Guest post by Richard Berner, Executive-in-Residence (Center for Global Economy and Business) and Adjunct Professor, NYU Stern School of Business
In response to the fragility of LIBOR and other interest-rate benchmarks, regulators globally are working with industry to identify sturdy alternatives. Despite significant progress, concerns persist that the transition to these new reference rates will be disruptive.
While these concerns are legitimate (see Eclipsing LIBOR), both U.S. and global authorities and market participants have begun to address them in ways that should go a long way to managing the risks. In this post, we review why LIBOR’s persistent fragility makes reform critical, and examine progress on some of the ongoing reforms.
LIBOR (now known as ICE LIBOR, after its new administrator) is an interest-rate benchmark, or reference rate, that plays a central role in global financial markets and the economy. U.S. dollar-denominated LIBOR is the benchmark for some $200 trillion (gross notional value) in derivatives and more than $10 trillion in domestic loans to consumers and businesses (see here).
Both authorities and market participants have long understood that LIBOR is fragile for two reasons. First, the original intent was that LIBOR rates would reflect the (unsecured) benchmark cost of bank funding. But, published LIBOR rates are based on a survey of banks that voluntarily participate, rather than on actual transactions. Moreover, following the financial crisis, the already-tenuous link between the survey and a dwindling volume of transactions became more fragile, as unsecured lending virtually disappeared (see Financial Stability Oversight Council and Financial Stability Board). And, partly because of fines for manipulation, banks are wary of participating in the survey. Thus, even the recent improvements in the governance and processes for publishing ICE-LIBOR are insufficient to assure its viability. Given LIBOR’s continued widespread use, shocks to it can threaten both individual institutions and the stability of the financial system.
Fortunately, LIBOR’s fragility has promoted a variety of constructive responses. To help the transition, authorities globally have announced that they and participating banks will support LIBOR until 2021. While the publication of LIBOR may continue past this date, there is no guarantee. So it is important to identify alternatives (see Bailey and Powell and Giancarlo). In the United States, supported by the Fed and other U.S. authorities, the private-sector Alternative Reference Rates Committee (ARRC) has worked to identify viable alternatives to U.S. dollar LIBOR. Similar initiatives are underway in the euro area, the United Kingdom, Switzerland, and Japan.
Illustrating LIBOR’s fragility. Before we get to the solutions, it is worth reemphasizing that LIBOR is living on borrowed time. Two recent episodes—neither having to do with bank funding risk—underscore LIBOR’s fragility and reduced reliability as a benchmark. Both reflect the post-crisis shift in the supply of unsecured short-term funding in money markets. Specifically, a downshift in the willingness of banks to provide such funding, together with a reduction in the sensitivity of supply to price, has rendered LIBOR highly vulnerable to both temporary, one-time shocks to funding demand and to changes in the supply-demand balance for other money-market instruments. And, while the impact on spreads of each episode may prove to be transitory, each shock undermines confidence in LIBOR’s reliability a bit more.
Spread between U.S. dollar LIBOR and the overnight index swap (basis points), 28 Jun 2010-28 Mar 2018
Source: Bloomberg.
In the first episode, starting in late 2015, the spread of LIBOR over the overnight index swap (OIS) rate—a proxy for the expected risk-free rate—widened by 30 basis points (see the chart above). The trigger was first the anticipation, and then the October implementation, of money market mutual fund (MMMF) reform. The new regulations required institutional prime money funds—those that hold mainly corporate debt instruments—to move to a floating-rate net asset value (NAV). At the same time, the rules allowed government-only funds to retain a fixed NAV—that is, redemption at face value. As a result, investors shifted $1 trillion from prime funds that invested in risky assets to government-only funds that invested in Treasury or U.S. government agency instruments. Shifts out of prime money funds forced issuers of commercial paper and certificates of deposit temporarily to pay more for funding. Once the redemptions ended, the LIBOR-OIS spread reverted to its previous levels.
In the second, more recent episode, three factors led the LIBOR-OIS spread to widen by nearly 50 basis points since the beginning of this year:
- The tax law changes enacted in December 2017 triggered repatriation of cash that U.S. corporations had parked in short-term money-market instruments overseas, raising the cost of funding for the issuers of those instruments.
- The combination of lower tax revenue and increased spending (enacted in early February 2018) boosted Treasury borrowing, raising the volume of Treasury bills outstanding by $332 billion between February 8 and March 29.
- The interaction of these developments with the ongoing influence of money fund reform on investor demand has likely amplified the impact of these and other shocks on money-market rates.
Monthly data on MMMF assets, reported in the U.S. Money Market Fund Monitor from the Office of Financial Research, illustrate these shifts:
- Holdings of Treasuries by government funds are growing rapidly—up by 30% to more than $800 billion from June 2017 to end-February 2018.
- Conversely, MMMF investments in Treasury repos have declined from an all-time high of $724 billion in June 2017 to $567 billion at the end of February.
- MMMF utilization of the Fed’s overnight reverse repo (ON RRP) facility has dropped to the lowest level since its introduction in September 2013. (That’s largely because ON RRP now pays less than a Treasury bill―1.50% vs. 1.60%).
