ICE Term Risk Free Rates

I will start with my usual rant about "risk free". SONIA is based on unsecured transaction and is credit risky. Not risk free on the credit side. A term rate is, as its name says, fixed for a term and the value of a term instrument changes when the market changes during that term. Not risk free on the market side. I cannot understand why this incorrect terminology of "risk free" is so widely used while plenty of correct and more meaningful terminology are available: collateralised rate, overnight-linked rate, overnight-indexed rate, derivative term rate, alternative rate (I don't like alternative rate too much either, it opposes the current name to a previous approach instead of positively describing the present, but it is better than risk free),

IBA (ICE (InterContinental Exchange) Benchmark Administration; I like those acronyms based on acronyms) is launching a "ICE term RFR". The term rate is based on futures prices. The first version is SONIA-linked.

As a reference term rate, I would prefer, as I mentioned in my "Quant Perspective on IBOR Fallback Proposal", a benchmark based directly on OIS than one constructed, through a curve model, with futures. Contrarily to what is indicated in the ICE paper on term risk free rates, the transaction data do not need a "financial model to generate yield curves" if OIS rates of the correct tenor are directly used. The model is required mainly in the futures case.

Regarding OIS, the ICE paper indicates the prevalence of forward starting and MPC meeting OIS. It would be good to have clear statistics on how much is traded on 1, 3, 6-month tenors and other dates.

The futures daily volume is around 4bn. This is below the 40 to 50bn of ON depo that are used for SONIA itself. This is a decent number. The paper claim that the "data is available", but is it "freely available to all market participants and researchers"?  I don't thing so (at least I have not found the data yet). My request for benchmarks has been for a long time total transparency. To be totally useful and informative, the "ICE term portal" should provide the detailed futures data on which the term rate is based (and if possible without several layers of legal agreements).

Even if I started with not extremely positive remarks, to my opinion this is a good proposal. This is a benchmark that would be available in London morning (at least its intraday version). GBP LIBOR fallback would be improved if fallback rates are available shortly after 11:00 am London time, like LIBOR itself. In the waiting for a more robust OIS benchmark, this would be a good starting point. It could be used as the first waterfall step of a multi-stage fallback as proposed in the Section 6.11 of my perspective.

IBA analysis of the usefulness of a "term RFR" is similar to some extend to the one I have developed in my "quant perspective" (Section 3.5), except that they do not mention the LIBOR fallback issue in their note. To my opinion the fallback is where their proposed futures-based term rate could be useful as it would be immediately available. I think it cannot replace an actual OIS based benchmark for term rates on which the most liquid derivatives would be based as the subjective model part could have a large impact.

For the methodology itself, it is based on a quite standard curve calibration with overnight forward rate following a step function between MPC meetings. Such a mechanism is for example described in my multi-curve book in Section 5.11. On top of this there is an arbitrary 1st of month jump in months without MPC meetings. Any other day of the month would be as good in theory. In practice, a mid month would probably be better to avoid too short constant rate periods. When a MPC meeting is on 22th March, there are only 9 days to month-end. Even better than mid-month, it could be slightly earlier to be on average in between the two MPC meetings or month specific to be equal distance of the two surrounding meetings once the actual dates are known. The curve calibration process is a simplified version of what one would like to use for market making on short term OIS as it does not take into account the intra-month seasonality. For example in GBP SONIA, the year end and quarter end have usually rate fixing lower that surrounding dates. The year end has seen jumps up to 12 bps in the last years. With year-end being 4 days on 4 years out of 7, this is nor a negligible figure.

Wait and see if there is any interest for it from regulators or from ISDA for the IBOR definition fallback.


A Quant Perspective on IBOR Fallback Proposals - Version 1.1

I have received numerous feedbacks on my "Quant Perspective on IBOR Fallback Proposals" document. I have updated the document and a version 1.1 is now available on SSRN: https://ssrn.com/abstract=3226183

Don't hesitate to contact me for discussion about the subject.

Convexity adjustment for Option 1: Spot Overnight Rate. See the note for the details.


Six months of Fed Funds market expectation in ten seconds.

The US Fed has hiked rates several times in the last year. What was the market's expectation for those hikes six or twelve month before they happened? To estimate that, you can ask many market participants about their expectations every day and then average the answers. This is the polling approach.

Instead of asking where their mouth is, you can also look where their money is. Interest rate derivatives are the most liquid financial instruments, worth a lot more than many mouths. How do you read the market expectation in derivatives? The overnight-indexed swaps (OIS) are linked to the Effective Federal Fund Rates (EFFR). Those swaps are liquid and usually trade with standard tenors like 1, 2, 3-months. It is possible to extract from those instruments the market expectation about futures rates. This can be done in a systematic way using curve calibration on those OIS with interpolation mechanism that imply constant rates between FOMC meetings and allows for jumps on those dates.

I have done the curve calibration on a daily basis on a 6-month period. Each day the curve is calibrated to market OISs, taking into account the FOMC dates in the next year. The forward rates for each day over the year after the calibration is computed and displayed in a graph.

