INTERACTION BETWEEN REAL AND MONETARY SECTORS OF THE ECONOMY IN TERMS OF ECONOMIC INSTABILITY

Transcription

1 Vol. 2, No.2, Winer Published by JSES. INTERACTION BETWEEN REAL AND MONETARY SECTORS OF THE ECONOMY IN TERMS Ion PARTACHI a, Olga PANIS a Academy of Economic Sudies, Chisinau, Moldova Absrac This paper is a coninuaion of an aricle dedicaed o he analysis of moneary policy shocks in he conex of he ineres rae channel operaion in he Republic of Moldova during This work represens he elaboraion of he error correcion model, which includes he relaionship beween he base rae and moneary aggregae M3 and inflaion. The analysis, based on ineracion and equilibraion beween he indicaors in he real and moneary secor is relevan in he curren condiions of economic insabiliy. As inernaional pracice of advanced economies show, here is srong correlaion beween he level of developmen of he (moneary) financial secor and economic growh, which is associaed wih such phenomena as: economic cycliciy, inflaion. The presen paper proposes he analysis of some ransmission effecs of moneary shocks, in he conex of ineracion mechanism of real and moneary secor of he economy. The main analysis and conclusions are based on he implemenaion of correlaion ess, granger causaliy ess, analysis of variance decomposiion of forecas and error correcion model. Keywords: Economic insabiliy, Aggregae demand, Transmission moneary effecs, Poenial GDP, Impulse response, Vecor error correcion, Forecas error variance decomposiion JEL Classificaion: E31, P24, C32 Auhors Affiliaion a Academy of Economic Sudies, Chisinau, Moldova 7

2 Inroducion Moldova, like many counries in ransiion has faced srong inflaionary phenomenon ha remained a double-digi levels for 17 years. Period of 2003q3-2008q3 has recorded a srong inflaionary rend, which was suppored by he depreciaion of he naional currency agains he currencies of major rading parners and as a resul he average rae of inflaion for his period was abou 13 percen. According o he inernaional pracice of he advanced economies, here is a srong correlaion beween he level of developmen of he financial secor (moneary) and economic growh ha is associaed wih such phenomena as he economic cycliciy, unemploymen and inflaion. The financial crisis has caused economic insabiliy and has proofed ha undersanding he ineracion beween moneary and real secor of economy is more profound and complicaed. Balanced budge, modes inflaion, moneary policy and is objecives are he main elemens and deerminans of economic sabiliy in he shor and long. In he case of Moldova, he adopion of a new program - Moneary Policy Sraegy of he Naional Bank of Moldova for showed posiive effecs, so ha during hree years he inflaion rae has fallen o single-digi values (Sepember Sepember 2013 inflaion consiued 3.9 percen). According o his sraegy since 2010, he NBM sared implemening he moneary policy regime - inflaion argeing. Considering he moneary auhoriy insrumens available in he conex of Moldova's economic realiies, and he experience of oher cenral banks, inflaion argeing enables more efficien primary objecive of he NBM - ensuring and mainaining price sabiliy - sipulaed by he Law on Naional Bank of Moldova 2. Managed by cenral banks moneary policy (for RM NBM) has achieved is objecives hrough moneary ransmission channels, giving crucial role o he ineres rae in seering markes: moneary, credi and currency. Generally, he NBM base rae changes lead o changes in ineres raes on loans and deposis in he economy and he exchange rae of he naional currency. Moneary ransmission mechanism appears as a change in moneary condiions and influences he demand for goods and services, aiming mainly a ensuring and mainaining price sabiliy. 1 To ensure and mainain price sabiliy, he Bank ses inflaion arge - measured by he consumer price index published monhly by he Naional Bureau of Saisics (NBS) - a 5.0 percen for 2010 wih a possible deviaion of ± 1.0 percenage poins. In he period NBM ensure inflaion rae in mid-single digi 2 hp:// 8

