Streetwise Professor

May 8, 2018

Libor Was a Crappy Wrench. Here–Use This Beautiful New Hammer Instead!

Filed under: Derivatives,Economics,Exchanges,Financial crisis,Regulation — The Professor @ 8:02 pm

When discussing the 1864 election, Lincoln mused that it was unwise to swap horses in midstream.  (Lincoln used a variant of this phrase many times during the campaign.) The New York Fed and the Board of Governors are proposing to do that nonetheless when it comes to interest rates.  They want to transition from reliance on Libor to a new Secured Overnight Financing Rate (SOFR, because you can never have enough acronyms), despite the fact that there are trillions of dollars of notional in outstanding derivatives and more trillions in loans with payments tied to Libor.

There are at least two issues here.  The first is if Libor fades away, dies, or is murdered, what is to be done with the outstanding contracts that it is written into? Renegotiations of contracts (even if possible) would be intense, costly, and protracted, because any adjustment to contracts to replace Libor could result in the transfer of tens of billions of dollars among the parties to these contracts.  This is particularly like because of the stark differences between Libor and SOFR.  How would you value the difference between a stream of cash flows based on a flawed mechanism intended to reflect term rates on unsecured borrowings with a stream of cash flows based on overnight secured borrowings?  Apples to oranges doesn’t come close to describing the difference.

Seriously: how would you determine the value so that you could adjust contracts?  A conventional answer is to hold some sort of auction (such as that used to determine CDS payoffs in a default), and then settle all outstanding contracts based on the clearing price in the auction (again like a CDS auction).  But I can’t see how that would work here.

Let’s say you have a contract entitling you to receive a set of payoffs tied to Libor.  You participate in an auction where you bid an amount that you would be willing to pay/receive to give up that set of payoffs for a set of SOFR payoffs.  What would you bid?  Well, in a conventional auction your bid would be based on the value of holding onto the item you would give up (here, the Libor payments).  But if Libor is going to go away, how would you determine that opportunity cost?

Not to mention that there is an immense variety of payoff formulae based on Libor, meaning that there would have to be an immense variety of (impractical) auctions.

So it will come down to bruising negotiations, which given the amounts at stake, would consume large amounts of real resources.

The second issue is whether the SOFR rate will perform the same function as well as Libor did.  Market participants always had the choice to use some other rate to determine floating rates in swaps–T-bill rates, O/N repo rates, what have you.  They settled on Libor pretty quickly because Libor hedged the risks that swap users faced better than the alternatives.  A creditworthy bank that borrowed unsecured for 1, 3, 6, or 12 month terms could hedge its funding costs pretty well by using a Libor-based swap: a swap based on some alternative (like an O/N secured rate) would have been a dirtier hedge.  Similarly, another way that banks hedged interest rate risk was to lend at rates tied to their funding cost–which varied closely with Libor.  Well, the borrowers (e.g., corporates) could swap those floating rate loans into fixed by using Libor-based swaps.

That is, Libor-based swaps and other derivatives came to dominate because they were better hedges for interest rate risks faced by banks and corporates than alternatives would have been.  There was an element of reflexivity here too: the availability of Libor-based hedging instruments made it desirable to enter into borrowing and lending transactions based on Libor, because you could hedge them. This positive feedback mechanism created the vexing situation faced today, where there are immense sums of contracts that embed Libor in one way or another.

SOFR will not have this desirable feature–unless the Fed wants to drive banks to do all their funding secured overnight! That is, there will be a mismatch between the new rate that is intended replace Libor as a benchmark in derivatives and loan transactions, and the risks that that market participants want to hedge.

In essence, the Fed identified the problem with Libor–its vulnerability to manipulation because it was not based on transactions–and says that it has fixed it by creating a benchmark based on a lot of transactions.  The problem is that the benchmark that is “better” in some respects (less vulnerable to a certain kind of manipulation) is worse in others (matching the risk that market participants want to hedge).  In a near obsessive quest to fix one flaw, the Fed totally overlooked the purpose of the thing that they were trying to fix, and have created something of dubious utility because it does a poorer job of achieving that purpose.  In focusing on the details of the construction of the benchmark, they’ve lost sight of the big picture: what the benchmark is supposed to be used for.

It’s like the Fed has said: “Libor was one crappy wrench, so we’ve gone out and created this beautiful hammer. Use that instead!”

Or, to reprise an old standby, the Fed is like the drunk looking for his car keys under the lamppost, not because he lost them there, but because the light is better.  There is more light (transactions) in the O/N secured market, but that’s not where the market’s hedging keys are.

