April 22, 2013
Good afternoon. Thank you, Anthony, for that kind introduction. I’m honored to be joining you for City Week.
I’d like to talk about one of the most significant risks facing the capital markets today. That is the risk to market integrity as well as financial stability of the continued use of LIBOR, Euribor and similar benchmark interest rates.
Given their fundamental role in the capital markets and our economy, such benchmark rates must be based on facts, not fiction.
Coordinating with the FCA
The U.K. Financial Conduct Authority (FCA) (along with its predecessor the Financial Services Authority (FSA)) and Martin Wheatley have been valued partners of the U.S. Commodity Futures Trading Commission (CFTC) on this matter.
The CFTC initiated an investigation in 2008 related to the London Interbank Offered Rate (LIBOR). It is the reference rate for 70 percent of the U.S. futures market. It is also referenced by over half of the swaps market, which the CFTC was recently tasked to oversee.
The FCA has been instrumental in the CFTC’s investigations, leading to charges against Barclays and other banks for manipulative conduct regarding LIBOR and similar benchmarks.
Following the Barclays announcement, the international community asked Martin Wheatley and me to co-chair the International Organization of Securities Commissions (IOSCO) Task Force on Financial Market Benchmarks.
Last week, the task force published its second consultation paper outlining a set of international principles to enhance the integrity, reliability and oversight of benchmarks.
The IOSCO principles state that for benchmarks to be robust and reliable, among other things, they must have two essential elements: be anchored in observable transactions and supported by appropriate governance structures.
The IOSCO report further notes that in order to provide confidence that the price discovery system is reliable, benchmarks must be based on prices and rates formed by the competitive forces of supply and demand entered into at arm’s length between buyers and sellers in the market.
Unsecured, Interbank Market: Essentially Nonexistent
LIBOR, Euribor and similar interest rate benchmarks purport to represent the rate at which unsecured borrowing occurs between large banks.
The challenge we face, however, is that banks simply are not lending to each other as they once did. As Mervyn King, governor of the Bank of England, said in 2008 of LIBOR: “It is, in many ways, the rate at which banks do not lend to each other.” He went on further to say: “[I]t is not a rate at which anyone is actually borrowing.”
The lack of transactions in the unsecured, interbank lending market along with weak governance structures for related benchmarks undermines market integrity.
The dearth of transactions in this market is a result of many factors: the 2008 crisis, the continuing European debt crisis, the downgrading of large banks’ credit ratings, as well as central banks providing significant funding directly to banks.
There has been a significant structural shift in how financial market participants finance their balance sheets and trading positions. There is an increasing move from borrowing unsecured (without posting collateral) toward borrowings that are secured by posting collateral. In particular, this shift has occurred within the funding markets between banks.
In the aftermath of the financial crisis, for understandable reasons, banks have been hesitant to take on each other’s credit risk.
Recent changes to Basel capital rules further suggest that banks are unlikely to return to interbank lending on an unsecured basis.
Basel III includes a new asset correlation factor, which requires additional capital when a bank is exposed to another bank. This was included in the new standards to reduce financial system interconnectedness.
Basel III also includes a new requirement called the liquidity coverage ratio (LCR). Banks will have to hold a sufficient amount of high quality liquid assets to cover their projected net outflows over 30 days.
A number of major banks have indicated that this new LCR requirement alone would make it prohibitively expensive for banks to lend to each other in the interbank market for tenors greater than 30 days. Thus, it is unlikely that banks will return to the days when they would lend to each other for three months, six months or a year.
The shift away from banks funding each other in the unsecured market has led to a scarcity or outright absence of actual transactions underpinning LIBOR and other interest rate benchmarks.
This situation – having benchmark rates that are not anchored in actual transactions – undermines market integrity and leaves the financial system with benchmarks that are prone to misconduct.
Further, significant incentives for misconduct exist when hundreds of trillions of dollars of financial instruments reference benchmarks based on essentially nonexistent markets.
