In the mid-1980s, banks in London wanted a uniform benchmark to apply across their new and innovative offering of financial products: interest rate swaps, currency options, forward rate agreements, to name a few. And so the London Interbank Offered Rate – LIBOR – was born.
LIBOR has, until relatively recently, enjoyed great success and global favour. Deals worth an estimated USD 350 trillion determine floating interest rates with reference to the LIBOR benchmark. It is the primary worldwide benchmark for short-term interest rates. Banks, insurers, pension funds, corporates and individuals are all familiar with LIBOR, the three-month US dollar rate being the cornerstone of the floating rate interest market. LIBOR also feeds into other indices. It is the most commonly used index for mortgages in the US, according to the Consumer Financial Protection Bureau.
LIBOR works, in very basic terms, by panel banks submitting to an administrator the rates at which they would lend money to one another, in different currencies and different maturities. From this data, averages are produced, which are then reported as a set of indices.
LIBOR hit stormy waters when in 2012, interest rate ‘rigging’ came under the regulatory spotlight. Banks have so far been fined USD 9 billion for offences involving fraud, widespread collusion and manipulation of LIBOR rates. Barclays’ tab alone stands at £290 million. The entire story has been widely reported in every echelon of the press.
The FCA’s Martin Wheatley published an independent post-review report in September 2012. Significant reforms were implemented in 2013, including: a requirement for panel banks to use actual trades and keep records to substantiate submissions; criminal sanctions for LIBOR manipulation; a drastic cull of quoted rates, down to only 35 from 150. Almost immediately, the British Bankers’ Association transferred regulatory oversight of LIBOR to the FCA. Then in early 2014, the BBA bowed out completely, handing over to a new administrator, the Intercontinental Exchange (ICE).
Still, it looks as though these measures were not enough to restore public confidence. ‘A good reputation is like the cypress; once cut, it never puts forth leaf again’ (Francesco Guicciardini).
The FCA’s Andrew Bailey has now indicated that LIBOR will probably cease to exist by the end of 2021.
Whilst no one is stopping panel banks from quoting, or ICE from producing the rates, the FCA has stated that post-2021, it will no longer ‘shore up’ the current system by persuading reluctant panel banks to contribute. Realistically, that makes the LIBOR benchmark unsustainable.
Leaving aside LIBOR’s tarnished image, the FCA has pointed to a ‘thin’ market and insufficient trades to provide the raw data for the indices. The underlying market that LIBOR seeks to measure - the market for unsecured wholesale term lending to banks - ‘is no longer sufficiently active’ (Andrew Bailey).
A reasonable target date has been set to allow time to find a suitable successor and, if possible, to facilitate an orderly shift.
The Sterling Overnight Index Average (SONIA) has been widely tipped as LIBOR’s heir apparent. This is an interest rate benchmark reflecting unsecured short-term transactions which are tied to the pound. Indeed, a couple of days after Mr Bailey’s speech, the Bank of England Working Group on Sterling Risk-Free Reference Rates published a White Paper on the proposed adoption of SONIA in the sterling markets.
The SONIA benchmark was introduced in 1997. The Bank of England is currently reshaping it, widening it to include overnight unsecured transactions negotiated bilaterally as well as via brokers. The proposal is for the Bank to collect information for interest rates on loans in excess of £25 million. Once a trimmed average has been produced, SONIA rates will be published the following day, at 9 am.
SONIA has its critics. Unsurprisingly, one of these is ICE, the US financial services company that administers LIBOR. ICE believes it has done an excellent job of restoring confidence in LIBOR, referring to ‘its long term sustainable future’.
Others have pointed out SONIA’s inadequacies:
- LIBOR covers five different major currencies; SONIA only covers sterling
- SONIA only measures the overnight rate, whereas LIBOR rates cover seven maturities, ranging from overnight to 12 months
- As such, LIBOR will of necessity be replaced by multiple benchmarks across different countries, leading to a more fractured market and greater complexity for negotiating parties.
Nevertheless, given the FCA’s stance, LIBOR’s demise at the end of 2021 looks certain. To echo Andrew Bailey’s advice, ‘Work must therefore begin in earnest on planning transition to alternative reference rates that are based firmly on transactions’.
What can be done about loans extending into 2022 and onwards? Is it possible to avoid lengthy negotiations and a protracted repapering exercise?
Currently, the LMA loan documentation sets out a cascade of alternatives in the event that the relevant LIBOR screen rate is unavailable. However, the drafting does not really contemplate the entire disappearance of the LIBOR benchmark. In that case, depending on if and how the parties have amended the LMA standard form provisions, there may be reliance on reference bank rates; or as a final option, the cost of funds to lenders. One can envisage that neither of these latter alternatives will be acceptable to the parties on a long-term basis.
The LMA have stated that they are ‘closely monitoring the situation’. Until there is further clarity, however, wording to promulgate a successor rate cannot be published.
A point to consider is whether or not to facilitate the amendment of facility agreements to reference the right successor benchmark, when the time comes.
In our view, it is worth reviewing and reflecting on the amendments and waivers provisions in loan agreements being entered into now, if those loans extend beyond 2021. The LMA news release following Bailey’s speech points to its optional wording, released in 2014, which provides for a fall-back in case of a full discontinuance of a benchmark. The fall-back helpfully provides for a replacement rate agreed upon by the borrower and Majority Lenders. This might well be preferable to requiring all lender consent for the relevant amendments.