The Financial and Macroeconomic Effects of the OMT Announcements

Transcription

1 The Financial and Macroeconomic Effecs of he OMT Announcemens Carlo Alavilla Domenico Giannone Michele Lenza ECB Luiss, Ecares-ULB and CEPR ECB and Ecares-ULB ECB-IMF Conference on "Inernaional Dimensions of Convenional and Unconvenional Moneary Policy Frankfur am Main, 29 April 214 The views expressed in his paper are hose of he auhors and do no necessarily reflec he views of he European Cenral Bank and he Eurosysem.

2 Ourigh moneary ransacions - OMTs 1. July 26, 212, during a conference in London, Presiden Draghi said ha he ECB was ready o do whaever i akes o preserve he euro wihin he limis of is mandae. 2. Augus 2, 212, during he press conference afer he Governing Council meeing, Presiden Draghi announced ha ECB may underake ourigh open marke operaions. 3. Sepember 6, 212, he ECB's Governing Council announced echnical feaures. Key feaures Condiionaliy: only sovereign bonds of counries ha have enered an agreemen wih he EA rescue vehicles (EFSF and ESM). No arges for he yields are se The ECB s holdings do no have senioriy over privae crediors. Mauriy of up o 3 years 2

3 This paper: quaniaive impac of OMTs announcemens Scenario Analysis used o assess he impac of he OMT Announcemens 1. Assess he high-frequency impac of OMT on sovereign bond yields of he Big4 Even-sudy Possible caveas 2. Measuring he macro Impac of OMT muli-counry BVAR model as difference beween a Policy and a No-policy scenario No-Policy scenario: Uncondiional Forecas of he enire model (baseline) Policy Scenario: imposing he esimaed high-frequency impac 3

4 Lieraure U.S. Modigliani and Such (1966, 1967) Gagnon e al. (211) Swanson (211) Krishnamurhy and Vissing-Jorgensen (211) D Amico and King (213) Chen, Curdia, and Ferrero (212) Chung, Lafore, Reifschneider, and Williams (212) Baumeiser and Benai (213) OT QE QE QE QE Macro impac Macro impac Macro impac Euro Area Eser and Schwaab, 212; Ghysels, Idier, Manganelli, and Vergoe, 212; Rivola, 212; Szczerbowicz, 212 Krishnamurhy and Vissing-Jorgensen (213) Lenza, Pill, and Reichlin (21) SMP SMP SMP SMP SMP, LTRO, OMT Macro Impac 4

5 Ouline 1 The impac of he OMTs on he yield curve 2 Macroeconomic Effecs 3 Conclusions EB seminar 11 March 214: Exended Period and Forward Guidance 5

6 Even sudy Analysis y n i i, i 1 SMP LTRO OMT FG 1-day 2-day 1-day 2-day 1-day 2-day 1-day 2-day 1-year Germany France Ialy Spain year Germany France Ialy Spain year CDS Germany France Ialy Spain D

7 Even sudy Analysis Some Shorcomings: Hp: policy changes are immediaely incorporaed in prices and heir effecs are persisen. Disored if oher evens influenced he arge variable on announcemen days (evens should be dominan evens for he idenified even day. If oher significan economic news arrives during he period and poenially creaes measuremen error problems for he even sudy.) Inabiliy of capuring possible lagged effecs and reversals For deails see Alavilla-Giannone (214): 7 Markes Beliefs abou he Effeciveness of Non-sandard Moneary Policy Measures