Putting together those graphs produces a 10 seconds movie representing the market expectation over a 6-month period (from June 2017 to January 2018) of the next 12-month on calibration date.

In the graphs, the actual EFFR fixing as published by the Fed are represented in red. They represent the realized rates with full insight. For each date, the expectation for each day in the next 12-month is represented in blue. We insist that the two colors represent very different things, the red is backward looking after the date (realized path), the blue is forward looking for the next 12 months and updated each day.

With this representation you can see the market expectation changing (slightly) each day and in most cases converging to the FOMC decisions.

The movie can be found on YouTube at https://youtu.be/pqv3sIVjz7c

Don't hesitate to contact me for production implementation of those curve calibration methods. The technical description of the method can be found in Section 5.11 of my Interest Rate Modelling in the Multi-curve Framework book.

Note: The movie was produced with is a simplified curve description template. It does not include month-end expected behavior. Exogenous intra-month seasonality can be added to the curve to obtain an even more realistic forward behavior (Section 5.9.4 in the book). 


Risk-based overnight-linked futures

Some years ago I proposed an innovative design for risk-based swap futures.

I worked with ASX to adapt it to the AUD BBSW swap market and a product based on that design has been traded on the exchange since 2016 (even if it has not attracted a lot of trading activity).

With the new importance of overnight benchmarks, the ETD market has to find a product that could be used for price discovery and risk management of the full term structure of interest rate.

I have detailed a version of the generic futures design to match the overnight conventions and OTC markets. The product has been presented to the main interest rate futures exchanges in Europe and the USA. We will see if it takes a new life.

The technical details are now available in the form of a note. The document can be downloaded from SSRN at https://ssrn.com/abstract=3238640.

In the graph below I have displayed the convexity adjustments between the ETD futures and the OTC swaps that can be obtained with this design (in red) and the ones with the current design of overnight-linked futures. See the paper for the exact description of the graph.

Consultation on IBOR fallbacks: Question 3

Game of Benchmarks: Season 2

Game of Benchmarks: Season 2 - Episode 3: Timing


Question related to: Description of Adjusted RFR - Fixing timing

The simplest option for adjusted RFR in the ISDA IBOR fallback consultation is Option 1: Spot Overnight Rate. The option simply replaces an IBOR fixing on a given date by an Overnight fixing on the same date.

Even this simplest option has to be assessed carefully in term of practical achievability. The overnight rate are known only at the end of the fixing date or the next day in the morning while the IBOR fixing are known shortly after 11:00 am. Is there some provisions in some derivative contracts that would require to know the fixing by 11:00 am? For example the cap/floor are exercised — even if it is automatic exercise — at 11:00 am. Can the exercise mechanism be delayed to the end of the day or even the next day? The fallback provision should describe what would happen to the contracts that have a tied schedule linked to the fixing time. The GBP FRA may be one of those types of contracts where knowing the fixing result at 11:00 am is important as they have to be paid on the same day. What would happen to the fallback provision if the derivative contract is written in such a way that obtaining the fixing by 11:00 am or shortly thereafter is important?

A list of FAQ is maintained by ISDA.

Season 2: The questions


Question 1: Question related to Option 3: Compounded Setting in Arrears Rate

Question 2: Question related to: Description of Fallbacks - Triggers

Question 3Description of Adjusted RFR - Timing


Short note on long-term repos

The market infrastructure for interbank lending and derivatives has dramatically changed over the last ten years. The interbank lending is done now mainly on a secured basis. Derivatives are margined daily with variation margin guaranteeing the full present value of the trades. The world of interest rate benchmarks is also rapidly changing with the possible discontinuation of the IBOR benchmarks that have reigned on the benchmark kingdom over the last 30 years and the emergence of secured rate overnight benchmarks.

All derivative users are now well aware of the difference between overnight-indexed swap (OIS) rates and term LIBOR deposit rates even if its discovery may have been sudden to some market participants in 2007.

When we combined the secured term lending, the collateralised derivatives and the secured benchmark, what is left of the OIS-term deposit difference? This is the question I try to answer in a brief note now available on SSRN at https://ssrn.com/abstract=3258690

Using simple no-arbitrage strategy with daily hedging I prove that the collateralised OIS fix rate when the underlying benchmark is an overnight repo is equal to the term repo rate on the same period.


Term RFRs

At a recent ISDA meeting, a FCA representative, Edwin Schooling-Latter, positively commented on the achievability of term versions of RFRs. His comments are presented in a recent Risk article titled ''Term versions of RFRs will work – FCA official''.

This comments goes in the direction of the opinion of many (but not all) market participants but is in opposition to many regulators public comments (in particular the FSB OSSG) and against the indication in the FAQ associated to the ISDA consultation on IBOR fallback. The term rates are not even an option in the ISDA consultation.

There are two different issues associated to the "term rates" question: the IBORs fallback and the standard for new trades.

For new trades, you can change the rules and do what you want/is the most convenient in the circumstances.

