Interest rate derivatives house of the year: Goldman Sachs

Risk Awards 2020: US bank leads the way on SOFR, and gets creative to facilitate US insurer hedging

Josh Shiffrin, Jill Borst, Guillaume Heile
L–R: Josh Schiffrin, Jill Borst, Guillaume Helie
Photo: Alex Towle

There are now three certainties in life, according to John Williams, president of the New York Federal Reserve – death, taxes and the end of Libor.

But with just over two years left until the earliest point at which the Libor family of benchmarks could stop being published, the US swap market doesn’t seem to have got the message. Volumes in swaps referencing the secured overnight financing rate (SOFR) – the replacement for US dollar Libor – have been low, with trade repository data showing only 1,022 trades done as of November 8, of which 87% have a tenor of less than five years.

The main problem has been a lack of liquidity. This is expected to ease when clearing houses change their discounting rates from fed funds to SOFR next year. But some market participants can’t wait that long, and need hedges for their repo financing or their SOFR-linked notes now.

Enter Goldman Sachs, which played a key role in getting the market off the ground, being a counterparty in the first interdealer SOFR swap, and the first cross-currency swap against SOFR.

Large issuers describe Goldman as their go-to dealer for SOFR swaps: “I don’t think it’s close on the SOFR market-making front,” says a treasury source at one large financial institution. “We’ve basically bid out everything in the SOFR space as it’s not very transparent. They’ve won 90% of our enquiries.”

A treasury source at one government-sponsored entity says the bank is “one of the strongest in the SOFR market”, particularly in longer swap maturities. A source at one SOFR-issuing financial institution agrees, saying that while dealers have the capability to do SOFR, what separates Goldman from the others is its ability to price particularly tight on longer tenors.

It’s not just issuers. Over the course of the second and third quarters of this year, Goldman helped a mortgage real estate investment trust, which used Libor swaps to hedge its assets, to transition to a mix of federal funds and SOFR. This not only allowed the REIT to move positions off Libor, but, given it funds itself via repo, helped better align its asset and liability profile – SOFR is an overnight rate, derived from transactions in three segments of the US dollar repo market.

We’ve basically bid out everything in the SOFR space as it’s not very transparent
Treasury source at a large financial institution

Goldman Sachs says the key to its ability to take down risk in the shorter maturity buckets is the co-location of its swaps, repo and short macro trading desks. On the surface, these may look like very different products, but Goldman’s traders argue the differences are cosmetic and the degree of correlation is typically high.

For example, when US repo hit 10% on September 16, the desk was able to provide liquidity by matching off risk between total return swaps and SOFR futures. Repo and even cross-currency basis trades can also be used as a SOFR offset if the timing is right.

Jill Borst
Jill Borst

At the longer end of the curve, the trick is still to line up flows coming from SOFR note issuers and institutional hedgers.

“The unified leadership is really helping us tap cross-market liquidity and source orthogonal risk for our franchise across businesses that span everything from collateral to convexity and volatility,” says Jill Borst, head of North America interest rate products sales and co-head of global short-term interest rates and repo at Goldman Sachs.

The bank also draws plaudits from a number of clients for the quality of its content around the Libor changes, for instance analysis of how the industry’s Libor fallback consultations might impact basis spread levels.

Tricky vol

On the swaptions side, like other dealers Goldman Sachs has found sourcing US dollar rates volatility trickier since the end of 2018. First, the fall in swap rates meant the inventory of dealers’ callable bonds had shrunk, effectively reducing supply of receiver swaptions. Second, the Fed rate-setting meeting in December 2018 – and subsequent rates rally – signalled a swift move from one correlation regime to another, depressing supply from the market as regular participants readjusted their portfolios.

“What that essentially means is during the timing difference between the market establishing a new correlation regime and liquidity populating in macro markets, the dealer is the liquidity provider,” says Anshul Sehgal, co-head of US interest rate trading at Goldman Sachs.

This was a problem in the first half of 2019, as US insurers were clamouring for low-strike receiver swaptions to protect their portfolios against a fall in swap rates. It struggled to source options to hedge its various non-linear risks, known as greeks, and upon hearing this news, insurers actually accelerated their hedging plans.

During the timing difference between the market establishing a new correlation regime and liquidity populating in macro markets, the dealer is the liquidity provider
Anshul Sehgal, Goldman Sachs

To meet the demand from insurers, Goldman had to cook up an unconventional risk management approach, hedging the sold low-strike receivers – that delivered into long-dated swaps – with products that delivered into shorter tenors. To cover the risk of a curve-flattening affecting the shorter-dated options more than the longer-dated positions, Goldman bought curve options – products linked to movements in relative positions on the US swap curve – from its exotics desk.

“During normal periods, you can manage your portfolio by hedging away every Greek. But you have to find creative ways to risk manage the portfolio, bound your maximum downside and be willing to wear the mark-to-market volatility of having large positions – essentially basis positions – between different assets,” says Sehgal.

Anshul-Sehgal
Anshul Sehgal

As a result, of the $53.8 billion notional of interest rate options purchased by US insurers in the first half of 2019 – mainly low-strike receivers – Goldman Sachs had a 20% market share, according to the insurers’ regulatory filings.

