SONET Sterling Secured Overnight Executed Transactions

Transcription

1 SONET Sterling Secured Overnight Executed Transactions Update: February 2017 Contacts: Oliver Huggins & Chris Woods Phone:

2 Disclaimer This document has been provided to you for informational purposes only and is intended as a broad overview of certain aspects of the proposed SONET rate. All information and opinions of whatsoever nature contained herein are preliminary and remain at all times subject to definitive documentation and regulatory review, where applicable. This document does not, and does not purport to, contain a detailed description of any aspect of the SONET rate or any other topics discussed in this document, and it has not been prepared for any specific person. This document does not, and does not seek to, constitute advice of any nature. You may not rely upon the contents of this document under any circumstance and should seek your own independent legal, investment, tax and other advice. The information and any opinion contained in this document do not constitute a recommendation or offer with respect to any derivative contract, financial instrument, security or service. Neither the London Stock Exchange Group plc nor any of its affiliates makes any representation, warranty or guarantee (whether express or implied) that the contents of this document are accurate, complete or up-to-date, and make no commitment to offer any particular product or service. Neither the London Stock Exchange Group plc nor any of its affiliates shall have any liability for any losses, claims, demands, actions, proceedings, damages, costs or expenses arising out of, or in any way connected with, the information contained in this document, other than liability that cannot be excluded by applicable law. Copyright in the contents of this document belongs to the London Stock Exchange Group plc. The contents of this presentation are based on advanced plans as developed to date, but subject in all cases to definitive documentation and regulatory review, where applicable. FTSE Russell is not an investment firm and this presentation is not advice about any investment activity. None of the information in this presentation or reference to a FTSE Russell index constitutes an offer to buy or sell, or a promotion of, a security. This presentation is solely for informational purposes. Accordingly, nothing contained in this presentation is intended to constitute legal, tax, securities, or investment advice, nor an opinion regarding the appropriateness of making any investment through our indexes. About FTSE Russell FTSE Russell is a leading global provider of benchmarking, analytics and data solutions for investors, giving them a precise view of the market relevant to their investment process. A comprehensive range of reliable and accurate indexes provides investors worldwide with the tools they require to measure and benchmark markets across asset classes, styles or strategies. FTSE Russell index expertise and products are used extensively by institutional and retail investors globally. For over 30 years, leading asset owners, asset managers, ETF providers and investment banks have chosen FTSE Russell indexes to benchmark their investment performance and create ETFs, structured products and index-based derivatives. FTSE Russell is focused on applying the highest industry standards in index design and governance, employing transparent rules-based methodology informed by independent committees of leading market participants. FTSE Russell fully embraces the IOSCO Principles and its Statement of Compliance has received independent assurance. Index innovation is driven by client needs and customer partnerships, allowing FTSE Russell to continually enhance the breadth, depth and reach of its offering. FTSE Russell is wholly owned by London Stock Exchange Group. For more information, visit Page 2

3 Contents Executive summary Page 4 Developments since November 2016 Page 5 SONET: Risk-Free Rate Page 6 Appendix Page 20 Page 3

4 SONET: Executive summary Background and Timing The FSB published its report on interest rate benchmark reform in July 2014 (following direction from the G20) 1 In the UK, the Financial Conduct Authority is overseeing the reform of LIBOR and the Bank of England is overseeing the development of sterling risk free rates (RFRs). The BoE has convened a Working Group to look at alternative RFRs The Bank of England have written a letter to the Working Group expressing a desire for a recommendation on the sterling risk free rate in Q Risk Free Rate Candidates A number of alternative rates have been considered by the Bank of England Working Group including: Reformed SONIA, an unsecured rate to be administered by the BoE SONET, FTSE Russell s secured rate candidate The LSEG SONET team presented to the BoE Working Group on 2 nd November Following this, LSEG has worked closely with a WG delegation to test and refine SONET SONET reflects the dynamics of the secured funding market and captures a significant portion of daily activity (up to 85 billion per day) Current Activities The RFR must be able to support an OIS transition that can plausibly take place over two to three years 2 FTSE Russell plans to publish an indicative version of SONET in Q1 2017, in line with BoE timelines The SONET team is now engaging market participants on the features and use of SONET and possible OIS transition path LSEG welcomes any feedback or thoughts on SONET Market participants are encouraged to provide feedback via SONET@FTSERussell.com [1] [2] Page 4

