consn nd rend LOIL -2.9 (2) (2) *** () -.66 *** () LIPIUS (2) -.87 (2) *** () *** () FFR *** (2) *** (2) ** (2) *** (2) LIPIM -.57 (2) (2) *** (2) *** (2) INF *** () *** () *** (2) *** (2) LM.433 () -.82 (9) *** (2) *** (2) IBOR * (6) * (7) *** (7) *** (7) LNEER (3) (3) *** (7) *** (2) LKLCI -2.8 (3) (3) -.79 *** (2) -.7 *** (2) Noe: *** Denoes significn he % level, ** significn 5% level nd * significn % level which rejec of he null hypohesis on non-sionry. Criicl vlue obin from Fuller (976) for consn bu no ime rend is -3.45, nd for %, 5% nd % significn level respecively, nd he criicl vlue for consn nd ime rend is -3.98, nd -3.3 for %, 5% nd % significn level respecively. Number in brcke is he opimum lgged bsed on Akike Informion Crierion (AIC).