This is an object lesson in how governments and other large bureaucracies go astray.  The details of a particular problem receive outsized attention, and all efforts are focused on fixing that problem without considering the larger context, and the potential unintended consequences of the “fix.” Government is especially vulnerable to this given the tendency to focus on scandal and controversy and the inevitable narrative simplification and decontextualization that scandal creates.

The current ‘bor administrator–ICE–is striving to keep it alive.  These efforts deserve support.  Secured overnight rate-based benchmarks are ill-suited to serve as the basis for interest rate derivatives that are used to hedge the transactions that Libor-based derivatives do.

Print Friendly, PDF & Email


  1. According to Craig Phillips ( they are motivated by the ongoing marginalization of LIBOR, and their goal is to avoid serious disruptions should LIBOR ceases to be a reliable reflection of the market sentiment. They explicitly say they are looking for a market-based transition, and specifically refer to the fact that “[a]n estimated 82% of LIBOR exposures will mature or roll off by the end of 2021”. Is it all wrong?

    Comment by Boris Lvin — May 9, 2018 @ 12:49 am

  2. As long as LIBOR continues to be published, contracts written against it will be settlable, no?

    The market seems quite happy with LIBOR – open interest records in March and volume up 37% yoy:

    Comment by Green as Grass — May 9, 2018 @ 2:35 am

  3. If Libor continues, and SOFR joins it, can’t it be left to the market to decide which to use?

    Comment by dearieme — May 9, 2018 @ 5:38 am

  4. I loved the commentary by the Streetwise Professor, but I still don’t think it quite gets to the main point. LIBOR is what it is in the derivatives world not because banks hedge their borrowing costs with LIBOR. It’s because the floating rate corporate loan is the mostly widely executed financial transaction in the world. The banking industry has pegged most of its corporate loans to LIBOR. It then offers swaps to corporates to convert those loans to fixed rate loans. Having bought back the floating rate risk, bank swap dealers then hedge using the CME Eurodollar futures contract, which has barely suffered a dent from the LIBOR crisis. No government agency or trade association can simply tell an industry to change its practices unless a law is being broken. Most of the discussion on the LIBOR crisis focuses on the derivatives and is missing the real story. The derivatives exist for corporates to hedge loans. If you want to change how things are done, you’ll have to change the rate at which banks peg their loans from LIBOR to something else. Instead, the regulators do nothing but talk about changing the way derivatives are pegged. I am fairly confident LIBOR will survive. ICE and the CME have too much vested in LIBOR, and as of now, it is working.

    Comment by Don Chance — May 9, 2018 @ 8:47 am

  5. @Don–Welcome to SWP. Good to hear from you.

    I did mention LIBOR-linked loans, and the swapping from floating to fixed. It’s sort of a chicken-and-egg thing, which is what I meant by reflexivity. The symbiotic relationship between the funding, corporate loan, and derivatives markets all interacted to make the IBORs the preeminent benchmarks.

    Indeed, all of the problems are shared by these instruments. An overnight secured rate is a lousy index for corporate loans.

    Re survival–yes, ICE and CME have a big stake in that. The question is how to keep banks participating. They face big potential legal risks. In some markets (e.g., natural gas reporting in the US) these risks have discouraged participation. Rate/price reporting is a public good in the technical sense of the term, and it is a challenge to incentivize market participants to contribute to that public good, especially given the substantial private costs that they incur to do so.

    The ProfessorComment by The Professor — May 9, 2018 @ 11:23 am

  6. @Boris–(1) The roll-off presumes that no more LIBOR exposures are created. As @Green as Grass notes, IBOR-based open interest in listed derivatives is increasing, and I have little doubt this is the case OTC as well. (2) The roll-off issue is only one of the two I mentioned. The other is whether the supposed replacement is a good substitute, which I do not believe it is.

    The “market-based transition” begs important questions. Notably, there is a big coordination issue here. Furthermore, as mentioned in response to @Don Chance’s comment, the public good nature of LIBOR raises the potential for market failure.

    Hence, I find Phillips’ remarks to be superficial.

    The ProfessorComment by The Professor — May 9, 2018 @ 11:29 am

  7. Given the rising CME and ICE open interest *and* volumes you have to wonder who exactly still thinks there’s a problem with ‘IBORs because it certainly ain’t the trade.

    CFTC commitments of traders data show ever increasing user numbers too on the CME as I recollect. It’s not like a whole crowd quit and those left are somehow trading more.

    These contracts all look like they’re in rude health on the usual measures.

    Comment by Green as Grass — May 10, 2018 @ 2:45 am

RSS feed for comments on this post. TrackBack URI

Leave a comment

Powered by WordPress