5 SONET: Developments since November 2016 In November and December 2016, the SONET team worked closely with the Working Group to scrutinise and refine the SONET methodology. Analysis covered: Provenance and dynamics of rate spikes Relative dynamics of SONET and SONIA, especially in periods of monetary policy activity (e.g. Quantitative Easing) Dynamics of over a dozen different calculation methodologies focussing on spike days, Bank Rate cut days and IMM days Effect on the SONET rate of the sub-components that make up the repo dataset: DBV, GC, Specifics and Special transactions; cleared and uncleared See the appendix for a sample of the analysis performed As a result we have refined the Methodology of SONET: SONET will be calculated using a Volume Weighted Average with 25% iterative filtering across all trades The Methodology will be reviewed periodically by the SONET Advisory Committee in the future, to ensure SONET continues to capture the market FTSE Russell is in the process of setting up the SONET Advisory Committee. The first meeting is expected to take place in Q The focus of the SONET team is now on engaging stakeholders on the utility of SONET and how the OIS market may transition from an unsecured benchmark to a secured benchmark, should a secured benchmark be chosen LSEG welcomes any feedback or thoughts on SONET Market participants are encouraged to provide feedback via SONET@FTSERussell.com Page 5

6 SONET: Risk Free Rate Page 6

7 SONET: Overview SONET is a measure of sterling risk free overnight funding For use as the risk free rate for sterling markets To benchmark the daily funding activity of commercial banks and other participants To underpin the discounting of long term sterling cash flows As a reference rate for use in derivative and other contracts Priced to encourage widespread adoption Historical data will be published to encourage participant engagement Made widely available throughout the industry on a FRAND basis Resilience to manipulation through multiple lines of defence By design: 25% iteratively filtered Volume Weighted Average SONET will benefit from FTSE Russell s comprehensive governance structure SONET Advisory Committee to include market participants (buy and sell side) Compliance with EU Benchmark Regulation and IOSCO Will benefit from the support of LSEG s range of capabilities Benefitting from the full suite of capabilities within LSEG: Index Administration, Execution Facilities and Post Trade In partnership with market participants and other infrastructure providers through our Open Access model SONET will launch in Q Page 7

8 SONET: Benefits for market participants SONET as a secured overnight risk free benchmark Represents actual activity Continuing relevance in the future Resilient volumes in stressed markets Proxy for derivatives funding Up to 85 billion daily Secured funding and commercial lending Wide participant base Secured funding highly likely to remain widely used SONET Advisory Committee role Secured lending volumes more likely to remain stable than unsecured Aligns to actual funding practice Over 1 trillion 1 of GBP derivatives daily Initial buy-side feedback indicates that a secured rate is the most appropriate choice for Sterling Risk Free Rate SONET aligns to actual daily derivatives funding practice and is opposable, unlike LIBOR SONET captures a significant portion of the secured market (up to 85 billion in daily volume; c65% market coverage) SONET methodology is designed to be robust against risks of manipulation, and the governance model ensures integrity and ongoing relevance Given the ultimate objective of the Working Group to move a significant portion of derivatives from LIBOR to the RFR 2, it is imperative to make the correct choice of RFR in the first instance [1] Source: [2] Source: Page 8

9 SONET: Definition The rate is defined in two parts, to facilitate the evolution of the benchmark in the future without disruption to contracts and processes that use it: The Underlying Interest is an enduring statement of the economic concept that SONET seeks to measure The Methodology describes how the UI is currently measured, including inputs and calculation This structure allows for the future evolution of the Methodology, in consultation with the SONET Advisory Committee, without requiring a change in the Underlying Interest SONET Definition Underlying Interest Methodology Monitoring and Maintenance Sterling SONET is a measure of the sterling risk free reference rate for secured overnight funds Iteratively filtered 1 (removing 25% volume from all trades) Volume Weighted Average of the interest rates on: Sterling cleared and uncleared repo transactions secured against government securities With one business day maturity To settle on the day that is used in the SONET index calculation There is no minimum transaction size The Methodology has been determined in conjunction with expert market practitioners. It is representative of the repo market, in particular on dates where some differentiation is apparent between DBV, GC and Specific Repos The Methodology will regularly be reviewed under the direction of the FTSE Russell SONET Advisory Committee, who may then recommend methodology adjustments to ensure SONET continues to be an appropriate reflection of the sterling repo market [1] All data is aggregated and transactions ranked by rate in descending order (highest to lowest) prior to calculating. VWA is calculated for the entire dataset. The transaction with repo rate furthest from the calculated VWA is then removed from the data set. The new VWA of the remaining trades is calculated. The process is repeated, and transactions removed until 25 percent of the original volume has been removed. Page 9

