It is 23.59 on December 31, 2021. In a minute’s time the London Interbank Offered rate, the interest rate on which trillions of dollars-worth of financial instruments are currently based, will cease to exist. The chance that all of these deals will have all been smoothly transitioned to the various alternatives that are now being developed is vanishingly small.
So, what will happen? At the most basic level any loan, bond or derivative that has not been transitioned will simply fall back to the last Libor rate before 2022 and stay there, jolting everything that is floating rate into a fixed rate instrument. At a broader level there will be widespread confusion, not helped by opportunistic players taking advantage of the chaos. This will spread from the capital markets to affect every part of the financial sector.
Will the regulators really let that happen?
“It is genuinely difficult to explain the full range of risks that exposure to a Libor contract after 2021 represents.” Edwin Schooling Latter, the urbane director of markets and wholesale policy at the UK’s Financial Conduct Authority (FCA), has been trying to hammer this message home at many recent industry events as that end date draws ever closer.
He has his work cut out. Libor underpins financial contracts between banks, asset managers, insurers and corporates estimated to be worth $350 trillion globally on a gross notional basis and it is about to disappear. The FCA has stated that it will no longer compel the panel banks that submit the data on which the interest rate is based to do so after 2021.
That figure of $350 trillion is so dizzyingly large that observers have struggled to convey the enormity of the task at hand. Professor Darrell Duffie, Dean Witter distinguished professor of finance at Stanford University’s Graduate School of Business, simply calls it “the largest financial engineering project in history”.
It is genuinely difficult to explain the full range of risks that exposure to a Libor contract after 2021 represents – Edwin Schooling Latter, FCA
Most contracts based on Libor today need to transition to one of the new overnight risk-free rates and they need to do it as a matter of urgency. As much as 90% of contracts that price off Libor are short term in nature, which makes their transition relatively straightforward.
But many aren’t and that creates a problem. Even if 90% of $350 trillion is easily transitioned, that still leaves $35 trillion that isn’t. “If there are still tens or hundreds of trillions of instruments referencing Libor at the end of 2021, will regulators let it stop?” asks Serge Gwynne, partner in Oliver Wyman’s corporate and institutional banking practice in London. “I don’t know, but I do know they will do all they can to avoid this from happening.”
Libor transition in the derivative markets is well on track, but it is a very different story in the cash bond and loan space. In the US, 72 bond deals, worth more than $80 billion, have been issued referencing the new secured overnight financing rate (Sofr) and in the UK there has not been a new unsecured listed public bond referencing sterling Libor and fixing past end-2021 since October 2018.
In the loan market, however, it is a very different story. “There is a big pricing challenge in commercial and corporate lending space and banks are at risk of losing more than they gain,” explains Gwynne. “Individual corporates will have hundreds of transactions that are Libor sensitive and banks will have to look at all of their exposure to that client. They need to get process and governance up and running before you can determine the impact of the transition. You need a minimum of six to 12 months on that negotiating process and you can’t even start to negotiate until there is a stable market in the rate that you want to transition to.”
And they have under two and a half years to do so.
Are overnight interest rates the future for term lending?
Not content with pressing ahead with this breakneck timetable, the regulators also seem to be trying to engineer a change in the way that people borrow. They want to see the term bond and loan markets move away from a term rate such as Libor and rely instead on the new alternatives, which are overnight rates.
While admitting that some form of term rate is needed, in a speech to the Securities Industry and Financial Markets Association (Sifma) in July, FCA chief executive Andrew Bailey declared: “These term rates cannot and will not be the primary avenue to transition. The risk-free rates themselves, Sonia [sterling overnight index average] and Sofr, should serve that purpose.”
At the same meeting, John Williams, president and chief executive of the Federal Reserve Bank of New York (FRBNY), agreed. “We are still some time off from a point at which a robust, IOSCO [International Organization of Securities Commissions]-compliant term rate can be created and use of such a term rate should be limited to certain segments of the loan market and to fallbacks for new contracts.”
For some, this is a step too far. Regulatory interference in the rate at which private individuals decide to lend to and borrow from each other is seen by many as blatant overreach. They understand that Libor has to go but baulk at the idea that they have to borrow money based on an overnight rate because the regulator has decided it wants them to.
Three-month Libor and compounded Sofr (estimated)
Source: FRBNY and Bloomberg
In the UK, the Bank of England is strenuously extolling the virtues of the concept, however. “Many issuers have found it easier than had been initially expected simply to use [the replacement risk free rate] compounded realized Sonia as a reference rate,” declared Andrew Hauser, executive director, markets at the Bank of England, in June.
“This has rapidly become a convention for a range of sterling floating rate notes and securitizations; and recent announcements suggest similar approaches are also feasible for a range of corporate bonds and loans.”
Benedict James, a partner at law firm Linklaters in London, who specializes in advising financial institutions on the prudential and structural regulatory landscape, recently pointed out that this is the first time that the UK regulator has felt the need to interfere in interest rates since the Usury Laws were repealed in the 1850s. He seems less than delighted that they have chosen so to do.
“The vast majority of borrowers could say that they are just trying to run a business and don’t have time to deal with this,” he tells Euromoney. “So, when Libor disappears, they could end up with a fallback to the bank’s cost of funds.” This he sees as the worst of all worlds. “We will have the bizarre situation where the regulators’ attempts to get a more objective rate end up with borrowers using the least objective rate of all – cost of funds – that is worse than Libor.”
The risks associated with the failure to launch an alternative rate need to be part of the conversation and the industry needs to be lobbying the regulator to say whether we need an extension – Robin Creswell, Payden & Rygel
In the US, there is also rising concern over the lack of a robust, term alternative to Libor. “In the UK the market is moving to the overnight Sonia rate, but in the US the feedback we have received from market participants is that it is more challenging to transition lending arrangements in US dollars to an overnight rate, given the global scale and volume of US dollar lending arrangements and associated asset-liability risks,” says Tim Bowler, president of ICE Benchmark Administration, which took over the administration of Libor from the British Bankers’ Association in 2014.
There is now a real sense that, while the transition away from Libor is starting to receive the attention and airtime across the market that it most definitely deserves, the risk of a term rate for the cash market simply not being available is one that needs to be taken very seriously.
“There is a high risk attached to the wrong outcome,” warns Robin Creswell, managing principal at asset manager Payden & Rygel. “The risks associated with the failure to launch an alternative rate need to be part of the conversation and the industry needs to be lobbying the regulator to say whether we need an extension. Other new regulations such as Mifid II and Solvency II were granted extensions and everyone acted in good faith.”
Will it come to that? Is the FCA really ready to pull the plug on Libor without there being stable and established term rates to replace it?
Input sources for current Libor tenors
Alternatives to Libor
Libor is produced in five different currencies (dollar, euro, sterling, Swiss franc and yen) and seven different tenors (overnight, one week, one month, two months, three months, six months and 12 months). This results in 35 separate rates that are published at 11.55 am London time every business day. Its production now relies overwhelmingly on expert judgement by a panel of submitting banks, with the volume of actual transactions that term Libor is based on now down to around $500 million on a typical day.
The flaws in this model have been brutally exposed since the financial crisis and the need for new reference rates, which unbundle the risk-free and bank sector credit risk elements of the interest rate curve, is beyond dispute. “Many Libor-linked instruments no longer directly involve banks, yet remain exposed to variations in the perception of bank risk seen in the past decade,” observed Hauser in June.
“That does not reflect an efficient allocation of risk in the economy, and poses material economic and distributional risks at times of stress.”
We will have the bizarre situation where the regulators’ attempts to get a more objective rate end up with borrowers using the least objective rate of all – cost of funds – that is worse than Libor – Benedict James, Linklaters
However, as the markets hurtle towards the year-end 2021 date at which Libor is due to disappear, the various replacements are a very long way from ready. In the UK, the pre-existing Sonia rate is well developed, with swaps established at both the short and long end: the notional of outstanding cleared Sonia swaps now exceeds £10 trillion. In Sonia futures open interest had reached £129 billion by the end of June. In the cash markets there was £28 billion of new issuance referencing Sonia between June 2018 and July 2019.
In Europe, neither the euro interbank offered rate (Euribor) nor the euro overnight index average (Eonia) comply with EU Benchmarks Regulation. Euribor is, therefore, being reformed to adopt a hybrid methodology and the European Central Bank is poised to launch its new euro unsecured overnight interest rate, the euro short-term rate (€STR), in October. The situation is helped by the fact that Euribor and Eonia will continue to exist in some form, but the task is still a huge one.
|
|
|
Cornelia |
“Every jurisdiction is struggling with its own problems and it will take a few years to be solved,” Cornelia Holthausen, deputy director general, market operations at the European Central Bank (ECB), tells Euromoney. “The euro area is urgently working on the transition to €STR. Europe is a different situation to the UK in that the short-term rate is starting from scratch, but the fact that a recalibrated Eonia, linked to €STR, will continue to exist for some time should help the transition.”
Sofr not so good
But the biggest challenge in this process is to transition the world’s largest capital market – the US – away from US dollar Libor. In 2018 the Federal Reserve Board and the New York Fed convened the Alternative Reference Rates Committee (ARRC), a group of private-market participants, to manage the transition away from US dollar Libor to the brand new Sofr.
By July, open interest in Sofr futures had grown to nearly half a trillion dollars but Libor still dominates swaps markets in the US. Issuance of US dollar bonds referencing Sofr had reached $135 billion by the end of June, but that is not nearly enough and is dominated by sovereigns, supranationals and agencies.
The illiquidity in Sofr swaps is a problem that will only be resolved by a wholesale switch by central clearing counterparties (CCPs) away from Libor. This is now due to take place only in the second half of 2020. “Not enough progress is happening on Sofr swaps,” says Gwynne.
“If CCPs aren’t planning to switch until the second half of next year it is difficult to see other triggers for this in the meantime. We would then only have a year and a bit before the deadline. The problem is that there are no cash products to hedge, so where is the economic incentive?”
It is a good question and one which the market, as it tries to get used to Sofr, is not rushing to address. However, Larry Manis, portfolio manager at Payden & Rygel in Los Angeles, says that it is just a case of adapting. “We have been active in the Sofr cash market. Operational adjustments are required to calculate accrued interest – it is a change, but it was achievable. It is confusing calculating the economics and it requires a shift, but you can get there.”
|
|
|
Larry Manis, |
Manis is a portfolio manager on the firm’s Global Low Duration Bond Fund, which invests with an average duration of 1.5 to 2.5 years and uses a one to three-year benchmark. “The repo rate is very different from Libor,” he explains. “It solves the issue of having robust underlying market data, but it behaves differently to Libor and the market needs to learn how it behaves. There are spikes, but the Fed can introduce tools to reduce volatility such as a repo facility. The Fed is very much aware of this. But in and of itself, Sofr will be more volatile.”
In Europe, Sonia and €STR are unsecured rates, but Sofr, being secured, is more unpredictable. On the last day of 2018, amid intense market stress, it shot up 147% to an intraday peak of 7.25%.
The impact on derivative and cash instruments
Using overnight rates has profoundly different implications in the cash and derivative markets. The International Swaps and Derivatives Association (Isda) has a protocol system for amending legacy contracts, which will be published in 2020 and which should make the process of moving to a risk-free rate far smoother.
“Universal changes to derivatives contracts will take out about 95% of the exposure [to Libor]. If the market signs up to the Isda protocol when it’s published early next year, it will be a huge step in the right direction,” declared the New York Fed’s Williams at the Sifma meeting in July.
Tom Wipf, vice chairman of institutional securities at Morgan Stanley and chair of ARRC, agrees, telling Euromoney: “Once people have secured market risk through the Isda protocol we have a line of sight to 2021.”
If 95% of it is solved in this way, why is there still a problem? Because regulators’ desire to see all financial markets transition to overnight rates as standard is not shared in many parts of the cash market. Many borrowers want term rates and those lending to them almost certainly do as well. “Lenders do not want to give investors economics that they don’t want,” says Manis. “It is possible to get a term rate in place before Libor ceases, but it is an aggressive timetable.”
