A C T A U N I V E R S I T A T I S L O D Z I E N S I S FOLIA OECONOMICA 192, 2005
C h ristia n M e r k l* , L ú c io V in h a s d e S o u z a * *
T H E R U S S IA N C E N T R A L BANK AS A M O N E T A R Y T A R G E T E R ? AN E M P IR IC A L ANA LY SIS***
Abstract. The paper reviews the recent conduct o f monetary policy and the central bank’s rule-based behavior in Russia. Using different policy rules, we test whether the central bank in Russia reacts to changes in inflation, output gap and the exchange rate in a consistent and predictable manner. Our results indicate that during the period o f 1993-2002 the Bank of Russia has used monetary aggregates as a main policy instrument in conducting monetary policy.
Keywords: monetary policy rules, exchange rate, central bank, Russia. JKL Classification: E52, E61, F33, F41.
1. INTRODUCTION
T h e last 10 years have witnessed an upsurge in research o n m o netary policy rule evaluation, m otivated by the sem inal p ap e r o f T a y lo r (1993). F ollow ing this study, a great num ber o f researchers have investigated the F ederal R eserve’s b ehavior using either a sim ple T a y lo r rule o r som e simple v ariations thereof. O verall, for th e US o r o th er developed co un tries, the T ay lo r rule explains ra th e r well the behavior o f central banks. M o st o f the
* M .A. Institute for World Economic (IfW), Kiel University.
** Ph.D., Research Associate, Kiel Institute for World Economics, and Head o f the “Russia/Belarus Desk, DG-ECFIN, European Commission. The views expressed here are exclusively those o f the authors and do not necessarily reflect the official views o f the European Commission. All usual disclaimers apply.
*** We thank the suggestions o f Felix Hammermann, IfW, Thomas Kick, FAU Nuremberg, Franziska Schobert, Deutsche Bundesbank, Elena Rumyantseva, Central Bank o f Russia, Rainer Schweickert, IfW, Oleg Zamulin, NES, an anonymous referee. Furthermore we acknowledge the comments o f the participants of seminars held in Kiel at the IfW, at the Bank o f Finland, at the NES/CEFIR, Moscow, the FindEcon-conference at the University o f Łódź, the Deka-Bank workshop in Francfort, and the UACES conference in Birmingham. All usual disclaimers apply.
tim e they stabilize deviations either from a targ e t level in flation o r o u tp u t gap, using an interest ra te instrum ent.
H ow ever, in the case o f developing countries and em erging m ark ets, the findings o f m o n etary policy rule evaluations are som ew hat inconsistent, w ith results changing, depending upon tim e span and m odel specification (M o h a n ty and K la u 2003). T his can be explained by several facts: given the specific n atu re o f m ark ets in em erging econom ies, the ad e q u ate policy in stru m en t could no t only be the short-term interest rate, b u t also the m o n etary base (a M cC allum rule).
O ver the p ast few years a n um ber o f studies have investigated m onetary policy rules in em erging m arkets, finding th a t even w ith som e shortcom ings, central b an k s in em erging m ark e ts follow also som e rule-based m onetary policy, and th a t an open-econom y version o f the T ay lo r rule can describe m uch o f the v ariatio n in short-term interest rates (C ald ero n an d Schm idt- H ebbel 2003, M inella et al. 2003, M o h an ty and K lau 2003, T ay lo r 2001, T o rres G a rcia 2003).
It is, how ever, n o t clear w hether this applies to tran sitio n econom ies, w here financial m ark e ts are even less developed and w here the im plem en ta tio n o f a m oney-based m o n etary policy m ay face institution al problem s. B ecause o f even g re ater m odel specification difficulties and problem s associated with collecting reliable d a ta , very little research has been done on m o n etary policy rules in tran sitio n econom ies. T his study is one o f the first a ttem p ts to fill this gap, as it exam ines the co n d u ct o f m onetary policy in R ussia d u rin g the period o f 1993-2002. T h e em pirical estim ation o f altern ativ e rules fo r m o netary policy allows a test o f the statem en t th at in financially less developed econom ies, m o n e ta ry ta rg e tin g rules can prov ide an effective description of the behavior o f the m o n etary a u t horities - an d, in the case o f R ussia, o f its stated objectives (cf. T ay lor 2000).
