ECB policies involving government bond purchases: Impacts and channels

Transcription

1 ECB policies involving government bond purchases: Impacts and channels Arvind Krishnamurthy, Northwestern University Stefan Nagel, University of Michigan Annette Vissing- Jorgensen, University of California Berkeley Preliminary version November 2013 This research was supported by a grant from the Global Markets Institute at Goldman Sachs. Any opinions, findings, conclusions, or recommendations expressed in this paper are those of the authors and do not necessarily reflect the views of Goldman Sachs or the Global Markets Institute.

2 Objective During the European debt crisis GIIPS countries faced large increases in govt. yields. Figure Year Government Bond Yields, Jan 2010 to July 2013

3 Evaluation of ECB policies: Preview ECB aimed to lower yields, by purchasing bonds directly (SMP, OMT) or lending to banks in expectation that they would buy govt. bonds. (LTRO) Q1: Did policies succeed in lowering government bond yields? Yes Q2: How did the policies lower government bond yields? r!! = 1 T! E i! dt! + TermPremium!!!!!! + P!"#,! 1 λ!"#,! + P!"#,! 1 λ!"#,! + SEG! Lowered default risk of government bonds by alleviating multiple equilibrium problems or ``overpaying for bonds Market segmentation/price pressure effects (SMP, LTROs) Some reduction in redenomination risk (LTRO) Q3: Did policies have positive spillovers to the economies more broadly? Yes

4 (1) Securities Markets Programme (SMP) May 10, 2010: ``In view of the current exceptional circumstances prevailing in the market, the Governing Council decided [.] To conduct interventions in the euro area public and private debt securities markets (Securities Markets Programme) to ensure depth and liquidity in those market segments which are dysfunctional. The objective of this programme is to address the malfunctioning of securities markets and restore an appropriate monetary policy transmission mechanism. Aug 7, 2011: ``The Governing Council of the European Central Bank (ECB) welcomes the announcements made by the governments of Italy and Spain concerning new measures and reforms in the areas of fiscal and structural policies. [.] It is on the basis of the above assessments that the ECB will actively implement its Securities Markets Programme. This programme has been designed to help restoring a better transmission of our monetary policy decisions taking account of dysfunctional market segments and therefore to ensure price stability in the euro area.

5 (1) Securities Markets Programme (SMP) Figure 2. ECB bond holdings under the Securities Markets Programme (SMP) - Only government bonds were purchased. - It s our understanding that up to 2 nd announcement purchases were of Greece, Portugal, and Ireland bonds.

6 (1) Securities Markets Programme (SMP) Table 1. Amounts of sovereign bond purchases due to ECB policies Panel A. Securities markets programme, country distribution and average maturity of ECB holdings as of Feb. 21, 2013 Issuer country ECB holdings (EUR billion), principal value Average remaining maturity (years) Italy Spain Portugal Ireland Greece Source: ECB press release. Note: Average maturities at purchase were up to 3 years higher than those stated above since purchased took place from May 2010 to February 2012.

7 (2) Outright Monetary Transactions (OMT) July 26, 2012: Draghi statement at a conference that ``Within our mandate, the ECB is ready to do whatever it takes to preserve the euro. And believe me, it will be enough. August 2, 2012 (ECB press conference): ``The Governing Council, within its mandate to maintain price stability over the medium term and in observance of its independence in determining monetary policy, may undertake outright open market operations of a size adequate to reach its objective. [ ] Furthermore, the Governing Council may consider undertaking further non- standard monetary policy measures according to what is required to repair monetary policy transmission. Over the coming weeks, we will design the appropriate modalities for such policy measures. September 6, 2012: Details announced regarding the OMT: Transactions will be focused on the shorter part of the yield curve, and in particular on sovereign bonds with a maturity of between one and three years. No ex ante quantitative limits are set on the size of Outright Monetary Transactions. As of today, no bonds purchased yet under the OMT program.

8 (2) Outright Monetary Transactions (OMT) OMT was motivated by perceived redenomination risk: Draghi on OMT, Dec 6, 2012: we said that the main aim of the OMT is to remove tail risk to overcome monetary and financial fragmentation of the euro area that would stem from a redenomination risk

9 (3) Three- year Long- Term Refinancing Operations (LTRO) December 1, 2011: Draghi speech to the European Parliament we are aware of the scarcity of eligible collateral [for banks] and suggests that the the most important thing for the ECB is to repair the credit channel. The Financial Times reports that markets interpreted the Draghi speech to indicate an expansion of SMP or a three- year LTRO. December 8, 2011: ECB announcement ``The Governing Council of the European Central Bank (ECB) has today decided on additional enhanced credit support measures to support bank lending and liquidity in the euro area money market. In particular, the Governing Council has decided: To conduct two longer- term refinancing operations (LTROs) with a maturity of 36 months and the option of early repayment after one year.

