Listen to the latest episode of Credit Exchange with Lisa Lee
Published in London & New York
10 Queen Street Place, London
1345 Avenue of the Americas, New York
Creditflux is an
company
© Creditflux Ltd 2025. All rights reserved. Available by subscription only.
Analysis CLOs
Don’t fall foul of bad language
by Bradley Pearman, Fair Oaks Capital
A few years ago, maturity amendments were a major focus. But waning attention to European CLO documents is leaving some investors exposed to extension risk
CLO documents define the structural and protective framework for each transaction. While the overall layout is similar across vehicles, many nuances exist which create opportunities for CLO lifespans to vary considerably.
These differences impact returns across the debt stack. Lower mezzanine buyers, which typically purchase bonds at a discount, require a pull-to-par to realise their predicted returns. Meanwhile, for investment grade buyers — which typically buy closer to par — any delay to receiving principal repayments often come with an opportunity cost relating to not being able to reinvest proceeds in new issues at wider spreads.
Distribution of triple A factors for European CLOs with reinvestment period ending in 2022 (%)
Source: lntex and Fair Oaks. Data as at 2 May 2025. Sample includes euro denominated, floating rate triple A tranches with over €150m in original face value outstanding.
Despite this, it seems investors’ attention to certain sections of CLO documentation is inconsistent. One such section is that governing how a CLO manager agrees to extend the maturity of an underlying debt obligation.
In 2022-23, when the European loan market was seeing a large number of extensions, these clauses were much discussed. However, with that wave of maturity amendments receding, investor attention has waned. Investors now place less importance on these amendments in negotiations and do not address weaknesses in the language.
With no cap, extension capacity can be recycled indefinitely
While there may be low levels of maturity amendments currently, unaddressed weaknesses in documentation will leave some investors exposed to extension risk if maturity amendments become common again.
The chart (above) shows variability in triple A factors of deals with a similar vintage, highlighting significant differences in amortisation patterns among European CLOs. Greater standardisation in CLO documentation would help reduce this dispersion and improve predictability for investors.
Removal of the cumulative cap
In 2022-23 European CLO investors added ‘anti-snooze’ provisions to CLO documentation. These force the manager to vote either for or against all maturity amendments (if permitted to do so within that section). This led to some managers increasing the size of maturity amendment buckets, which allow for a certain number of asset maturity amendments while the weighted average life (WAL) test of a deal was failing. Prior to the increases in these limits, the market standard was 5% of the target par amount.
An example from a CLO document is as follows: “The Weighted Average Life Test is satisfied, provided that, notwithstanding the above, the Issuer (or the Collateral Manager on the Issuer’s behalf) may vote in favour of a Maturity Amendment that is a Credit Amendment that does not meet this condition (b) if, after giving effect to such Maturity Amendment, (i) not more than 2.5 per cent of the Collateral Principal Amount as at the relevant date of determination and (ii) not more than 5.0 per cent of the Target Par Amount cumulatively since the Issue Date, consists, in each case, of Collateral Obligations where the Issuer (or the Collateral Manager on the Issuer’s behalf) voted in favour of a Maturity Amendment where this condition (b) was not met.”
Consent fees can allow managers to bypass standard constraints
This language applies a 2.5% ‘point-in-time’ limit (applicable to assets held in the portfolio at the time) and a cumulative 5% limit (applicable to all assets, whether currently held or not).
In addition to the cumulative buckets being increased in size, some recent CLOs have included only a point-in-time maturity amendment limit, with no cumulative cap. The lack of a cumulative limit gives significant flexibility to the manager to use maturity amendments to extend the WAL of their deal while the WAL test is failing. A point-in-time limit will only restrict the manager based on the extended assets currently held “in the portfolio”. As a result, a manager could extend a loan, sell it (thereby emptying the bucket), and reinvest in another long-dated asset. This process effectively recycles extension capacity indefinitely, even while the WAL test is failing.
Other extension loopholes
In addition to the point-in-time-only method, managers have used other techniques to enable them to consent to maturity amendments, sometimes on an unlimited basis.
Extend-and-sell flexibility: In some European CLO documentation, language has been introduced which allows the manager to consent to — or ‘abstain’ on — a maturity amendment, regardless of the cumulative limit or whether the WAL test is passing, provided that the extended asset is sold within a certain number of days (and held at a lower value if not sold within that period). This again allows the manager to consent (or abstain and be ‘dragged’) to an unlimited number of maturity amendments.
To address the extension risk that arises from this, some CLOs have included clauses that allow the manager to consent to an extension of an asset they plan to sell, provided that if they do sell the asset, they replace it with an asset which has a maturity no longer than that of the extended asset prior to the maturity extension. This prevents the manager extending the WAL of the deal using this method but still allows it to participate in maturity amendments that may result in better liquidity.
Fee-driven consent loophole: Some CLOs have included language that allows a manager to consent to a maturity amendment, subject to no limit, if the issuer is paid a fee for consenting. This language is particularly loose and could allow a manager to extend any loan in the portfolio if a small fee is paid to lenders, irrespective of whether the WAL test is failing. Use of such language on a broader scale could incentivise widespread use of (potentially very small) consent fees simply to bypass standard CLO constraints.
Majority-consent-based voting language: European CLOs have included language that allows the manager to consent to a maturity amendment if “lenders or debt-holders, as the case may be, that constitute the required lenders or debt-holders for approval of a maturity amendment have already consented (or are expected to consent)”. This language again allows essentially unlimited maturity extensions, irrespective of whether the deal has a WAL test that is failing.
Weakening anti-snooze language
The above provisos relate to allowing the manager to consent to maturity amendments under certain circumstances. However ‘anti-snooze’ language has also sometimes been weakened and overcomplicated, so that it allows the manager to abstain (‘snooze’) under a very broad range of scenarios, despite the language looking as though it adequately restricts such activity.
Allowing abstentions from loan maturity amendments while the WAL test is failing can have the same effect as allowing consent (with respect to extending the WAL of a deal).
For example, some anti-snooze language forces the manager to vote against a maturity amendment if “such maturity amendment would result in limbs (a) and (b) above (to the extent applicable) not being satisfied”, where (a) is the usual pre-note maturity requirement and where (b) is the WAL test requirement, which includes the carve-out for a 5% cumulative limit for maturity amendments.
This language, because of the presence of ‘and’, means that the manager is forced to vote against an extension only in cases both in which the extension is beyond the maturity of the notes (which would be a very limited number of cases unless the deal is very far into its life-cycle) and the WAL test is failing and the cumulative bucket is full.
This allows extensions even if the deal is failing its WAL test
Essentially, the manager would not be forced to vote against, and so could snooze, provided the extended maturity is within the CLO maturity or there is any room in the cumulative bucket. This language therefore only restricts extensions in a very limited number of cases.
Over time, with the prevalence of ‘snooze-drags’ in the loan market increasing, we have seen recognition of this in loan docs. Borrowers or their agents often add an ‘abstain’ box to loan maturity extension response forms. Lenders may tick these to signify that although they are not opting for or against such an amendment, they will be deemed to have consented. One may therefore argue that opting to ‘abstain’ in such a fashion is not actually an abstention at all.