T h e rest o f the p ap e r is organized as follows. Section 2 specifies different em pirical m odels to be used in evaluating m o n etary policy rules, while Section 3 presents the results o f o u r em pirical estim ations. F inally, Section 4 draw s som e conclusions.
2. SPECIFICATION OF THE EMPIRICAL MODEL
Since 1991 the R ussian econom y has experienced b o th sh arp fluctuations in m ain m acroeconom ic variables and deep stru ctu ra l changes. G iven this unstable natu re o f the economic environm ent in R ussia, the task o f estim ating a m o n etary policy rule is com plicated and no single policy rule equation
m ight fully ca p tu re all aspects o f the central b an k b eh av io r d u rin g this period. T h erefo re, we will estim ate d ifferent types o f rules, described below.
T h e recent literature on m onetary policy rules prim arily distinguishes tw o types o f instrum ent rules: interest ra te based in stru m en t rules and m o n etary based instrum ent rules, referred to as the T ay lo r rule and the M cC allum (1988) rule, respectively. T h e key difference in these rules involves the choice o f the instrum ent in central b a n k ’s reaction fu n ctio n in response to changes in m acroeconom ic conditions. W hile the T ay lo r rule, which uses a sh o rt-term nom inal interest ra te as an instrum en t, is widely used in m o n etary policy estim ations because o f its sim plicity, the M cC allum rule uses the grow th ra te o f m o netary base as an in stru m en t, which figured p rom inently in m o n etary policy form ulatio n before the 1990s.
O riginally, b o th rules were designed to be used in the ev alu atio n o f the m o n etary policy in large industrial countries, and m an y observers expressed concerns regarding the effectiveness o f this basic policy rules in evaluating the co n d u c t o f m o n etary policy in em erging econom ies. T h is concern raises the q u estio n as to w hat kind o f m odifications are needed to fit b etter the realities o f em erging econom ies, with underdeveloped financial m ark ets, dependence on prim ary com m odity exports, sh arp swings in productivity and relative prices, and high exposure to volatile capital flows.
T o ad dress adequately this question, researchers use m odified versions o f these instru m en t rules. O ne general consensus in this regard is th at m o n etary policy m akers in em erging econom ies are m o re concerned ab o u t exchange rate m ovem ents th a n those in m a tu re econom ies, am ong other reasons due to the degree o f exchange ra te p ass-thro ug h to prices. Hence, the exchange ra te has been incorp o rated , resulting in the open econom y version o f the central b a n k ’s reactio n function.
In his sem inal w ork, T ay lo r (1993) proposed the follow ing, now well- know n, policy rule to describe the F e d ’s behavior in setting th e sh o rt term interest rates:
(1) i = л + 0.5y + 0 .5 (я -2 .0 ) + 2.0,
w here i is the sh o rt term interest rate, n is th e inflation over the four previous q u arte rs, у is the percent deviation o f real G D P from a targ et (or “ o u tp u t g ap ” ). T h e inflation targ et and the equilibrium real interest ra te are set at 2.0 and assum ed as co n stan t over tim e. T h e “ policy m a k e r” is here assum ed to care, w ith equal weights, a b o u t d ev iatio n o f inflation and o u tp u t from target.
T his sim ple eq u atio n ca n n o t be estim ated in the original form in the case o f R ussia, since a relatively stable long-run average in flatio n does no t
exist. T h e only way to estim ate eq u atio n (1) is to assum e th a t there is a c o n sta n t intercept and estim ate the coefficients by ru n n in g a simple regression w ithout specifying the param eters o f the m odel (apart from inflation and o u tp u t gap). We calculated the o u tp u t gap by the trad itio n al H od rick- P rescott (H P ) filter.
