Monetary and Fiscal Responses during the Financial Crisis in the Developing and Emerging Economies

Transcription

1 Inernaonal Journal of Economcs and Fnance; Vol. 5, No. 9; 2013 ISSN X E-ISSN Publshed by Canadan Cener of Scence and Educaon Moneary and Fscal Responses durng he Fnancal Crss n he Developng and Emergng Economes Besnk Tap Fea 1 1 Deparmen of Economcs, Faculy of Busness and Economcs, Souh Eas European Unversy, Teova, Republc of Macedona Correspondence: Besnk Tap Fea, Deparmen of Economcs, Faculy of Busness and Economcs, Souh Eas European Unversy, Teova, Republc of Macedona. E-mal: b.fea@seeu.edu.mk Receved: June 14, 2013 Acceped: July 17, 2013 Onlne Publshed: Augus 26, 2013 do: /jef.v5n9p110 URL: hp://dx.do.org/ /jef.v5n9p110 Absrac The sudy provdes emprcal analyses of he role of moneary and fscal polcy on economc growh durng he fnancal crss n developng and emergng economes. I nvesgae 72 epsodes of fnancal crss n developng and emergng counres, n order o assess he effec of moneary and fscal polcy on oupu cos over he fnancal crss. I fnd ou ha effec of moneary and fscal ghenng wll ncrease oupu cos durng he fnancal crss. The resuls show ha fscal polcy has been more effecve ools n dealng wh fnancal crss, han he effec of moneary polcy. In addon, he resul suggess ha he coordnaon wh an expansonary fscal polcy and a neural moneary polcy wll reduce oupu cos durng he fnancal crss n developng and emergng counres. Keywords: moneary polcy, fscal polcy, fnancal crss, economc growh 1. Inroducon The am of hs paper s o assess he effec of moneary and fscal polcy on economc growh durng he fnancal crss n developng and emergng counres. The economc downurn caused by he global fnancal crss n 2007 has posed agan dscusson among he researchers regardng he mpac of fnancal crss on oupu growh. There are que few sudes ha nvesgae he effecveness of moneary and fscal polcy on oupu growh durng he fnancal crss. However, he queson of he suable moneary and fscal measures has become more pronounced especally durng he global fnancal crss. Furhermore, here s no consensus among he researchers regardng moneary and fscal polcy mx. To address hs queson, I examne 72 epsodes of he fnancal crss n developng and emergng counres, n order o measure he effec of moneary and fscal polcy on oupu cos durng he fnancal crss. In addon, dfferen moneary and fscal sraeges have been appled n advanced economes and emergng and developng counres n order o preven furher progress of he fnancal crss and smoohng economc recesson. Mos of advanced economes he governmen has been more focused boh n expansonary moneary polces by Cenral Bank s neres rae cu and fscal smulus packages, supporng fnancal and real economc acvy. Regardng emergng and developng counres he fscal and moneary measures have been dfferen from he developed counres for he reason ha hey beleve ha hose counres have small room n erms of applyng expansonary moneary and fscal polcy smulus. Durng he fnancal crses he polcymakers of he moneary polcy n developng and emergng counres have been more neresed n mananng hgher neres raes and admnsrave lendng conrols n order o keep he nflaon under conrol and o preven capal ouflows. However, some of he developng and emergng counres have adoped somehow an expansonary fscal polcy by changes of he budge srucure, cung curren expendure n favor o capal spendng, some of hem nroducng a cu n publc admnsraon coss. In he leraure, mos of he sudes ague ha fscal polcy s more effecve han moneary polcy durng he fnancal crss and herefore fscal expanson can reduce oupu cos or oupu loss (IMF repor, 2008a and 2008b). As for moneary polcy he repor shows ha counercyclcal moneary polcy can suppor shorenng of economc recesson, however s effcency s lmed durng he crss. Baldacc a al., (2009) examne effec of fscal polcy on real oupu durng he fnancal crss and hey fnd ou ha governmen consumpon can shoren duraon of he fnancal crss and such measure s more effecve han polcy supporng publc nvesmen or 110

