Evolution of the BBSW Methodology
In October 2015, the Council of Financial Regulators (CFR) sought views on the evolution of the methodology for the bank bill swap rate (BBSW) benchmark. The consultation closed on 3 December 2015. On 9 February 2016, the CFR released a discussion paper on the evolution of the BBSW methodology.
BBSW is a key financial benchmark in Australia and is administered by the Australian Financial Markets Association (AFMA). BBSW interest rates measure where the interbank market trades bank accepted bills (BABs) and negotiable certificates of deposit (NCDs) with terms from 1 to 6 months issued by the ‘Prime Banks’. BBSW rates serve as reference rates for pricing many debt securities and lending transactions. They are also used to determine payment obligations on a range of derivatives such as interest rate swaps, cross-currency swaps and forward rate agreements, and are influential in the pricing of bank bill futures. Given its wide usage, BBSW is considered to be a financial benchmark of systemic importance.
International regulatory bodies have developed standards to be used in financial benchmark construction. In particular, the International Organization of Securities Commission’s (IOSCO) Principles for Financial Benchmarks, which was issued in July 2013, recognises the use of executable bids or offers as a means to construct financial benchmarks where anchored in an observable market. Consistent with this, the administration of BBSW was reformed in 2013 with the intention of improving its reliability by moving from a submissions-based to a market data-based benchmark.
To ensure that BBSW remains a trusted, reliable and robust financial benchmark, the CFR recommended that consultation be undertaken with market participants on the methodology for BBSW. This consultation paper presented options and invited views on how the BBSW methodology could evolve going forward.
2. The BBSW calculation methodology
AFMA, as the administrator of BBSW, is responsible for the methodology used in calculating BBSW. In response to international developments and following the withdrawal of some financial institutions from the BBSW submission process, AFMA, in consultation with ASIC and the RBA, adopted the current method for calculating BBSW in September 2013. BBSW is calculated using live and executable prices from trading venues for BABs and NCDs. AFMA’s adoption of an automated process that extracts these rates directly from observable prices removed the need for submissions and was designed to ensure that BBSW remains underpinned by an actively traded market. The key characteristics of the BBSW methodology are:
- Eligible securities: the securities currently eligible for BBSW are BABs and NCDs issued by ‘Prime Banks’. BBSW rates span terms from 1 to 6 months, with most trading occurring in the 1, 3 and 6 month tenors
- Prime Banks: AFMA Prime Banks are a designated sub-set of the banks operating in Australia whose short-term securities trade as a homogeneous asset class and are recognised as being of the highest quality with regard to liquidity, credit and consistency of relative yield. These institutions are designated by AFMA as having met the relevant eligibility criteria and conditions. All Prime Banks are subject to an annual test of their ongoing eligibility as a Prime Bank. Banks that meet the following eligibility criteria may nominate to be considered a Prime Bank:
- Be an Authorised Deposit-taking Institution (ADI) as defined by APRA
- Satisfy a credit rating benchmark set by AFMA, which is currently a Standard & Poor’s short term rating of A1+ and a long term rating for senior unsecured debt obligations of at least AA-
- Have securities eligible for use in the RBA’s open market operations and standing facilities.
Banks must also achieve a minimum level of support from significant investors in bank short-term paper for their inclusion on the Prime Bank list, as measured in an annual survey conducted by AFMA. For several years, the four major Australian banks have been the only Prime Banks.
- Price collection: AFMA calculates BBSW benchmark rates as the midpoint of the nationally observed best bid and best offer (NBBO) for Prime Bank Eligible Securities. The rate set process uses live and executable bid and offer prices sourced from interbank trading venues approved by AFMA, which are currently ICAP, Tullett Prebon and Yieldbroker. The bids and offers are sourced from three samples around 10am each day (i.e. 9.59am, 10.00am and 10.01am), and from each approved venue. As an ongoing condition of accepting Prime Bank status, Prime Banks agree to support approved venues underlying BBSW by quoting bids and offers.
3. Turnover in the market for BABs and NCDs
The outstanding stock of BABs and NCDs issued by the Prime Banks has increased since 2013 to around $140 billion (Graph 1). However, trading activity during the daily BBSW rate set has declined over recent years. To provide additional transparency of market activity during the time of the BBSW rate set process, AFMA publishes (with a one-month lag) the daily aggregate volume of transactions at the time of the rate set. Historically, since BBSW has been set around 10am, trading activity in BABs and NCDs has been most active at this time, with liquidity concentrated in the three-month maturity segment.