A trader at one US insurer praises Goldman’s constancy: “Liquidity had dried up. Our trades are bigger than average, and Goldman comes in for multiple billions at a time. They never flinch, always give tight prices and are very consistent,” says a derivatives trader at one US insurer.

A senior trader at a second US insurer confirms Goldman was one of its biggest counterparties for swaptions trading this year.

The alternative hedge strategy performed “incredibly well” in the first half of the year, says Sehgal, but suffered in the bout of volatility in August. Despite that, the position never had a negative present value, he says.

Larger risk transfer trades have been Goldman’s bread and butter for years, given clients often share flow business among their relationship banks. But in an environment with less high-margin business available – and Goldman making a broader strategic push to cover all aspects of a client’s business – flow swaps trading has become more of a focus.

The philosophy, though, has changed, with flow being seen more as a source of information, rather than as a standalone money spinner.

Liquidity had dried up. Our trades are bigger than average, and Goldman comes in for multiple billions at a time. They never flinch, always give tight prices and are very consistent
Derivatives trader at a US insurer

“I think the concept of pure bid/offer is evaporating in terms of the ability to actually say ‘on this ticket, we’re making this much’. It’s much more about trying to traffic as much flow through your pipes as possible, to better inform you in terms of your inventory management,” says Nirubhan Pathmanabhan, co-head of global interest rate product trading at Goldman Sachs.

He adds: “And also there’s that sort of positive loop back in terms of providing your clients with feedback in terms of what is driving the markets at any point in time.”

This year, the bank increased its flow size for swaps, bonds and even swaptions, capturing trades that generate up to $200,000 of sales credits. This means 64% of global sales credits are now driven by flow. The bank has maintained a top four position in US dollar swaps on the main trading platforms by DV01, which is the sensitivity to a one basis point move in underlying rates. In euros, Goldman has been number one for euro swaps by DV01 on Tradeweb over the past two years.

A derivatives trader at one large European asset manager was very positive about Goldman’s swaps flow prowess: “Last year they were very good and this year they were again exceptional. They’re very focused on e-trading, which we’re very happy about.”

SPV success

On the structured side, from 2018 to November 2019, Goldman Sachs was the number one overall dollar-denominated rates and inflation-linked structured products issuer, with 218 medium term notes totalling $2.58 billion – with the most common product being callable notes issued under its own name.

This year also saw the launch of the bank’s new special-purpose vehicle, Luminis, designed to issue smaller rates and equity repack deals – with an average size of $5million–$7 million – for institutional and retail investors.

L to R: Hamza Hoummady, Thalia Chryssikou and David Wade
L to R: Hamza Hoummady, Thalia Chryssikou and David Wade

By using standardised legal documentation, the time it takes to take a trade idea to market has gone from a minimum of two weeks under its old repack platform, Signum, to more like two or three days. Issuance costs have been slashed.

“We changed the prospectus of our vehicle to make it more modular, which means every time we have a new investment, we don’t have to recreate the document from scratch. We had a one-off investment in the beginning to modularise it, but term-sheet creation is now much faster,” says Thalia Chryssikou, the bank’s co-head of global sales strats and structuring for equities and fixed income, currencies and commodities.

Luminis has already traded 10 times more tickets in a year than Signum did in 2018. The higher ticket count and shorter time to market has also given the Goldman rates trading desk another outlet to distribute its inventory.

GS Quant for Python

Like other banks, Goldman has been offering data and analytics to customers – a way to extend the client relationship at a time when trading revenue is under pressure. But this year, the bank has begun offering application programming interfaces (API) that enable clients to access the bank’s own interest rate models, marks and historical price data.

With the tool, called GS Quant, clients with Python skills can call and manipulate the data as required. For instance, a user could backtest potential strategies, run scenario analysis on exotic options portfolios, or pull up 20 years of hard-to-find data such as historical swaptions volatility skew.

GS Quant already receives around 200,000 API requests per month. For less sure-footed coders, Goldman also offers a simple interface with a set of menus.

One of the earliest users of GS Quant was Danish pension fund ATP, which partnered with Goldman to develop the tool. The fund is building a lot of its infrastructure in Python and uses the data in GS Quant to backtest interest rate derivatives strategies and create bespoke risk reports.

“The vast majority of our infrastructure is built in Python. Operationally it is therefore a significant advantage that we can access third-party tools directly with the coding language,” says Christian Kjær, head of liquid markets at ATP.

“When we approached Goldman with the idea it turned out they had similar considerations, so it was a great opportunity for a partnership. Being part of the development phase gave us the chance to influence the design to suit our needs in the best possible way. We now have a super-flexible tool that integrates seamlessly in our existing platform to complement our internal pricing tools.”

  • LinkedIn  
  • Save this article
  • Print this page  

Only users who have a paid subscription or are part of a corporate subscription are able to print or copy content.

To access these options, along with all other subscription benefits, please contact [email protected] or view our subscription options here: http://subscriptions.risk.net/subscribe

You are currently unable to copy this content. Please contact [email protected] to find out more.

You need to sign in to use this feature. If you don’t have a Risk.net account, please register for a trial.

Sign in
You are currently on corporate access.

To use this feature you will need an individual account. If you have one already please sign in.

Sign in.

Alternatively you can request an individual account here: