Following the credit crisis of 2008, tranche trading all but disappeared; it is now back with gusto. For example, bespoke tranche trading reached $80 billion issuance in 2018, and continues to grow rapidly. Although a far cry from pre-crisis level, there are encouraging signs for the market’s revival. In the first of this blog series, Kurt Koschnitzke, Executive Director, structured credit Trading, Nomura and Gaurav Tejwani, Portfolio Manager, Brigade Capital Management outline the different aspects of tranche trading. This blog is taken from the Quantifi webinar ‘Trends in Structured Credit Markets’, moderated by Dmitry Pugachevsky, Director of Research, Quantifi.
‘The views and opinions expressed in this blog are those of the individual and not of the companies they represent. As this blog has been transcribed from the webinar recording, there may be minor differences’
In 2018, trading volume of index tranches reached $215 billion, while bespokes were only $80 billion. What are your views on which tranches are more popular now; index or bespokes?
Gaurav: In any market, liquid products always attract more trading volume or attention and in our market - that is also the case. This is not just a matter of liquidity or lower transaction cost but also with the transparency that comes along with index tranches. Pricing transparency and transparency on risks, for example the exact deltas of the tranches or other risk measures, are well-known. What also makes index tranches more popular is the ability to hedge, either with single names, index or options on index. For these reasons, I think index tranches will continue to stay more popular than bespoke tranches. The other aspect that makes index tranches more popular is the ability to create unique long short profiles and that is something that can be done in bespokes also but it's much easier to create a structure such as long equity short mezz tranche on the same exact portfolio vice versa.
So all in index tranches continue to be more popular, that being said, there's only so much risk that investors would like to continue to take on the same 100-125 names over and over again so there is a need for bespokes and we're starting to see liquidity in bespokes gradually improve. In addition for the most part we've been in an environment where it defaults have been few and far between, so as and when we see more trouble in certain names and people start developing very specific views on either certain sectors or certain names we might start to see some more interest in bespoke tranches relatively speaking.
Figure 1: Tranche Volumes
Kurt: I think liquid tranches are more popular volume-wise. From my seat, as shown on the graph (figure 1) the exponential growth is also true for the bespoke market too. At Nomura, our run rate from last year to this year shows this growth trend continuing even in bespokes, especially if you take into account some kind of CSO 1 weighted volume. What we sometimes find is that the headline volume numbers are down a little bit but that's because my volumes have gone from 2 year and now they are split between two and five years. While I do not expect the bespoke market to ever approach the volumes of the index market in tranches, I do think that this bespoke market will continue to exhibit the same exponential growth.
Until recently a most popular tenor for bespoke was two years but now it seems like five year becomes more and more liquid. Do you think it's a stable trend and how do you compare trading two year versus five year?
Kurt: I do think it is a stable trend. Had you asked me at the beginning of this year I would have said that five year was going to eclipse two year but that trend has levelled off somewhat in my opinion but my flows are still split. When I look at the trading that I do, it’s certainly easier for me to find equity investors in the two and two and a half year bucket - clearly that's where it's easier for credit pickers to have a view on a name.
I have also done a few bespoke shorts, those are in the shorter tenor buckets as well as all of my secondary bespoke trading has been in the two and two and a half year bucket, so that's alive and well. Five-year activity seems to be taking a breather over the summer and I think absolute spread levels played into that but that is the space where I think there is more going on in terms of innovation. I know ourselves at Nomura, and other banks have been working on the first managed deals as well as possibly looking at getting some of these tranches rated. So I think they're both going to co-exist.
Figure 2: Popular Bespoke Tenors
Gaurav: The debate between two year and five-year for someone on the buy side is that you can argue either way. The benefit of two year, as Kurt said, that there is visibility on the likelihood of defaults on the portfolio, but at the same time, the curves on many of the names are so steep that there has to be a price at which investors should be happy to switch from two years to five. I think the argument on defaults, fair as is it may be, applies quite well to some of the distressed names. However when you look at good quality investment grade corporate entities, for example a name trading at say 50 basis points or 100 basis points out to five years I think it's very hard to make the argument that one can have visibility on default into two years but not five years. Often such names, if they default or become distressed, it is because of either a fraud or some exogenous shock that no one saw coming. So while the argument about defaults does explain the greater amount of issuance in two year can't alone explain it and I think it has to do with asset liability mismatch. I think because bespokes are not necessarily seen as liquid enough to be able to easily get in and out the fact that many participants feel like they have visibility on sufficient capital to two year horizon. This is either because they have locked up funding or because they have gates which prevent investors from immediately withdrawing money overnight etc. Many investors feel they have they have faith in being able to carry these trades for two years without being forced to unwind them and I think that is an important determinant. As we go through this phase where a five-year liquidity is starting to improve, I am hoping that more and more players will look at the five year tenor because there are a lot more interesting long short trades that can be constructed when you go out to five years.