The search for a term solution
Even the regulators have been forced to accept that they can’t force everyone to use an overnight rate. “The prevailing view on our Risk-Free Rate Working Group is that overnight Sonia, compounded in arrears will and should become the norm in bilateral and syndicated loan markets too,” the FCA’s Bailey insisted in his Sifma speech, conceding that: “Some borrowers may prefer precise cash flow certainty months in advance even if it would be less costly to use the overnight rate in arrears.”
Summary of RFR liquidity across products
Source: Oliver Wyman
The solution that has been proposed is to derive term rates from compound or simple averages of observed rates, using end-of-day futures prices rather than intra-day transaction prices. As long as there is sufficient liquidity in the underlying, then term rates should be possible. The urgent question, however, is just how long this is going to take.
“Sofr is a perfectly valid index for all the reasons that Libor has its challenges,” says Glenn Havlicek, chief executive and co-founder of GLMX, a financial technology firm specializing in money markets trading, liquidity management and reporting.
“Repo is a $15 trillion market, so there is some pretty good action that can provide an overnight level – it is a rate which it is possible to determine with certainty. There are far more variables in the repo rate than the Eurodollar deposit rate, but they can be overcome. There certainly is a significant term repo market. However, information related to trading activity is hard to come by.”
Glenn Havlicek, chief executive and co-founder of GLMX
He argues that there is already a non-trivial term repo market. “On any day we at GLMX can see 50/50 overnight/term split, so there definitely is term market activity there. How large and consistent is term trading and is there enough of it to create reliable one, two, three, six month term indices every day? Probably, but the ability to mine that information is limited.”
Many borrowers want a term rate not only because the market needs to be able to build a yield curve and they need to be able to budget in advance but also for logistical reasons. For instance, many corporate systems require rates to be entered in advance.
“Small to medium-sized businesses rely on term interest rate benchmark settings in their lending agreements to manage their interest rate risk, and also for operational cash flow for budgeting purposes,” says ICE’s Bowler. “Although overnight compounding of risk-free rates works exceptionally well for derivatives traders, it is more challenging for real economy businesses to use these rates in their day-to-day operations.”
Using compounding in arrears profoundly changes the dynamic between lender and borrower. “Libor is simple and binary. Sofr tries to replicate this but has the disadvantage of being backward looking. The manifestation of lending in a Sofr world is that you don’t know how the daily average rate, looking back three months from now, will compare with the setting on day one,” explains Havlicek.
If CCPs aren’t planning to switch until the second half of next year it is difficult to see other triggers for this in the meantime. We would then only have a year and a bit before the deadline – Serge Gwynne, Oliver Wyman
Before founding GLMX, Havlicek was managing director of global liquidity management at JPMorgan. “Traditionally, the lender assumed the risk that their funding costs could vary from the fixed Libor-based lending return. In a backward-looking Sofr environment, the risk of that variability shifts to the borrower. As a result, there is little incentive for the borrower to lock in a rate for longer.”
That, surely, can’t be good news for the banks.
James at Linklaters certainly doesn’t think it will prove popular with lenders. “If you have a compounding market you won’t have break costs: If it is an overnight market the theology of break costs goes out of the window,” he explains. “If you let people have a three- month loan and they can pay it back whenever they want that is an overdraft – a series of one day loans. If corporates are offered this, they will bite your arm off. Wholesale lenders don’t like it very much though.”
He says that a better solution is for banks themselves to price the term rate. “Regulators are saying don’t borrow term, borrow overnight and then swap it into whatever you want,” he explains. “The sensible thing is to allow people to price on the OIS [overnight index swaps] market. Banks are much better placed to do this so they will base the term price from the OIS market. The Bank of England should make incentives for a deep and liquid OIS swap market to price the term market from.”
Banks want a credit-sensitive benchmark
In the US, ICE has proposed the introduction of a bank yield index, which could address the problem of Sofr volatility and provide lenders with a credit-sensitive benchmark to lend against.
|
|
|
Tim Bowler, |
“We are working to provide a solution to this challenge with the US Dollar ICE Bank Yield Index, which is essentially a lending benchmark that incorporates a term risk free rate setting plus a credit spread that reflects bank borrowing costs relative to risk free rates. This potential benchmark would mirror the economics of the lending rates that banks and businesses use today and would potentially help in the overall transition away from Libor,” Bowler tells Euromoney. “We have got consistent feedback from the market that there is a need for a credit sensitive US dollar benchmark.”