Indeed, as law enforcement actions brought by the CFTC, the FCA and the U.S. Justice Department, among others, have shown, LIBOR and other benchmark rates have been readily and pervasively rigged.
Barclays, UBS and RBS paid fines of approximately $2.5 billion for manipulative conduct relating to these rates.
At each bank, the misconduct spanned many years.
At each bank it took place in offices in several cities around the globe.
At each bank it included numerous people – sometimes dozens, among them senior management.
Each case involved multiple benchmark rates and currencies. In one case, there were over 2,000 instances of misconduct during a six-year period.
And in each case, there was evidence of collusion with other banks.
In the UBS and RBS cases, one or more inter-dealer brokers painted false pictures to influence submissions of other banks, i.e., to spread the falsehoods more widely.
Barclays and UBS also were reporting falsely low borrowing rates in an effort to protect their reputations.
Thus we find ourselves in a situation where there are both the incentives and ability to manipulate a critical rate in our markets.
Beyond these cases, there is a significant amount of publicly available market data that calls into question the integrity of LIBOR today.
Let’s take a look at what happened just in the last few weeks as the Cyprus crisis infected the Eurozone. Here is a view of one Eurozone bank’s one-year credit default swap (CDS) spread versus that same bank’s daily submissions to the U.S. dollar LIBOR panel (Slide 1).
The bank’s CDS spread, one market measure of its credit risk, widened dramatically. The bank, however, didn’t change its submission as to where it could borrow from other banks. Though CDS trade in a different market and are for a bank’s holding company, the disconnect, as shown in this slide, raises questions about the credibility of LIBOR.
In Slide 2, we look at the average of all five Eurozone banks that submit to LIBOR. The picture is similar.
Next, let’s turn to the volatility of three-month U.S. dollar LIBOR in comparison with the volatility of other short-term interest rates. LIBOR, the blue line, is far more stable than any other comparable rate. Other short-term rates have much higher volatility (Slide 3).
Also of note, is that the 18 banks submitting to U.S. dollar LIBOR collectively did not change their submissions on 85 percent of the 252 submission days in 2012. You can see in Slide 4 just how few times the banks actually changed their submissions over the course of last year.
In fact, some of the banks didn’t change their submissions for four to five straight months. This was during a period when there were a number of uncertainties in the market driven by elections, changing economic outlook and other events. And yet somehow these banks said they could still borrow at exactly the same rate for four to five months. Slide 5 represents the longest consecutive period last year that the submissions remained unchanged.
Taking another look at CDS spreads versus LIBOR submissions, this time over the last three years, highlights another query. As we see in Slides 6 and 7, during significant market upheavals in the second half of 2011, the market’s views of these two banks’ credit risk changed dramatically. Yet their LIBOR submissions moved only modestly.
While we’re done with slides today, two last points reflected in market data:
There is a well-known concept in finance called interest rate parity, basically that currency forward rates will align with interest rates in two different economies. Since the financial crisis, that has not been the case, whether looking at the dollar versus the euro, sterling or yen. Theory hasn’t been aligning with practice. The borrowing rate implied in the currency markets is quite different than LIBOR.
Lastly, why are the results of two leading interbank benchmark surveys – one done for LIBOR and the other for Euribor – so different when each asks about U.S, dollar borrowing? The same difference occurs in the surveys for euro borrowing. These rates are calculated on the basis of the banks’ answers to roughly the same question. For LIBOR, a bank is asked at what rate it thinks it can borrow, while for Euribor, a bank is asked at what rate it thinks other banks are able to borrow.
Promoting Market Integrity and Financial Stability
Whether we consider the broad structural shift away from unsecured, interbank lending; the recent enforcement actions; or questions that arise from current market data, I believe that LIBOR, Euribor and other similar interest rate benchmarks are unsustainable in the long run.
These benchmarks – referencing markets with insufficient transactions, particularly in longer tenors – undermine market integrity and threaten financial stability.
For capital and risk to be efficiently allocated within the economy, interest rate benchmarks should reflect actual price discovery anchored in observable transactions.