2 Inernaonal Journal of Economcs and Fnance Vol. 5, No. 9; 2013 ax cus. On he oher hand, L J., and Tang L., (2010) analyze he effecveness of moneary and fscal polcy response wn crss for 72 epsodes durng n 57 emergng and developng counres. They fnd ou ha moneary expanson (conracon) can decrease (ncrease) oupu cos, whereas fscal expanson (conracon) has no effec on boh bankng and currency crss. They conclude ha polcy mx has o be coordnaed by dscreonary moneary expanson wh a neural fscal polcy durng he fnancal crss, snce fscal expanson or conracon has no effec on oupu cos. On he oher hand he sudy by Huchson a al. (2010) nvesgae he effec of moneary and fscal polcy over he fnancal crss n emergng and developng economes and hey conclude ha fscal expanson s more effecve han moneary expanson. They fnd ou ha expansonary fscal polcy s relaed wh lower oupu cos durng he fnancal crses, whereas he effecs of expansonary moneary polcy have no been denfed. Goldfan and Gupa (2003) analyses a fnancal crss n 80 counres for he perod , and hey fnd ou ha f he economes have currency and bankng crss he moneary and fscal polcy are neffecve. Moreover, lle emprcal evdence has addressed o he queson regardng opmal macroeconomc polcy mx durng he fnancal crss. I ry o fll hs gap n he leraure. Therefore he man objecve of hs paper s o examne he mpac of he fnancal crss on real oupu for developng and emergng counres and wha knd of macroeconomc measure should be used n he developng and emergng counres durng he economc crss n order o allevae economc recesson. For hs purpose, I analyze 72 epsodes of fnancal crss ha have been occurred over n developng and emergng counres n order o measure he effec of moneary and fscal polcy on oupu growh durng he fnancal crses. I employ cross-seconal mehodology and followng mehodology adoped by Gupa e al. (2007). The remnder paper s organzed as a follows: Secon II Economerc analyss of he mpac of moneary and fscal polcy measure on oupu cos; Secon III Daa descrpon; Secon IV emprcal resul and Secon V conclusons 2. Economerc Analyss of he Effec of Moneary and Fscal Polcy on Oupu Cos durng he Fnancal Crss To nvesgae he effec of moneary and fscal polcy on oupu cos durng he fnancal crss I employ benchmark emprcal model ha conan a sandard se of varables. I follow he mehodology by Je (2013) and Huchson a al., (2010), who nvesgaes he effec of moneary and fscal polcy on oupu cos. The benchmark model of oupu cos or oupu-loss ncludes mporan conrol varables n he regresson n order o measure margnal effec of macroeconomcs varables and avodng omed-varables bas. The specfcaon of economercs model s as follows: Cos B 0 B X 1 B 2 D fsc Where oupu-cos s he cos of oupu assocaed wh fnancal crss, D are bnary ndcaors for I mon expansonary and conraconary changes n fscal polcy sance, X s a vecor of conrol varables, D are bnary ndcaors for expansonary and conraconary moneary polcy. I measure moneary polcy by he changes n he nernaonal reserves and n he dscoun rae as moneary ndcaors. Fscal polcy s measured by changes of fscal sance ha are ndependen of he busness cycle. The consrucons of moneary and fscal ndcaors are explaned n deal n he followng secon. 