F ollow ing T a y lo r (2001), we estim ate the m odified open econom y T ay lo r rule below where the lagged interest rate and th e exchange ra te are included to co n tro l for au to co rrelatio n problem s and the reaction o f the central b an k with regard to the exchange rate.
(2) i, = + + ß 2y, + ß 3xr, + /?4x r,_ t + ß i it - i + и„
where x t, is the grow th o f th e real effective exchange rate, u, is a white noise e rro r term and t — 1 indicates the past values o f th e variables. T he rem aining variables arc the same as in th e eq u a tio n (1). T h e expected signs o f the p aram eters are as follows: ß 0, ß 2, ß 5 > 0 , ß l (l — ß 5) > 1, ß 3 < 0, and /У4 < 0.
T h e M cC allum rule can be expressed as follows:
(3) A bt = Ax* — Av, + 0.5(Ax* — Ax,_ t ) + /*,,
w here Ab, is the rate o f grow th o f the m on etary base in percent per year, Ax* is the targ e t rate o f grow th o f nom inal G D P , in percent per year, Av, the ra te o f grow th o f base velocity, in p ercent p er year, and averaged over th e previous 4 years in the original M cC allum estim ation, and Ax is rate o f grow th o f nom inal G D P in percent per year. In this rule th e targ et value o f nom inal G D P grow th is calculated as the sum o f the targ e t inflation rate and the long-run average ra te o f grow th of real G D P .
Instead o f the m o n etary base as proposed by M cC allu m , we will use the m o n etary aggregate M l as a policy in stru m ent for m o n e ta ry policy in Russia (cf. Section 3 for further explanations), although there m ay be problems associated with the direct co n tro l o f this aggregate and with significant fluctuations in m oney velocity. We are also aw are o f the fact th a t som e existing studies attem p t to explain inflation dynam ics by the grow th of m o n etary aggregates (e.g. Pesonen and K o rh o n e n 1998, D ąb ro w sk i et al. 2003) using those as an explanatory variable. However, o u r G ran g er causality tests indicate th a t a t least in the sh o rt-ru n - u p to seven m o n th s - there is only a G ra n g e r causality from prices to m o n etary aggregates, and not the o th er way aro u n d .
It is widely accepted th a t the tim e series d a ta usually suffer som e level o f au to co rrelatio n , and if it is no t corrected the estim atio n results cann ot
be trea ted as reliable. T o correct for the au to co rrelatio n prob lem s, we will use differences ra th e r th a n levels and ad d several lags, acco rd in g to in fo rm atio n criteria and statistical significances o f the coefficients. Finally, to address the econom etric problem caused by several possible structural breaks in the R ussian econom y d u rin g the period 1993-2002, we use dum m y variables.
3. EMPIRICAL RESULTS
3.1. Data and Methodology
T h e availability o f R ussian d a ta is lim ited and p h en o m en a such as d o llarizatio n an d the b arter econom y m ay lead to a som ew hat biased picture. Som e au th o rs (e.g. F alcetti et al. 2000) also believe th a t the decline in o u tp u t was overestim ated d u rin g the first years o f the tra n sitio n period. In o u r em pirical estim ations we use m onthly d a ta covering th e tim e span 1993-2002. T his period has been chosen for d a ta availability reasons. T he sources o f the d a ta are the In tern atio n al M o n etary F u n d ’s In tern atio n al F in a n c ia l S tatistics d a ta b a se , th e w ebsite o f th e B ank o f R ussia, the m o n th ly d a ta b a s e o f th e V ienna In stitu te fo r In te rn a tio n a l E co n o m ic Studies (W IIW ), and the R ussian E u ro p ean C en tre fo r E conom ic Policy (R E C E P ). F o r o ur purposes, we need d a ta on sh ort-term interest rates (refin an cin g ra tes), consum er price in flation , m o n e ta ry ag gregates, the o u tp u t gap, different exchange rate m easures (d ollar exchange rate, nom inal effective exchange rate, and real effective exchange rale). We use o u tp u t n u m bers from R E C E P and W IIW (industrial p ro d u c tio n n um b ers) and d eflate them by the m onthly consum er price inflation, due to the lack of a m o n th ly G D P d eflator.