10 (3) Three- year Long- Term Refinancing Operations (LTRO) Terms: Fixed- rate full allotment with realized loan rate=realized average MRO rate over loan period. New: The commitment to lend over a long period. The ECB had already in Oct 2008 moved to fixed rate full allotment for its regular lending operations. 1 st allotment date (Dec. 21, 2011): 335 billion euros borrowed (by banks, from the ECB) 2nd allotment date (Feb. 29, 2012) 448 billion euros borrowed (by banks, from the ECB)

11 (3) Three- year Long- Term Refinancing Operations (LTRO) A lot of the money borrowed was likely used by banks to buy government bonds. FT: (Dec 9, 2011) French President Nicolas Sarkozy said the ECB s increased provision of funds meant governments in countries like Italy and Spain could look to their countries banks to buy their bonds. This means that each state can turn to its banks, which will have liquidity at their disposal, Sarkozy told reporters at the summit in Brussels. WSJ: (Feb 29, 2012) quotes Intesa s Chairman Andrea Beltratti,: The new funds, which come with a 1% interest rate, will be used in part for a profitable trading strategy regarding Italian government bonds. He said the bank would mainly purchase Italian government bonds with maturities of three years or less, so that they match with the three- year duration of the ECB loans. Estimate of amounts of sovereign bond purchases by banks using 3- year LTRO money: Around 86 billion euros for Italy, 66 billion euros for Spain, not much (if any) for Ireland and Greece and no 2012 data to sort this out for Portugal.

12 Channels for policy impact on government yields: Identification 1) Signaling: Future monetary stance of the ECB (Important channel for US QE) 2) Duration risk: Lower macro risk - > lower duration risk premia 3) Default risk: a. Signal of future fiscal transfers? b. Multiple equilibria: Policy may ``pick the good equilibrium c. Yield change via other channels lowers borrowing costs and thus default risk d. Yield change via other channels may improve macro situation (e.g., less needed austerity or better bank health) 4) Market segmentation: Price pressure leading to violations of standard pricing relations. Policies may alleviate this or induce their own price pressure effects. 5) Redenomination risk: Policy may affect risk of leaving euro/euro breakup

13 Channels for policy impact on government yields: Identification r!! = 1 T! E i! dt! + TermPremium!!!!!! + P!"#,! 1 λ!"#,! + P!"#,! 1 λ!"#,! + SEG! Signaling+duration risk: S! =!!!! E i! dt + TermPremium! EUR swap rate (OIS, i.e. fixed rate in interest rate swap based on EONIA) Default, redenomination, market segmentation: Domestic bond yields (net of EUR swap rate) Foreign- law USD bond yields (net of USD swap rate) [Illiquidity? Less/more default?] CDS rates [payoff in restructuring?] Instruments load differently on unobservable default, redenomination, segmentation factors: Extract with state- space model/kalman filter.

14 Channels for policy impact on government yields: State- space model Unobserved components: x = (Default, redenomination, mkt. seg.) with AR(1) dynamics where F is diagonal, and u! ~N(0, Q). x! = a + Fx!!! + u!, Observable yields and CDS rates: y! = Hx! + w! where w! ~N(0, V), with diagonal V, and where H is a known matrix. Kalman filter & smoother to extract x from y

15 Identifying the channels: Sovereign CDS and redenomination Identification differs between Italy and other countries: For Italy, CDS contracts don t cover redenomination ISDA Master Agreement for CDS contracts: CDS contracts must pay if there is. ``any change in the currency or composition of any payment of interest or principal to any currency which is not a Permitted Currency.. ``Permitted currency means (1) the legal tender of any Group of 7 country.or (2) the legal tender of any country which.has a local currency long- term debt rating of either Aaa or higher Italy is G7. This was also the common perception in the market. For example, Credit Suisse (2010) states that `` by definition (of being a G- 7 member) Italy,, can re- denominate its government debt out of Euros into a new currency without triggering a restructuring, whereas Spain cannot (except in the extremely unlikely event it can keep a AAA rating through the process).

16 Table 2. Identification assumptions Identifying the channels: Assumptions ITALY Sov. Default Corp. Default Redenom. Domestic Mkt. Seg. Foreign Mkt. Seg. Sovereign Yield - Swap Sovereign CDS Sovereign USD Yield - USD Swap Corporate Yield - Swap Corporate CDS SPAIN and PORTUGAL Sov. Default Italy Sov. Default Redenom. Domestic Mkt. Seg. Sovereign Yield - Swap Sovereign CDS Sovereign USD Yield - USD Swap b C ITL Sov. USD Yield - USD Swap ITL Sovereign CDS Foreign Mkt. Seg.

