Regulators have called time on Libor, the world’s most-cited reference rate. But its demise is not a sure thing. Danielle Myles assesses Libor’s chances of survival, and looks at what could take its place.

Liobor cover

It is regarded as the world’s most important number and the DNA of global finance. The London Interbank Offered Rate – better known as Libor – is the most cited pricing benchmark for financial contracts the world over. It determines the interest paid on more than $200,000bn of instruments, from the most complex derivatives and leveraged financings to small business loans and mortgages. If something is exposed to a floating interest rate, chances are it is based on Libor.

But its days could be numbered. The UK’s Financial Conduct Authority (FCA), cheered on by its US counterparts, has announced it will stop supporting Libor after 2021. The regulatory community is adamant that it must and will be replaced by new benchmarks, and working groups have been busily devising alternatives. But there are huge challenges involved in replacing one of the most ubiquitous features of the business world, not least convincing thousands of market participants to ditch it in favour of a new rate. Libor’s disappearance is not a forgone conclusion, but the stage has been set for a showdown over its future.

Shortfalls in judgement

Libor, of course, has a troubled past. It fell into disrepute after revelations in 2012 that traders had manipulated the benchmark for personal gain, resulting in $9bn in fines and no fewer than five prison sentences. A series of reforms means regulators are no longer worried about misconduct, but they are still uncomfortable with how Libor is set. This process sees a bank panel submit every morning to ICE Benchmark Administration (IBA) the rates at which they could borrow in the London money markets across seven tenors and five currencies: sterling, US dollar, euro, Swiss franc and Japanese yen. Before midday, the IBA publishes an average for each of the 35 rates. Thus, Libor is created.

What troubles the FCA is that these rates do not accurately reflect the industry’s cost of funding. This is because the number of submitting banks has fallen (partially due to heightened regulatory risk following the rigging scandal), and, more importantly, because the interbank markets for some Libor rates are almost non-existent. In one currency-tenor combination there were just 15 wholesale lending transactions throughout the whole of 2016. In situations such as this, daily Libor rates are based solely on expert judgement.

These concerns are not the FCA’s alone. They are consistent with the Financial Stability Board’s (FSB's) findings in 2014 about submission-based benchmarks more broadly. The UK regulator has also followed FSB recommendations in suggesting Libor be replaced by so-called risk-free rates (RFRs). They are called risk-free as they are overnight rates, and so do not reflect counterparty credit risk. RFRs are based on independently sourced transaction data, and the depth and liquidity of overnight markets means there is no risk of a transaction shortage. Libor’s five currencies have their own RFR, set in their local money market and overseen by a local authority. Some are reformed versions of existing overnight rates, such as the UK’s Sterling Overnight Index Average (Sonia), while others, such as the US’s Secured Overnight Financing Rate (SOFR), are new creations.

Administrator speaks out

But the FCA’s actions do not necessarily mean Libor will cease to exist. Its chief, Andrew Bailey, has simply said that after 2021 it will not compel banks to make submissions, something the regulator claims it has done for some years. It is plausible that Libor will continue after that date, and administrator IBA has been working hard to ensure that happens. It has proposed replacing the submission question with a so-called waterfall methodology. At the first level, panel banks that have borrowed in the wholesale markets from two different sources over the past 24 hours must submit those transactions’ weighted-average rate. A bank that has borrowed two to 10 days prior must submit the weighted average of those transactions, after adjusting it to reflect changes in the RFR (level two submission). As a last resort, banks can make a level three submission, which requires expert judgement based on market data and is reached in accordance with an IBA-approved framework.

The goal is to produce a Libor rate that is, to the greatest extent possible, anchored in wholesale funding transactions. Panel banks tested the process in 2017 and it will soon go live. “We expect to have all panel banks transitioned to waterfall methodology no later than the end of the first quarter of 2019,” says IBA president Tim Bowler. However, it is not clear whether all 35 rates will survive. “We are working with the banking industry over the course of the coming months to identify the Libor currency-tenor pairs that are most critical to their clients, to design a path for their continued publication on a voluntary basis beyond 2021,” explains Mr Bowler.