For legacy trades, the fallback changes the reference rate when the contractual one is not available, but does not change the other aspects of the contracts. For those, a term rate is a requirement, not merely one of the options. The choice is between a term rate and chaos (with different levels of chaos possible). It seems there is a willingness to move the main benchmarks from Interbank (IBOR) to pseudo-risk-free rate (RFRs). If this is the case also for IBOR fallback, the choice is between term versions of RFRs and chaos.

My interpretation of the backing by FCA of term RFR is the following: "better to have a meaningful number which make sense than to have market chaos".

My equivalent interpretation of refusal of term RFR by other regulators is: "better to have chaos than to change the wording of a previous opinions based on a political agenda for the market".

There was manipulation of benchmarks in the past, and it is better to avoid it as far as possible in the future. Should it be at the cost of destroying the present or should the future be constructed in a orderly and efficient way taking into account the present?

All of the above are personal interpretations, opinions and questions!


EUR Overnight Benchmark - ESTER

The ECB has announced the result of the public consultation about the alternative EUR overnight benchmark. As expected, the ESTER rate, to be published by the same ECB, has been recommended by the Working Group on 13 September 2018 be used as the risk-free rate for the euro area.

I still don't like the name risk-free rate (RFR). As described on the ECB ESTER page, "ESTER will reflect the wholesale euro unsecured overnight borrowing costs of euro area banks". The rate is for unsecured and credit-risky transactions.

The world is now divided in two groups, one where the reference overnight benchmarks are secured overnight rates - e.g. USD and CHF - and one where the reference overnight benchmarks are unsecured overnight rates - e.g. EUR and GBP. That can only increase the cross-currency basis and give more work to quants.

On the secured rate front, we will soon have two competing term rates: OIS-like derivatives based on repo overnight rates and term repos. What is the link between them? In the unsecured world, there was a clear difference between the OIS rate based on unsecured overnight rates and term unsecured deposits (IBORs). What happen to this clear difference in the secured world, where repos are collateralised (usually by bonds) and derivatives are collateralised (usually by cash paying repo-based benchmark rates)? Note on that to follow shortly.


More overnight products

Another overnight-linked futures. CME is planning to launch SONIA futures. The futures would come in two flavours: Quarterly IMM dates and MPC meeting dates. More details can be found on CME website at https://www.cmegroup.com/trading/interest-rates/sonia-futures.html. To my knowledge, this is the first time that a central bank meeting date futures is launched. The central bank meeting date futures is one of the options I had proposed in my design of overnight-linked futures.

Other financial institutions issuing SOFR linked paper.

Credit Suisse has issued a six-month certificate of deposit linked to SOFR. The paper pays SOFR+35bps. More details in the FT article "Credit Suisse becomes first bank to issue debt tied to Sofr"(subscription required).

Barclays has issued commercial paper linked to SOFR. See the Bloomberg article "Libor Challenger Embraced in Debut Commercial Paper Transaction".


A Quant Perspective on IBOR Fallback Proposals

A month ago, ISDA launched a consultation on IBOR fallbacks. The question of the fallback in case of a benchmark discontinuation has obviously legal background. Parties to the derivatives contracts have to ask themselves: What is the meaning of what I signed? Is it really what I want? The answer to the last question is probably: "no, it is not what I want!" This is why most of the derivatives users agree that a change in the fallback language is required.

Once you are convinced that what you have is not what you want, you have to review the alternatives. I have published a note with a personal review of the alternatives from a "quant" perspective. Even if personally I would prefer to the called it a "qualitative analysis" as before assessing the quantities associated to the alternatives, you have to check their qualities against a set of qualitative criteria.

The note gives some background for the Season 2 of Game of Benchmarks: The Questions!

The note is available on SSRN:

A Quant Perspective on IBOR Fallback Proposals


With the increased expectation of some IBORs discontinuation and the increasing regulatory requirements related to benchmarks, a more robust fallback provision for benchmark-linked derivatives is becoming paramount for the interest rate market. Several options for such a fallback have been proposed. This note describes and analyses some of those options. The focus is on the quantitative finance impacts. None of the options that have been proposed fits all of the criteria for a good fallback provision. It appears that some of the options that have gained traction failed even the achievability criterion. The note concludes with the author's personal preference.

Season 2: The questions


Question 1: Question related to Option 3: Compounded Setting in Arrears Rate

Question 2: Question related to: Description of Fallbacks - Triggers

More SOFR products (II)

More and more SOFR-linked products appear. This time it is the World Bank which issued a SOFR-linked floating rate note (ISIN: US459058GK33), with a two-year maturity and a coupon of SOFR+22bps paid quarterly with a 4 days lockout. The notional issued was 1 billion notional. This is the longest maturity I have seen so far for a SOFR-linked instrument. Once more a lockout period which transforms a compounded setting in arrears into a partially setting in advance rate. The press release can be found here.

It appears that on the same date very large SOFR v LIBOR swaps were traded. The total notional of the swaps was USD 920 millions. More details are provided in the Risk article World Bank completes first SOFR bond hedge.

muRisQ Advisory

Over the past couple of years, in parallel to my work as Head of Quantitative Research at OpenGamma,  I have been working as a freelance advisor on a couple of projects. Those projects include designing of a new interest rate futures, presenting multiple executive training, advising hedge funds on the multi-curve and collateral framework, advising on CSA, Variation and Initial Margin frameworks and commenting on regulations.