17 Identifying the channels: Spain and Portugal Because CDS capture redenomination and default risk, comparison CDS USD bonds does not reveal foreign market segmentation component Assumption: Market segmentation component of Spanish and Portuguese USD bonds is multiple of that for Italian USD bonds The multiple, b C, must be estimated We impose priors to identify b C (through penalty in likelihood) Redenomination risk component 0 Correlation of redenomination risk and default risk component observed correlation for Italy

18 Identifying the channels: Potential issues with CDS rates Concerns that CDS may not pay off in a debt restructuring that exposes bond holders to losses: In this case CDS rates do not fully reflect default risk Likely mainly an issue in fall 2011/early 2012: o On October 27, 2011 the ISDA determined that the European Union s agreement with banks for a ``voluntary 50 percent write down on their Greek bond holdings would be unlikely to trigger CDS payments. o ISDA later, in March 2012, declared a Greek credit event. Consequence within our identification scheme Italy: Underestimation of the default risk component Spain/Portugal: Underestimation (negative bias) for redenomination risk

19 Identifying the channels: Accounting for non- synchronicity Illiquidity induces non- synchronicity in reaction of asset prices to latent component shocks Especially USD bonds tend to lag the other instruments in reaction to news State- space model modification: o Let yields/cds rates load on lags of latent components, too o with asset- specific geometrically decaying weights o truncated at lag q (to be estimated) o scaled to sum to one

20 Identifying the channels: Event study approach 2- day changes (end of day before to end of day after) in asset prices and extracted latent components (default, redenomination, segmentation) around event dates. SMP: May 10, 2010 and August 7, 2011 OMT: July 26, 2012, August 2, 2012, and September 6, LTRO: December 1, 2011 and December 8, Caveats Other news around LTRO dates (e.g., Monti austerity budget on Dec 5, 2011) Programs may be anticipated to some extent and effectiveness not clear upon announcement Initial financial market reaction not necessarily equals ultimate impact

21 Evidence on overall impact: Bond yields Table 3. Impacts of policies on sovereign bond yields, by policy and country, 2- day changes (bps) Avg. yield 6mo 2yr 5yr 10yr Avg. yield 6mo 2yr 5yr 10yr Policy Ann. date ITALY SPAIN SMP May 10, ND Aug 7, ND Total ND OMT Jul 26, Aug 2, Sep 6, Total year Dec 1, LTROs Dec 8, Total Avg. yields are from Barclay's indices. 2, 5, and 10- year yields are mid- yields from Bloomberg. ND=No data. Bold indicates significance at the 5% level.

22 Evidence on overall impact: Bond yields Avg. yield 6mo 2yr 5yr 10yr Avg. yield 6mo 2yr 5yr 10yr Policy Ann. date PORTUGAL IRELAND May 10, SMP ND ND ND Aug 7, ND ND Total ND ND ND OMT Jul 26, 2012 ND ND ND ND Aug 2, 2012 ND ND ND ND Sep 6, 2012 ND ND ND 5-49 ND Total ND ND ND ND 3- year Dec 1, 2011 ND ND ND LTROs Dec 8, 2011 ND ND ND Total ND ND ND ND

23 Evidence on overall impact: Bond yields Avg. yield 6mo 2yr 5yr 10yr Policy Ann. date GREECE SMP May 10, 2010 ND Aug 7, 2011 ND ND Total ND ND OMT Jul 26, 2012 ND ND ND ND - 78 Aug 2, 2012 ND ND ND ND - 67 Sep 6, 2012 ND ND ND ND - 36 Total ND ND ND ND year Dec 1, 2011 ND ND LTROs Dec 8, 2011 ND ND Total ND ND

24 Evidence on channels: Signaling and duration risk Table 4. Impacts of policies on euro swap (EONIA OIS), 2- day changes (bps) Policy Ann. Date 2yr 5yr 10yr SMP May 10, Aug 7, Total OMT Jul 26, Aug 2, Sep 6, Total year Dec 1, LTROs Dec 8, Total

25 State- space model input: Example of Italy Domestic and foreign- currency sovereign bond yields; CDS rates (maturity 2015): Plus ENI domestic bond and ENI CDS rate (maturity 2015).

