Q1 2019: A CLO Vintage To Avoid?
January has been a very quiet month for the European CLO market, with spreads tightening but still relatively wide. While we know there are around 30 different CLO warehouses open – some of which are for deals that couldn’t be priced in December and were postponed – the pipeline is struggling to materialise. So far only CSAM has managed to price a new deal, and in the process showed how difficult it has become to make the arbitrage to work for equity investors (see our recent blog CLO Arbitrage Worst Since 2013).
Currently there are a number of CLOs in the pipeline, with Carlyle, Guggenheim, GSO, ICG and Spire likely to be the first five to test the market next. The Japanese bid for AAA-rated CLO debt has not gone away (though the bid has widened by 10-20bp), but there is likely a fair spread basis between those deals anchored by Japanese buyers and the deals that are more widely syndicated. Secondary AAA spreads are wrapped around Euribor+105-125bp (with Euribor floored at 0%), while the single-Bs will show a large range of roughly Euribor+8.5-10%, as the performance will have a much bigger impact so low in the capital structure.
European CLO buyers already had around 45 managers to choose from in 2018. There are another 10 or so managers waiting to price their first CLO in Europe, according to a Citi report last week, alongside 17 managers looking to price their second or third CLO in 2019. New managers include names like Insight Investment, Fair Oaks, Sound Point, CIFC and Angelo Gordon, while according to a recent Bloomberg report Capital Four is working with Goldman Sachs on its debut deal. Underwriting new managers is always difficult, and it is made even more so when investors are still trying to find out how the new managers from 2018 are performing. Just last week we met a number of new CLO managers, some of whom have large CLO platforms in the US. Some of these managers have only had a European CLO team in London for the last six months, with zero track record except what their portfolio managers have done at their previous firms – not appealing to us as an investor. Let’s not forget that the European leveraged loan market is dwarfed by its US counterpart, and one might wonder why we need more managers at all.
Even though pricing doesn’t always reflect this, we see a lot of tiering between managers. Highly respected managers like PGIM, Barings and Alcentra will run CLOs where the collateral has a worse average rating (including CCC exposures of 6-9%) and lower-priced collateral well into the double digits, in an attempt to boost the equity yield. Managers like CVC, Natixis AM and KKR Credit, meanwhile, typically have lower exposure to credit intensive names, lower spread and higher ratings. Obviously we as debt investors underwrite CLO managers for a totally different reason to equity investors; we want to get paid timely interest and principal. We focus on managers that have small exposures to the retail and automotive sectors, as well as names like Douglas, Flint, Dummen Orange, Deoleo and CMC, to name a few.
So far we have passed on pretty much everything in the Q1 2019 pipeline and we see better value in the secondary market. Though the documentation looks reasonable, for us the leverage is too high.
Higher leverage is evident in the lower-rated bonds, particularly at the BBB/BB/B levels, where the subordination has been reduced due to the lower yield on the loan pool, which means the arbitrage has to be tweaked to work for equity investors. The result is that ratings are also stretched, with what was once a BBB rated tranche now a BBB- rated tranche. On the face of it this doesn’t look like a big difference, but around 2% less subordination is material and actually equates to an extra 6% of defaults at 33% loss severity, and these minus ratings also carry a higher risk of dropping into the next rating bucket.
Lower ratings also increase liquidity risk, in our view. For European investment banks that trade CLOs, BBB- rated paper carries a 33% higher capital charge than BBB rated paper, making them far less likely to hold such assets for trading purposes, so from a future liquidity standpoint we would avoid these.
Different CLO vintages tend to perform very differently, and documentation and arbitrage also vary significantly depending on the period in which a deal was priced. And studying the Q1 2019 vintage, this might be one to forget pretty quickly.