For those projects, I'm working under the structure of an independent advisory firm called muRisQ Advisory. Its (concise) website can be found at http://murisq.com/

Don't hesitate to contact me regarding its services or for training, model validation, product design and risk management strategies.


Benchmark and CSA

The following quotes are from a recent article in Risk titled "Esma: Eonia can be used in CSAs after 2020".

Jakobus Feldkamp, senior policy officer for market integrity at the Paris-based European Securities and Markets Authority, tells Risk.net that CSAs will not be dragged into the BMR.

“Esma agrees that it can be argued that a reference to Eonia in a bilateral agreement on an individual exchange of collateral under an OTC derivative is not strictly ‘use of a benchmark’ in the sense of the BMR,” says Feldkamp.

The Article 3 (1) (7) of the European Benchmarks Regulation (BMR), refers to
"determination of the amount payable under a financial instrument or a financial contract by referencing an index or a combination of indices".  The regulation enters in full force on 1 January 2020. The question behind the interpretation of this sentence is to know if CSA referring to EONIA can still be legally used in Europe after that date. Not a minor issue certainly.

The quote from Feldkamp says "Esma agrees that it can be argued [...]". It does not say that it is ESMA's position that CSAs are not financial instruments, only that ESMA agrees that someone else can make that argument, a very different meaning.

But I'm not a lawyer, so what do I know about the meaning of a sentence?

I'm not a lawyer, but I'm a financial engineer (or at least I can claim that I'm one as the title is not protected, see the list of regulated professions in Europe). What about the following situation. I draw a derivative contract with a counterparty. The contract is a fixed v fixed swap that pays net 1 (thousand, million, billion, chose the amount according to your wealth) EUR in one-year time. The contract is under CSA with EONIA collateral. What is paid under such a contract? EONIA collateral rate, compounded over one year. Miracle, this is exactly the payoff, up to the notional payment, of the floating leg of an OIS. This is pure coincidence, I swear it! To make the things clean, I have to remove the collateral initial and final payments, but that can be easily done with a fixed amount payments. I write the fixed amounts contract as a loan and not a derivative, so it does not need collateral. (I can provide an exact term sheet if you hire me as a consultant ;) )  We put in place a netting agreement between the derivatives and the loans to avoid economic credit risk. This does not affect the collateral as the margin regulation explicitly prohibit to take those cross-products netting agreement into consideration for the computation of the collateral on derivatives. I have just created a legal synthetic OIS in EUR using derivatives, CSA and loans when a simple OIS would be illegal!

What is the goal of the regulation? Make the risk management of financial risk for people that need to manage it more difficult and requiring financial engineering or to make the market safer and more efficient? I see more of the former here, but maybe my eyesight is getting poor.

It is very good that the press asked this important question and was able to get an answer. That what the press should do and I congratulate the journalist for that. But personally, I would have preferred that such an announcement, which amounts to a regulatory decision, was done publicly, for example on the ESMA website and not in a private for-profit news magazine. If I was not a subscriber of the magazine, I would not know about this new ESMA policy.

The list of benchmark administrators registered under the new regulation can be found on the ESMA website at https://www.esma.europa.eu/benchmarks-register


More SOFR products

A floating rate note and a futures, this is what this week brings us on the SOFR front.

Fannie Mae has launched SOFR-indexed notes. The issuance is described in a Risk article: Investors cheer debut Fannie SOFR note launch. Three notes with maturities of 6, 12 and 18 months are issues, with spread of 8, 12 and 16 basis points above compounded SOFR. The accrual periods ends with a four-day lock-out period, similar in some way to the two-day lock-out for the Fed Fund swaps. Once more this means that the debate between forward-looking term rates and compounding backward looking rates is wide open.

In the same week, the Intercontinental Exchange has announced October 1 Launch of ICE One and Three Month SOFR Futures. This extends the ICE overnight offering beyond the SONIA futures for which 100 billions notional have been traded and competes with the similar products at CME.


Consultation on IBOR fallbacks: Question 2

Game of Benchmarks: Season 2

Game of Benchmarks: Season 2 - Episode 2: Triggers


Question related to: Description of Fallbacks - Triggers

In the triggers description, the benchmark to which a fallback would be applied is called “the relevant IBOR”. The list of those relevant IBOR is “GBP LIBOR”, “CHF LIBOR”, “JPY LIBOR”, etc. The relevant IBOR does therefore not include the tenor.

It is not clear from the text if the trigger applies on a tenor by tenor basis or on a full “family". Could there be a situation where one tenor, e.g. GBP-LIBOR-12M, is discontinued but not the others in the same family, e.g. GBP-LIBOR-3M. Would the discontinuation of one tenor trigger the fallback for all of them?

Could you clarify the tenor/family issue in the FAQ and potentially adapt the trigger wording?