3 The objecive of his work is o esimae he economic-economeric relaions beween he base rae, as he main insrumen of moneary policy in Moldova; moneary aggregae - M3 3, inflaion level and also some Taylor rule elemens ha appear like equilibrium condiions in he economy, where moneary policy insrumens are presened as a funcion of forwardlooking variables Survey mehodology Some of he earlies and mos successful empirical works in he field mean o subsaniae he quaniaive logic of economic ransmission were works 5 of B. Bernanke, M. M. Gerler, F. Mishkin, J. Taylor. Tradiionally, he diagnosis of he moneary effecs on he real secor is performed by VAR and VECM models developed by Sims (1980) o esimae he presence of shor and long -erm connecions, based on economic logic of connecions beween facors. In developing he performed esimaions were used monhly series convered ino quarerly daa; he period aken ino consideraion was he firs quarer of 2001 he hird quarer of The daa series used were fixed based, firs quarer of , based on series of monhly saisical bulleins of he Naional Bureau of Saisics and he Naional Bank of Moldova. Quarerly daa used in he model are herefore affeced by he seasonaliy phenomenon and possibly he non-saionary one, which required he use of some ess in order o check he series, saring wih he Dickey, Hasza and Fuller saionary ess (1984), modified by Osborn e aler (1988 ) and Franses (1991). Iniially were analyzed and esimaed five macroeconomic variables: GDP, poenial GDP, inflaion, moneary aggregae M3, he base rae of he Naional Bank. The VECM model and he forecas variance decomposiion were also consruced based on inflaion, M3 aggregae and he base rae. The model assumes he presence of saisical ineracion beween he real and moneary secors (he influence of moneary facors on he real secor indicaors). The radiional approach over ransmission processes of moneary effecs is based on he ransmission effecs of based rae shocks in he shor-erm on he moneary secor he real secor of economy. The 3 M3 choice for esimaes is subsaniaed by invesigaing he abiliy of he indicaor, which signed risks o price sabiliy over he medium o long erm because of he componens included in M3 4 The rule of "anicipaion" as a response of ineres raes o inflaion and GDP deviaion 5 Bernanke Ben S., Gerler Mark (1995). Inside he black box: he credi channel of moneary policy ransmission. Naional Bureau of Economic Research. Cambridge 6 fixed he base rae sar wih I quarer

4 classical scheme of he moneary ransmission effecs can be presened as follows: i basa i marke MM D I NFL where i basa he base rae of cenral bank, i marke - he rae of marke, MM money supply, D (oupu) aggregae demand, I nfl inflaion. The presen work analyzes he ineracion beween indicaors of real and moneary secor, herefore invesigaion begins wih he analysis of Taylor rule, because heir pracical and heoreical aspecs are used o conrol inflaion, which ends o accelerae wih consan increase in prices of oil, gas, increasing he volume of consumer loans. I is worh menioning ha he sudy of he ransmission mechanisms hrough he ineres rae channel in he Republic of Moldova, aking ino accoun he specific ransiional insabil economy, is curren and relevan, imporan because i allows drawing conclusions abou he effeciveness of moneary policy and developing differen scenarios of macroeconomic indicaors depending on he decisions aken by he Naional Bank. The explanaion of all moneary rules concepualize he cenral bank adjusing a shorerm nominal ineres rae in linear response o he deviaion of inflaion from arge level and o he deviaion of he real oupu from some arge level. Moneary policy rules which reac o corren values of inflaion and oupu deviaion could easily induse equilibrum 7. Bank signals o marke paricipans abou resricing or loosening moneary policy, which affecs shor-erm ineres raes and he environmen, hus, influencing inflaion by economy demand. Therefore he response o moneary policy shocks is analyzed firsly, in erms of influence on inflaion. Raising raes is causing higher prices of money, reduced demand for loans, ighening moneary dimensions ha lead o currency appreciaion. The simple form * * of Taylor rule: i baza i 0.5( y y ) 0.5( ) (1) where : i baza - he base rae; i - real ineres rae; - average inflaion rae for he las four quarers; y - real GDP; 7 Verejan O., Crasovschi A. (2012). Poliica Moneară si cursul de schimb. Analiza saisică cazul Republicii Moldova. Academie de Sudii Economice din Moldova. 10

5 * y - real GDP rend. The rule describes he siuaion when he ime coincides wih he arge inflaion (refernce) moneary auhoriies, and GDP is produced in is poenial. Oherwise, sar by gap coefficiens plays inflaion and GDP gap. This also is Taylor principle, which indicaes ha he cenral bank employs more aggressive on inflaion and aggregae demand. Gap inflaion is included in he Taylor rule ou of simple economic reasons. If inflaion in an absrac counry for some ime has been, for example, 5%, hen he loan rae can no be less han 5%, as oherwise he credior has losses. In fac, permanen inflaion flucuaions occur wih is long-erm arge. Therefore is needed o correc his flucuaion rae wih cerain weigh M M02 Inflaion Inflaion-arge Gap-Inflaion 2012M M M M M M08 Fig. 1 Inflaion and arge - Inflaion Source: BNM, BNS, auhor s calculaions 2012M M M M M M M M M M M M M09 Gap GDP eners in he Taylor rule ou of 2 reasons 8 : i is an indicaor of fuure inflaion and allows dividing he price shock (supply shock) and demand-side shock. In he presen work, poenial GDP was esimaed by he Hodrick-Presco filer 9. I is no a heoreical mehod, saisical, which exracs a smooh componen ime series like GDP, penalizing frequencies from occurring flucuaions like business cycles. The filer is parameerized (parameer, for which Presco suggess he value of Esimaed GDP gap is shown on he image 9 Hodrick R.J and E.C.Presco (1980). Pos-war U.S. business cycles: An empirical invesigaion. Carnegie Mellon Universiy 11

6 x x rend cycle 2 min ( y ) x x T 1 rend T 1 2 ( 1 ) ( 1 ) 2 (2) where x log( y ), in his case y is real GDP acual GDP poenial GDP Gap - GDP Q1 2010Q2 2010Q3 2010Q4 2011Q1 2011Q2 2011Q3 2011Q4 2012Q1 2012Q2 2012Q3 2012Q4 2013Q1 2013Q2 Fig. 2 Esimaion of poenial GDP by Hodrick-Presco Filer (LAMDA = 1600) Source: BNM, BNS, auhor s calculaions The analysis of acual GDP and poenial GDP shows ha he las quarer is greaer han poenial GDP as acual GDP, indicaing he absence of inflaionary pressures from he demand for hose las moneary policy decision aken by he Council of Adminisraion was diminishing base rae up o 3.5 percen. 12

7 Sep.00 Sep.01 Sep.02 Sep.03 Sep.04 Sep.05 Sep.06 Sep.07 Sep.08 Sep.09 Sep.10 Sep.11 Sep.12 Sep.13 Ion PARTACHI, Olga PANIS INTERACTION BETWEEN REAL AND MONETARY SECTOR OF THE ECONOMY UN TERMS The base rae of NBM The rae on loans up o one year Fig 3. The base rae of NBM and he rae on loans up o one year Source: BNM, BNS, auhor s calculaions 2. Economeric esimaions Diagnosis of ineracion beween real and moneary secor was made by correlaion marix, Granger Causaliy es, Johanson coinegraion es ha allowed consrucion of VEC model for predicing of shor and long-erm ime VECM o esimae he presence of shor and long-erm. In developing he performed esimaion was used quarerly daa convered o monhly series, he period considered was he firs quarer second quarer Daa ses used were fixed base, firs quarer , based on series of monhly saisical bulleins of he Naional Bureau of Saisics and he Naional Bank of Moldova. Quarerly daa used in he model are herefore affeced by he phenomenon - he seasonaliy and possibly nonsaionary, which required he use of ess o check he series since saionariy ess Dickey, Hasza and Fuller (1984), modified by Osborn e aler (1988) and Franses (1991). The seps in developing he model were: - analysis of descripive elemens of variables; - saionariy es series; - esing he influence of variables hrough Granger causaliy es; - consrucion of equaions (VECM); - ess, presenaion of resuls and heir inerpreaion. 10 fixed he base rae sar wih I quarer

8 Daa Quarerly and monhly daa for analysis and esimaion was aken from he elecronic daabase of he Naional Bureau of Saisics of he Republic of Moldova, he Naional Bank of Moldova for he period January Sepember Table 1. The correlaion coefficiens beween he variables GDP_SA GDP_POT_SA M3_SA CPI_SA RATA_BASA GDP_SA GDP_POT_SA M3_SA CPI_SA RATA_BASA Source: BNM, BNS, auhor s calculaions The able reveals ha he correlaion beween variables is srong and he signs of he coefficiens reflec he logical connecion beween he facors: domesic demand growh (GDP growh) causes increased demand for money, and herefore, of inflaion, which leads o increased he base rae, which in urn affecs he economic aciviy growh (aggregae demand) followed by reducion of demand inflaionary processes. Table 2. Saionary es Dickey - Fuller Variable ADF - (level) ADF - (firs diferences) RATA_BASA M3_SA CPI_SA GDP_SA GDP_POT_SA filer Hodrick-Presco Tes criical value 1% Source: BNM, BNS, auhor s calculaions 5% %

