In their rush to rid the financial world of Libor rates, regulators glossed over a stubborn problem: an untold number of so-called tough legacy contracts are hardwired to the discredited benchmark and cannot be re-hitched to alternatives. With less than 18 months until Libor’s scheduled phase-out, regulators are finally addressing the issue head-on.

“The whole ballgame here is tough legacy and there’s no simple solution for it,” says Anne Beaumont, a partner at law firm Friedman Kaplan.

On June 23, the UK government came out swinging. It proposed legislation that would give the Financial Conduct Authority (FCA) sweeping powers to create so-called synthetic Libor rates for use in tough legacy contracts.

On paper, it’s a neat solution. A modified, formula-based rate with the Libor name would continue to be published after the end of 2021, when panel banks will no longer be compelled to make submissions.

In practice, the UK’s proposed solution may further complicate transition efforts in the US, where regulators are taking a different approach to the problem. Legislative proposals released in March by the Federal Reserve-convened Alternative Reference Rates Committee (ARRC) would provide safe harbour to force tough legacy contracts into regulatory-preferred alternatives upon Libor’s demise.

The parallel efforts are vastly different in scope. The US safe harbour would apply only to contracts governed by New York law, while the UK’s safety net rates could cover all five Libor currencies, including US dollars, Swiss francs and yen.

They are also different in substance. The US proposals would force tough legacy contracts onto new risk-free rates (RFRs); the UK solution gives them the option of using a modified version of Libor – a classic conflict of laws.

Diego Ballon Ossio, senior associate at Clifford Chance, says the UK’s stance is justified. “On the one hand, you don’t want to mess up other people’s work on this, but if you think about the regulatory objectives of the FCA, that’s not currency specific,” he says. “If the intention is to give stability to the regulated sector in the UK, you need to acknowledge the fact they don’t just use sterling.”

Others are uneasy about the territorial issues. “It’s not like the US courts can just defer to UK legislation, so there are always questions around whether people will accept it,” says a securitisation expert in the US.

The whole ballgame here is tough legacy and there’s no simple solution for it

Anne Beaumont, Friedman Kaplan

The problem cuts both ways. The FCA wants synthetic Libor to be a last resort for contracts that cannot be transitioned to overnight RFRs. But it will have little control over how its creation is used in other jurisdictions. “There’s nothing to stop people in Japan or the US using it for other types of contracts. So unless you’re able to embed its modified uses within the benchmark, it’s quite difficult,” says Guy Usher, co-head of financial markets and products at Fieldfisher. 

The bigger question is whether it will even be possible to create a synthetic version of Libor for currencies other than sterling. The FCA has said the formula-based benchmarks will be built using forward-looking term versions of replacement risk-free rates. These are unlikely to be available for Swiss francs and remain a distant prospect in the US market. 

“The fact it’s on the government’s agenda can only be seen as positive, but there is no solution here that will be easy, simple and straightforward – it’s fraught with complexity,” says Davide Barzilai, partner at Norton Rose Fulbright in London.

Forward thinking

On the surface, synthetic Libor seems to be the simplest solution to the tough legacy problem. Many of these contracts link interest payments to rates published on trader screens, such as Reuters’ Libor01. That means they should be able to track whatever number appears on those screens, irrespective of how it is calculated. But a substantive change – for instance, introducing a fixed credit spread on top of a term RFR – could still invite legal challenges.

“On the one hand, there’s an arbitrariness to the number and it doesn’t matter how you calculate it – you could put last night’s Yankees score on the screen,” says Beaumont at Friedman Kaplan. “On the other hand, it is supposed to represent something real and it goes back to the financial realities of whether you have a mismatch between your assets and liabilities.”

“It might be the case your contract just tells you to go to the screen and grab a screen rate, but your counterparty could still argue that’s not Libor, and you end up in court over that, as it fundamentally changed what you bought,” says the securitisation expert.

The final methodology could be crucial in determining whether synthetic Libor disputes end up in the courts. That should become clear before the end of this year, when draft legislation detailing the FCA’s powers to create the rate will be published by the Treasury, as part of the 2020 Financial Services Bill.

It might be the case your contract just tells you to go to the screen and grab a screen rate, but your counterparty could still argue that’s not Libor, and you end up in court over that

Securitisation expert

The FCA will draft a policy statement on the use of its new powers in consultation with the market. The final measures could be in place by the end of this year.

At an industry event on July 14, Edwin Schooling Latter, the FCA’s director for markets and wholesale policy, sketched out how a synthetic Libor rate could be constructed.

“The most obvious building block for a Libor methodology change is of course a forward-looking version of the relevant term rate,” he said. A fixed spread would then be layered on top of the term RFR to reflect the bank credit risk inherent in Libor. This would likely be calculated at the point when Libor’s cessation is announced – possibly during “the final weeks” of 2020, according to Schooling Latter.

That still leaves a lot up in the air. For starters, authorised term RFRs don’t currently exist for any Libor currencies and there is no guarantee they will be available any time soon.

The sterling market is furthest along. In June, FTSE Russell and Ice Benchmark Administration began publishing beta versions of the Secured Overnight Index Average (Sonia) in tenors from one to 12 months. The terms are based on daily snapshots of executable Sonia swap prices streamed to the central limit order books of select trading venues. Refinitiv and IHS Markit are also producing term versions of Sonia.   

Even that is cutting it fine. The sterling RFR group’s January 2020 roadmap envisaged beta testing beginning at the end of the first quarter, with live publication as early as the third quarter. Assuming a six-month trial period, live publication could be pushed back until the start of 2021.  

US efforts to build a term version of the Secured Overnight Financing Rate (SOFR) are even further behind. The ARRC will begin reviewing aspiring rate providers in September, with publication slated to begin in the first half of 2021.