Edit on 18-Aug-2018:

I have received an answer from ISDA regarding the above question:

If the discontinuation is of one tenor only, then it is likely that market participants will follow the interpolation approach that they followed when certain LIBOR tenors were discontinued several years ago.  The fallbacks we are implementing contemplate discontinuation of all tenors (although this is not yet a hard and fast rule because we are still confirming that there are no scenarios in which our fallbacks should apply to a tenor discontinuation - but I think that would be unlikely).

A list of FAQ is maintained by ISDA.

Season 2: The questions


Question 1: Question related to Option 3: Compounded Setting in Arrears Rate

Question 2: Question related to: Description of Fallbacks - Triggers

Question 3Description of Adjusted RFR - Timing


Consultation on IBOR fallbacks: Question 1

On 12 July 2018, ISDA has published a Consultation on Certain Aspects of Fallbacks for Derivatives Referencing GBP LIBOR,1 CHF LIBOR, JPY LIBOR, TIBOR, Euroyen TIBOR and BBSW.

I'm preparing a quant perspective on IBOR fallback proposals that I will publish in the coming days.

In the mean time, by reading the consultation document, I have a certain number of questions related to the clarification of the wording or the formulas proposed. I'm sending the questions to the ISDA email associated to the consultation and I'm also posting them here. I will update the post when I have an answer or a clarification. This is

Game of Benchmarks: Season 2

Game of Benchmarks: Season 2 - Episode 1: Compounded Setting in Arrears Rate question


Question related to Option 3: Compounded Setting in Arrears Rate


The three dates that characterise a IBOR fixing are its fixing date, the effective date of the underlying deposit and the maturity date of the same deposit. When used in derivatives, the effective and maturity dates of the underlying deposit are theoretical dates as no actual deposit takes place. Derivatives payments are themselves characterised by four dates: the fixing date, the start accrual date, the end accrual date and the payment date.

The consultation document indicates "The fallback could be to the relevant RFR observed over the relevant IBOR tenor and compounded daily during that period.” A formula is provided in Annexe A.

It is not clear if the “period” referred in the text (T to T+f in the annexe) is the IBOR theoretical deposit period or the derivative period.

If the period refers to the IBOR theoretical deposit period: How would the RFR for the period been know if the payment takes place before the end of the said deposit period. This would be the case for FRA with “FRA discounting” settlement (settlement in advance), for IBOR with fixing in-arrear or for vanilla floating coupons on coupons after a period with end accrual date on a non-good business day.

If the period refers to the derivative period: Does that mean that the same IBOR fixing will fallback on different values depending in which derivative it is referred and the consistency between rates fixing on the same date will be broken? What would be the period for a FRA that settles on the fixing date (e.g. GBP-LIBOR)?

Edit on 6-Oct-2018:

I have received an answer from ISDA regarding the above question:
This is an issue that we are going to need to look into for each of the relevant currencies as we move to implementation (if we move to implement the compounded in arrears rate).  The likely answer is the IBOR theoretical period with an adjustment for the payment date but it would be helpful if you could point out this issue in your response and explain the implications.

Season 2: The questions


Question 1: Question related to Option 3: Compounded Setting in Arrears Rate

Question 2: Question related to: Description of Fallbacks - Triggers

Question 3Description of Adjusted RFR - Timing


First cleared SOFR trades

Last week, on Wednesday 18 July, LCH has announced the first cleared SOFR linked swaps.

According to a Risk article (subscription required), the first trade was a SOFR V EFFR basis swap.

Now the real fun start. Compute all the basis and check how they behave!


Risk-based futures

Financial Fiction Episode 3: Risk-based futures

For linear interest rate derivatives, risk and DV01 are considered as very similar expressions.

This is also what we can deduce from the name of the latest new futures launch by NASDAQ.

Some five years ago, I proposed a new design for interest rate futures that I called risk-based futures. I worked with ASX in 2014 to adapt the design for the AUD market. At the end of 2015, ASX launched a swap futures based on that design. The ASX product has an extra feature of having a variable tick value, but the central feature of the design is the same: a futures price representing a rate or yield and the physical delivery of an ATM trade on the futures expiry.

NASDAQ will be launching soon its DV01 Treasury Futures. According to NASDAQ, the new product will be available for trading on Thursday, July 19, 2018, pending regulatory approval.

The name "risk-based futures"(1) has been changed to "DV01 futures", but the central idea is the same: future style margin on the price multiplied by a conventional DV01/PVBP and physical settlement at expiry into an at-the-money trade.

In my generic design the delivery mechanism can take several variants. NASDAQ has selected the simplest one where the notional of the delivered trade is fixed in advance. This create a jump in risk at delivery. The risk jumps from the futures term-sheet DV01 (e.g. 850 USD by basis point for the U.S.10-yr DV01 Treasury futures) to the actual DV01 of the delivered instrument. The variant is to deliver at expiry a trade with a notional adapted to smooth the risk.

The design of the futures I proposed is very versatile. As mentioned above, ASX has proposed a version; the underlying is a BBSW swap. The NASDAQ is now proposing a version with US Treasury underlying. I have proposed a version for the overnight-indexed market. The overnight-linked market is expected to become more important with the increased importance of SOFR, SONIA and probably a new EUR overnight benchmark. I will publish the details of the OIS based version in the coming weeks.