9 Table 3. Granger causaliy es Null Hypohesis: Obs F-Saisic Prob. L_M3_SA does no Granger Cause L_CPI_SA L_CPI_SA does no Granger Cause L_M3_SA L_GDP_POT_SA does no Granger Cause L_CPI_SA L_CPI_SA does no Granger Cause L_GDP_POT_SA L_GDP_SA does no Granger Cause L_CPI_SA L_CPI_SA does no Granger Cause L_GDP_SA L_R_BASA does no Granger Cause L_CPI_SA L_CPI_SA does no Granger Cause L_R_BASA L_GDP_POT_SA does no Granger Cause L_M3_SA L_M3_SA does no Granger Cause L_GDP_POT_SA L_GDP_SA does no Granger Cause L_M3_SA L_M3_SA does no Granger Cause L_GDP_SA L_R_BASA does no Granger Cause L_M3_SA L_M3_SA does no Granger Cause L_R_BASA L_GDP_SA does no Granger Cause L_GDP_POT_SA L_GDP_POT_SA does no Granger Cause L_GDP_SA L_R_BASA does no Granger Cause L_GDP_POT_SA L_GDP_POT_SA does no Granger Cause L_R_BASA L_R_BASA does no Granger Cause L_GDP_SA L_GDP_SA does no Granger Cause L_R_BASA Source: BNM, BNS, auhor s calculaions The analysis of he relaionship beween variables is appropriae and does no conradic wih economic logic. Base rae affec money supply ha in urn affec inflaion. For some variables Granger causaliy es revealed muual causaliy. In order o illusrae quaniaively he exisence of hese effecs of influence beween he variables, were buil he VAR and VECM models wih differen lags making IRF calculaions by Cholesky mehod. Since all variables are inegraed of order I, we applied he coinegraion Iohansan es, which showed ha he 1% significance es (Trace) and 5 % - significance es (Maximum eigenvalue) of variables base rae, moneary aggregae M3 and inflaion here is a coinegraion relaionship. Iohansen posiive es resul allows us o build a VECM. Long erm elasiciy coefficiens of he coinegraion relaionship and error correcion coefficiens are presened below: 15

10 Table 4. Long-erm elasiciies, normalized agains inflaion and error correcion parameers Variabils The coefficien of elasiciy The coefficien of correcive L_CPI_SA L_M3_SA L_RATA_BASA Source: BNM, BNS, auhor s calculaions The adjusmen coefficien is negaive and according o he significaive, nonzero es, he VECM model helps o mainain long-erm equilibrium inflaion. The coinegraion relaionship indicaes ha in long-erm he prices respond posiively o increased aggregae М3 and negaive o base rae. Error correcion coefficiens describe he speed of adjusmen of he endogenous variables in long-erm equilibrium, hus, in he given case, he ime necessary o reduce he gap beween he curren level of inflaion and he seady sae (deleed by shock) is abou 1.5 quarers. Response of CPI o CPI Response of CPI o M3 Response of CPI o Raa base Fig 4. Shock response funcion in a sandard deviaion. Deviaion afer Choleski 11 Source: BNM, BNS, auhor s calculaions From he Choleski impulse response funcion we can noe ha inflaion, is bes described by iself, beginning wih he firs period, he auoregression effec aenuaes during he fourh 11 Picure shows values - answers forecased inflaion in sandard deviaion residues M3 and base rae 16