10 SONET: Market coverage SONET market coverage: average daily volume and number of trades From 1 July to 16 September : SONET has a peak transaction volume of 68bn The mean daily transaction volume was 56bn, with an average of 1200 daily trades We estimate SONET covers 64% of the total market 2 SONET Reformed SONIA Mean daily transaction volume ( bn) Mean trade size ( m) Median trade size ( m) Mean daily number of trades Source: LSEG internal analysis for period 01/07/16 16/09/16 80 bn Volume Daily Transaction Volumes 3 No. of trades 2, bn 1, bn 1,600 1, bn 1, bn 1, bn bn bn 200 bn Feb 2016 Apr 2016 Jun 2016 Aug 2016 Oct 2016 Dec SONET (LHS) Reformed SONIA (LHS) Reformed SONIA (RHS) SONET (RHS) Source: BoE October 2016 paper The reform of SONIA and supporting data; and LSEG internal analysis. Reformed SONIA volumes contain full population contributing to SMM from 1 st July 2016 [1] This period was chosen to be in line with BoE The reform of SONIA consultation paper. More history is available. [2] As at 9 th December See appendix for further details. [3] Reformed SONIA data only available to 16/09/16 Page 10

11 SONET: Administration SONET will be administered by FTSE International Ltd Responsible for administration and calculation FTSE Russell will agree terms with data suppliers (including codes of conduct where appropriate) Maintain oversight arrangements Calculate the rate and ensure appropriate contingency arrangements are in place Consistent with IOSCO Principles for Financial Benchmarks FTSE Russell is fully IOSCO compliant, and SONET will be administered according to these principles FTSE Russell IOSCO statement can be viewed at: Consistent with European benchmark regulation FTSE Russell will apply for FTSE International Ltd to be authorised as a benchmark administrator under the European Benchmark Regulation These regulations can be viewed at: Benefit from external assurance FTSE Russell routinely commissions externally assured reports on its internal controls; SONET will be included in such reports Page 11

12 SONET: Governance FTSE Russell Governance Board: Benefits from many years experience overseeing some 80 index families FTSE Russell Governance Board Provides ultimate oversight of methodologies and approval of proposed enhancements, following input from internal working groups and external advisory committees Index methodology group Methodology & procedures implementation group Security events advisory group External Committees Internal Working Groups: Extensive operations experience and in-house research teams to analyse and advise Consider operational aspects of the maintenance of the benchmark including the design of controls and procedures Evaluate the operational feasibility of proposed methodology enhancements FTSE Russell Policy Advisory Board Equity Advisory Committees Americas EMEA Asia Pac Other Committees Nationality Country Classification ICB Fixed Income Committees Americas EMEA Asia Pac Partner Advisory Committees Bursa Malaysia FTSE JSE FTSE ST External Advisory Committees: SONET Advisory Committee to benefit from the continued involvement of BoE Working Group members, users, and key stakeholders (the BoE to be invited in an observer capacity) Chaired by a senior independent practitioner Two way conversation between Committee and FTSE Russell Governance Board The Committee will ensure that benchmarks are managed and calculated according to the published methodology, and maintain the methodology under review to ensure that SONET continues to meet market needs and accurately reflects the underlying market SONET FTSE EPRA NAREIT External FTSE NAREIT FTSE ESG Internal The SONET Advisory Committee will meet quarterly and discuss any changes that may be required Page 12