26 State- space model parameter estimates: Non- synchronicity Table 5. Weights on contemporaneous and lagged latent factors (up to lag 2) ITALY (q = 7) Lag 0 Lag 1 Lag 2 Sovereign Yield - Swap Sovereign CDS Sovereign USD Yield - USD Swap Corporate USD Yield - Swap Corporate CDS

27 State- space model parameter estimates: Non- synchronicity Table 5. Weights on contemporaneous and lagged latent factors (up to lag 2) SPAIN (q = 9) Lag 0 Lag 1 Lag 2 Sovereign Yield - Swap Sovereign CDS Sovereign USD Yield - USD Swap Italian Sov. USD Yield - USD Swap Italian Sovereign CDS PORTUGAL (q = 9) Lag 0 Lag 1 Lag 2 Sovereign Yield - Swap Sovereign CDS Sovereign USD Yield - USD Swap Italian Sov. USD Yield - USD Swap Italian Sovereign CDS

28 Evidence on channels: Default, redenomination, market segmentation Table 6. Latent yield components, 2- day changes (bps) ITALY SPAIN PORTUGAL Dom Dom Mkt Sov. Mkt Sov. Seg. Def. Red Seg. Def. Red Policy Ann. date Sov. Def. Red SMP May 10, Aug 7, Total OMT Jul 26, Aug 2, Sep 6, Total y LTRO Dec 1, Dec 8, Total Dom Mkt Seg.

29 Evidence on channels: Italy Spread jan jan jan jan2013 Date Default Redenomation Segmentation

30 Evidence on channels: Spain Spread jan jan jan jan201 Date Default Redenomation Segmentation

31 Evidence on channels: Portugal Spread jan jan jan jan2013 Date Default Redenomation Segmentation

32 Additional evidence on channels Corporate CDS rates mostly above swap- adj. corporate yields in Italy: Not much indication of redenomination risk

33 Additional evidence on channels Two- year vs. 10- year swap- adj. sovereign yields: Point to post- LTRO market segmentation/price pressure effects on bonds with maturity < 3 years Italy: Redenomination plus Segmentation jan jul jan jul jan jul jan jul2013 Yield-swap-CDS, 2-year Yield-swap-CDS, 10-year

34 Additional evidence on channels Spain: Segmentation jan jul jan jul jan jul jan jul2013 Yield-swap-CDS, 2-year Yield-swap-CDS, 10-year

35 Additional evidence on channels Portugal: Segmentation jan jul jan jul jan jul jan jul2013 Yield-swap-CDS, 2-year Yield-swap-CDS, 10-year

36 Macroeconomic Impact Strongly positive stock market reaction in GIIPS and core to SMP and OMT, not much to LTRO Table 7: Impact of ECB policies on stock market values, 2- day changes (pct) Panel A. All stocks GIIPS Policy Ann. date Italy Spain Portugal Ireland Greece SMP May 10, Aug 7, Total OMT Jul 26, Aug 2, Sep 6, Total year Dec 1, LTROs Dec 8, Total

37 Macroeconomic Impact Core EMU Policy Ann. date Ger- many France Nether- lands Bel- gium Austria Finland SMP May 10, Aug 7, Total OMT Jul 26, Aug 2, Sep 6, Total year Dec 1, LTROs Dec 8, Total

38 Macroeconomic Impact Financial sector stocks did the best, but even non- financial stock returns are high on SMP and OMT event dates. Suggests spillovers of policies into the broader economy. For many countries, only a small fraction of the large financial sector returns can be explained by the increase in market value of GIIPS sovereign bonds held by financial institutions. Exceptions are Portugal, Greece, Germany and Belgium. Some spillovers to corporate bonds, but much smaller effects than for sovereigns. Very little movement in long- run inflation expectations (as measured by 10- year inflation swaps), suggesting that the ECB has not lost its ``inflation- fighting reputation

39 Macroeconomic Impact: Back- of- the envelope calculation Policy benefits: At least 917B based on market value increase in GIIPS govt bonds and EMU stocks Some double counting via a fraction f of GIIPS sovereign bonds owned by publicly traded banks: Overall market value increase as (1- f)* 251 billion billion= 917 billion. We estimate f 1/3 using resident bank holdings from Bruegel database adjusted and 2011 EBA stress test data Plus (not measured): Market value changes for corporate debt and non- publicly traded firms, along with wage gains via reduced economic contraction.

40 Macroeconomic Impact: Back- of- the envelope calculation Policy cost to ECB/Taxpayers: At most 251B If the fundamental problem is solvency risk: Market value increase in GIIPS govt debt ( 251B) due to transfer (ECB ``over- paying for the bonds or perceived as signal about future fiscal transfers from core countries) If underlying problem is multiple equilibria or market segmentation o Zero for OMT (they won t need to buy if they are credible) o <0 for SMP, as ECB makes a profit on purchases Qualifiers to this large net welfare benefit: We re not accounting for any potential subsidy to future debt (i.e. debt not currently issued). Moral hazard problem. We re not assessing distributional impact (but remember that even German stocks went up ).