LIBOR Benchmark Reform Around the World Transcript

Julia: Hello, I'm Julia Verlaine from J.P. Morgan, and welcome to the second episode of our LIBOR transition series. In our first episode, we discussed benchmark reform in the US, with SOFR having been selected as the alternative reference rate. But this is a global effort with 400 trillion dollars in assets across currencies, including the British pound, euro, and the Japanese yen to name a few, that are likely to migrate to new reference rates. Let's talk about benchmark reform across the globe. To do so, I am joined by Charles Bristow, Global Head of Rates, Fixed Income Financing, Credit Portfolio Trading, and Markets Resource Management at J.P. Morgan, Chris Palmer, who leads the firm-wide LIBOR transition program, and Cyprien Decoux, from rates structuring in EMEA. Basically, I'd like to set the stage with one question, which is, when is LIBOR actually going away, and what is the time frame for this across the globe? Let's just jump straight into that. Time of death, when is it really?

Charles: Well, that is the multi-trillion dollar question I guess, Julia, isn't it? I mean, everyone would like to know that. The reality is, there's a lot of pressure for there to have been a very material if not complete amount of migration by 2021. The banks that submit to the LIBOR panels have committed to continue submitting until that date, but beyond that, there's this huge degree of uncertainty. One of the issues plaguing LIBOR for many years has always this sort of uncertainty over its stability, but clearly as you approach that deadline, the uncertainty increases. You know, that's a big date. Whether or not it's reasonable for all transactions to have migrated by that point in time is hotly contested, and hence, at some point, I guess we'll speak about fall-back mechanisms, but you have to start to think about what happens beyond that date if the panel isn't any longer supported.

Cyprien: And if I can add to that, I think a lot of people are questioning whether it will actually happen. In my view, this reform will happen. People tend to think that this reform is sometimes driven by suspected wrongdoing, but the real fundamental issue behind this reform is actually the fact that the interbank market changed radically since the financial crisis. The number of transactions went down dramatically since the crisis. There's not a lot of active rates that are the background of this LIBOR rate. This is not going away. This is a real issue. We need to address that. That's why the SOFR was starting in 2014, and we are in the process of addressing this issue.

Julia: I guess the real question is, are reforms happening at the same time in different regions, and can we get into that? Maybe we should take it back and talk about what is happening in the UK with the selected reference rate and then Europe. But if reforms are happening at different times, then is there also risk of market fragmentation? That's what I'd like to ultimately get to with this timeline question.

Charles: First of all, I want to strongly agree with Cyprien in that the resolve to get this delivered is absolutely huge. If there's any doubts about the question of whether this will occur and whether at some point it will lose momentum, I think people have to get past that. This is not an if, it's a when problem. I think when you're involved in the working groups as I have been and as a number of people here at J.P. Morgan have been, you really see that. Whether that's from the other industry participants, the end users, politicians, regulators. Around the table in those working groups is a huge determination to make it happen. The reality is, there's a huge amount of inventory, and that migration process isn't super easy, and the questions aren't simple. But the resolve to make it happen is unwavering. If I talk about one of the working groups that I've worked on, which is the UK risk free working group, which was looking at the migration of sterling LIBOR, we were obviously focused on determining what's the appropriate alternative benchmark to succeed sterling LIBOR, or at least to base a sterling interest rate and other derivative market. As is well known, that group chose SONIA, or rather reform SONIA. Actually, that was an interesting one, in that in other parts of the world, people were choosing brand new indices. After looking at all of the options, what we found in the UK was that actually, the old SONIA wasn't a fundamentally flawed index. There was actually a reasonable number of transactions and the market was very stable. Given that SONIA was already an accepted index, reforming it to make it stronger and then just having it as an ongoing index that had more depth was viewed as a way to guarantee the migration, have more certainty and more confidence. That was really the driver behind picking it.

Julia: And Cyp, what about in Europe? I know they're pursuing a hybrid approach, EURIBOR is not going away. Do you want to fill us in there?

