This has been a banner year for collateralized loan obligations, as strong investor demand allowed managers to keep issuing new deals without pushing up funding costs.
Problem is, it was also a banner year for the leveraged loans that serve as collateral. As below investment grade companies continued to refinance their debt, CLO managers had to accept lower interest rates. That made it challenging to meet various performance metrics designed to protect their own investors.
Indeed, managers found that managing to one kind of metric often put them at risk of failing another one.
According to data from Wells Fargo and Intex, at least 61% of CLOs issued since 2012 are within 10 basis points of failing their minimum weighted average spread, or WAS, tests, or the minimum required under investor documentation.
Another 42% of portfolios are within 5 basis points of failing to meet required diversity ratios protecting against industry or obligor concentration.
At least 12% of CLOs are failing to keep up with minimum average ratings standards of the underlying loans.
Managing to these different metrics requires a great deal of flexibility, and that is one of the reasons that Bain Capital Credit, which refinanced or reset seven of its outstanding deals this year, sometimes chose to modify deal documents - even if that meant triggering a requirement to keep "skin in the game" of a deal, according to John Wright, the head of Bain's CLO/structured products business.
The global credit firm (founded as Sankaty Advisors) was one of the first CLO managers that opted not to take advantage of a workaround that allowed deals grandfathered from risk retention rules to avoid triggering compliance. In an interview with Asset Securitization Report, Wright said that there were other reasons to forgo the workaround, as well. At certain times of the year, it was also very attractive to lock in longer term financing by extending the maturities of the notes or raise additional funds by upsizing a deal.
An edited transcript follows.
This was an active year in terms of volume for Bain Capital Credit. What made the market the most attractive for the firm and its investors?
Over the course of the year we have seen two consistent themes. First, loan spreads have been tightening, and the second is CLO liability spreads have been tightening. Strong CLO new issue activity shows the relationship between those two things is still healthy, which drives CLO equity demand. In addition, there’s been an unprecedented amount of activity around refinancing liabilities or resetting CLOs to keep the liability cost in line with loan market spreads.
Bain Capital Credit was mostly active in refinancing/resetting older deals. Was that a function of the market conditions, or the coincidence of so many deals exiting non-call and reinvestment periods?
That was largely driven by market conditions. In each of our deals, we worked with the equity investors to decide whether it made sense either to do a straight refinancing or a reset transaction where we refinanced the liabilities and extended the life of the deal.
Did it make sense in some instances to do a reset instead of ramping up a new primary deal with the acquired loan assets?
The two are not mutually exclusive. In some resets, we upsized portfolios to add new assets.
Bain Capital Credit only took advantage of the “Crescent” exemption in some of its eligible refinancings this year. How did Bain Capital Credit make the decision to maintain the exemption or take on risk retention in the existing deal?
If you put aside the question of risk retention and look at a straight refi versus a reset, there have been times in the market when the cost of issuing longer liabilities was not much of a concession relative to the cost of doing a straight refi with shorter liabilities. One of the drawbacks of doing a Crescent refi is the inability to modify deal documents. In some instances, equity investors were more interested in locking in liabilities for longer, and we were able to build in some management flexibility, which we would not have been able to do under a straight refi. If the last hurdle is making the deal retention compliant, as manager we have been willing to support deals to get the best outcome.
Are there any deals left that are eligible for a Crescent refi?
The Crescent exemption applies to pre-2015 vintage deals, for up to four years after issuance. Most 2014 deals are out of their non-call period, and the four-year window will close during 2018, so we’re largely through that Crescent wave.
Have WAS eased on deals here in the latter part of the year? How has Bain approached keeping deals passing their tests?
Yes, given the strong loan repricing environment, we have seen some pressure on weighted average spread. One benefit to having a large and experienced credit team is the ability to cover a wide range of names in the loan market. This ensures we have the most options available to manage tests, but also keep the focus on credit quality.