11 quarer. The maximum shock from moneary aggregae М3 on inflaion is reached during quarers IV - VI, which does no conradic he classical heories abou he medium and long erm links beween inflaion and moneary facors. The shock influence on he inflaion rae is negaively affecing i in he firs quarer, which is logical and allows us o conclude ha he Naional Bank of Moldova can influence inflaion in he shor erm. Table 5. Forecas error variance decomposiion % Horizon (quarers) CPI M3 Base rae Inflaion Source: BNM, BNS, auhor s calculaions The analysis of variance decomposiion of forecas errors showed significan conribuion of he moneary aggregae and he base rae in explaining he inflaion variaion. During he firs period he inflaion auoregression describes very well is behavior, furher descripion of oher facors of inflaion dispersion begins in he second quarer. The influence of base rae beer describes he behavior of inflaion; his can be explained by he fac ha he base rae is a shor-erm insrumen, bu is influence increase hroughou he forecas period. The sronges influence begins in he fourh quarer, bu peaks a he end of he inerval. Conclusions Taking ino consideraion he inabiliy o obain significan saisical esimaes of he influence on macroeconomic variables (based on he classical represenaion of he ransmission mechanism of moneary policy) due o he heerogeneiy of processes, as well as he lack of hisorical long daa series hrough he VECM model, only hree facors were 17

12 esimaed. Using he VECM conceps o analyze he deerminaion of he moneary effecs associaed wih cerain difficulies caused by changes in he moneary policy regime, he behavior of economic agens and exernal facors. However, from he analyzed variables are presened long-erm connecions, which influence he shor-erm relaionships, hus, confirming he iniial hypohesis abou he presence of he link beween he real and moneary secors. The analysis based on ineracion, he balance beween he real and moneary secor indicaors is boh relevan and opical under he curren economic insabiliy condiions. According o he resuls obained hrough he VECM model, he adjusmen speed of inflaion (speed of response) is abou 2 quarers, his indicaor is quie good, i shows he cumulaive influence of he M3 and he basic rae, he las shows he reacion of he moneary auhoriy on he changes he base rae of inflaion. The obained coefficiens in he analysis are saisically significan, for example beween inflaion and moneary aggregae were found srong long-erm links; he base rae also demonsraed saisically significan resuls in relaion o inflaion, which proves ha he relaionship beween variables is due and effec, i should be noed ha in he fuure i can mainain and will reach a posiive value (his is seen in he response funcion o a sandard deviaion shock afer Choleski), his can be inerpreed as moneary policy decisions (he base rae) have a forecasing adapive characer, i.e. he base rae is adjused depending on he curren level of inflaion. In his case his means ha he moneary auhoriies can no only focus on mainaining sable levels of inflaion, bu also o ry o opimize he cyclical flucuaions of oupu in achieving he inflaion arge. Forecas error variance decomposiion allows o make conclusions ha inflaion in Moldova, a presen, is a moneary phenomenon, because he effec of shock ransmission from he base rae is more significan owards he response o inflaion shock in he moneary aggregae, bu is influence grew during he periods, i.e. he major par of he variaion in inflaion in he consruced model was explained by he base rae. In his case we can alk abou cos-push inflaion caused by monopoly or high level of sensiiviy of he economy o exernal and inernal shocks, based on economic dependence on naural resources and energy. However, he overall resuls reflec general rends in relaions beween he analyzed variables, and due o he emergence of hisorical daa and improvemen of pracical aspecs of building economeric models in he fuure will arise he possibiliy o analyze more deailed he effecs of ransmission of moneary policy shocks no only on inflaion bu also on he res of macroeconomic indicaors ha are of paricular ineres for proper undersanding and decision-making ha lead o growh and sabiliy in he fuure. 18

13 References Bernanke Ben S., Gerler M. (1995). Inside he black box: he credi channel of moneary policy ransmission. Naional Bureau of Economic Research. Cambridge Cozmanca B.O. (2008). Mecanisme de ransmisie a poliicii moneare. Lucrare dizeraţie, ASE, Bucureşi; Cuhal R. (2010). Poliica moneară în asigurarea creşerii economice. Chişinău. Mishkin F. S. (2001). The Transmission Mechanism and he Role of Asse Prices in Moneary Policy. Naional Bureau of Economic Reserch. Cambridge Alăr M. (2000). Moneary Macroeconomics. MSc Course Hodrick R.J and E.C.Presco (1980). Pos-war U.S. business cycles: An empirical invesigaion. Carnegie Mellon Universiy Šulicova V. (2011). The inernaional ransmission of inflaion in cenral and easern European counries: CO-INTEGRATION ANALYSIS. Inernaional Scienific Conference YOUNG SCIENTISTS Verejan O., Crasovschi A. (2012). Poliica Moneară si cursul de schimb. Analiza saisică cazul Republicii Moldova. Academie de Sudii Economice a Moldovei. ECONOMICA nr. 4 (82) hp:// (Accessed 25 Augus 2013) 19