Cyprien: Yeah, yeah, yeah, sure. I think in continental Europe, the starting point was that the European commission did put a specific EU benchmark regulation in place. This regulation created some set of criteria in order to select whether the current benchmark, which are EURIBOR and EONIA, were compliant or not. What they immediately found is that these two benchmarks were not compliant, and work need to be done in order to reform or change these indices. Let's take these two indices separately, because they led to different outcomes. Starting, at the moment with EURIBOR, EMMI, which is an administrator of EURIBOR, decided to reform this index. What they did effectively is to create a waterfall methodology. This waterfall methodology is supposed to use much more transaction based indices, rather than judgment and panel based indices. On the other side, you have EONIA. Very early in the process, EMMI, which is also administrating EONIA, decided that EONIA would not be benchmark compliant, and they didn't even try to reform it. The ECB at the same time created a working group they selected a new index which is called ESTER, which stands for Euro Short-Term Rate. In terms of timing, because that was part of your question, ESTER will be launched on the 2nd of October. One very important thing about this index is that it will be linked to EONIA for the foreseeable future. This means that once ESTER is launched, EONIA will track ESTER with a fixed spread, and the liquidity will completely move from EONIA to ESTER relatively quickly. I mean, this is something Charlie can probably comment more on. But it should really help the liquidity of the ESTER market.

Charles: Yeah. It's a type of migration mechanism, because effectively by putting them in lockstep, it's analogous with pegging a currency. You're somewhat turning them into the same thing from a moment in time.

Julia: And in terms of liquidity, I mean, on how that's evolving in these two different regions, is the UK more advanced?

Charles: The UK is definitely more advanced. In credit to Europe, they probably come from behind in their preparedness. I think Europe went through a longer discussion phase, and then has in quite quick succession been making decisions and moving into an implementation phase. In the UK, those were going side-by-side for a long period of time. The UK has the advantage in that SONIA is a preexisting index so there's already stock in it, there's already daily need. People are trading it. There is already liquidity. I mean, I think if you look at for instance issuance markets, there's not a huge market in issuance of floating rate bonds, but every floating rate bond so far this year that's been issued in sterling has been referencing SONIA. So clearly you can see goodwill in that market towards migrating towards the use of SONIA wholesale. We have a very stable large market in SONIA based swaps, particularly at the short end of the curve, and there is liquidity at the long end of the curve, but for now, because the inventory remains on LIBOR index, the maintenance of that industry still stays there. The real challenge for the industry and for the stakeholders in this is to get to the next stage where you begin to transition the stock of inventory into a SONIA base, so that the maintenance of it starts to be SONIA based.

Julia: Where are both regions at in terms of the plan and in terms of developing a term rate? Is Europe on the same page as the UK? Or do we have different solutions?

Charles: Ok so I'm going to answer that question in a second. First of all, I just want to say something on term rates, because actually, I think it's important to define what we mean. LIBOR in reality was a family of term rates, i.e. 3 months LIBOR, 6 months LIBOR, 12 month LIBOR represented different things. They were credit for different terms. As such, 12 month LIBOR wasn't the expectation of four periods of three months, it was actually a different thing. When we talk about term rates for overnight indices, they're not term rates in the same way as LIBOR was. They're not actually measuring term of credit. All they're doing is saying at the beginning of a period, this is our expectation of the average level of the overnight over the coming period. It's a technical point, but it really matters, because the point is, whether you observe it monthly, quarterly, semi-annually, annually, ultimately, you're still economically trading the same thing. It's just an expectation of the period to come. And as such, what you're helping is more with an administrative burden. You're not actually fundamentally changing the nature of the product. I just wanted to get that off my chest. Term rate means two different things to two different people.

Cyprien: The big question you're asking is, do you actually need these term rates and who will need these term rates? I think the answer is a little bit complicated, because it depends on who you're asking. They did a survey for example in the UK, the working group did, and the consultation was relatively clear that people who are cash market participants, so people who are using loans, bonds, floating rate notes, et cetera, need the term rate.