Another financial fiction becoming reality! See Financial Fiction 1 and Financial Fiction 2.

(1) I used that name at the The 3rd Interest Rate Conference (London, UK) in March 2015 in a presentation with the title Deliverable Swap Futures: a risk-based design.


Triple basis

In a recent Risk article (Clearers diverge on SOFR swaps discounting - subscription required) it was indicated that, contrarily to previous announcements, when CME will start to clear SOFR linked swaps, the interest used for Price Alignment Interest (PAI) will be SOFR itself and not the EFFR which is used for other USD OTC derivatives. This is also a departure from the initial ARRC suggestion. Regarding LCH, the current indication is that it will clear SOFR linked swaps with EFFR PAI.

This means that out of the six variations of overnight-linked derivatives discussed in a previous blog, five will be available in Q3. It is not certain that the sith variations will ever be really traded, but we can expect that if SOFR is becoming the benchmark benchmark (I'm not sure it is the correct term, but it seems appropriate here) at some stage some legacy EFFR OIS will switch to SOFR for collateral.

From a long term perspective, it seems logical to clear SOFR based swaps using SOFR collateral interest. Starting with EFFR collateral would mean that at some stage in the future, the CCP rule or the master agreement would need to be changed to switch to the new rate.

Now a trick questions: When will the basis swaps EFFR v SOFR start to trade/clear? What will be the PAI on those swaps?

Over the recent years we have seen the appearance of a basis between swaps traded at different CCPs. With this announcement we will see the appearance of a triple basis: on top of the CCP basis, we will have the discounting basis (SOFR v EFFR) and the "convexity adjustment" basis (on SOFR forwards). It will be impossible to untangle those basis as there will be for the moment only a limited number of traded swap types out of the multiple combinations and no option market yet.


EIB SONIA-linked bond

The move to the use of overnight benchmarks as the main interest rate benchmarks is progressing. After the wave of new overnight-linked futures in the last months by CME, ICE and CurveGlobal, we now have a large overnight-linked bond issuance. The EIB has successfully issued a one billion 5-year SONIA-linked bond.

The coupon payment is based on the backward-looking overnight SONIA composition plus a spread of 35 basis points (Maturity 2023-06-29, quarterly payments, ISIN XS1848770407). Note that a small 2 millions bond had been issued in January (quarterly SONIA + 25 bps, maturity 2023-03-20, ISIN XS1889459713). Thanks to an acute observer of the market for pointing that to me.

Backward-looking overnight is viable. Certainly for some issuers and investors. There was never a doubt about it for some market participants. This does not prove it is viable for all market participants, and this is where the question lies. Note that the coupon will be paid with a five-day lag after the last overnight benchmark fixing. This is a serious discrepancy with the current overnight-linked derivative market which pays with a zero day lag in Sterling (and 2 in EUR and USD). Some elements of the issuance are discussed in a Risk article 'EIB shrugs off term RFR worries with Sonia bond plan' (subscription required) and a Financial Times article 'EIB to offer Libor alternative with Sonia-based bond' (subscription required).

The question of backward-looking or forward-looking — i.e. OIS benchmark — overnight-linked coupons is heavily discussed in the potential LIBOR discontinuation and fallback provisions. The EIB issuance, even if linked to backward-looking overnight benchmark, does not bring any substantial positive information regarding the fallback discussion. If anything, the five-day lag expresses the difficulty of that approach. The backward looking payments have been used for many years in the derivatives market and there is not doubt that they are working for products designed with its requirements in mind.

Regarding the LIBOR fallback, the products linked to LIBOR have been designed with the term rate at their core. Can we replace a term rate by a backward-looking rate? That is the question. The answer is: in some case yes and in some cases no. The obvious 'no' are FRA with FRA discounting settlement and LIBOR in-arrear that require a payment at the start of the term period. As the fallback procedure should be the same for all LIBOR usages in derivatives, the global answer to the fallback procedure question has to be 'no, a backward-looking fallback is not possible'. The above answer is not only a personal opinion, it is a physical impossibility, except if one can realize backward time travel.

I will provide more technical details and opinions about the LIBOR fallback in a forthcoming notes. Hopefully I will have time to finish it by mid-July.


Multi-curve framework: double or quits?

This blog entry is a reaction to the column by Bakkar and Brigo in Risk titled Model risk in the transition to risk-free rates. While I agree with most of the column, there is a secondary issue, which is close to my heart, I would like to react to. And I certainly concur with the last sentence of the column which starts with "This should involve the quant community in an active role" as this has also been my opinion for a long time.

I would nevertheless like to present a different perspective on some side elements of their argument. To my opinion, the expanded use of overnight benchmarks will (should) not reduce the multi-curve complexity.

The pre-2007 single curve "simplicity" was coming from a combination of error and domination. Error in confusing two effects, the (pseudo-)discounting and the reference benchmark. This error was helped by the domination of a rate family: the IBOR benchmarks — the LIBOR king. The crisis emphasised the difference between the overnight benchmark used for collateral remuneration and the IBOR benchmarks and lead to the overnight discounting. A second crisis impact is that the king is dying.