8 Narraive Evidence: 151 variables Euro Area France Germany Ialy Spain Business Climae Ind. Bank of France Bus. Senimen Budge (% of GDP) Budge Balance (Year o dae) Adj. Real Re. Sales YoY ECB Ineres Raes Business Confidence Indicaor Capial Invesmen Business Confidence CPI (MoM) Curren Accoun SA Cenral Gov. Balance (Euros) Consrucion Invesmen Consumer Conf. Ind. sa CPI (YoY) Consumer Conf. Consumer Confidence Indicaor CPI (MoM) CPI (NIC incl. obacco, MoM) CPI (Core Index) (MoM) CPI Core (YoY) CPI (MoM) CPI (YoY) CPI (NIC incl. obacco, YoY) CPI (Core Index) (YoY) CPI Esimae (YoY) CPI (YoY) Curren Accoun (EURO) Defici o GDP CPI (EU Harm.) (MoM) Curren Accoun nsa Consumer Spending (MoM) Domesic Demand Governmen Spending CPI (EU Harm.) (YoY) Economic Conf. Consumer Spending (YoY) Expors Hourly Wages (MoM) GDP (Consan SA) (QoQ) GDP s.a. (QoQ) CPI Expors SA (MoM) Hourly Wages (YoY) GDP (Consan SA) (YoY) GDP s.a. (YoY) CPI Facory Orders MoM (sa) Impors House Price Index QoQ Gov Deb/GDP Raio CPI Ex Tobacco Index Facory Orders YoY (nsa) Indusrial Orders n.s.a. (YoY) House Price Index YoY Gov Expend (QoQ) France Reail PMI GDP nsa (YoY) Indusrial Orders s.a. (MoM) Ind. Oupu WDA (YoY) Gross Fix Cap (QoQ) GDP (QoQ) GDP s.a. (QOQ) Ind. Prod. nsa(yoy) PPI (MoM) Household Cons (QoQ) GDP (YoY) GDP wda (YoY) Ind. Prod. sa (MoM) PPI (YoY) Ind. Prod. sa (MoM) Housing Perm. 3M YoY% Chg. GfK Cons. Conf. Survey Ind. Prod. wda(yoy) Real Re. Sales (YoY) Ind. Prod. wda (YoY) Housing Sars 3M YOY% Chg. Governmen Spending Ind. Sales n.s.a. (YoY) Cons. Confidence Indus. Conf. ILO Mainland Unempl. Rae IFO Business Climae Ind. Sales s.a. (MoM) Trade Balance (Mln Euros) Labour Coss (YoY) ILO Unemploymen Rae Impor Price Index (MoM) PMI Manufacuring Unempl. MoM Ne ( s) M3 s.a. (YoY) Impors (QoQ) Impor Price Index (YoY) PMI Services Unempl. Rae (Survey) M3 s.a. 3 mh ave. Ind. Prod. (MoM) Impors PPI (MoM) PPI (MoM) Ind. Prod. (YoY) Impors SA (MoM) PPI (YoY) PPI (YoY) Mainland Unemp. Chg. (s) Ind. Prod. YoY (nsa wda) Privae Consumpion Re. Sales (MoM) Manuf. Prod. (MoM) Ind. Prod. (YoY) Reail Sales (YoY) Re. Sales (YoY) Manuf. Prod. (YoY) Ind. Prod. MoM (sa) Reail Sales s.a. (MoM) Services Conf. Non Farm Payrolls (QoQ) PMI Manufacuring Reailers Confid. General Trade Balance Own Company Prod. Oulook PMI Services Toal invesmens Trade Balance sa PMI Manufacuring Privae Consumpion Trade Balance (Toal) (Euros) Unempl. Rae PMI Services Producer Prices (MoM) Trade Balance Eu (Euros) Ind. New Ord. NSA (YoY) PPI (MoM) Producer Prices (YoY) Trade Balance Non Eu (Euros) Ind. New Ord. SA (MoM) PPI (YoY) Reail Sales (MoM) Unempl. Rae PMI Composie Producion Oulook Indicaor Reail Sales (YoY) Unempl. Rae (s.a) PMI Manuf. Toal Jobseekers Trade Balance Unempl. Rae (SA) PMI Services Trade Balance (Euros) Unempl. 8 Chg. ( s) ZEW Survey (Econ. Sen.) Wages (QoQ) Unempl. Rae (s.a)

9 Narraive Evidence Evens: Ourigh Moneary Transacions Draghi s speech Announcemen Technical feauures Classical: y n i 1 D i i, 26 Jul 12 2 Aug 12 6 Sep 12 Daily Variables 2 day Even Window Counry Variable Classical Conrolled DE Bond 2 year.8.1 FR Bond 2 year.4.1 IT Bond 2 year 1.99 *** 1.75 *** ES Bond 2 year 2.34 *** 2.9 *** DE Bond 1 year.23 *.29 * FR Bond 1 year.9.4 IT Bond 1 year.82 ***.63 *** ES Bond 1 year 1.15 ***.96 *** y Conrolled n i 1 D i i, News i, 9 n= number of evens Sample: 2-Jan-28-1-Feb-214 9

10 Lagged effecs? IT: Bond 2-year ES: Bond 2-year /7/12 1/1/12 1/1/13 1/7/12 1/1/12 1/1/13 DE: Bond 2-year FR: Bond 2-year /7/12 1/1/12 1/1/13 1/7/12 1/1/12 1/1/13

11 Even sudy Analysis: size of he window 26 July Augus Sepember Spain - ask Spain - bid Ialy - ask Ialy - bid 11

12 Ouline 1 The impac of he OMTs on he yield curve 2 Macroeconomic Effecs 3 Conclusions EB seminar 11 March 214: Exended Period and Forward Guidance 12

13 Muli-counry BVAR Counries : Ialy Spain Germany France Variables: Counry variables Real GDP GDP Deflaor M3 ~ N Loans o Privae secor (NFCs+HHs) Ineres rae on 2 year Gov. Bonds Ineres rae on 1 year Gov. Bonds Euro area variables EONIA rae Fuure Implied bond marke volailiy y C B, 1 y 1... B p y - p Real GDP IT ES DE FR 17,3% 1,9% 28,2% 21,% % of EMU 77,3% Esimaion uncerainy may make he model unsable/unreliable. Curse of dimensionaliy? Need o limi variabiliy owing o esimaion error: bayesian shrinkage (De Mol e al. 28, Banbura e al. 21) Hyerarchical approach o se he degree of shrinkage (Giannone e al. 212) 13