3.2. Results for the Taylor Rule
W hen we estim ate an open econom y version o f the T a y lo r rule — in levels and in differences, the estim ated coefficient o f inflation is only significant in one specification. T he estim ated coefficient o f the o u tp u t gap does no t show the expected sign or is insignificant fo r th e estim atio ns in levels (oth er proxies o f the o u tp u t gap show also un satisfacto ry results). T h e estim ated coefficients o f the exchange ra te variables are insignificant. T h e estim ated coefficient o f the lagged interest ra te is equal to 0.9 and rem ains relatively stable over the different m odel specifications, indicating
th a t the interest rate in a new period is ab o u t 90% o f th e old interest rate plus the effect o f the o th er independent variables (in the level es tim ations). T he long-run response o f the central b an k can be calculated as follows:
w here ß LR is th e lo ng-run response o n in flatio n , is th e estim ated coefficient for year-to-year inflation and ß 5 is the estim ated coefficient of the interest ra te one year before, as defined in eq u atio n (2). We get a lon g-ru n response o f ab o u t 0.3 and thus the T a y lo r principle (/iLR> 1) does n o t hold. T his m eans th a t according to o u r estim ation s the central bank reacts to a one percent increase o f inflation w ith less th a n a one percent increase in the short-term nom inal interest ra te (decrease in real interest rate).
T h e unsatisfactory result o f the o u tp u t gap m ig ht be caused by the facts th a t the objective o f the B ank o f R ussia was limited to inflation and exchange ra te stabilization o r th a t the real tim e d a ta significantly differed from the ex-post d a ta so th a t we get a biased picture in o u r estim ations (e.g. O rp h an id es 2001). Overall, the estim ation results suggest th a t a simple T ay lo r rule and its m odifications do n o t describe well interest ra te setting b ehaviou r o f th e B ank o f R ussia.1
3.3. Results for the McCallum Rule
Because o f d a ta availability problem s for the M l series, som e m issing points have been recovered by using the М 2 series, since these tw o series are highly correlated (over 9 5 % ).2 W e deflated the appro x im ated m onetary aggregates series with the m onthly consum er price index. We expect th at the signs o f the estim ated coefficients will be reversed, as a decrease in M 1 m eans a m o n etary contractio n and a decrease in the interest rate a m onetary expansion.
T h e estim ated coefficients are statistically insignificant, indicating a poor p erform ance o f the original M cC allum rule as specified in eq u atio n (3). M oreo ver, this regression specification has an o th er statistical disadvantage; as it requires discarding a large num b er o f observ atio ns in o rd e r to average the velocity o f m oney over the four-year period. Because o f this draw back,
1 We do not show here these results, but they are available from the authors on request. 2 The monetary base has is also highly correlated with M l (89%) and its use does not change the results.
we decided to estim ate a m odified M cC allum rule, where the interest rate in stru m en t (o f a T ay lo r type rule) is substituted by a deflated m on etary aggregate. A ssum ing th a t the Bank o f R ussia was indeed concerned with the o u tp u t stabilization d u rin g this period, we constructed a real-tim e series to correct the bias in d ata. As the regression results indicate, in general a m odified M cC allum rule perform s m uch b etter in explaining the behaviour o f the Bank o f R ussia th an simple interest rate based rules o r th e original T ay lo r rule. T he estim ated coefficients show the expected signs, but the m easure o f the o u tp u t gap is statistically insignificant.
H ow ever, the m o n etary aggregates series is n o n -statio n ary and this casts som e d o u b t to the validity o f the results. W hen we correct this statistical problem by differencing (cf. T able A l , first colum n), the regression results m ostly rem ain unchanged, even th o u g h the m agn itu d e o f th e p o in t estim ates was som ew hat different. In addition, we include seasonal dum m ies for D ecem ber and J a n u a ry (which are highly significant, b u t n o t show n in the T ab le A l) , as the R ussian m oney supply shows seasonal spikes d urin g these m o n th s. A ccording to D ąbrow ski et al. (2002) this effect is p robably a ttrib u ta b le to technical and accounting m easures.
