‘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’
Are there still incentives for banks to issue CLOs in times of low interest rates and increased tranche risk weights (BCBS 374)?
Kurt: I noticed the question is on CLOs – to be clear, I am not directly involved in Nomura’s CLO trading effort, but given that the whole capital structure is being placed from the bank I can’t see that risk weights make much of an impact on this. That is very similar to my market; the risk weighted to capital gains transfer do not come into play much for me as I close out all my capital structures within a few weeks.
Are bespoke portfolios across dealers or deals very similar?
Gaurav: Yes, I would say there is a significant amount of overlap in the names that we see in bespokes – of course, not every day will have the same exact names – but there is fair amount of overlap. I think there are a couple of reasons for this and the primary reason is that because issuances is mostly in a two-year to three year a part of the curve you are forced to pick names that have liquidity up to two years. While at the five year point there might be many market participants that are willing to buy or sell protection on a name, at the two-year point the number of buyers of protection is limited. Often at the two-year point buyer of protection is looking to hedge for defaults, rather than just for spread moves and that limits the number of names that are available. The other aspect I’m guessing is that the number of participants involved in the equity tranche is a handful, half a dozen at best, and there might be certain set of names that they’re comfortable underwriting and taking default risk on. As a result, the number of names that show up in the portfolios is limited. One of the reasons for having bespokes is the ability to go outside of the names within the index and hence have a more diversified exposure both from the long and the short side. While bespokes are serving that purpose unfortunately, it is not as expanded as it should be.
Kurt: The one thing I’ll add is from a macro perspective these portfolios from my view all look similar – eighty-five percent of the portfolio is the same but you do see significant differences in the fifteen or twenty percent that drive the value of the bespoke or add most of the spread etc. I do see my equity counterparties varying quite a bit on that. There will be different heavy sector concentrations in that fifteen to twenty percent of high yield names but broadly the remaining 80 plus names that fell at the portfolio end up being selected for the criteria Gaurav mentioned, liquidity etc., and ends up being the same names.
Gaurav: I think my overall observation is that you tend to see the widest names in the IG universe and the tighter names in the high yield universe. I think names that traded 10 or 20 basis points no one has the incentive to put them in a portfolio because they do not have enough juice or spread. While the most distressed names might not be something that some of the investor up the stack might want, even if the equity tranche holder is comfortable with it, so at the end of the day you end up with the best names in high yield and the worst names in IG.
Do you remember the overall volume of index and bespoke tranches compared to the pre-crisis levels?
Kurt: I cannot remember exact pre-crisis volumes but the comparison in my mind is almost not valid. When I was trading bespoke in 2004 -2007 it was all single tranche issue deals, so comparing those volumes to today’s fully placed volumes I don’t think make much sense. Certainly, I am not doing as many deals currently but the growth trend is alive and we are definitely going to have year-over-year increased volume especially on a CSO1 weighted basis.
Gaurav: I think Kurt’s absolutely right – if you look at pre-crisis, most of the issuance volumes that people measured were based on the rated deals because that was easily publicly available information as to how much got rated. Many of the unrated transactions that were done in swap form were not necessarily tracked as cleanly. At that time, regarding the order of magnitude, the issuance looks similar in CLOs and in bespokes, the order of magnitude was 100 billion in each and kept growing in bespokes. However, what Kurt said is a very important point, in that it doesn’t tell you much. It doesn’t tell you much because the typical issuance then, involves ten bespokes, so let’s say a two to three percent slice or a three to five percent slice on a five-year, seven-year or ten year part of the trunk. In terms of notional, the order of magnitude might look the same today but in terms of the average Delta, by definition if you are placing the entire cap structure right now the average Delta is one. In those days, the average Delta was three, four, maybe even five, then on top of that you adjust for the duration and the volumes then – risk adjusted – were probably at least ten times bigger if not more, so it’s very hard to compare those two. I think that minimum ratio of ten is probably on the conservative side.
What volumes do you see in the secondary market bespoke market and is it reasonably priced?
Kurt: Largely, my secondary volume is senior tranches that have rolled down and the investors no longer have the return for the capital required to hold that, so they trade out of it. The execution is pretty good because there’s plenty of players willing to take this short dated senior coupons presumably to fund other trades that they like, so that’s a highlight of the bespoke market liquidity.
Gaurav: We all have our biases when we answer that question – Kurt said liquidity is good because he sits on the sell-side and if you ask anyone on the buy-side they will say that secondary liquidity is not that great – we have those inherent biases. The fact is that if secondary transactions take place that means buyer and a seller agreed to transact at that price whether they liked it or not, which means it was a fair price. Which in turn means the liquidity was acceptable to both sides, given the market environment at that point in time. There can be no excuses around it, if you traded, that was the fair price. A better question to ask is what percent of bespokes that are issued end up being held to maturity versus what percent are unwound versus what percent are novated? Sitting on the buy-side it’s harder for me to have that information precisely, maybe some of the active participants on the sell-side have better sense. My guess is at least two-thirds, if not more, are held to maturity – especially on the equity part of the of the cap structure. It is possible that some of the mezz does transact in the secondary market. As far as novation is concerned, yes it can be done as nothing stops a person from novating but more often than not the original counterparty has the best price because they know the portfolio and the other players in that exact transaction. That being said novations do happen and I’ve seen it multiple times over the last few years. I would say that less than a quarter of the deals get novated remaining if they are traded at back with the original counterparty.
What formats of risk usually trade in the bespoke market, for example if index market has K equal hundred how does bespoke trade?
Kurt: We see all the risk formats coming back from pre-crisis, all the recent equity I have traded has traded at least part of the deal in Principle Only (PO) format on the equity. Almost always equity investor has a convexity in mind in the low dollar price long equity is the trade they want to get into. We’ve also traded all running spread equity and the standard strike 500 equity. PO is probably the largest bid offer simply because that leaves me as a dealer with the largest coupon annuity risk but certainly it isn’t a hindrance to getting a deal done and we will trade all the formats. For the balance of the capital structure, I think all the junior mezz I have traded is strike 100 and super senior I largely trade fair spread, so not a standardized coupon on that.
What is the barrier to entry for new players and what does one need to start trading these products?
Gaurav: I think the biggest barrier to entry is the ability to have ISDAs and that means there is a certain size below which one cannot trade the product and it rules out a fairly large number of investors in the market. Even if you have organizations that have the ability to do it, if they have smaller funds or separately managed accounts then unless they are a certain size it may not be worth the time and effort involved in signing ISDAs, and that I think is the biggest bottleneck. The next bottleneck is the question of complexity that often gets attached to the product, and whether fair or not one. One of the simpler products in terms of cash flows is seen as more complex than a cash flow structured product like a CLO and that is the way it has been and therefore for now it is seen as complex I think many organizations are nervous about being involved in complex derivatives. Last but not least I think there is this dearth of publicly available high-quality data on tranches on single names etc. You can probably acquire some of that data but it is expensive and I think that will continue to be a bottleneck.
Kurt: I really like Gaurav’s point around comparative complexities between CSOs and CLOs, I think that one warrants more thought because I agree with his views, in my opinion the CSO is the simpler cash flow product yet many organizations get caught up in modelling complexities around it.