Market participants suggest that, as was the case following the MMMF regulatory reforms, the LIBOR-OIS spread should narrow as the Treasury’s bill issuance surge abates (a process that now seems underway) and markets adjust to new patterns of corporates investing repatriated cash. But episodes like these doubtless create expectations that the next funding surprise will again lead to wider spreads, keeping concerns about LIBOR’s reliability at the simmer.
Identifying and implementing alternatives. Relying on the International Organization of Securities Commissions’ (IOSCO) Principles for Financial Benchmarks, the ARRC developed criteria against which to evaluate the alternatives to LIBOR as reliable measures of market activity. Among others, these include quality based on market liquidity, transaction volume and resilience to shocks; processes and governance to ensure compliance with the principles and benchmark integrity; and ease of implementation. In short, it is critical that alternatives “are anchored in active underlying markets, are based on a comprehensive set of transactions that incorporate controls to mitigate risk of errors and manipulation, and are capable of being published every day, even in adverse circumstances” (Lorie Logan, The Role of the New York Fed as Administrator and Producer of Reference Rates).
After considering several alternatives, the ARRC settled on the secured overnight financing rate (SOFR)― the rate based on the deepest and strongest underlying market, with the largest daily volume of transactions (shown in the following chart). As its name implies, SOFR blends secured transactions from the various segments of the U.S. repurchase agreement market (triparty, General Collateral Financing (GCF) and bilateral) into a composite. Last week, in cooperation with the Office of Financial Research, the New York Fed began daily publication of SOFR.
Daily Volumes in U.S. Money Markets (Billions of U.S. Dollars), 1H 2017 and August-September 2017
Notes: Average volumes for the first half of 2017, except for 3-month T-bills, which are preliminary estimates from FINRA Trade Reporting and Compliance Engine (TRACE) data over August and September 2017. Source: Second Report of the Alternative Reference Rates Committee, Figure 3, pg. 11.
The launch of SOFR is only the first step in providing alternatives to LIBOR. The transition to an alternative rate will require acceptance by the end users of derivatives who “cannot be expected to choose or transition to trading a benchmark that does not have at least a threshold level of liquidity” (ARRC Interim report). To this end, last October, the Committee adopted a “paced” transition plan aimed at providing this threshold level of liquidity.
The immediate challenge is to create sufficient liquidity for derivatives contracts based on the new reference rate, SOFR. ARRC foresees building the infrastructure for futures and/or overnight index swap (OIS) trading in SOFR and gradually widening the coverage to other derivatives referencing SOFR. The plan further envisions steps to have central counterparties (CCPs) clear swaps, enabling users to make the transition from LIBOR and the Effective Federal Funds Rate (EFFR) to SOFR at progressively longer maturities. Finally, the plan contemplates the creation of a term reference rate (with a maturity to be determined) based on these derivatives. A key challenge is whether there will be enough volume in term derivatives referencing SOFR to form a robust basis for rates at maturities greater than overnight that may serve as a reference in certain cash products.
The good news is that these changes are ahead of schedule. Originally projected for the latter half of 2018, the Chicago Mercantile Exchange (CME) announced plans to launch one- and three-month futures contracts in SOFR on May 7. CME and LCH (the London CCP) will begin clearing SOFR-OIS, SOFR-LIBOR, and SOFR-EFFR basis swaps in the third quarter of this year. These should all add liquidity to SOFR and facilitate the transition to it from LIBOR and from the EFFR. Nevertheless, there is much more to do―in collaboration with end users―to move a significant portion of the derivatives markets away from LIBOR and to the new reference rate.
In addition to executing the ARRC’s plan, there are two further challenges to making a smooth transition to SOFR. First, relevant market groups—like the International Swap and Derivatives Association (ISDA)—must develop better contact language for both derivatives and cash products to allow for an economically appropriate successor rate if LIBOR stops. There is progress here, too. Most important, ISDA is drafting fallback arrangements for new LIBOR derivatives contracts that will specify a robust mechanism for determining payment obligations if LIBOR ceases to exist, and ISDA will offer a protocol to amend legacy contracts to include this new language. Market participants can and should play a significant role in helping to develop such protocols.
Second, unlike LIBOR, as a secured rate SOFR does not reflect bank funding (counterparty) risk. Some market participants want a broad measure that reflects such risk. In principle, the Financial Stability Board’s (FSB) Market Participants Group (MPG) proposal for an exclusively transactions-based LIBOR+ sought to provide such a rate. In practice, however, there probably is not enough demand for, and are not enough transactions in, unsecured funding to support LIBOR+. In fact, median daily volume for all unsecured transactions (including LIBOR and other unsecured instruments) for the 30 global systemically important (G-SIB) banks is only slightly over $1 billion at the one-month maturity, and the LIBOR panel is half that size (see ARRC Second Report, Figure 1, Page 3 using data from the FR2420 Report of Selected Money Market Rates and DTCC).
Overall, the reform momentum makes the outlook positive: while the LIBOR reform process still has a long way to go, both regulators and market participants are responding to the urgency of completing the transition before they end support for LIBOR. Sustaining the rapid progress now underway will be critical.
Acknowledgements: The author thanks Viktoria Baklanova, David Bowman and Matthew McCormick for helpful discussions and comments.