Did you have to make "deal matrix" changes to satisfy WAS tests when spreads were compressing?
Yes, the matrix provides some flexibility to lower the minimum weighted average spread by trading off ratings or diversity. The difficult thing is that those grids are set when the deal is issued, so unless you have amended your deal documents, you’re trying to manage around a grid that may be outdated relative to where the loan market is today. It is really beneficial for a manager like us with the scale we have to look at a broad number of companies for the most investment options. But given how strong the loan market repricing has been, managing the matrix continues to be a challenge in instances where we were not able to change the matrix as part of a transaction.
Can you provide an example of what new terms or deal structures were sought in a refinancing to soften the pressure on the WAS tests?
A key benefit to a new deal, or a deal with reset documents, is a fresh matrix that reflects the current asset environment. Also, for deals later in their life, if a CLO is reset, this will provide additional room on the weighted average life (WAL) test, which can be challenging to manage as a deal approaches the end of the reinvestment period.
How are managers approaching the ramp-up period for deals now? Is the demand so great for loans that you have to issue more deal with partial identified portfolios?
The ramp strategy for a deal should depend on market conditions. In today’s market, with a majority of secondary loans trading above par, we prefer deals with warehouses that have benefited from accumulating new issue loans over time. This can be supplemented with assets sourced in the secondary market and an allocation to unidentified new-issue loans. The proportion and economics of unidentified loans should reflect the current market and the time horizon required to complete the ramp.
Cov-lite loan issuance is at an all-time high. Is that somewhat problematic for CLOs given the low caps many deals have right now? Does that raise the price for existing loans in the market with maintenance covenants?
Covenant lite is really a question of how it is defined. Depending on the definition in a CLO document, it will make the test more or less difficult to pass. In the deals we manage today, we are not particularly pressured on the covenant lite bucket, but market conditions can change.
Speaking to marketwide challenges, if some portfolios are facing stress from the covenant lite restrictions in shopping for loans, how do they remedy that? Through reconfiguring the matrix of a deal?
As lenders we would always rather have covenants than not. However, we would also rather have the ability to pick the credits we think are better credits than be limited to those borrowers with covenants. The matrix doesn’t typically offer flexibility in the covenant lite bucket. Despite this, we have been able to manage the bucket, in part because we have structured the buckets to reflect market conditions. The weighted average life test has been more of a constraint for deals approaching the end of the reinvestment period. It becomes difficult to participate in new issue deals given long maturity, and the secondary market does not offer sufficient short maturity opportunities at or below par.
Is the definition of “covenant lite” becoming less restrictive – is that perhaps why we’re seeing more loans that are being classified as such? Any examples of how that’s changing?
The definition of covenant lite varies by deal. In the current market, with strong demand for CLO liabilities, more deals have been using a more flexible definition. This is also a function of what the loan market is offering as investment options for CLOs.
If leveraged lending guidance is lifted, would you be concerned about credit quality metrics on CLOs?
If leveraged lending guidance is lifted, I think you will see an increase in new deal leverage. However, my hope would be that the increase in reported leverage would reduce some of the aggressiveness we have seen in pro forma projections. While this may make the reported leverage on new deals look higher, we would have more confidence in the ability of companies to reduce leverage over time as they are not starting with such an aggressive EBITDA projection.
What do you think lies ahead for CLOs in 2018?
We believe CLOs are an asset class that performs really well through the cycle. In today’s market, we have seen so much spread tightening in loans and in CLO liabilities. We like CLO equity today because we can lock in post-crisis tight liabilities and have upside if there is volatility in the loan market. And while the market has not had a lot of volatility over the last year or so, it is difficult to predict. So the idea of locking in tight borrowing costs, earning an attractive base case yield with upside in a more volatile environment, that is attractive to us today. We believe we will continue to see robust CLO formation and activity next year.
Would you put a number to that?
It would be difficult to replicate this year’s combined volume (new issue and refi/reset) because you are unlikely to have the same number of refis as this year.