3. Descrpon of Daa 3.1 Defnon of Fnancal Crss In hs par I explan he characerscs of fnancal crss boh bankng crss and currency crss. I ulzed he daabase calculaed by Laeven and Valenca (2008, 2010) (LV-henceforh) and hey denfy 124 sysemac bankng crss and 208 bankng crss. They defne bankng crss as a corporae and fnancal secors experence a large number of defauls and fnancal nsuons and corporaons face grea dffcules repayng conracs on me. The currences crss s defned as a nomnal deprecaons of currency of a leas 30% percen ha s also a 10 percen ncrease n he rae of deprecaon compared o he year before. The sample epsodes nclude 72 counres over he perod from 1977 o I denoe he sarng of a boh crses n perod, as a bankng crss, assocaed wh currences crss over he perod [-3, +3]. The deals of he epsodes and daa sources are repored n Appendx A and B. In Table 1, I dsplay frequency of boh crses such as bankng crss and currency crss. As seen from he Table 1, n perod of 1970, bankng and currency epsodes are nfrequen, whch s 0.2 on averages per year, whereas fsc B 3 D mon u (1) 111

3 Inernaonal Journal of Economcs and Fnance Vol. 5, No. 9; 2013 from 1980, he frequency of boh crss are consderably ncreased from 2.2 on average per year o 3.43 on average per year. Snce 1980, an ncrease of boh crses, (bankng crss and currences crss), perhaps could be as resul of fnancal lberalzaon (Kamnsky and Renhar, 1999). In addon, boh crses are larger han sngle crss, whch ndcae ha bankng crss can lead o a currency crss or afer he currency crss. Thus, he polcy makers have o ake no accoun boh crses should no consder separaely. Table 1. Frequency of bankng and currency crses oal average oal average oal average oal average bankng crses boh crses epsodes currency crses Noe: boh crses epsodes are begnnng daa of a bankng crss wh currency crses over (-3, +3). Average s average per year. Source: Auhor s calculaon. 3.2 Defnon of Varables n Emprcal Research a. Oupu-loss or oupu cos There s several ways o measure oupu-cos assocaed wh fnancal crss. Followng Laeven e al., (2008, 2010), I consruc he daa for oupu cos by calculang he daa (pre-crss) for average GDP growh rae rend for gven counres -3 o -1, s sarng crss and (pos-crss) GDP growh rae +1 o +3, unl GDP growh rae reurn back o s rend. Therefore, he dfference beween real GDP growh rae rend (pre-crss) and acual real GDP growh (pos-crss) represen he oupu-cos for each gven counres. b. Fscal polcy I m neresed o measure dscreonary fscal polcy response o oupu cos. As he budge- balance can move wh he same pah wh rae of economc growh, I have o decompose budge-balance no her srucural and cyclcal componen n order o assess dscreonary fscal measure durng fnancal crss. I employ sandard mehod used by Blanchard, (1990), Je (2013) and Huchson a al., (2010), n order o ake ou boh rend and cyclcal componen from budge-balance. The dscreonary fscal polcy I calculae from he resdual of each counry based on he followng equaon. Ths s sandard measure for fscal sance whch allows us o fnd dscreonary fscal measure. The model for esmang fscal ndcaor s as follows: BB B y B y (2) where BB s budge balance n percen of GDP of each counres, y denoes he real GDP for each counres, denoes he me rend and denoes he resduals n he regresson. Then I esmae he dscreonary measure of fscal polcy such as: f ˆ ˆ (3) 1 Where s he calculaed he resduals from equaon (2). By hs esmaon I elmnae smulaney bas of fscal sance wh oupu