Kurt: Gaurav touched on something important I think. Part of what drives where these deals are getting done are the technicals in the market. For example the CDS market is again developing bespoke technicals, for example, we had a couple of two-year deals ready to go and the dealers were almost unwilling to bid two year CDS at that point. The bespoke machine had been up and running for two years strong by this summer and that bucket in their books was already very short and I think that is part of what brought the five-year into play. I think as the bespoke machine comes back up to full hum, it's going to continue to drive dynamics in the single name market and where we're able to get deals issued and where it makes sense to get them issued. Hedge costs became such a large part of two-year deals for me that it didn't make sense to do the deal, that's part of what drove us to five years.
We saw that volumes in both index and bespoke tranches have increased and are increasing exponentially but still there is the feeling that there is some kind of a bottleneck for the market increasing in size. What do you think are the reasons for this bottleneck?
Gaurav: I think there are two questions really; one is ‘why do we have less players in bespokes than we have in index tranches’? The other is ‘Why don’t we just have more players in index tranches?’ I will focus on the bespoke part of the equation since that's the question you ask first. I would argue that lack of single name liquidity, especially to two years, is probably the driving factor for this because not only does that mean that the cost of creating a new transaction is high, but it also means from the perspective of an investor the cost of hedging it over time, or the ability to unwind, is lower. I think until we have good liquidity in two or three year CDS that is going to be the case in terms of the bottleneck.
When it comes to index tranches, in general I think the reason we have less participants is partly because we have not had too much volatility in the market especially in terms of defaults. If you think of what the tranche market really does, it separates the spread and default component within the index such that the equity tranches take a more leveraged view on defaults while senior tranches are taking predominantly spread or mark-to-market risk. As a result, if you have an environment like today where defaults are few and far between, there are not as many players who want to express those views through tranches.
Those I would think are the other two primary factors, one additional factor which is outside of fundamentals has to do with operational constraints and a lack of data or analytics. For example, if you want to trade a CLO, it's fairly straightforward it's a CUSIP you have all the information available on Intex or Bloomberg and you go ahead and create it, but in order to trade tranches you require ISDAs with multiple counterparties and that's something which operationally is not easy for every buy-side player involved. The other factor to consider is that data on single names and all of the nuances that go into pricing a tranche is not as easy to find in the public domain. This is especially true if you want high quality data and then there are a lot of costs associated with which I think is another reason for the bottleneck, but you would hope that as time goes by data gets easier and easier to access.
Kurt: I have a couple of things I would add in terms of putting a check on growth. The first is since the rebirth of bespokes in the full capital structure trade, clients I talk to most in the marketplace agree that this is less of a correlation product and more of a levered credit product, so I see fewer guys trading bespoke tranches against Delta on single names and things like this for a real correlation play and it's just an absolute spread play. With that in mind, the current credit environment is too tight for some those trades that make sense and it has made it hard for me to place junior mezz, so absolute credit spreads are in my way a little bit there.
The other thing I would say that clients mentioned to me as something that that gives them pause is just the underlying CDS market. I know ISDA is working on clarifying technical defaults and I think as that market gets cleaned up it paves the way for more structured products on top of it. However, I do think there are people on the side-lines waiting for these highly tailored, high recovery defaults to be eliminated. Finally, when I look at the market and what used to make the market work back when it started in the mid-2000s, you had real money players involved and those guys are largely missing from the market now. To bring those players back it's going to be with a rated product and I think that piece is missing. It needs to be put in place before we get back to the volumes we once had.
Part 2: In the second blog in this series the panellists compare CSO vs CLOs, short trading for CSOs and trading of whole capital structures.
Part 3: In the final blog in the series the panellists answer questions from the audience on CLOs, bespoke portfolios and more