Turnover at the rate set has declined to currently be very low. In the first twelve months since AFMA started publishing data in September 2013, the average daily turnover in BABs and NCDs during the rate set was around $500 million, and on around 10 per cent of days there was no turnover at any tenor during the rate set. Over 2015 to date, average turnover has fallen to around $225 million, and on a third of days there has been no turnover during the rate set (Graph 2). The decline in 1-month turnover largely reflected banks lengthening their liability profile ahead of the liquidity coverage ratio (LCR) requirements being implemented on 1 January 2015.
Activity on the ‘approved venues’ at the time of the rate set is interbank trading only. Bank holdings of NCDs have declined significantly over recent years. With fewer holdings, it is to be expected that interbank trading activity would decline. However, this trend appears to have been exacerbated by much of the remaining interbank activity shifting away from the period of the rate set, with the average volume of eligible securities traded during the rate set as a proportion of total outstanding paper having fallen to very low levels (Graph 3).
IOSCO’s Principles for Financial Benchmarks highlight that a benchmark’s design should take into account the size and liquidity of the relevant market, the degree of market concentration, the adequacy of the sample used to represent the market, and the broader market dynamics. While BBSW remains one of the main benchmarks in Australia, the low turnover in the interbank market raises the risk that market participants may at some point be less willing to use BBSW for pricing other short-term debt securities or for determining cash flow obligations on derivatives.
Some preliminary data collected from the Prime Banks indicate that there is substantially more activity in the NCD market than is being measured at the rate set, with the activity mainly outside the interbank market. Over 2015 to date, the total turnover of NCDs issued by the Prime Banks at the BBSW tenors has been significantly larger than the turnover on the approved venues at the rate set. Furthermore, across the Prime Banks, at least $100 million in NCDs were bought or sold on almost all days that BBSW was calculated, with activity almost entirely at the 1-, 3- and 6- month tenors. However, the non-bank participants that buy and sell NCDs tend to transact bilaterally with the issuing bank, with the price struck at BBSW (or at a spread to BBSW for non-Prime Bank issuers), rather than at a directly negotiated rate (outright yield). If these participants could be encouraged to buy and sell NCDs at outright yields, then these transactions may have the potential to underpin the BBSW benchmark.
Given the decline in trading activity during the BBSW rate set, this is an opportune time to consider whether any adjustments need to be made to the BBSW methodology. The CFR invited views on the following options for the evolution of the BBSW methodology:
Option 1: Continue with the current NBBO calculation methodology
The first option is to continue with the current NBBO calculation methodology for BBSW, where the administrator collects live and executable prices from approved venues in the market for Prime Bank Eligible Securities. This would involve accepting that on a high proportion of days there will be little trading activity in the market at directly negotiated rates upon which to base BBSW.
Option 2: Anchor NBBO calculation methodology to a broader set of NCD market transactions
The second option would be to continue with the current NBBO calculation methodology, but to underpin the live and executable prices collected from the approved venues with a broader set of NCD market transactions contracted up to the time of the rate set. By more clearly anchoring the BBSW benchmark to observable transactions entered into at arm’s length between buyers and sellers in the market, this may ensure that the benchmark remains a credible indicator of rates in the market.
For this option to be feasible, it would be necessary for Prime Banks to directly negotiate the interest rates on Prime Bank NCDs with third parties, rather than linking the rate to BBSW. This would require a change to the existing market practice, which is for the interest rate on NCDs to be directly linked to BBSW. Were this to occur, the live and executable bid and offer prices that the Prime Banks provide to the interbank trading venues during the rate set could then be informed by observable transactions in the underlying market.
To ensure transparency of market activity, each Prime Bank would be required to submit daily data to the administrator on their NCD transactions. This would be consistent with existing reporting arrangements, as Prime Banks are already required under the AFMA Prime Bank Conventions to disclose their trading volumes in Prime Bank NCDs and BABs on all trading days. The key change to these arrangements would be for the interest rates reported by Prime Banks to have been directly negotiated with counterparties rather than linked to BBSW. This information would provide confidence that the executable bids and offers used to determine the benchmark are representative of the underlying market.
The Prime Banks would still need to exercise some expert judgment in forming their bids and offers under circumstances where there is low liquidity in the underlying market. This would particularly be the case for the BBSW tenors that do not coincide with the bulk of activity in the underlying market, which is at the 1-, 3- and 6- month tenors.