13 SONET: Publication, republication & contingency 00:00 18:00 T 0 T 0 21:00 T 0 09:00 T +1 Daily transaction window Data validation 21:00 SONET publication 09:00 Republication if required Daily Publication SONET will be published on the same day at 21:00 GMT SONET rate will include transactions between 00:00 and 18:00 GMT Alongside the SONET rate, FTSE Russell will publish additional information: Total volume & number of transactions; and Additional rate statistics Surveillance & Data Validation Ex-ante data checks, validation & market surveillance will be undertaken FTSE Russell will perform end-of-day analysis on data provided and investigate any anomalies as required Contingency Arrangements FTSE Russell will publish a daily rate in all circumstances, and has in place robust fallback solutions for calculation and data sourcing Where circumstances require, FTSE Russell will publish a short term contingency rate following the published contingency plan methodology, to be determined in consultation with the SONET Advisory Committee (including market participants) If the underlying economic reality changes such that SONET is no longer reflective of the underlying market, FTSE Russell will work alongside the SONET Advisory Committee to determine the appropriate evolution of the SONET methodology Republication Arrangements Where an error is identified in the SONET published rate, FTSE Russell propose to republish only if correction of the error results in a material rate move of 2 basis points or more In all circumstances, any republished rate would be made available at 09:00 GMT the following day SONET will benefit from FTSE Russell 24 hour global client service and operational support Page 13

14 SONET: Surveillance SONET is difficult to influence by design 80 bn 70 bn 60 bn 50 bn 40 bn 30 bn Trade size required to move rate 1 bp SONET is difficult to influence by design. As a result, a significant volume of manipulative trades are required to shift the rate: On average over 2016, a trade of 21bn would be required to move the rate by 1 basis point. This is over 800x the median daily trade size 20 bn 10 bn 0 bn 01/16 02/16 03/16 04/16 05/16 06/16 07/16 08/16 09/16 10/16 11/16 Over 2016, the minimum trade size required to move the rate by 1 basis point is 2.6bn, 100x the median trade size for that day Source: LSEG internal analysis Supported by targeted and robust surveillance frameworks CONTRIBUTOR Ex-ante checks Contributors are best placed to perform exante checks on the integrity of the data: Earlier sight Counterparty information Timestamps Checks that may be performed: Nil / 0 rates (may be valid) Trades cancelled or amended after feed snap Fat finger errors DATA TRANSFER Robust Control Framework Audited control frameworks around creation and delivery of data file Documentation and training of individuals involved in the submissions process ADMINISTRATOR Ex-Ante and Ex-Post Analysis Analysis of combined data set: time series trend analysis daily cross-sectional analysis FTSE Russell will also perform daily analysis of: Min, Max, Mean, SD Total Volume Basis Outliers Any data that fails checks will be flagged and investigated Page 14

15 SONET: Pricing SONET will be: Priced to encourage widespread adoption Made widely available throughout the industry Pricing Strategy SONET will be made available on a FRAND basis Charges will be structured to encourage adoption Partners will be incentivised to contribute and make the rate viable Usage will be licensed, with incentives to encourage early adoption of the benchmark rate Pricing will be structured to encourage adoption of SONET and drive liquidity Page 15

16 SONET: Market dynamics Feb 2016 Jan 2017 A direct comparison of SONET, Reformed SONIA (trimmed VWA), Current SONIA and the Bank Rate, between February 2016 January Rate % SONIA vs SONET vs Base Rate 77 Bp Basis Feb 2016 Mar 2016 Apr 2016 May 2016 Jun 2016 Jul 2016 Aug 2016 Sep 2016 Oct 2016 Nov 2016 Dec 2016 Jan Basis Reformed SONIA vs SONET (RHS) SONET (LHS) Reformed SONIA (LHS) Current SONIA (LHS) Bank Rate (LHS) Source: BoE February 2017 paper The reform of SONIA and supporting data; Bloomberg; and LSEG internal analysis. Reformed SONIA volumes contain full population contributing to SMM from 1st July 2016 Page 16

17 SONET: Three-month compounded rate Nov 2011 Jan 2017 A comparison of three-month compounded SONET and current SONIA 0.80 Rate % Three-month compounded SONIA vs SONET Bp Basis SONET SONIA Correlation of daily change Annualised daily volatility (%) Average absolute daily basis (bps) (5.00) Nov 2011 May 2012 Nov 2012 May 2013 Nov 2013 May 2014 Nov 2014 May 2015 Nov 2015 May 2016 Nov 2016 (10.00) Basis (RHS) 3 month Compounded SONIA (LHS) Source: Bloomberg and LSEG internal analysis. Page 17

18 SONET: Transition A feasible, market led transition plan is required 1 Initial buy-side feedback indicates that a secured rate is the correct choice for Sterling Risk Free Rate Given the ultimate objective of the Working Group to move a significant portion of derivatives from LIBOR to the RFR 1, it is imperative to make the correct choice of RFR in the first instance The requirement is for an OIS transition that would plausibly succeed over 2-3 years 1 LSEG can seek to support the process of a smooth, market-led migration of the OIS market from SONIA to SONET With Group s full suite of capabilities through our Open Access model; Combined with adoption by market participants and support from regulators; and A viable transition roadmap We have developed a path to transition which could plausibly succeed over 2-3 years [1] Source: Page 18