T h e estim ated coefficient o f the o u tp u t g ap 3 is insignificant, co n tradictin g predictions from theory. We used the yearly o u tp u t d a ta published in the annual rep o rts o f the Bank o f R ussia, and on the basis o f them constructed a m onthly series, interpolating and re-basing the available industrial production m onthly series from the W IIW . W hen we run regressions using th e forw ard interp o lated “ real-tim e” o u tp u t gap, the estim ated coefficients show always the expected signs and are statistically significant for the period from 1 9 9 4 -2 0 0 2 /
O verall, the estim ation results allow us to conclude th a t the B ank of R ussia has been targeting m o n etary aggregates in its policy decisions. A t tim es o f high in flation pressure, o r a positive o u tp u t gap calculated on the basis o f the co nstructed real-tim e d a ta , the B ank o f R u ssia responded by reducing m o n etary aggregates in real term s, while a t tim es o f exchange rate ap p reciatio n th e policy response was an ex pansio nary m o n e ta ry policy. M oreover, these results are n o t sensitive to the m odel specification and th ere are n o m a jo r statistical problem s.
G iven the absence o f explicit inflation targetin g in R ussia we estim ate a “ gap m o d el” as defined in M o h an ty and K la u (2003). T h e ad v an tag e of
3 When we use nominal and real G D P as an alternative to the output gap, the estimated coefficients show no sign o f improvement.
4 Because o f the used approximation to real-time data, the results do not necessarily mean that the CBR was concerned with output stabilization, but they indicate that this may have been the case. Further evidence can only be obtained with actual real-time data, which was not available to us.
this m odel is th a t it allow s us to use an H P m easure o f trend inflation instead o f a targeted level.
(5) A lo g (M l) = /10 + ß y(C P l — C P I trend) + ß 2y, + /?3(xr, — xrtrend) -f + ß^(xrt - i - xrtrend,^ i) + ^ jA lo g C M l,-!) + u„
w here M l is the deflated m on etary aggregate M l , C P Itrend is the H P filter o f the inflation ra te and xrtrend is a log o f the H P filter o f the exchange rate change. We add anoth er lag to inflation to control for the autocorrelation problem s. W e again include seasonal dum m ies for D ecem ber and Jan u a ry , and a n o th e r dum m y for the period before M ay 1998 is added since the C how test indicates a stru ctu ra l break a t this point. T h e results are on T a b le A l (second colu m n ). T h e regression results are sim ilar to th e specification before.
3.4. Testing Responses During Different Time Periods
T h e R ussian econom y has experienced different shocks d u rin g different tim e periods, and it w ould be insightful to see w hether the B ank of R ussia has responded differently in different periods. F irs t o f all, we se p a ra te the p erio d before and afte r 1995, as C h o w b re a k p o in t tests indicate a stru ctu ra l b reak a t this tim e (but, peculiarly, n o t in A ugust 1998). W e use for this purpose the equation (“ full m o d el” , cf. T ab le A l) o f the follow ing type:
(6) A lo g (M l) = ß 0 + ß j n f , - ß 2d inf, + /?3inf«_ x + ß ^yt — ß 5d o llo rxr,+ + ß 6ddollarxr, — ß ^ o lla r x r ,- ! -I-/ľ8A lo g (M l,_ j) + + ß gd 1 + ß io d 2 + ß n d + u„
where á is a dum m y variable th a t is one for the period before 1995 and zero otherw ise, an d d l and d 2 are seasonal dum m ies fo r D ecem ber and Ja n u a ry over the sam ple period, respectively.
T h e estim atio n results clearly suggest th a t the B ank o f R ussia conducted different m o n etary policies before and after 1995. T h e estim ated coefficients indicate th a t before 1995 the B ank o f R ussia was m o re concerned with reducing in flatio n ,5 while after 1995 priorities have shifted to w ard s exchange rate stabilization. These findings are consistent with the official announcem ents o f the B ank o f Russia.
