The collateral scarcity premia in EU repo markets

Transcription

1 26 February 2016 Milan The collateral scarcity premia in EU repo markets

2 Background Collateral plays a very important role in financial markets: dealers need collateral to finance their market-making activities market participants need collateral to enter into derivatives contracts increasingly significant role after the crisis partly due to regulatory reforms Without good access to high-quality collateral, dealers and market participants would find it more costly to trade negative impact on financial stability, through: reduced liquidity across instruments increased frictions in bond market financing SFTs have a central role in supporting collateral fluidity SFT Regulation will address data gaps 2

3 Our paper We provide evidence of the link between EU repo markets and EU securities lending market, using a new dataset We test empirically the relevance of short selling for repo markets We introduce a novel measure for collateral reuse We analyse the impact of recent changes in monetary policy on collateral scarcity premia 3

4 Overview of EU SFT markets EU repo market Reverse repos Repos Note: Gross value of repos and reverse repos outstanding from around 65 European institutions, in EUR tn. Includes sell-buy backs and buy-sell backs. Given that transactions between counterparties are not netted out, some double counting is likely. Sources: ICMA, ESMA. EU securities lending market cash non-cash cash/non-cash (rhs) Note: Outstanding value of European government bonds on loans against cash/non cash collateral, EUR billion. Ratio of cash/non-cash collateral on secondary axis. Sources: Markit, ESMA The size of the European repo market was around EUR 5.6tn as of June 2015 (possible doublecounting) Repos are increasingly securitydriven transactions As of June 2015, the total value of EU securities on loan was around EUR 470bn, including EUR 294bn for government bonds Increased securities lending activity over the last two years, mainly collateralised with non-cash 4

5 The repo market: General, Specific and Special General collateral (GC) repos are liquidity-driven transactions where the choice of security to be delivered is made after the trade (from collateral pools) DEALERS Cash GC Cash SC In specific collateral repos, the collateral is known before the trade is executed. When the collateral has characteristics that are in high demand, these trades become specials (SC) GC repo SC repo In special trades, buyers bid competitively for the collateral, driving the SC rate lower. 5

6 The collateral scarcity premium We focus on the drivers of the cost of obtaining high-quality collateral, measured as the degree of specialness in government bond repos Specialness = GC rate - SC rate Specialness is a proxy of the scarcity premium to be paid for procuring a specific security (Duffie, 1996) Specialness reflects securities markets dynamics: short selling, liquidity 6

7 The data The sample period (7 March 2013 to 21 September 2015) includes key changes in monetary policy: The introduction of a negative rate on the deposit facility (11 June 2014) The announcement of a QE programme (22 January 2015) The beginning of the ECB securities lending programme (9 April 2015) Data for 7 EA countries (AT, BE, DE, FI, FR, IT, NL): Repo: ICAP RepoFundsRate Securities lending: Markit Securities Finance Bond characteristics and market variables: Datastream 7

8 Specialness evolution GC/SC rates and specialness (bps) Mar-13 Sep-13 Mar-14 Sep-14 Mar-15 Sep-15 GC rate SC rate Specialness Note: Average specialness, GC and SC rates (country means). Weekly data in basis points. The vertical lines indicate, respectively, the introduction of negative deposit rate (grey) and the launch of ECB QE (blue). Sources: RepoFunds Rate (BrokerTec, MTS, ICAP), ESMA. Very low interest rates and moderate volatility Following monetary policy changes, sharp movements in both GC and SC rates

9 Traditional securities lending channel Lenders Borrowers Investors Agents Securities loaned Pension funds Mutual funds Insurance companies Custodian banks Asset managers Collateral Prime brokers Securities dealers Sovereign wealth funds Specialist lenders Banks Central Banks Fees The traditional securities lending channel connects a wide range of buy-side investors with broker/dealers

10 Securities lending: collateral reuse and utilisation rate Lender quantity on loan vs. broker-to-broker activity Mar-13 Jul-13 Nov-13 Mar-14 Jul-14 Nov-14 Mar-15 Jul-15 Broker-to-broker activity Lender quantity on loan Quantity on loan Note: Quantity of European government debt securities on loan, in billion of securities. Utilisation rate: proxy for short selling Mar-13 Sep-13 Mar-14 Sep-14 Mar-15 Sep-15 AT BE DE FI FR IT NL Note: Lender value on loan divided by lendable value, EA government bonds, in %. Broker-to-broker activity = Quantity on loan Lender quantity on loan Utilisation rate = Lender quantity on loan/lendable quantity