Charles: You know, there are a number of people that need them. If you're investing in a note, you want coupon certainty. If you're a corporate treasury, you're actually at the beginning of accrual periods, really need to get some sense of your cost and expensive profiles over that coming period. There are a number of real reasons for it. The other point is that, if you want to find a way to comfortably migrate legacy inventory, which was based on LIBOR fixing, the fix at the beginning of the period. Obviously that transition path from an old LIBOR fixing on to the new say overnight based rates, is going to be a lot easier if you can observe some sort of fixing that lets you know what the accrual rate is at the beginning of the period. That's a big problem that needs solving. I think there is in the working groups now, a lot of focus on working out what the right thing to do is.

Cyprien: I think also if I can just add one point, derivative are also practically much easier to change, because most of them are governed by ISDA. ISDA is working, you know, with the market participant to find solution. When this is done, it's done in one go, you know, for a lot of trades happening at the same time. If you look at the cash market or the retail market, it's a completely different game to actually restructure, you've got a small number of trade, or a small number of notional. It takes 20, 50 times effort because there's not the same legal framework. It's using different systems, it's much more directional. Sometimes the volume can be misleading and the vast majority of trades using LIBOR are derivatives, but they don't always represent the biggest issue to overcome. It's very important, because what the market is trying to do is to help on the operational side, on the predictability of cash flows, but we don't want to go back to the initial issue that LIBOR was. We are not trying to re-put some credit element into the index.

Charles: Exactly. Cyprien's nailed a really important point, which is that, as the wholesale market changed, the reality is that banks stopped needing term credit from one another. The nature of how banks were trading, were borrowing from each other changed. That was really what underpinned the sudden reduction in the number of transactions underpinning LIBOR. If we talk about the regions now to actually answer your question, the different regions are approaching it slightly differently. Now, in Europe, there is a willingness to, for some period of time, reform EURIBOR, which is a credit term index, so analogous to LIBOR, to make it usable for a good period of time and to make the transition mechanism smoother, whilst at the same time bringing in ESTER, and then also creating a term fixing based on ESTER. In Europe, you can envisage there being two parallel markets, obviously with the bulk of the transactions and the interbank transactions, so the exchange of interest rate sensitivity between banks taking place on ESTER, but holding open the ability for people to sustain EURIBOR trades, and sensitivity to EURIBOR where that's the right risk management instrument. I'd say in the UK, there's a reluctance to go fully down the path of these term fixings, but a recognition that it may in the short term aid transition, and that, it's a conversation that's ongoing now. Obviously the point about these term fixings of overnight indices is they do require a market that's trading and can be observed to determine the fair value. Cyprien, did you want to talk about the US? Or Chris.

Chris: With respect to term rates, the Fed has launched indicative term rates now. There's an expectation that those term rates will become available to trade not until at least the second half of 2021. They're posting them now to show how the methodology works to give people an indication of what they may look like. But as far as them being represented in any contracts, we're still at least 18 months away from that happening. They are based off futures, so that's one of the other discussion topics, is around the different methodologies to create that term structure around the globe. I'll spend a bit of time now talking about Japan. They have multiple rates at the moment. Their overnight rate is TONAR. They are looking to launch futures next year, and potentially develop a term rate off the back of that in the future. But they are also going to continue to keep TIBOR. That hasn't been confirmed yet, but there's a consultation that is coming out to talk through that. If you compare that to Switzerland, Switzerland's overnight rate is SARON. They've moved that. That market has already openly said that there will be no forward looking term structure created off SARON, because they don't believe they have enough transactions in the market to create an IOSCO compliant forward looking term rate. It does show that there are some differences around the globe in the approach to term rates.

Julia: So, I mean, you've just thrown around a lot of acronyms. So what I'm wondering is, if I am a client and I am going to have to adapt to these changes and understand as well, if I'm trading in multiple regions, how do I wrap my head around all of this, and where are there going to be divergences?