The idea is for the remaining Libor rates to co-exist alongside RFRs, but this depends on panel banks’ support. “We are committed to trying to find a path, but ultimately it’s up to the banking industry and their customers if they want to keep it,” says Mr Bowler. “We need the input from them to calculate and publish it.” The Banker understands panel banks are happy to adopt the waterfall methodology, and a number have reportedly accepted that they will submit to a reformed Libor post-2021.

The benchmark battle

These changes would be the latest in a long sequence of transitions that Libor has endured over its 30-plus year history. But regulators believe its time is up. In July, Mr Bailey spoke of “misplaced confidence” in its survival and warned that the “discontinuation of Libor should not be considered a remote probability ‘black swan’ event… firms should treat it as something that will happen”. In May, William Dudley, president of the Federal Reserve Bank of New York, said he was “sceptical about whether Libor can ever be adequately transaction-based” and that “Libor is likely to go away”. Chairman of the US Commodity Futures Trading Commission Christopher Giancarlo went further, declaring: “The discontinuation of Libor is not a possibility. It is a certainty.” 

Within the market, opinion is divided. Dr Luca Cazzulani, UniCredit’s deputy head of fixed-income strategy, concedes that submission-based methodologies in their current format are flawed. “The rate is essentially a price, and for a price to be as informative as possible, it should be based on supply and demand,” he says. “Hence, it should be based on a transaction rather than something that is more difficult to measure and lacking objectivity.” But while RFRs are undoubtedly on the rise, Mr Cazzulani believes that the complete disappearance of contribution-based rates “still isn’t a done deal”.

Charles Cochrane, a partner at Clifford Chance, agrees. “I don’t think it’s a fait accompli,” he says. “If the market doesn’t have a solution by the end of 2021, can you envisage a world where Libor administered by ICE still exists in some shape or form? Yes, I think you can.”

For borrowers and lenders, a more robust version of submissions-based rates is the cleanest solution to regulators’ concerns. “The problem with Libor is the huge reputational hit. But if it can be transitioned into something that is transaction-based and set in a credible way, then I think it has a future. It would be difficult to get the market away from it,” says Peter Chatwell, Mizuho’s head of European rates strategy. Benchmark users would be forgiven for favouring the path of least resistance. Certainly for corporates, which have not, by and large, been affected by the rigging scandal or submission shortcomings, Libor has always worked.

A case of inertia

While RFRs in Libor’s five currencies are now in operation, the market is clinging to their predecessor. Except for certain derivatives and a single £1bn ($1.3bn) bond sold by the European Investment Bank (EIB) that references Sonia, market participants have shunned regulators’ advice and continued to write business referencing Libor or Euribor, its European sister benchmark, which could also soon disappear (see box on page 25). Indeed, a recent industrywide report revealed there is a high level of awareness of benchmark reform, but a lack of concrete steps being taken to adopt RFRs.

Attuned to this, in July the FCA warned about the risks of inertia and the transition not yet being fast enough. But there are legitimate reasons for this. Some market participants, buoyed by the IBA’s reforms, are in denial over Libor’s potential disappearance. There is also significant uncertainty about how a world of RFRs will operate. One banker compared the situation to Brexit: authorities have announced a change to take effect in a few years’ time, and the market is left scratching its head over what it means and how to adapt. The FCA’s comments are understandable given that RFRs are pointless until the private sector starts using them. “Trade associations and regulators are working hard on this, but the market must determine how we do it,” says Mr Cochrane. “They might need a nudge from the working groups, but at the end of the day, it must be market-led.”

This raises the question of which parties must kick-start adoption. When it comes to loans, Clare Dawson, chief executive of the Loan Market Association (LMA), says: “Banks look to provide a product that the borrower wants. So it’s probably more likely to be driven by borrowers requesting a product based on a new benchmark.” Sarah Boyce, associate director of policy and technical at the Association of Corporate Treasurers, is aware of corporates with syndicated loans that wanted to renegotiate them past 2021, but banks were not yet in a position to offer a Sonia-based product. This has created a classic chicken-and-egg scenario whereby banks will create RFR-referencing products when they achieve threshold liquidity (indeed, Libor is so popular because it is highly liquid), but to reach that point, products must be created in the first place.  

Yet the main reason Libor is lingering is because including RFRs in new and outstanding contracts is riddled with problems.  

Overnight headaches

RFRs are fundamentally different to Libor in two ways. They are purely overnight, meaning they do not include rates for one month, three months, six months, and so on. They are also backward looking, which means the interest to be paid is not known when the contract starts. For derivatives, these features do not pose huge challenges as most trade overnight. Futures referencing SOFR have started trading in the US while Sonia-based futures are doing the same in the UK.

But corporate loans (which are predominantly floating-rate), securitisations and floating-rate bonds rely on Libor’s longer maturities and cannot function without a term rate. Overnight rates require interest to be calculated from scratch every day, and as they are backward looking, neither borrower nor lender knows how much interest will be paid until the end of the period. Using Libor, they know their exposure when the contract starts and can plan their cashflows and hedging accordingly.

The only cash instrument to date that references an RFR is a £1bn Sonia-linked bond sold by the EIB in June. Some have hailed it as a ground-breaking transaction which creates a template for other issuers. Others are more sceptical. The EIB’s bond took more than a year to come to market, reportedly due to modelling and technology changes needed to accommodate the daily calculation of interest. The FCA has acknowledged that some firms may need system changes to manage bonds and loans referencing backward-looking rates.

A multilateral can accommodate these liquidity requirements and operational changes, but most market participants cannot. “A lot of corporates, particularly the smaller and lower margin organisations, they don’t have that flexibility to borrow money on demand to settle an interest bill,” says Ms Boyce. She also notes that corporates are different to the vast majority of market participants in that they are not speculative, saying: “[They] almost exclusively operate in the financial markets to manage their risk… Not having term makes that much more difficult.” Backward-looking rates also present obstacles for bonds, which trade on the basis of known interest payments.

Of the many hurdles to migrating from Libor to RFRs, Richard Hopkin, head of fixed income at the Association for Financial Markets in Europe (Afme), considers the lack of different maturities to be the biggest. “If we could figure out how to get to a good term rate – for one, three, six and 12 months – that would be a big help,” he says. “For me, that is the next big thing.” UK and US working groups are looking at how to develop these based on their respective RFRs. However all the proposed alternatives are based on trading in derivatives that reference, respectively, Sonia or SOFR. Given that regulators want to make benchmarks more robust and verifiable, it seems illogical to opt for new rates that are complex and lack transparency. Some warn it risks putting the global reform initiative back to square one.

Outstanding contracts

Some bankers believe new RFR business is not the biggest headache. “At the beginning, the pool of contracts will probably be low and liquidity will need to be established, but this is not really different from any new product once it is launched,” says Mr Cazzulani. “The real issue is how you would deal with legacy contracts.”

For most asset classes, Libor must be switched to RFRs contract by contract. This is particularly problematic for bonds given that they have so many counterparties, and the beneficial owners (which may include individuals) are often hard to track down. A bondholder meeting would be needed, which involves great expense and effort to persuade sufficient bondholders to attend to reach quorum. Furthermore, changing payment terms usually requires great, if not unanimous, consent.

Derivatives find themselves in a different situation. They involve just two counterparties, and the International Swaps and Derivatives Association (ISDA) can implement industry-wide changes to ISDA-standard contracts by publishing protocols. Some observers suggest this means derivatives will be easier to change than cash instruments. But while protocols can amend legacy contracts as a technical matter, the ISDA’s assistant general counsel, Ann Battle, stresses that benchmark rates are economic terms that must be negotiated between the parties, particularly while some form of Libor still exists. 

“Absent of a regulatory mandate, amending a contract’s reference rate when the rate currently referenced still exists is a commercial decision between two counterparties” she says. “There are many issues for counterparties to consider other than simply whether the words in the documentation can be changed.” Protocols are a mechanism to efficiently amend multiple derivatives, but they are not a proxy for the commercial negotiation needed to amend any financial agreement.

New conduct risks

This problem highlights that transitioning from Libor to RFRs involves more than a clean switch of benchmarks. RFRs are overnight rates and so do not reflect counterparty credit risk. It means RFRs are likely to be lower than Libor. To maintain parties’ original bargain, an additional spread must be added. This may be relatively straightforward for bigger firms, but most players that use Libor lack the financial knowledge and skills to confidently renegotiate payment terms stemming from such technical changes.

In this sense, the potential disappearance of Euribor presents an even bigger problem as it is included in many retail products. Management consulting firm Oliver Wyman estimates that 5 million individuals collectively hold up to $800bn of Euribor-linked mortgages. This includes more than 90% of mortgage lending in Finland and about 60% across Spain, Portugal and Austria. The firm warns that banks could be exposed to considerable conduct risk in renegotiating these contracts, and therefore must communicate carefully with these clients and manage them through the transition.

Saving Libor

With significant uncertainty over Libor’s future and the development of workable RFRs, market participants have put in place some temporary fixes. So-called 'fallback provisions' are included in loans, bonds and derivatives to ensure a rate is available if Libor is not published on any given day. Though useful, these are not designed for a benchmark’s discontinuation. Afme has helped ease concerns within the securitisation industry by developing model wording for new contracts to provide a mechanism to migrate from Libor to RFRs in their final form. “It essentially removes the need for a bondholders’ meeting,” says Mr Hopkin, who describes it as “a small, but practical step towards stopping the problem becoming even bigger”. The LMA and UK working group dedicated to bonds are looking at something similar.

For many market participants, including Libor’s administrator the IBA, the ideal outcome would be a reformed Libor co-existing with overnight RFRs. Cash instruments that need term rates could continue using Libor while derivatives could, by and large, switch to RFRs. The exception is derivatives used for hedging, which must reference the same benchmark as the cash instrument to which they relate. Given that derivatives account for some 80% of Libor-based contracts, this should go some way to appeasing regulators by significantly reducing reliance on submission-based rates. Yet what is peculiar about this entire situation is that officials do not have the final say. By engaging with RFR working groups and the IBA – and for panel banks, continuing to submit to the latter – the market can determine benchmarks’ future. This is a rare opportunity, and one the market should not pass up.

Euribor’s fate

Libor’s plight is symbolic of a broader shift away from submissions-based rates happening around the world. The other major example is Euribor, the main benchmark for euro-denominated business, which determines the price in more than $150,000bn-worth of contracts. Over the past six years, its bank panel has shrunk from 44 to 20, and its administrator, the European Money Markets Institute, is testing a hybrid submissions process that is similar to Libor’s new waterfall methodology.

The adoption of this new hybrid methodology will determine whether Euribor complies with an incoming EU regulation which requires benchmarks be anchored, to the greatest extent possible, in wholesale market transactions. If it is not adopted, Euribor will cease to exist come 2020. Its future is far from guaranteed, especially because, unlike Libor, its panel banks have not struck an agreement to keep making submissions until 2020. “There is a sense that Euribor will end up being replaced, but the uncertainty regarding if, when and how is still enormous,” says UniCredit’s Dr Luca Cazzulani.

PLEASE ENTER YOUR DETAILS TO WATCH THIS VIDEO

All fields are mandatory

The Banker is a service from the Financial Times. The Financial Times Ltd takes your privacy seriously.

Choose how you want us to contact you.

Invites and Offers from The Banker

Receive exclusive personalised event invitations, carefully curated offers and promotions from The Banker



For more information about how we use your data, please refer to our privacy and cookie policies.

Terms and conditions

Join our community

The Banker on Twitter