Option 3: Amend the calculation methodology to be based on submissions of the aggregate cost of wholesale fund raising
The third option would be to amend the calculation methodology, so that Prime Banks would submit data on the aggregate cost of their wholesale fund raising at each BBSW tenor to the administrator. The administrator would then use these submissions to calculate BBSW, instead of using the NBBO method. This option is similar to the proposed calculation methodology for LIBOR, the main interest rate benchmark for the US dollar, pound sterling, and Japanese yen.
LIBOR is currently based on submissions from contributing banks on the lowest perceived rate at which they could go into the London interbank money market and obtain funding in reasonable market size for a given maturity and currency. In July 2014 the Financial Stability Board (FSB) published a report recommending that IBOR rates should be underpinned to the greatest extent possible with transactions data, consistent with the IOSCO Principles for Financial Benchmarks. In response, the administrator of LIBOR, ICE Benchmark Administration (IBA), is proposing a number of enhancements to anchor LIBOR to the greatest extent possible in transactions. The proposed changes most relevant for the Australian market are:
- Expand the scope of eligible transactions to include submitters’ unsecured wholesale funding deposits, commercial paper and primary issuance of certificates of deposit to wholesale counterparties
- Expand the panel of submitting banks
- Timing of transactions: benchmark submitters would include all of their eligible transactions within a pre-determined window, with the window potentially as wide as 24 hours
- Submission of the aggregate cost of wholesale fund raising: submitters would provide an aggregate measure of the cost of their wholesale fund raising to the benchmark administrator based on observable market transactions where there is adequate market activity. A waterfall of methodologies would be used where there are insufficient transactions. Submissions would be based on techniques for interpolation and extrapolation, and expert judgement when it remains necessary, so that LIBOR rates can be published in all market circumstances.
Option 4: Amend the calculation methodology to be based on submissions of individual wholesale funding transactions
The fourth option is to amend the calculation methodology, so that Prime Banks would submit data on their individual wholesale funding transactions at each BBSW tenor to the administrator. The administrator would then use these transactions to calculate BBSW. This option is similar to the proposed calculation methodology for EURIBOR, the main interest rate benchmark for the euro.
EURIBOR is currently based on submissions from a representative panel of banks on the perceived rate at which euro interbank term deposits are offered by one Prime Bank to another Prime Bank. The administrator of EURIBOR, the European Money Markets Institute (EMMI), has launched the EURIBOR+ Project, with the objective of developing a transaction-based unsecured market benchmark as a possible evolution of EURIBOR. The proposed changes most relevant for the Australian market are:
- Expand the scope of eligible transactions by moving from the interbank market to broader sources of wholesale borrowing by banks
- Expand the panel of submitting banks
- Submission of wholesale funding transactions: the key difference with the proposed methodology for LIBOR is that participating banks in EURIBOR would submit all the in-scope transactions to the benchmark administrator, which would then be responsible for the aggregation of these transactions to calculate the benchmark. A fall-back mechanism will be put in place for circumstances where there are insufficient transactions (this could involve the use of smoothing techniques or expert judgement).
5. Consultation questions
Given the volume of contracts that reference BBSW, it is important when considering potential amendments to the methodology that the fundamental properties of BBSW as a homogeneous measure of bank funding costs be maintained to ensure a seamless transition. With this in mind, views were invited on the following consultation questions relating to the key characteristics of BBSW:
- Eligible transactions: BBSW’s status as a benchmark reflects the fact that it is regarded as a reasonable proxy for banks’ short-term borrowing costs. That said, it covers only one aspect of bank funding, namely BABs and NCDs traded in the interbank market. A broader measure of the cost of Prime Banks’ fund raising that covered all wholesale funding with a broader range of counterparties would be a possible alternative. The measure could include non-tradable sources of funding, such as unsecured wholesale term deposits, as well as tradable sources of funding such as BABs and NCDs. Wholesale funding transactions that are currently being priced at a spread to BBSW would need to shift to being priced independently from BBSW for these transactions to be eligible for use in the calculation of BBSW.
- Q1. The market underlying the BBSW benchmark is currently the interbank market at the time of the rate set. Should the definition of the underlying market be broadened to include all funding transactions with wholesale counterparties, such as pension funds, and non-financial corporations?
- Q2. Should the eligible securities for calculating BBSW be broadened beyond BABs and NCDs to include other financial products such as term deposits that reflect the cost of the Prime Banks’ total wholesale fund raising at the relevant maturities?
- Q3. What should be the minimum size for wholesale funding transactions to be in scope for BBSW?
- Q4. Should offshore Australian dollar denominated wholesale funding transactions be included, or only transactions undertaken in Australia?
- Q5. If the NBBO method were not to be used, what should be the fall-back mechanism for Prime Banks and the administrator in the event that there are insufficient transactions for calculating BBSW?
- Prime Banks: if the range of eligible transactions were to be expanded, the set of Prime Banks could potentially be larger than the four major banks, as a number of other institutions of similar rating have wholesale issuance programs of reasonable size.
- Q6. Should the set of Prime Banks be larger than the four major banks, and how could the existing criteria for Prime Banks be amended to achieve this?
- Timing of transactions: options to increase the number of transactions underlying BBSW include: widening the period of the rate set; or delaying the time of the rate set. However, if the period was widened, users of BBSW would face the risk that prices could move significantly during the period (i.e. event risk), with the spreads between the BBSW tenors being distorted by transactions having taken place at different times rather than simultaneously at the rate set (i.e. basis risk). Another factor to consider is the potential for transactions to migrate away even from a widened or delayed rate set period. If there were to be an insufficient number of transactions, it would be more challenging for Prime Banks to use their expert judgment to estimate their funding rate over a period rather than at a particular point in time.
- Q7. Should the timing of the transactions in scope for calculating BBSW be changed? Three options to consider are:
- including all transactions taking place up to the time of the rate set (i.e. morning transactions prior to 10am)
- including all transactions taking place over the 24 hours prior to the rate set
- moving the rate set to a later time (e.g. 11am) to provide a wider window for the transactions underlying BBSW to be contracted up to the time of the rate set.
6. Consultation Process
In October 2015, the CFR sought views on the options presented in Section 4 for the evolution of the BBSW methodology. A number of questions were also posed in Section 5. The consultation closed on 3 December 2015.
All submissions and correspondence received (including any contact names or other personal information) have been made public on the CFR website, unless it was specifically requested that the whole or any part of a submission be treated as confidential.
7. Submissions Received
27 November 2015
3 December 2015
Finance and Treasury Association
9 December 2015
The CFR received 12 confidential submissions.
8. Discussion Paper
On 9 February 2016, the CFR released a discussion paper that summarises the feedback received from the submissions, and sets out a proposal for the evolution of the BBSW methodology for discussion with AFMA and market participants.
- Prior to September 2013, the methodology involved a panel submitting their view of the mid-rate for BABs and NCDs of varying maturities to AFMA by 10.05am each business day.
- AFMA (Australian Financial Markets Association) (2015), A Guide to the Bank Bill Swap (BBSW) Benchmark Rate, October 2015. Available at: <http://www.afma.com.au/data/bbsw/BBSW%20Guide.pdf>
- AFMA (2015), AFMA Prime Bank Conventions, October 2015. Available at: <http://www.afma.com.au/standards/market-conventions/Prime%20Bank%20Conventions.pdf>.
- AFMA (2015), Bank Bill Swap (BBSW) Benchmark Rate Conventions, 13 April 2015. Available at: <http://www.afma.com.au/standards/market-conventions/Bank%20Bill%20Swap%20(BBSW)%20Benchmark%20Rate%20Conventions.pdf>
- ASIC (Australian Securities and Investments Commission) (2015), Financial Benchmarks, Report 440, July 2015. Available at: <http://download.asic.gov.au/media/3285136/rep440-published-8-july-2015.pdf>
- FSB (Financial Stability Board) (2014), Reforming Major Interest Rate Benchmarks, 22 July 2014. Available at: <http://www.financialstabilityboard.org/wp-content/uploads/r_140722.pdf>
- FSB (2015), Progress in Reforming Major Interest Rate Benchmarks, 9 July 2015. Available at: <http://www.financialstabilityboard.org/wp-content/uploads/OSSG-interest-rate-benchmarks-progress-report-July-2015.pdf>
- ICE Benchmark Administration Limited (IBA) (2015), Second Position Paper on the Evolution of ICE LIBOR, 31 July 2015. Available at: <https://www.theice.com/publicdocs/ICE_LIBOR_Second_Position_Paper.pdf>
- IOSCO (International Organization of Securities Commissions) (2013), Principles for Financial Benchmarks, July 2013. Available at: <https://www.iosco.org/library/pubdocs/pdf/IOSCOPD415.pdf>