That still seems optimistic. Liquidity in the underlying SOFR swaps market has lagged expectations and is currently insufficient to support a term rate. “There is a dependency on the derivatives market moving more actively to SOFR,” Tom Wipf, chair of the ARRC and vice-chairman of institutional securities at Morgan Stanley, said on July 13. “We’re not going to create a term curve by any means that doesn’t have robust transactions underneath it.”

The most obvious building block for a Libor methodology change is of course a forward-looking version of the relevant term rate

Edwin Schooling Latter, FCA

The UK markets saw $8 trillion notional equivalent of Sonia swaps traded in the first quarter, more than double the amount linked to sterling Libor. By contrast, only around $180 billion notional of SOFR swaps changed hands over the same period – less than 1% of the $35.9 trillion of US dollar Libor contracts traded, according to data compiled by the International Swaps and Derivatives Association.

There are some reasons for optimism. Since June 2019, CME has been publishing term SOFR rates using futures prices. The indicative rates are based on $187 billion of daily transactions and the methodology can be adapted to include SOFR swaps data from available venues. 

Swap volumes could also see an uptick after the ‘big bang’ in October, when LCH and CME will begin using SOFR to discount future cashflows on cleared US dollar transactions.

Other jurisdictions have shelved plans for term rates altogether. The Swiss National Bank has ruled out the creation of a term Saron, citing limited liquidity in Swiss franc swap markets. Saron swaps traded just $7.5 billion equivalent during the first quarter, less than 5% of the total traded on Swiss franc Libor, Isda data shows.

At the July 12 event, the FCA’s Schooling Latter alluded to the problem of absent term rates. “Even if it would be desirable to use the powers, this might not be possible for all circumstances or for all Libor currencies,” he said. “For example, where the inputs necessary for an alternative methodology are not available at all for the relevant currency, or are not available on appropriate terms to the Libor administrator.”

Control problem

Even if a synthetic Libor can be created, the FCA must perform a delicate balancing act to ensure it does not obstruct fallbacks or otherwise hamper transition to new RFRs. It is using both carrots and sticks to move the market in the desired direction.

If Libor users want their safety net, they must first deal with contracts that can be moved to alternative rates. At the July 14 event, Schooling Latter linked the creation of synthetic Libor rates to take-up of Isda’s fallback protocol, due to be published later this month.

“It’s possible, if there has not been wide take-up of the protocol, we might actually feel less able to use those powers to address tough legacy needs in the cash market,” he said.  “That, I think, would be unfortunate.”

At the FCA’s insistence, the Isda swaps fallbacks include pre-cessation triggers that will begin the process of switching contracts to new RFRs as soon as a Libor rate is deemed non-representative. US legislative proposals would also be activated at that point.

The Isda protocol takes effect in November, clearing the way for the FCA to make announcements about the discontinuation of Libor rates before the year ends. Attention would then turn to the methodology for creating synthetic versions of the discontinued rates.   

Without a pre-cessation trigger, you could get sucked into synthetic Libor – or you could very deliberately want to get into it

Industry lawyer

But not all derivatives will be amended via the protocol. Some swaps users may choose to skip the new clauses, or elect to bilaterally alter contracts with only permanent cessation triggers.

“Without a pre-cessation trigger, you could get sucked into synthetic Libor – or you could very deliberately want to get into it,” says an industry lawyer.

The FCA’s goal is to flatten Libor exposures to an “irreducible minimum” by the end of 2021. UK regulated firms would be banned from using the synthetic rate in all new contracts and the vast majority of legacy instruments, with carve-outs only for specific circumstances. That would leave an estimated 10% of legacy debt the regulator deems to be unsuited to backward-looking RFRs.

Examples include contracts with securities involving complex registration systems, or bonds set to flip to the last available rate on Libor’s demise, effectively becoming fixed until maturity. Some of these could work on compounded-in arrears RFRs, but high consent hurdles make that practically impossible.

Elsewhere, the pool of tough legacy loans could be much larger. “We think of tough legacy as those contracts that can’t be changed or don’t get changed for some other reason, which might just be inertia, or just sleeping on the job,” says one US lawyer.

Outside the UK, the FCA will have little control over the use of synthetic rates in these contracts. “The FCA makes it very clear the use of any such rate will be only for very specific circumstances – legacy contracts you’ve tried to amend but can’t. If you’re in the UK, that’s all well and good, and to some extent in the EU, because of BMR regulations. But outside of Europe, why would they be bound? Can they continue to use this rate if it’s there on the screen?” says Barzilai at Norton Rose Fulbright.

This could lead to forum shopping and regulatory arbitrage as contracts find their way to the most rate-friendly regimes. “If banks in Asia were able to use that rate going forward, would that create some change in the market? If you’ve got a whole load of legacy loans, you might then sell those off to Asian investors just because they can continue to use them on Libor rates,” says Barzilai.

Overseas regulators could pre-empt this by slapping similar restrictions on the use of synthetic rates. In the complex US regulatory regime, this would require unified action by prudential, banking and markets regulators – a remote prospect at the best of times.  

“It’s a little easier in the UK, as benchmark regulation regulates usage. In the US, there are a whole cast of regulators covering the banks and it’s not even clear who can tell the corporates not to do something. Congress could pass a law but it’s not something they do often,” says the securitisation expert.

To some, all this suggests synthetic Libor is only a viable solution for the UK market. The FCA may find it difficult to pull the plug on other Libor currencies until local regulators find a way to tackle the tough legacy problem.

“I think sterling Libor will go away on schedule because it’s a smaller market and not used in retail mortgages, but we may see others stick around for a bit longer with different rates disappearing at different times,” says Friedman Kaplan’s Beaumont.

This story originally appeared in Central Banking’s sister publication, Risk.net.