W e o b ta in a sim ilar result w hen we use a dum m y variable for the craw lin g peg p e rio d , fro m Ju ly 1995 th ro u g h A u g u st 1998. A s o ne would expect, the com m itm ent to react to changes in the exchange rate was greater d u rin g th a t period. D u rin g the high in flatio n period, the Bank o f R ussia attached a greater priority to in flation, while at times o f relatively low inflation the m ain concern was exchange ra te stabi lization.
4. CONCLUDING REMARKS
T his p aper exam ined the conduct o f m o n etary policy in R ussia du ring the period o f 1993-2002. We estim ated tw o sets o f m o n etary policy rules, the T ay lo r rule, and the M cC allum rule, using b o th m o n th ly an d qu arterly d ata. T h e regression results indicate th a t a sim ple T ay lo r rule and its different variatio n s describe poorly the interest ra te setting b ehavior o f the B ank o f R ussia.
T he M cC allum rule, where the policy instrum ent is a m on etary aggregate, fits best the d a ta . A gain, given th a t the b an k o f R ussia officially ad o p ts the m oney supply as an interm ediate an ch o r to policy an d th a t, even to day , its m a in actual instru m en t o f m on etary policy are d epo sit au ctio ns, this is a consistent result.
N evertheless, this is in sharp co n tra st w ith the recent experience o f other ad v anced em erging m a rk e ts, w ere in terest ra te rules p ro d u c e a good d escription o f the policy setting behaviour o f the m o n etary au th o rity . T h e estim ated coefficients are significant and rem ain unchanged across different eq u a tio n specifications. T he results indicate th a t d u rin g th e period o f 1993-2002 the B ank o f R ussia has used m o n etary aggregates as a m ain policy instrum ent in conducting m onetary policy. F u rth erm o re , the presented results also suggest th a t before 1995 the Bank o f R ussia was m ore concerned w ith in flation reduction, while after 1995 the prim ary objective was exchange ra te stabilization.
T h e results on o u r estim ations are backw ard looking, in the sense th at they represent the relationships th a t existed so far in the d a ta . A s the experience o f o th er advanced em erging m arkets show , the p ro m o tio n o f forw ard looking behavior am ong R ussian econom ic agents, aided by the developm ent o f stronger institutions - especially by th e stren gth ening o f the credibility o f th e B ank o f R ussia and the developm ent o f its policy instrum ents, as indicated by the late 2002 reform s, plus the deepening of R u ssia’s financial m arkets, shall, in tim e, enable the im p lem en tatio n of a successful interest ra te policy rule, coupled w ith in flatio n targ etin g and
a floatin g exchange ra te regim e, which shall also reduce the G D P costs o f disinflation (as M inella et al. 2003, show for the F ed eral R epublic o f Brazil).®
APPENDIX
Table A l. Testing a McCallum rule for Russia, 1993-2002
Independent variable Difference model Gap model Full Model
Intercept 0.02 0.01 0.01 (0.00)*** (0.00)** (0.00)*** Quarter-to-quarter-inflation -0.28 -0.26 (0.06)*** (0.07)*** Quarter-to-quarter-inflation (-1) 0.22 0.12 (0.05)*** (0.08)* Monthly inflation -0.17 (0.15)
Dummy (for period before 1995)* monthly -0.50
inflation (0.20)**
Monthly inflation (-1) -0.13
(0.07)*
Output gap (ex post data) -0.01 -0.13 0.10
(0.08) (0.08) (0.09)
Growth in bilateral dollar exchange rate -0.23 -0.26 -0.29 (0.05)*** (0.05)*** (0.08)***
Dummy for period before 1995 growth 0.27
in bilateral dollar exchange rate (0.13)**
Growth in bilateral dollar exchange rate (-1) 0.13 0.30 0.11 (0.05)*** (0.05)*** (0.05) **
Growth rate o f M l( - l) 0.29 0.18 0.28
(0.07)*** (0.07)*** (0.07)***
R square (adjusted) 0.74 (0.72) 0.74 (0.72) 0.76 (0.74)
Durbin Watson statistics 2.02 1.70 1.97
Breusch-Godfrey test No rejection N o rejection N o rejection
Note: 1% and 1% change o f the variables used for the estimations are scaled to 0.01; the Breusch-Godfrey serial correlation LM-test (with no autocorrelation as a null hypothesis) was conducted for twelve lags; (-1) indicates a first lag; the effective sample period is 1993:3 - 2002:12 since we lose two months because o f lags and differences; in the case o f gap model (third column) we deduct the HP-trend from quarter-to-quarter inflation and the growth in the dollar exchange rate; standard errors are in parentheses, the asterisks indicate levels of significance at the 10 (*), 5(**) or 1 (***) percent level.
6 As a sign of this, Taylor rule regressions run only for the period after 2000, do show the expected signs for the variables, but most of them are non-significant.
REFERENCES
Calderon, C. and Schmidt-Hebbel, K. (2003), “Macroeconomic Policies and Performance in Latin America”, Journal o f International Money and Finance, 27(7), 895-924.
Dąbrowski, M., Paczyński, W. and Rawdanowicz, L. (2002), “Inflation and Monetary Policy in Russia: Transition Experience and Future Recommendations”, CASE: Studies and Analyses, Warsaw.
Falcctti, E., Raiser, M. and Sanfey, P. (2000), “Defying the Odds: Initial Conditions, Reforms and Growth in the First Decade of Transition”, EBRD: Working Paper, 55.
McCallum, B. (1988), “Robustness Properties of a Rule for Monetary Policy”, Carnegie-Rochester Conference Series on Public Society, 29, 173-204.
Minella, A ., Springer de Freitas, P., Goldfajn, L, and Kfoury Muinhos, M. (2003), “Inflation Targeting in Brazil: Constructing Credibility under Exchange Rate Volatility” , Journal o f International Money and Finance, 27(7), 1015-1040.
Mohanty, M. and Klau, M. (2003), “Monetary Policy Rules in Emerging Market Economies, Issues and Evidence”, Paper presented al the Kiel Workshop for Monetary Policy and Macroeconomic Stabilization in Latin America, Kiel: Institute for World Economics, September.
Orphanides, A. (2001), “Monetary Policy Rules Based on Real-time Data. Division o f Monetary AlTairs” , American Economic Review, 91(4), 964-985.
Personen, H. and Korhonen, 1. (1998), “The Short and Variable lags of Russian Monetary Policy” , Review o f Economies in Transition, 4, 59-72.
Taylor, J. (1993), “Discretion versus Policy Rules in Practice”, Carnegie-Rochester Conference on Public Policy, 39, 195-214.
Taylor, J. (2000), “ Using Monetary Policy Rules in Emerging Market Economies” , Paper presented at the 75th anniversary conference at the Banco de Mexico.
Taylor, J. (2001), “The Rules o f the Exchange Rate in Monetary Policy Rules, American Economic Review, 91, (2), 263-267.
Torres Garcia, A. (2003), “Monetary Policy and Interest Rates: Evidence from Mexico”, Paper presented at the Kiel Workshop for Monetary Policy and Macroeconomic Stabilization in Latin America, Kiel: Institute for World Economics, September.
C hristian M erk l, Lúcio Vinhas de S ou za
CZY BANK CENTRALNY ROSJI STOSUJE KRYTERIUM MONETARNE? ANALIZA EMPIRYCZNA
(Streszczenie)
W artykule dokonaliśmy przeglądu zasad polityki monetarnej prowadzonej przez bank centralny Rosji. Stosując różne podejścia do tych zasad, testujemy czy bank centralny Rosji reaguje na zmiany inflacji, luki produkcyjnej oraz kursu walutowego w przewidywalny sposób. Nasze badania prowadzą do wniosku, że w latach 1993-2002 bank centralny Rosji przyjmował za kryterium agregat pieniężny jako główny instrument prowadzonej polityki monetarnej.