Without transactions in the underlying market, the situation is similar to trying to buy a house, when your estate agent can’t give you comparable transaction prices in the neighborhood – because no houses were sold in the neighborhood in years.
As IOSCO notes, a benchmark should derive its value from the competitive forces of buyers and sellers meeting in an underlying cash market.
Derivatives derive their value from an underlying cash market. Market integrity dictates that whether that underlying cash market is oil, corn or the rate at which banks are borrowing, it must be based on something that is real. It should be anchored in observable transactions.
Further, these rates were readily and pervasively rigged in the past, and incentives for and ability to rig it in the future remain.
When market integrity is compromised, this also undermines the public’s confidence in the financial system.
The financial system’s reliance on interest rate benchmarks, such as LIBOR and Euribor, leaves the system in a fragile state.
Further, continuing to support LIBOR and Euribor in the name of stability may have the opposite effect. Using benchmarks that threaten market integrity may create more instability in the long run.
Given the structural changes in the interbank market, a number of banks have withdrawn from Euribor and some other interest rate benchmarks. Though IOSCO’s task force recommends that users of benchmarks have robust fallback provisions in contracts, many contracts do not currently have such fallback provisions. Thus, there is a risk to financial stability absent a planned, smooth and orderly transition.
I believe to promote market integrity as well as financial stability, we must move forward in a coordinated global effort to identify alternative interest rate benchmarks anchored in observable transactions and plan a smooth and orderly transition from benchmarks referencing unsecured, interbank markets.
There is no doubt there will be challenges to transitioning from these rates.
But the market does have experience with transitioning from benchmarks that have become obsolete in the past. When the euro was created, a number of interest rate benchmarks were discontinued. How many of you remember PIBOR, RIBOR, MIBOR and FIBOR? Transitions also have occurred for energy and shipping rate benchmarks.
Further Canadian dollar LIBOR and Australian dollar LIBOR will be discontinued this year, leading to necessary transitions in those markets.
The basic components of past transitions include: first, identifying a new and reliable benchmark, one that is anchored in transactions.
Market participants and regulators are currently considering alternative interest rate benchmarks anchored in observable transactions. For instance, the Bank of International Settlements’ (BIS) Economic Consultative Committee’s March report lists possible alternatives anchored in observable transactions: the overnight swaps rate (OIS) and short-term collateralized financing rates such as general collateral repo rates (GC repo).
Second, the new and existing benchmarks run in parallel for a period of time allowing market participants to see and compare the price or rate of the alternative benchmark versus the soon-to-be-discontinued benchmark. This period of the two benchmarks running in parallel has been used to facilitate a smooth transition.
Third, a date is announced well in advance of when the old or obsolete benchmark will be discontinued.
While ongoing international efforts targeting benchmarks have focused on governance principles, these efforts cannot address the central vulnerability of LIBOR, Euribor and similar interest rate benchmarks: the lack of transactions in the underlying market.
Given the known issues with these benchmarks, their scale and effect on market integrity, it is critical that international regulators work with market participants to promptly identify alternative interest rate benchmarks anchored in observable transactions with appropriate governance, as well as determine how to best smoothly transition to such alternatives.
Just as Canadian dollar LIBOR and Australian dollar LIBOR are being discontinued due to the lack of an underlying interbank market, U.S. dollar, sterling, yen and euro LIBOR face similar underlying market challenges. The scope, as we all know, is bigger.
But it’s best that we not fall prey to accepting that LIBOR or any benchmark is “too big to replace.”
Just imagine if the quality and integrity of the drinking water across the globe had been compromised. Then, the official sector and the utilities react by addressing the problem in Australia and Canada, but not in England or the United States. They say that there are too many people relying on the current drinking water in those countries.
If this were to occur, how could the public be confident in continuing to drink this water?
I believe market participants and regulators around the globe do have the ability and the ingenuity to tackle the challenges of benchmark interest rates, even in the face of their scale, to restore integrity and promote financial stability.
Last Updated: April 22, 2013