14 Elasiciy Rubric o a change in he policy rae: moneary policy shock We idenify he moneary policy shock by means of a recursive scheme which generalizes, in a cross-counry framework, he scheme of Chrisiano e al. (1999) Idenificaion resricions: - Policy variable is he euro area shor-erm ineres rae - The macro block (GDP and GDP deflaor, for each counry) is no allowed o reac o he change in he policy rae in he quarer in which he change happens - The financial block (bond raes, money and credi aggregaes, for each counry) is allowed o reac o he change in he policy rae in he quarer of he shock The policy rae is only a funcion of GDP and GDP deflaor a ime and he lags of all he variables (a sor of generalized Taylor rule). 14

15 Transmission Mechanism: Real GDP Ialy Spain Germany France Noes: Solid line indicaes he median response, he doed lines refer o he 16h and 84h percenile of he poserior disribuion of he IRFs.

16 Transmission Mechanism: Consumer prices Rubric.1 Ialy.1 Spain Germany.1 France Noes: Solid line indicaes he median response, he doed lines refer o he 16h and 84h percenile of he poserior disribuion of he IRFs. 16

17 Scenario Analysis Scenario analysis is designed o address he following quesion: Given he knowledge of he economy a ime ( Ω ) wha is he prediced pah of fuure ime series ( X+h ) condiional on he fuure policy change? ( z * ) To answer his quesion we compue he condiional expecaion (see Banbura e al. 214 for he algorihm): X * E X h h z, No-policy (uncondiional, baseline) forecass: X h E X h 17

18 Measuring he impac The difference in wo condiional expecaions ha differs for he informaion se. Y * X z EX OMT E h h, h Noe ha he variable z could in principle be consrained for differen ime periods depending on he policy pah. IMPORTANT REMARKS: The algorihm exracs he mos likely combinaion of shocks ha, given pas regulariies, could have generaed he scenario pahs (an alernaive would be o pick specific idenified shocks) All he scenarios assume ha he srucure of he economy (refleced in he esimaed coefficiens) and he naure and he relaive imporance of differen shocks (refleced in he esimaed covariance marix of he shocks) remain he same as in he pas. 18

19 From Narraive Evidence o Macro-Impac Rubric The OMT scenario Impac: IT,ES,DE,FR GDP IT,ES,DE,FR Price Pah: IT-2y ES -2y DE -2y FR -2y EA policy rae Projecion horizon: 3 years 19

20 The impac of OMT Announcemens Variables Effec (pp) Probabiliy of Posiive Effec GDP,34,6 Germany Price,28,67 Loans 1,8,9 France Ialy Spain GDP Price Loans GDP Price Loans GDP Price Loans,46,64,28, ,22 1,5,81 1,21,86 3,58,82 2,1,8,74,75 2,31,9 2

21 The Rubricimpac of OMT Announcemens Variables Effec (pp) Probabiliy of Posiive Effec GDP,34,6 Germany Price,28,67 Loans 1,8,9 France Ialy Spain GDP Price Loans GDP Price Loans GDP Price Loans,46,64,28, ,22 1,5,81 1,21,86 3,58,82 2,1,8,74,75 2,31,9 21

22 The impac of OMT Announcemens Variables Effec (pp) Probabiliy of Posiive Effec GDP,34,6 Germany Price,28,67 Loans 1,8,9 France Ialy Spain GDP Price Loans GDP Price Loans GDP Price Loans,46,64,28, ,22 1,5,81 1,21,86 3,58,82 2,1,8,74,75 2,31,9 22

23 Impac on GDP and HICP for IT and ES over 3 years Rubric 4 Ialy - GDP growh 4 Spain - GDP growh % 2 2.% quarers quarers 4 Ialy - Price Level 4 Spain - Price Level % 2 1.7% quarers OMT announcemen idenificaion No impac effec on GDP and prices No long-erm effec on EONIA, German and French 2y bond yields Effec on Ialian and Spanish yields: -1.75% for IT, -2.9% for ES quarers

24 GDP for IT and ES Ialy Spain OMT announcemen idenificaion No impac effec on GDP and prices No long-erm effec on EONIA, German and French 2y bond yields Effec on Ialian and Spanish yields: -1.75% for IT, -2.9% for ES 24

25 The impac of OMT announcemens: 1-year bond raes IT.5 ES DE 1 FR Noes: Black line = Median; Shaded area refer o he 25 h -75 h and 16 h 84 h percenile of he poserior disribuion of he IRFs. 25

26 The impac of OMT announcemens: bond volailiy Noes: Black line = Median; Shaded area refer o he 25 h -75 h and 16 h 84 h percenile of he poserior disribuion of he IRFs. 26

27 Conclusions OMT announcemens associaed wih a posiive and saisically significan economic effec Increase in economic aciviy (abou 2% higher GDP) and inflaion Relaively robus increase in Credi, up o 3% Transmission over he enire yield curve (1-year Bond raes fall abou 1%) Significan decrease in bond marke volailiy 27