Cyprien: It's a very good point, because we have a lot of clients that are as you said multi-currencies. They are trading in different markets. I think what you get from the conversation earlier is that the reform is happening everywhere in different form. There are still a few countries and those that are a little bit behind, but they are now waking up to the issue, and they are starting to launch their own working group and reform process. All these region are facing the same issues, whether it's, should we go for unsecured versus secured? Should we push for term rate or purely base it on overnight? Everybody's coming up with a different conclusion, but the question is always the same. The difficulty for some clients is to say the timing is different, and the nature of the index is different. For example, you will have some indices that will be secured, and also are unsecured. We hear sometimes, clients are saying, "It's an issue for me, because there are two different instruments. If I for example trade a cross-currency swap, and one leg is secured and the other leg is unsecured, it's an issue for me." I think their reasoning is that in a crisis, a secured index will behave differently from an unsecured index. The first one is that all these indices are overnight. An unsecured overnight rate, yes, those have a little bit of credit, but it's a very small amount of credit, because you only take a few on the overnight basis. The second thing is that the differences in methodology already happened before. EURIBOR and LIBOR were fundamentally differently defined at a different number of panel banks. There were different instruments. People were trading euro dollar cross-currency swap without even questioning it.

Julia: Do we want to dive in to the secured versus unsecured?

Charles: Yeah. Cyprien's right. These are issues that people need to understand. They won't manifest themselves very often. The point is though, maybe I'm thinking with a trader head here, the only thing that really matters is how these things behave in a crisis. You've got to be prepared for the rainy day, and a crisis isn't necessarily a credit crisis. Cyprien's absolutely right. The one day credit of secured versus unsecured, even in a pretty big crisis, is generally not worth a huge amount of basis points. But there are other factors that can cause deviation between secured and unsecured. To go back to your question, you have to know whether the index you're trading references the unsecured world or the secured world. Secured means it looks at transactions to repo trades, where banks are borrowing or lending against securities, usually bonds, as collateral. Unsecured obviously as it suggests where they're just lending to each other, they do mean a huge amount. In general, you would say, yes, secured always is better credit and therefore should always trade under unsecured. But actually, if you look at the history, there are plenty of occasions where secured rates have traded over unsecured, because actually, some of the things driving it are dynamics related to banks' balance sheets, or capacity to deploy cash, that can actually have very unexpected effects on where they would actually find it rational to transact in those underlying products. It's really important that people understand, if they're going to transact in any financial instrument, that they understand actually what underpins it. You're going to have a couple of different families, secured and unsecured, and I think it is an unfortunate consequence of the fact that each of these indices is being reformed within its own geography, that there will be some international products that will ultimately contain one flavor and the other flavor.

Cyprien: I mean, the biggest example of secured rate is SOFR. SOFR is an average of repo rates, and what you can see is that its trade rates are really technically, so I don't think we need to go into the details, but you need to understand the repo market to understand how SOFR works. If you want to understand the repo market, you need to understand how banks allocate balance sheet and what the banking regulations is, et cetera. A lot of people that used to trade derivatives in and out for hedging purposes don't actually want to understand that. The market needs to adapt, people need to understand it. All these issues needs to be inside the brain of the traders or inside the brain of the people who are doing hedging.

Julia: We're touching upon something important here, which is, what I'd like to know is the transition program that J.P. Morgan is doing. Chris, you're running that, and this sounds like we're dealing with lots of different currencies, reforms, systems. Educating clients seems to be the main thing that Cyprien is touching upon. Could you talk about the J.P. Morgan transition program and what we're doing?