19 SONET: Transition summary timeline Potential roadmap (in consultation with market participants and BoE): SONET Launch : Spot SONET market usage develops 2: Term extension & OIS Transition SONET OIS to clear from inception, supporting liquidity build SONET OIS market sufficiently liquid to support IRS discounting An OIS market Transition will progress in four stages; 1. Building SONET spot market usage: Use of the overnight index among market participants 2. Term extension & OIS Transition: A liquid market in the short term OTC OIS and futures markets develops. Clearing of SONET OIS products could facilitate transition of SONIA OIS liquidity to SONET OIS. Depth in the spot market could stimulate demand for a term hedging market 3. Move IRS discounting to SONET: Transition the discounting of cleared sterling instruments to SONET 4. LIBOR alternative: Compound overnight SONET and/or Term SONET as the reference rate for contracts Timely initial liquidity build in SONET OIS is crucial to establish clearing eligibility within the prescribed timeframe 3: Transition IRS discounting to SONET 4: LIBOR alternative A detailed plausible pathway to transition has been developed and the SONET team are engaging market participants [1] Indicative SONET rate launched Q Page 19

20 Appendix Page 20

21 SONET: Market coverage Total Sterling Gilt Repo Market Size: c. 432 bn [1] Cleared ( bn outstanding 9th Dec 2015) Uncleared ( bn outstanding 9th Dec 2015) o/n Term Total o/n Term Total Relevant Market Size A. Cleared trades are counted gross : i.e. each bilaterally traded contract is counted as 2 separate repos Total value traded in cleared o/n is therefore: 35.5bn (71.5/2) Sp fic Platform Direct [2] [3] Platform 5 5 B. Uncleared Direct DBV figure includes approx. 50% stock lending transactions on average, resulting in c. 25.5bn (51/2) uncleared direct DBV Therefore, the total value traded in uncleared o/n repo is c. 49.5bn ( ) Conclusion DBV Direct [4] Triparty Direct [5] Total 71 [A] [B] As at 9 th December 2015, the total estimated daily execution volume for the Sterling Gilt Repo Market is c. 85bn ( ) Index volume on this date was 54.7bn representing c.64% of the total relevant market Key: Sources and notes: 1. Derived from ICMA Euro market survey. Represents total value of repo contracts outstanding as at Dec 9th Survey quotes total market size as EUR 5,608bn, with 10.6% of that against collateral issued by the UK govt. Converted into GBP at an exchange rate of this gives a total market size of c. 432bn 2. LSEG best estimate = Relevant flow; within our dataset = Potentially relevant flow; not within our dataset 3. Figure estimated by taking all other known data and assumptions in terms of cleared repo, uncleared DBV and triparty and subtracting from the total estimate. Assumption made that all uncleared business in specific gilt repo is for term based on anecdotal evidence from voice brokers and further assumption that majority of this business is bank-client rather than interbank. 4. Based on EUI data from 2015/16 all DBV repo contracts open as at 9 th December Estimated based on market-wide observation in the ICMA report which stated that 54% of all repo business was executed direct with counterparties of which 11% was triparty. Furthermore the report noted that some 17.2% of triparty repos were conducted on an O/N basis Page 21

22 SONET: Analysis of sub-component dynamics Summary of work undertaken in Q The purpose of the analysis was to determine which of the alternative methodologies consistently generated a SONET rate which most closely reflected where the market is trading In particular the aim was to manage the impact of Specials which may be manifest in the Specifics transactions, and might bias the rate We looked at the history of all the transactional data across the historical data set, breaking out the impact of GC, Specifics and DBV transactions We looked at a range of methodologies: Volume Weighted Median (VWM) Volume Weighted Average (VWA) VWA Specific-only Filtering a specified percentage of transaction volume is removed from the Specific sub-portfolio, then the VWA of the aggregate remaining data is taken. Filtering can either be on the downside of the specifics data only, or from both sides of the distribution according to an iterative algorithm. A range of percentages were considered. VWA All Filtering As above, but the filtering is applied to the entire aggregate dataset rather than just the specifics sub-portfolio. For almost all the days in the history there was very little basis between the methodologies, and little differentiation between the different sub-portfolios of transactions However, a small number of days in the five year history show some dispersion in the data. Those days were analysed in detail to determine which methodology provided the most appropriate rate On the following pages we show the detailed analysis for two of those days: 31/12/2015 and 30/12/2011 Page 22