Moving to overnight benchmark as the dominant benchmark may simplify the valuation of vanilla instruments by simplifying the vanilla contracts(1); both the collateral/discounting and the reference benchmark will be the same. But it will only simplify the bank balance sheets on the vanilla derivative side, there is no guarantee that it will simplify the balance sheet globally. The funding term structure of a bank will become the LIBOR of tomorrow but without its liquidity. The generalised VM/IM and its associated costs has pushed the funding issue into the valuation of derivatives (FVA, MVA). There is no need to see the LIBOR or the funding cost to appear directly in the instrument term sheet to need it in derivatives valuation.

I certainly concur with the authors that transition to new benchmark could be an opportunity to review models related to funding but I don't think it will reduce multi-curve complexities. It may lead to a name different from Multi-curve Framework, but the issue of different curves to be used for different purposes will continue to be relevant, maybe even more so with the introduction of new benchmarks. Maybe it will be called the Funding Framework(2).

Even if I concur with the authors that the transition could be an opportunity, I'm afraid that the market will have its back against the wall due to time and regulatory constraints — like in the margin changes case —, that only the minimal to survive will be realistically achieved and that the bank managements, market associations and regulators will not let the quant community play an active role.

Like for other similar issues, I sincerely hope that I will be proved wrong on the quants involvment. Don't hesitate to let me know that I'm wrong, I will enjoy it.

Don't hesitate to contact me for projects on the subject of benchmarks transition.

(1) Contract here as to be understood as a combination of the specific instruments term sheet and the generic master agreement, which govern the now mandatory associated daily variation margin.

(2) By the way, this was the name I used in my March 2007 paper on Discounting Irony, the first paper on what is today called multi-curve framework.


SOFR and clearing

The two main clearing houses for USD OTC interest rate contracts (LCH and CME) have announced that they will start clearing SOFR indexd swaps in the third quarter of 2018.

The announcement was discussed in the following Risk.net article: LCH and CME to start clearing SOFR swaps in third quarter.

Even if the clearing of SOFR swaps becomes available, users will have to wait a while for the full effects of the new rate to appear in the market. As envisaged by the Alternative Reference Rates Committee (ARRC), the Fed Funds rate will continue to be used to calculate the price alignment interest, i.e. to pay the interest on the variation margin.

This means that for the moment, one would not be able to price and risk manage SOFR OIS without having a full term structure of the Fed Funds OIS. The SOFR OIS will move from the status of non-existence to the status of second class citizen of the OTC market world, but not yet to first class citizen.

This is not a surprise as the transition from one benchmark to another is a very difficult task, the difficulty of which is well documented, in particular on this blog.

From a SOFR's derivative perspective, it means that there will be two type of derivatives, the one collateralized a zero rates (futures) and the one collateralized at fed fund rates (OTC OIS). And we will be waiting for the third type, collateralized at SOFR to appear soon.

In term of valuation, it means a multiplication of convexity adjustments. In a recent blog and working paper, I mentioned the adjustments between overnight futures and OIS collateralized at the OIS index. The situation proposed by the hybrid rate approach will require further adjustments. This may be the subject of further blogs and working papers. It will take time to develop as we need a good multi-curve model with multiple discounting and a meaningful understanding of the join dynamic of Fed Funds and SOFR.Very soon we will have, just for overnight rates in USD 2 forwards and 3 discounting. With a couple of weeks of history on the SOFR fixing and no related derivatives today, we are very far away from having that understanding. It is of note also that the secured rate SOFR is fixing above the unsecured rate Fed Fund and also that the spread has gone from 13 basis points to 4 basis points in a couple of weeks.

It will also soon be time to offer for trading the rate quoted overnight indexed futures that I proposed some years ago. If they were traded on SOFR and the OTC clearing was done based on SOFR collateral rate, they would allow to create a full term structure with a unique design from 1 month to 30 years. It would also have very little convexity adjustment wrt OTC products, a lot less than the current offering based on 3 month futures traded with expiry up to 5 years. Don't hesitate to contact me for more detail about that design.


Overnight indexed futures

With the recent changes in market infrastructure and in the regulatory framework the importance of overnight benchmarks has increased in the last years and is expected to increase further. With that increased importance, the market will look for source of liquidity for overnight based derivatives beyond the traditional OIS. In its document on SONIA as the RFR, the The Working Group on Sterling Risk-Free Reference Rates calls for the development and promotion of interest rate derivative products which reference the RFR, including the design of a futures contract.

At a couple of days interval, CurveGlobal and CME have announced their plans to launch new overnight benchmark based futures. In the case of CurveGlobal, the futures is called Three month SONIA futures. The launch is planned for Q2 2018. In the case of CME the futures is called CME Three-Month SOFR Futures. The launch is planned for 7 May 2018. The CME futures is based on the Secured Overnight Financing Rate (SOFR). The SOFR rate is published by the Federal Reserve Bank of New York in cooperation with the U.S. Office of Financial Research. The rate publication started on 2 April 2018 and was discussed briefly in two previous blogs (here and here).