He believes that sufficient underlying liquidity is still there. “The US dollar market is different to other markets because the US dollar is widely-used in cross-border commercial lending agreements and the vast majority of non-US banks have more dollar loans on the left hand side of their balance sheet than they have dollar transaction-related deposits on the right.”
He says that this would offer banks greater protection in times of stress. “Market participants are used to pricing lending agreements as a spread against Libor. There is concern from both sides of the table about credit provisioning. If banks have to lend without referencing a credit-sensitive lending benchmark, they might reduce unfunded and undrawn lines due to concerns that these could be drawn more rapidly during a period of stress if they are priced relative to a risk free rate.”
Unsurprisingly the regulators have been less than enthusiastic so far. “ARRC has no objection to any type of reference rate that meets IOSCO standards,” says Wipf. “But I don’t see where you get enough volume – enough supporting transactions – to create an index. If someone can meet that requirement then we have no objection: we don’t see that path but others may.”
The perceived weakness in the plan is that it relies on banks continuing to submit data. The bank yield index is simply not seen as a significant enough improvement to Libor to be the long-term replacement. In his July speech Bailey was fairly dismissive. “Today we see no prospect of the benchmark administrator being able to continue with a dynamic credit spread – the likely choice would be between a risk-free rate plus fixed spread, or nothing. In other words, this does not provide a route to making Libor representative again.”
For banks, however, this could be an attractive solution and they will, probably, get behind it. “We already have 13 of the 16 US dollar Libor panel banks providing data to IBA for testing the US Dollar ICE Bank Yield Index and we feel confident that banks will continue to provide data if we are able to produce a credit sensitive benchmark to lend against as a result,” says Bowler.
Perhaps the biggest risk is that, in focusing attention on a proposal that may or may not receive IOSCO approval, the Bank Yield Index could end up actually delaying the establishment of a much-needed stable term rate in the US before 2021 is upon us.
Traditionally, the lender assumed the risk that their funding costs could vary from the fixed Libor-based lending return. In a backward-looking Sofr environment, the risk of that variability shifts to the borrower – Glenn Havlicek, GLMX
Given the challenges some of them face in coming up with term rates, regulators remain steadfast in their belief that the cash markets can and should get used to overnight rates. And given how much of the $350 trillion Libor exposure is derivative-linked that position is not surprising.
“Once you have eliminated all other options then you can look at term forward-looking rates,” says Wipf. “There is the appeal of something that looks like Libor but, until you can get enough activity in Sofr futures and swaps, it is hard to create a curve that will meet IOSCO standards,” he admits. “But if you get derivatives and FRNs [floating rate notes] onto overnight rates in advance and arrears you can transition 90% of the market.”
Wipf concedes that different players want different things but says that Sofr is flexible enough to work for everyone. “We don’t want to have too many pricing conventions out there,” he warns. “Maybe smaller corporates and consumers who want to know what their payments will be will use Sofr compounded in advance. Maybe money market funds will use average Sofr. As long as we don’t have five or six conventions within this, I don’t think that we are missing any big parts of the curve.”
A key milestone for Libor transition in the UK took place on June 11, when Associated British Ports (ABP), the largest port operator in the country, secured agreement from bondholders to switch the base rate on an existing £65 million FRN from Libor to Sonia.
“My position hasn’t changed since 2017. I don’t see a need for a term benchmark,” declares the company’s treasurer, Shaun Kennedy. “Some people might want one, but nothing will ever be as good as referencing the overnight rate. It is also very important to use the same rate as the derivatives market.
Shaun Kennedy, treasurer, Associated British Ports
“We will be trying to transition all products before the end of 2021,” he tells Euromoney. “The only thing that will stop us is if counterparties on the other side don’t want to. We have been through processes before. It can be coordinated, but there has to be a lot of engagement. Processes can be difficult when not everyone gets around to voting but I don’t see any reason why it can’t be done.”