23 SONET: Analysis of sub-component dynamics Sample 1: 31 Dec % Transaction rates by repo type on 31/12/2015 Rate calculations 25% 20% 15% 10% Source: LSEG Internal Analysis DBV, GC & Spec DBV & GC DBV % Filtered VWM VWA Specific Filtering VWA All filtering % % <-50 <-40 <-30 <-20 <-10 <0 <10 <20 <30 <40 <50 <60 <70 <80 <90 <100 DBV Only VWM: VWA: The histogram shows the full dispersion of transaction rates for 31/12/2015, a date on which the market clearly spiked downwards to the 0-10bp level from a previously stable level of c.50bps It is apparent from the histogram and rate calculations that: There is some Specialness emerging within the Specific portfolio (seen in the long tail on the LHS) There is also noise in the DBV market: some of the transactions continue to take place at the 30-70bp level, even when the bulk of the market has clearly dropped to 0-20bps. This is reflected in the difference between the DBV only VWM and VWA calculations in the table Other dates were analysed in the same way and we observed that, The Specialness seen above emerged mostly in the latter period ( ) and was not particularly evident in the early years Conversely the bifurcation effect in the DBV market was very pronounced in some of the older sample dates (see next page) Page 23

24 SONET: Analysis of sub-component dynamics Sample 2: 30 Dec % Transaction rates by repo type on 30/12/2011 Rate calculations 25% 20% Source: LSEG Internal Analysis % Filtered VWM VWA Specific Filtering VWA All filtering 15% DBV, GC & Spec 10% 5% DBV & GC DBV % <0 <10 <20 <30 <40 <50 <60 <70 <80 <90 <100 DBV Only VWM: VWA: The histogram shows the full dispersion of transaction rates for 30/12/2011, a date on which the rate spiked down at year end. It is apparent from the histogram and rate calculations that: DBV shows clear bifurcation, with approximately half of the market trading at 0-30bps and the rest remaining at 40-60bps No particular Specialness tail is evident in the Specifics The VWA All 25% Filtering calculation generates a SONET rate of 16.7bps, which is representative of the GC transactions; the Specific transactions, and the DBV transactions that have responded to the drop in the repo market rates The conclusion from the analysis of this and other dates is that VWA All 25% iterative filtering is the most appropriate choice of methodology since it prunes both sides of the distribution. This pruning removes transactions that are significantly away from the bulk of the market, which may be evident in either the cleared or uncleared data. Page 24

25 Features Calculation & publication Definition & Risk Profile SONET: SONIA comparison Underlying interest SONET Sterling SONET is a measure of the sterling risk free reference rate for secured overnight funds Reformed SONIA A measure of the rate at which interest is paid on sterling overnight wholesale funds in circumstances where credit, liquidity and other risks are minimal. Methodology The 25% iteratively filtered volume weighted average of the interest rates on: Sterling cleared and uncleared repo transactions secured against government securities With one business day maturity To settle on the day that is used in the SONET index calculation There is no minimum transaction size The trimmed volume weighted average (excluding trimming the highest and lowest 25% by volume of transactions) of the interest rates on unsecured wholesale one business day maturity deposit transactions in sterling, settled same-day, as reported to the Bank s Sterling Money Market data collection, executed between 00:00 hours and 18:00 hours UK time, with a minimum transaction size of 25 million. Risk Profile Minimal credit risk: secured exposures Reflects the actual collateralised funding costs in the market Shows variance against the Bank Rate Embeds the credit riskiness of the participants Trades at a very stable spread to the Bank Rate What transactions are included? Scope of dataset: eligible transactions covered by cleared specific & GC repos and Euroclear settled uncleared DBV trades Maturity & settlement: one business day maturity; O/N,T/N, S/N settlement Execution window: executed between 00:00 18:00 UK time Minimum size: no minimum Scope of dataset: eligible transactions reported to Bank s Sterling Money Market (SMM) data collection Maturity & settlement: one business day maturity, same day settled (spot) Execution window: executed between 00:00 18:00 UK time Minimum size: minimum transaction of 25 million Geographic scope: worldwide Geographic scope: transactions booked in EEA branches of SMM reporting institutions How is the rate calculated? 25% iteratively filtered Volume Weighted Average 50% Trimmed Volume Weighted Median (excluding trimming the highest and lowest 25% by volume of transactions) What are the publication arrangements? SONET will be published same day at 21:00 Republication if change causes a 2 basis point or greater move in rate. Republication to occur at 09:00 GMT (t+1) Alongside the headline rate (to 4dp), FTSE Russell will publish volume of transactions and additional rate statistics Published 09:00 on following day Headline rate (to 4dp) alongside aggregate volumes & 10 th, 25 th, 75 th & 90 th percentiles published [tbc by Bank] Republished if change in the underlying distribution causes a 2 basis point or greater move in rate. Deadline for republication is 16:00. Contingency calculation methodology will use the Bank Rate with the addition of an average of the spread to SONIA over some recent period. Governance Benchmark will be administered in compliance with the IOSCO Principles and the forthcoming EU Benchmark Regulation SONET Advisory Committee to form integral part of Governance structure TBC Surveillance Ex-ante checks by contributors & ex-ante and ex-post analysis by FTSE Russell Surveillance of underlying SMM reported transactions Transition strategy See slides 18 & 19 Point in time switchover from SONIA to Reformed SONIA, in March/April Source: BoE October 2016 paper The reform of SONIA ; BoE February 2017 paper The reform of SONIA ; LSEG internal analysis Page 25

26 August Update: SONET Market coverage: January 2016 August bn Volume Daily Transaction Volumes No. of trades 2, bn 90 bn 1, bn 70 bn 1, bn 50 bn 40 bn bn 20 bn bn 0 bn (100) Feb 2016 Apr 2016 Jun 2016 Aug 2016 Oct 2016 Dec 2016 Feb 2017 Apr 2017 Jun 2017 Aug 2017 SONET (LHS) Reformed SONIA (LHS) Reformed SONIA (RHS) SONET (RHS) SONET data includes all overnight cleared repo (overnight, tomorrow next, spot next); and uncleared overnight direct DBV repo. Currently uncleared specific and triparty transactions are not included Source: BoE and LSEG internal analysis. Reformed SONIA volumes contain full population contributing to SMM from 1 st July Reformed SONIA data only available to 03/03/2017 Note: SONET data provider systems update as at 1 st January 2017 and 22 nd March 2017 Page 26

27 August Update: SONET Market coverage: January 2016 August bn Volumes Volumes: SONIA vs Reformed SONIA vs SONET 100bn 80bn 60bn 40bn 20bn 0bn Jan-16 Feb-16 Mar-16 Apr-16 May-16 Jun-16 Jul-16 Aug-16 Sep-16 Oct-16 Nov-16 Dec-16 Jan-17 Feb-17 Mar-17 Apr-17 May-17 Jun-17 Jul-17 Aug-17 SONET Current SONIA Reformed SONIA SONET data includes all overnight cleared repo (overnight, tomorrow next, spot next); and uncleared overnight direct DBV repo. Currently uncleared specific and triparty transactions are not included Source: BoE and LSEG internal analysis. Reformed SONIA volumes contain full population contributing to SMM from 1 st July Reformed SONIA data only available from 01/02/2016 to 03/03/2017 Note: SONET data provider systems update as at 1 st January 2017 and 22 nd March 2017 Page 27

28 August Update: SONET Market dynamics: January 2016 August Rate % A direct comparison of SONET, Reformed SONIA, Current SONIA and the Bank Rate, between January 2016 August 2017 Rates: SONIA vs Reformed SONIA vs SONET vs Base Rate Jan-16 Feb-16 Mar-16 Apr-16 May-16 Jun-16 Jul-16 Aug-16 Sep-16 Oct-16 Nov-16 Dec-16 Jan-17 Feb-17 Mar-17 Apr-17 May-17 Jun-17 Jul-17 Aug-17 SONET Current SONIA Reformed SONIA Bank Rate Source: BoE, Bloomberg, and LSEG internal analysis. Reformed SONIA volumes contain full population contributing to SMM from 1st July Reformed SONIA data only available from 01/02/2016 to 03/03/2017 Note: SONET data provider systems update as at 1 st January 2017 and 22 nd March 2017 Page 28