11 Empirical methodology Degree of specialness i,t = α + β(broker to broker) i,t +γ(short selling) i,t +δ(bond characteristics) i,t +z(controls) t + ε i,t Degree of specialness: GC SC Broker-to-broker activity: collateral reuse (-) Short selling proxied by utilisation rates (+) Liquidity premium: auction cycle (on-the-run, off-the-run dummies); time to maturity Alternatively time trend and other control variables (excess liquidity, VSTOXX, Eonia) 11

12 Empirical methodology Sample period: 7 March September Before negative deposit rate: 7 March June Between Neg. dep. Rates and QE: 11 June January After QE announcement: 22 January September 2015 Three samples: all bonds, on-the-run and off-the-run Panel regression with fixed effects using robust standard errors (cluster at ISIN level) Lagged values to address potential endogeneity 12

13 Results Baseline model results *** p<0.01, ** p<0.05, * p<0.1 Before Between After (1a) (1b) (2a) (2b) (3a) (3b) Broker-to-broker (-) (-) (+) (+) (-) (-) * * ** ** ** ** Utilisation (+) (+) (+) (+) (+) (+) *** *** *** *** *** ** Average size of (-) (-) (-) (-) (-) (-) *** *** * * *** *** 5y on-the-run (+) (+) (+) (+) (-) (-) * * First day off (-) (-) (+) (+) (+) (+) *** *** Time-to-maturity (-) (-) (-) (-) (+) (+) *** *** *** *** ** ** Time trend (-) (-) (-) (-) (-) (-) *** *** *** *** *** *** Borrower Concentration (-) (-) (-) *** ** Short selling is positively related to specialness; no more significant after ECB QE Broker-to-broker activity switches from negative to positive with negative deposit rates Liquidity premium associated with on-therun bonds is positively related to specialness Impact of negative deposit rates and QE to be analysed further 13 13

14 Results: on-the-run and off-the-run samples ESMA REGULAR USE *** p<0.01, ** p<0.05, * p<0.1 ON-THE-RUN SAMPLE Before Between After (1a) (1b) (2a) (2b) (3a) (3b) Broker-to-broker (-) (-) (-) (-) (+) (+) *** *** Utilisation (+) (+) (+) (+) (+) (+) *** *** *** *** OFF-THE-RUN SAMPLE Before Between After (1a) (1b) (2a) (2b) (3a) (3b) Broker-to-broker (-) (-) (+) (+) (-) (-) ** ** * Utilisation (+) (+) (+) (+) (+) (+) ** * *** *** ** * Short selling is positively related to specialness; no more significant after ECB QE for on-the-run sample Asymmetric impact of different monetary policy changes for on-the-run and off-the-run sample Broker-to-broker activity switches from negative to positive with negative deposit rates for off-therun bonds 14 14

15 Dynamic of specialness around changing policies Neg. Deposit rate ECB QE ECB Securities Lending Programme 1M b/a 1M b/a 2M b/a 1M b/a Broker-to-Broker (-) (+) (-) (-) * * Utilisation rate (+) (+) (+) (+) QE seems to increase specialness but the Securities Lending Programme reduces the effect Post (event) (-) (+) (-) (-) *** *** *** * Post * On-the-run (-) (+) (+) (-) *** * *** p<0.01, ** p<0.05, * p<

16 Conclusions In line with previous studies, our paper confirms empirically that the cost of obtaining high-quality collateral increases with short selling and liquidity premia This is the case even in our sample period, where benign financial conditions prevail. In bear market conditions - when good collateral is needed the most - this may lead to tensions in some asset market segments By reducing the scarcity premia, collateral reuse may under certain conditions alleviate these tensions, but requires transparency and monitoring due to risks associated ECB QE has a statistically significant positive impact on scarcity premia, which is reduced by ECB securities lending programme 16