movemen n our emprcal research. Fnally I esmae he bnary dummy varable of changes n he fscal surplus by arrangng he 56 observaon from small o large. The expansonary fscal polcy s provded from he frs 28 observaon and I denoe he counry/year wh 1 fscal expanson and 0 oherwse. The las 28 observaon represens conraconary fscal polcy and I denoe n he same manner counry/year wh 1 fscal conracon and 0 oherwse. Ths s sandard measure of fscal polcy sance see more Blanchard, (1990), Je (2013) and Huchson a al., (2010). c. Moneary polcy There are several way o measure moneary polcy, I follow Je (2013) and Huchson a al., (2010), Bag and Goldfajn (2001), Goldfajn and Gupa (2003) and hey consder changes of nernaonal reserves and dscoun rae. Accumulang nernaonal reserves s accompaned wh an ncrease of he moneary base whch s he nsrumen of moneary loosng. De-accumulang nernaonal reserve s accompaned wh a decrease of he moneary base whch s he nsrumens of moneary ghenng. In hs conex, I perform bnary varable for moneary expanson and conracon. Moneary expanson s calculaed by one or more changes n he reserve whch s hgher han wo sandard devaon from he counry mean, and I denoe wh value 1 moneary 112

4 Inernaonal Journal of Economcs and Fnance Vol. 5, No. 9; 2013 expanson and 0 oherwse. Moneary expanson s calculaed by changes n he reserve, whch s smaller han wo sandard devaon from counry mean and I denoe wh value 1 moneary ghenng and 0 oherwse. I have no nroduced nerbank ners rae as s no avalable measure n developng and emergng counres. In addon, he nerbank ners rae does no show marke behavor n hose counres, and s no under he conrol of he moneary auhory. Therefore, I nroduce he dscoun rae as s under he conrol of moneary auhory and show dscree polcy movemen. The dscoun rae s calculaed as a monhly ncrease of he dscoun rae whch s he nsrumen of moneary conracon and a monhly decrease of he dscoun rae whch s he nsrumen of moneary expanson. As resul, I consruc bnary dummy varable for moneary expanson and ghenng n order o lm he problem of endogeney. Moneary ghenng s calculaed by one or more changes n he dscoun rae, whch s hgher han wo sandard devaon from he counry mean, and I denoe wh value 1 moneary ghenng and 0 oherwse. Moneary expanson s calculaed by changes n he dscoun rae, whch s smaller han wo sandard devaon from counry mean and I denoe wh value 1 moneary expanson and 0 oherwse. d. Conrol varables I use domesc and nernaonal he conrol macroeconomc varables n mulple regresson n order o ake no he accoun omed-varables bas. The ls of conrol varables are based on he prevous leraure, parcularly, L and Tang (2010) and Clavo e al., (2004). The ls s mporan snce I m neresed o conrol for facors (unless moneary and fscal varables) whch may affec oupu growh durng he fnancal crss. The lss of varables ha I use n my emprcal research are rade openness, nflaon rae and degree of openness of he capal accoun. 4. Emprcal Resuls 4.1 Descrpve Sascs of Fnancal Crss, Fscal and Moneary Polces Table 2 shows he summary sascs of oupu cos (OC), moneary and fscal polcy ndcaors and conrol varables. I nclude varey fscal and moneary ndcaors such as: fscal expanson/ ghenng (Fscale/Fscal) and moneary expanson/ghenng (Dscoundec/Reservednc and Dscounnc/Reservedec) n order o provde more robus resul. Table 2. Daa descrpon for fscal and moneary polcy Varable Obs. Mean SE Mn Max OC FISCALE FISCALT DISCOUNTINC RESERVEINC DISCOUNTDEC RESERVEDEC TROP INFLATION KAOPEN Source: Aouhor s calculaon. Moreover, I nroduce he conrol varables n order o provde more accurae resul of he effec of moneary and fscal varables on oupu cos, durng he fnancal crss. For hs purpose, I nclude hree conrol varables rade openness (TROP), nflaon (INFLATION) and openness of he capal accoun (KAOPEN). 