In a recent working paper, I describe the detailed futures instrument cash-flow and propose the pricing, including the convexity adjustment, of the new instrument in the collateral framework using a Gaussian HJM-like model. The paper is called

The adjustment obtained is relatively similar to the one obtained for LIBOR futures (see for example Eurodollar Futures and Options: Convexity Adjustment in HJM One-Factor Model) when the stochastic nature of the LIBOR/OIS spread is ignored. Some extra small adjustment need to be added to take into account that the futures settles only at the end of the accrual period while the LIBOR futures settle on fixing at the start of the underlying deposit period.

This futures is a new products that may help to bridge the gap between OTC and ETD in the overnight indexed products. It has still several inconvenient features that makes it difficult to be used as a building block to build an overnight yield curve up to 30 years as required by the market. Most of those drawback would be solved if the market was to adopt a design closer to the one I proposed some years ago.

I will continue to preach for the alternative design and don’t hesitate to contact me if you are interested by my sermon!


SOFR one week on

The Secured Overnight Financing Rate (SOFR) has now been published by the Fed New York for a week. Time for an early analysis.

The data published does not only consists of the benchmark rate itself which is a median rate of the data collected, but also the 1, 25, 75 and 99% percentiles. From that information, we can see that 50% of the rates captured are outside a 10 basis points band around the published rate. For the Effective Fed Fund Rate (EFFR), the 50% band is only 2 bps.

What is the origin of this larger dispersion of rates for the SOFR rate? Maybe the daily publication of the figures will increase transparency in that particular market which will lead to the narrowing of the dispersion? We will see if this is the case in a couple of months.

Another type of explanation for a larger diversity of rates could be the existence of a bid-offer in the SOFR underlying data. The Effective Fed Fund Rate is calculated on overnight federal funds transactions, which means that the transactions used are interbank transactions, transaction between are relatively homogeneous set of participants. On the other side, the SOFR transactions also include trades with clients and market-takers. The set of participants is less homogeneous. Could this 10 basis point difference be explained by bid-offer?

On a different subject, note that between the 3 April and the 4 April, the SOFR went down by 9 basis points while the EFFR was unchanged; the spread between SOFR and EFFR decreased by 9 basis points.

The spread between SOFR and EFFR is currently at 6 basis points. The secured rate still well above the unsecured rate!

The page on the Fed New York website with data related to SOFR is:

The page on the Fed New York website with data related to Federal Funds Effective Rate is:


SOFR published

The first value of the Secured Overnight Financing Rate (SOFR) benchmark has been published today by the New York Fed. The page for the rate is: https://apps.newyorkfed.org/markets/autorates/sofr

A new benchmark that is expected to play a role in the Games of Benchmarks.

The first value of the benchmark, for overnight rates between Tuesday 2 April and Wednesday 3 April, has been fixed at 1.80%. For the same period, the Fed Fund Effective rate has been fixed at 1.68%.

This means that the secured rate is 12 basis points above the unsecured rate!

Note that like for some other overnight rates,  the new benchmark is based on rates for trade on the start of the overnight period but only published the next day. It is difficult to understand why, in this era of real time and supercomputers, it takes so long for the publication of the rate. Even LIBOR that is probably not a paragon of virtue was published as soon as computed, around 11:00 a.m. London time. Note that SONIA, when it will be moved to its reformed version on 23 April 2018, will also be only published on a next day basis. Personally I would prefer an immediate publication. But it seems that timely publication is not a requirement for benchmarks.


Variation margin in presence of trade cash flows

A couple of years ago, I have posted a blog called Continuous dividend v cash flows.

With the generalisation of Variation Margin (VM) collateral, the derivative world is not driven anymore by discrete cash flows but by continuous dividend. Due to practical constraints, the VM is paid with a one day delay. This delay reduces significantly the effectiveness of the margin process as credit risk exposure reduction mechanism around the trade cash flow payments. The above blog presented an efficient and simple approach to bring back the effectiveness of the VM process even around trade flows dates. The approach is based on the usage of a forward valuation in the VM computation process.

Since then, the research related to the spike of counterparty exposure has earned to its authors the Quant of the Year award. In the award winning paper (Does initial margin eliminate counterparty risk?), the authors have introduced my proposal in one of the conclusion paragraph.

In its Guide on assessment methodology for the IMM and A-CVA, the ECB has also put as special emphasise (Part 4, Chapter 7, 43.(3).(e)-(g)) on looking at this issue.

As the issue is gaining more attention, I though it would be useful to republish the blog, with minor updates, as a working paper. The document can now be found on SSRN: Variation margin in presence of trade cash flows.


Change of benchmark overnight index is a difficult task: 2017 year-end review

Is SARON, as of 1st January 2018, the CHF benchmark for all OIS trades and all CSAs? Has TOIS been discontinued at the same date?

If no, please read my blog from February 2017 and its follow-ups.

If yes, I apologize for my doubts and congratulate the CHF market for that amazing achievement that I though was impossible.

Blog written on 27 February 2017 and scheduled to be published on 1 January 2018.