Chris: Yeah, sure. You've hit the nail on the head in the sense that you have to educate the market in understanding all these intricacies of how everything's connected. The term I use is, it's not a race you can win. We can't put ourselves in the back and say, "We're all set up. We're organized. We've done our transition," Because the reality is, you can only move as fast as the industry is running. From that perspective, we spend a lot of time talking to clients, both in conferences, one-to-one meetings with clients. They really want to know how we've structured our program. We've broke it into five work streams. The first one is around implementation of new rates. Can you actually deal in the new rates? Charlie mentioned before around corporate treasures for example, wanting to make sure that they had forward cash flows. If we didn't have forward term rates, there's a lot of system changes that are required to be able to book a loan for example that compounds in arrears. There are infrastructure providers out in the market who need to change their systems. The second one's around fallback language and executed documents. So what's in your documents? That's a huge task, and actually it's seen as the biggest risk in the industry at the moment, is around trying to actually what language relating to fallbacks, be they securities, be they derivatives for the ISDA consultation, or on the cash side, what's actually written there, and then how can you amend that language. The third one's around measuring your exposure to LIBOR. Understanding what products you have in your inventory today that have LIBOR on them, and how do you regularly update that inventory to report that to your senior managers. What we've seen is the DCO letter last year was a really good indication that you had to be prepared to present that to a regulator. The fourth one's around risk management reduction, and Charlie and Cyprien have both mentioned the fact that we need to work with our clients to try and work out how we can reduce our exposure to LIBOR over time. As the alternate rates increase in liquidity, there'll clearly be a swap, and there's a transition for that. How do we proactively talk to our clients on that process? Then the fifth one is around client education communication. We spend a lot of time at conferences, and I referenced before talking through some of the specifics and actually, not just talking to the senior treasurers, but actually talking to the next level down of staff as well, talking to their operations team, the technology teams. The national working groups that have all been created have now started to create some infrastructure working groups as well.

Cyprien: I think there is one thing which is very important in what Chris is saying, is that this reform will happen, but also it effecting clients and very different groups. You need to have your legal team aware of it, you have to have your ops team aware of it, you have to have your IT and pricing team involved as well. This always take time. All these teams need to work together, but in parallel. If you have done all the legal analysis, if you are ready to change, but if your systems are not up to date, there's not much you can do.

Julia: Charlie, how are you thinking of this with your trader hat on?

Charles: In your question of, you know, are we going to be ready, the reality is, the derivatives market will be absolutely fine. The derivatives market already trades loads of currencies on to loads of different indices, we've got entire teams of quants that can go out and transition our tools. The traders are used to working with different indices and interpreting them and all the rest of it. I actually have no concern at all that whenever this transition happens, there'll be liquidity in whatever is the appropriate underlying index. I think that the problem is, is because the derivatives market based on LIBOR is so big, there's a real danger that people think it's a derivatives market problem. There's a call to arms, the answer is, like, it isn’t, people have to envisage for a moment, you've got to do that thought experiment. If you have a system that somehow pulls in a swap price or a fixing or a reference rate or loan rate that in some way references LIBOR, and you're running a company that does anything from selling cars and offering finance, or whatever you do... if one day that rate can't be plugged into your core central system, and that shuts your business down, then you really need to be engaging with it. Some of those solutions could be really, really simple, and some of them actually could be really taxing. Until people actually start really getting to the details of what does this mean for me actually, there's scope for lots and lots of small accidents. I look at it with a trader hat. I think we should probably talk a little bit about fallbacks. Obviously the big concern with LIBOR has always been, you've seen this sort of massive down slope in the number of volumes of transactions that underpin the index. Quite rightly, everyone is saying, hang on a second, you've got trillions and trillions of transactions referencing this index that now doesn't have a huge underpinning. Of course, quite rightly, people are worried, well, what happens if it's not there one day? We commented earlier about the fact that it's only committed until 2021. You need some kind of mechanism to say, well, if it's not there one day, what should we do? And actually, what's going on in the world of fallbacks is really important. Because in practical terms, if an index has a fallback, then what you're beginning to say is actually the path of least resistance, is to transition legacy inventory onto whatever the fallback proposes it should go to. For instance, recently, the fallback language that ISDA is discussing is taking a direction towards looking at, well, what's the recent average of the various tenor basis spreads, and saying if ever the LIBOR panel fails, you take those spreads and you project them forward forever. Inherently what that means is, the basis loses term structure. Forever, the same spread exists for term. The market is very quick at snapping to that and saying, now there's no term structure. I think people need to look at the fallback language, because the fallback language is giving you a clue as to how generally rational participants will think about transitioning.