4.2 Model Esmaes The resul from able 3 show nvesgaons of eq. 1, applyng sandard model for oupu cos for 72 epsodes of fnancal crss n developng and emergng counres. I nclude varey fscal and moneary ndcaors and hree conrol varables (rade openness, nflaon and openness of he capal accoun) n order o provde more robus resul of he effec of moneary and fscal varables on oupu cos, durng he fnancal crss. A posve value of he coeffcen of explanaory varables mean a decrease of oupu cos or oupu loss and negave value of he coeffcen of explanaory varables mean an ncrease of he oupu cos or oupu loss durng he fnancal crses. As seen from able 3, I fnd ou ha fscal and moneary ghenng wll shapely ncrease cos of crss and 113

5 Inernaonal Journal of Economcs and Fnance Vol. 5, No. 9; 2013 coeffcens are sacally sgnfcan (column (3.1) and (3.2)). Furhermore, he evdence shows ha he mpac of moneary expanson on oupu cos s no sacally sgnfcan (boh dscoun rae and nernaonal reserve), whle fscal expanson shows posve mpac on oupu cos and coeffcen s sascally sgnfcan. A one percenage ncrease n he fscal expendure wll decrease oupu cos or cos of he crss by 1.41 percenages. The 70 percenage he varaon oupu cos s explaned by explanaory varables. In he Column (3.2), I exclude he polcy varables ha are sascally nsgnfcan. As seen from (3.2), he number of observaon s reduce due o he mssng of varables for some counres, and he coeffcen of deermnaon s slghly ncrease by Almos I fnd he same resul, he fscal and moneary conracon has sgnfcan negave mpac on oupu cos assocaed wh crses and he coeffcens are sgnfcan. Fscal expanson has posve mpac on oupu cos durng he crss and he coeffcen s sascally sgnfcan. A one percenage ncrease fscal expendure reduces oupu cos by 1.38 percenages and he coeffcen s sgnfcan. Table 3. Regressons wh polcy ndcaors and conrol varables Varable (3.1) (3.2) Inercep ** (0.71) *** (0.83) FISCALE ** (2.44) * (2.50) FISCALT *** (-2.69) ** (2.79 DISCOUNTINC (0.13) RESERVEINC (-0.07) DISCOUNTDEC (0.85) RESERVEDEC ** (-2.23) *** (-2.32) INFLATION (3.48) (3.61) KAOPEN *** (0.84) ** (0.87) TROP (-0.32) (-0.34) R-squared F-es Obs Noe: The able repors oupu loss followng fnancal crses, dependan varables oupu loss o one percen polcy varables wh conrol varables (assocaed -sascs n parenhess), *,**,***, show he sgnfcance a 10, 5 and 1 percen respecvely. Fnally, I fnd ou ha fscal polcy s more effecve ools han moneary polcy durng he fnancal crss n he developng and emergng counres. My resul s conssen wh he resul of Huchson e al., (2010), where hey fnd ha fscal polcy s more effecve ools han moneary polcy. However, my resul s dfferen han he resul of Je (2013) where hey fnd ha moneary polcy s more effecve han fscal polcy. 5. Concluson The paper examnes he effec moneary and fscal polcy on oupu cos or loss durng he fnancal crss for 72 epsodes of fnancal crss n developng and emergng counres from 1980 o The resul suggess ha n developng and emergng counres fscal polcy s more effecve hen moneary polcy durng he fnancal crss. An ncrease of governmen expendure by one percenage reduces oupu cos by approxmaely 1.4 percenages durng he fnancal crss, whle he coeffcen of moneary expanson s sascally nsgnfcan. Moreover, I fnd ou ha moneary and fscal conracon ncrease sgnfcanly oupu cos. Therefore, he macroeconomc