Kennedy is certainly ahead of the curve and dismisses any concern that his new borrowing rate is backwards-looking. “It is helpful to know what you will be paying, but why do you need a forward-looking rate for that?” he asks. “If people are concerned about having cash flow available there is always a time lag. There is a lot of risk with Libor and a forward-looking rate would be similarly volatile and you are not getting cash flow certainty.”
Is it too late to wait?
Many corporate treasurers won’t agree with him, but they face a very difficult decision: move to an overnight rate or wait and risk getting to 2021 with no term rate to transition to. You don’t need to be an advocate for the former to understand that the longer it takes for a term rate to be developed, the greater the risk of huge market disruption when Libor disappears.
“In the lending market a lot of banks (in terms of the products they offer) and corporates seem to be waiting for a term rate. This is a reasonably risky strategy,” says Gwynne. “The timing of a term rate can’t be fully controlled by anyone and there is always a risk that it is delayed. If you start developing a term product and the term rate is only set at the end of Q2, 2020, then you still only have 18 months for the product to become established.
“If you agree that there will be large corporate demand for loans based off overnight rate then there is an argument for banks to develop their capabilities,” he continues. “That way, if the term rate is delayed you have a product. But a lot of banks rely on the lending infrastructure providers, which currently don’t support RFR [risk-free rate] lending. So, banks can do something tactical offline but not at scale.”
In my view, the biggest challenge isn’t liquidity or the creation of a term rate, it’s a willingness on the part of the market to stop using Libor – John Williams, FRBNY
In his July comments, New York Fed chief Williams made it clear how much of a risk the pace of transition in some parts of the market has become. “In my view, the biggest challenge isn’t liquidity or the creation of a term rate, it’s a willingness on the part of the market to stop using Libor,” he said. During research for this feature many interviewees claimed that there been a change in mood over the summer, a tipping point, and that momentum was now starting to build towards the deadline. That may be true, but it isn’t true across the board and certainly not in the loan markets.
“We have made a lot of progress with derivatives, we have done the FRN and will do another one and then we have loans and US private placements,” says Kennedy at ABP. “I am ready to transition these, but everyone is reluctant to be the first one to do it and you can’t force things to develop. We haven’t had standard docs put in place for loans yet.”
The situation in the loan market is a serious one. If the banks are not pushing lending against overnight rates, then it is hardly surprising that corporate borrowers aren’t rushing to embrace it either. The closer we get to 2021, the bigger a ticking time bomb this will become.
Holthausen at the ECB is acutely aware of the urgent need for better communication across the board. “In the short term the larger players are aware of the changes, but we are concerned about other users: smaller corporates and regional corporates,” she says. “ There is the potential for confusion, so we have to step up communication and outreach.”
There is also significant risk embedded in the varying levels of progress – and risk-free rate choices – that different markets are making. “Many participants don’t understand that Euribor and Libor may go different ways,” she points out. “I don’t see a problem if different countries take different paths, but what is most worrying is the cross-currency market. All the different working groups have to work very closely on this. What if one counterparty phases out an ‘ibor’ and the other one doesn’t? It is a very complex area and there is uncertainty.”
A legislative solution
Many believe that the sheer scale of the potential systemic risk that a disorderly end to Libor would unleash means that legislative intervention now needs to be seriously considered. “The regulator’s obligation is to help to maintain orderly markets,” point out Creswell. “It will be up to parliament to decide whether or not to legislate, although it will be on the regulator’s recommendation.” It is hardly an issue that many parliamentarians will be eager to get involved in. “Parliament might decide that legislation sounds like a government solution for what is a corporate markets problem.”
It certainly is a capital markets problem, but as the deadline draws nearer maybe legislation will have to move up the agenda. “A legislative solution would be helpful and reduce the burden, but we can’t go forward assuming that is the base case,” says Manis. In the US, the production of a stable term rate will only happen when there is enough underlying activity in swaps and futures, something that is dependent on their being enough cash demand for hedging.
So, with little more than two years to go, sticking to the 2021 Libor transition date could be a white-knuckle ride. “Sofr is imperfect, but it is the best alternative that we have and it can improve over time,” insists Havlicek. “Regulators need to identify a definitive transition date on which Sofr will replace Libor and then people will start to move.”