Julia: Chris, could you help define fallback language for us?

Chris: Fallback language is the language that's written into the existing documentation that would occur should LIBOR not be available. The problem with most fallback language in the documentation today is, it's written on the basis that LIBOR is unavailable for one day or two days, et cetera. It's not written on the context that LIBOR is not available forever. The initial push from the ISDA fallback language and to update the master agreements was to highlight that point. The language that's written in many of the documentation effectively turns floating rate products into fixed rate products.

Julia: Teasing out something before, why is there this misconception that this is a derivatives problem? Why are people overlooking the other side?

Charles: I think it's as simple as the fact that the optical numbers, the hundreds of trillions that people quote around derivatives are such awe inspiring numbers. That just becomes the big elephant in the corner of the room, and no one can take their eyes off it. When you're talking about hundreds of trillions, and somebody wants to talk to you about a hundred million structured mortgage book inside a regional lender, you think, oh, that doesn't really matter too much. The reality is, that mortgage product, that structured note, that retail investment is real, tangible money, and is actually extremely relevant to the individual, the entity that has it. The derivatives notionals, yeah, it's really exciting, but the reality in the derivatives market is, so much of it nets down, so much of it is a... it's a notional, that's why in derivatives we call it notional. It's a reference amount that talks about the amount of risk in the instrument. There are LIBOR based products that are real cash products, and the notion of those products are extremely relevant relative to the size of the holders. Just to put onto that point as well, practically, the system changes and the operational changes required to solve those cash problems are far greater than the derivative problems. As Charlie mentioned, from a processing perspective, from the technology point of view, the quants associated to the changing of the curves, derivatives businesses do this every day. That's just natural to them. The corporate market and the FRN market, that's a bigger change. The greatest move that we've seen so far that's been successful is the SONIA issuance of FRNs, and the fact that compounding and arrears, and investors have bought those, and that is a shift that's moved. That took some time, it was a bit slow at the beginning, but it has taken some time. You think about that, that's, we're 12 to 15 months into that process now. If clients haven't started that system change now, and you're 12 to 15 months to make that change, you're into 2021 now. I agree with that. Talking about 2021, humanity and markets especially have an amazing ability to sort of ignore something that's not just immediately around the corner. 2021 will seem like a long, long way away, a bit like Brexit. 2021 will see like a long way away, until it's the end of 2020. Then it's Q1 in 2021. At some point, people begin to look at their cash instruments or the derivative instruments and say hang on a second, what happens now? You tend to go in a completely irrational way between thinking something is next year's problem, to being right in the middle of a major crisis. I think people need to realize that if they've got a year or two's work to do, now's the time to start engaging. That's interesting, because this change is very different to CFTC we've had with Dodd-Frank, MIFID, et cetera. All of these programs, and the phrase I'll use is banks our size, we tend to run very fast to the finish line very well, and we make all these deadlines. They're regulatory deadlines. As of this date, you must have this new trading profile or this new customer outreach, whatever it may be. This is a completely different program, and it has to be run completely differently. You need to plan far, far ahead in comparison to previous programs that we have. We cannot wait until Brexit's over and we know what the answer to Brexit is before we throw resources at this program. If you do that, you will miss the boat.

Julia: That wraps up our discussion for today, so thank you for your time.

Chris: Thanks, Julia.

Cyprien: Thank you. Thanks.

Julia: We've covered a lot of ground here today, but unfortunately we need to leave it here. This is clearly an evolving topic, and our main goal is to continue educating our clients further. As such, we have articles on JPMorgan.com, this podcast series, along with regional client events and round tables. Thanks for listening.