关于“收益率曲线下的货币政策……”的讨论.pdf

1、DISCUSSION OF”MONETARYPOLICYALONGTHEYIELDCURVE.”Juan Pablo NicoliniFRB Minneapolis and Universidad Torcuato Di TellaINTRODUCTIONStudies a life-cycle model with price frictions.Real effects of monetary policy through the interactionbetween two ingredients.Tractable model soProve a series of results f

2、or particular cases.Use numerical solutions to study the general case.Very provocative paper:long run REAL effects of monetarypolicy.WHYPROVOCATIVE?Two Illustration QTM.Lucas 1980Two equations on three nominal variablesMt+1Mt,Pt+1Pt,it(Pt+1/Pt)=(Mt+1/Mt)gyt+1(Pt+1/Pt)=it+1rt+1Neutrality of money wit

3、h flexible prices.Mt+1Mtor itdetermined by policy.No need to take a stand on policy instrument.LUCAS1980LUCAS1980LUCAS1980WHYPROVOCATIVE?Inflation&Interest Rates in Brazil18/56ADDITIONALEVIDENCEFollowing Lucas(1980),we apply the filter to the data of otherOECD countries for the period 1960-2017.Use=

4、100 to filter all series in all countries.We“test”ONLY one of the two QTM implications for thosecountries.WHY=100?U.S.Nominal Interest Rates196019701980199020002010Year05101520%Data =6.5 =100Data and HP-filtered series196019701980199020002010Year-50510%=6.5 =100Cyclical componentIWe use=100for all s

5、eries and all countries.Medium run=4 years.29/56AUSTRALIA19601970198019902000Year-505101520%corr.=0.69titAUSTRALIAFILTERED19601970198019902000Year-505101520%corr.=0.89tit -rtUSCANADA19601970198019902000Year05101520%corr.=0.80titCANADAFILTERED19601970198019902000Year-505101520%corr.=0.97tit -rtUSSPAI

6、N19601970198019902000Year051015202530%corr.=0.50titSPAINFILTERED19601970198019902000Year-5051015202530%corr.=0.71tit -rtUSGREATBRITAIN19601970198019902000Year051015202530%corr.=0.68titGREATBRITAINFILTERED19601970198019902000Year-505101520%corr.=0.95tit -rtUSSO