polcy mx wh an expansonary fscal polcy wh a neural moneary polcy reduces oupu cos durng he fnancal crss n developng and emergng counres. References Bag, T., & Goldfajn, I. (2002). Moneary polcy n he afermah of a currency crses: The case of Asa. Revew of Inernaonal Economcs, 10(1), hp://dx.do.org/ / Baldacc, E., Gupa, S., & Mulas-Granados, C. (2009). How effecve s fscal polcy response n sysemac bankng crses. IMF workng paper. Rereved from hp:// Blanchard, O., Chouraqu, J. C., Hagemann, R. P., & Saror, N. (1990). The susanably of fscal polcy: New answers o an old queson OECD. Economc Sudes, 15, hp://dx.do.org/ / Calvo, G. (1998). Capal flow and capal marke crses: The sample economcs off sudden sops. Journal of Appled Economcs, 1(1), hp://dx.do.org/1903/

6 Inernaonal Journal of Economcs and Fnance Vol. 5, No. 9; 2013 Goldfajn, I., & Gupa, P. (2003). Does moneary polcy sablze exchange rae followng a currency crss? IMF saff papers, 50. Rereved from hp:// Gupa, P., Mshra, D., & Sahay, R. (2007). Behavor of he oupu durng he currency crss. Journal of Inernaonal Economes, 72, hp://dx.do.org/ /j.jneco Huchson, M., Noy, I., & Wang, L. (2010). Fscal polcy and moneary polcy and he cos of sudden sops. Journal of Inernaonal Money and Fnance. hp://dx.do.org/ /j.jmonfn IMF. (2008, Ocober). Fscal polcy as a counercyclcal ool. World Economc Oulook, (Chaper 5, pp ). Rereved from hp:// IMF. (2008, Ocober). From recesson o recovery: How soon and how srong? World Economc Oulook, (Chaper 3, pp ). Rereved from hp:// IMF. (2010). World economc oulook. Rereved from hp:// Je, L. (2013). The effecveness of fscal and moneary polcy responses o wn crss. Appled Economcs, 45, 27. hp://dx.do.org/ / Kamnsky, G., & Renhar, C. (1999). The wn crses: The causes of bankng and balance of paymen problems. Amercan Economc Revew, 89(3), hp://dx.do.org/ /aer Laeven, L., & Valenca, F. (2008). Sysemac bankng crss: A new Daabase. IMF workng paper, 08/224. Rereved from hp:// Laeven, L., & Valenca, F. (2010). Resoluon of Bankng crss: The good, he bad, and he Ugly. IMF workng paper, 10/146. Rereved from hp:// Appendx A. Fnancal crss epsodes n developng and emergng counres Albana 1994 Kenya 1992 Algera 1990 Korea 1997 Argenna 1980 Lebanon 1990 Argenna 1989 Malaysa 1997 Argenna 1989 Macedona 1993 Argenna 2001 Mexco 1981 Armena 1994 Mexco 1994 Azerbajan, Rep Morocco 1980 Belarus 1994 Mozambque 1987 Brazl 1994 Ncaragua 1990 Bulgara 1996 Ngera 1991 Cameroon 1994 Paraguay 1995 Cenral Afrcan Rep Peru 1983 Chad 1992 Phlppnes 1983 Chle 1981 Phlppnes 1997 Congo, Dem. Rep 1983 Russa 1998 Congo, Dem. Rep 1991 Prncpe 1992 Congo, Rep 1992 Serra Leone 1989 Domncan Repub Sweden 1991 Ecuador 1982 Tanzana 1987 Ecuador 1998 Thaland 1997 Egyp 1980 Togo 1993 Esona 1992 Turkey 2000 Fnland 1991 Ukrane 1998 Georga 1991 Uruguay 1981 Ghana 1982 Uruguay 2002 Gunea-Bssau 1994 Venezuela 1994 Ha 1994 Yemen 1995 Indonesa 1997 Zamba 1995 Jordan 1989 Source: Laeven and Valenca, Sysemac bankng crses: a new daabase, IMF, workng paper. 115

7 Inernaonal Journal of Economcs and Fnance Vol. 5, No. 9; 2013 Appendx B. Daa Sources Varables Daa Sources Real GDP growh rae WDI Dscoun rae/inernaonal reserves IMF, IFS Annual budge balance (% of GDP) IMF, GFS Trade openness WDI Inflaon WDI Capal accoun openness Chn and Io, 2006 Copyrghs Copyrgh for hs arcle s reaned by he auhor(s), wh frs publcaon rghs graned o he journal. Ths s an open-access arcle dsrbued under he erms and condons of he Creave Commons Arbuon lcense (hp://creavecommons.org/lcenses/by/3.0/). 116