Fair Oaks Reflections – Issue 4
28 February 2019
Much has been written over the last two years on the prevalence of “cov-lite” loans and, more recently, on other weaknesses in loan documentation. Below are what we believe to be the key points for investors considering exposure to the loan market.
Senior secured lenders have lost some of their control over borrowers through the loss of maintenance covenants, the weakening of incurrence covenants (covenants restricting certain borrower actions) and the related loosening of the definition of EBITDA. We believe that the impact of these developments on the loan default rate is likely to be mildly positive in the short to medium term but potentially negative over the long term as the increased flexibility in borrowing standards may result in higher leverage over time. In our view, the impact on loan recovery rates is likely to be mildly negative over the medium to long term.
Despite these negative developments, broadly syndicated senior secured loans remain the first ranking capital of sub-investment grade companies and are still likely to have relatively low default rates (e.g. 2019 defaults are expected to be 2.16%(1) vs. 1.6% in 2018(2)). Expected future recovery rates for bank loans are in our view likely to continue to compare favourably to those of high yield bonds.
The share of US and European broadly syndicated loans that are cov-lite has increased rapidly over the last two years to c.80%(3):
Covenant-lite share of outstandings:3
Cov-lite loans have incurrence covenants but lack maintenance covenants which require the borrower to report quarterly leverage and coverage ratios within certain thresholds. With maintenance covenants, a borrower that underperforms expectations can breach its covenants, which will then require it to obtain a waiver from its lenders to avoid default. The prevalence of cov-lite loans is likely to have three impacts:
While most defaults are triggered by liquidity crises rather than financial covenant breaches, companies that experience a temporary decline in performance may avoid a default that could have been triggered by maintenance covenants in their loans. We do not expect the prevalence of cov-lite loans to dramatically reduce the future default rate but we expect it to have a marginally beneficial effect in the short to medium term.
2.Lower recovery rates:
The potential for lower recovery rates has been widely noted, usually on the assumption that by triggering a default earlier in a declining business, the likelihood of a higher recovery will be greater. This assumption does not take into account the fact that recovery is usually received in the form of a (smaller) restructured loan rather than cash.
If a company is in a sustained decline, an early default and restructuring with a small loan write-down will often lead to a second default. Furthermore, our experience during the financial crisis was that while maintenance covenants did not often trigger defaults, they did divert management time away from maximising the value of the business to managing covenants and finding ways to adjust the reported numbers.
Nonetheless, we agree with the consensus that future loan recovery rates are likely to be below the historical average (Moody’s recently highlighted a decline in their average recovery expectations for senior loans to 61%(4)). This is partly because the defaults that are avoided due to the lack of maintenance covenants would likely have had above-average recovery rates and these will be removed from future calculations of average recovery rates.
3.Lower loan returns:
We believe that the biggest impact of cov-lite loans is likely to be lower loan returns and this is often overlooked. Historically, borrowers that breached maintenance covenants had to get lenders to agree to reset covenants rather than declare a default. This usually resulted in the payment of waiver fees and the imposition of higher loan margins. During the financial crisis we saw margin increases on a substantial proportion of US and European loans as borrowers were forced to negotiate covenant resets. Loan investors are unlikely to experience the same fee and margin improvements in the next cyclical downturn.
WEAKER INCURRENCE COVENANTS
Without maintenance covenants, lenders are reliant on incurrence covenants. Among other things, these typically place limitations on acquisitions, disposals, incremental borrowing and payments or transfers outside of the restricted borrower group. These covenants are particularly relevant to loan recoveries since they are intended to ensure that lenders retain their first priority claim over the value of the business. The same supply/demand imbalance that led to the disappearance of maintenance covenants has also led to some weakening of incurrence covenants. In particular, borrowers generally have more flexibility to increase debt than they had under previous loan agreements. We believe that this flexibility may lead to higher defaults and lower loan recovery rates in future as private equity owners take advantage of their ability to debt-fund add-on acquisitions, increasing leverage.
Despite the recent trends and with a few notable exceptions (e.g. J. Crew), loan documentation still effectively ensures senior lenders’ first-ranking position and priority claim on most of the value of the business. This means that investors should not expect the historical differential between loan recoveries and senior unsecured bond recoveries (75% and 33% respectively in 2017(5)) to change materially in the future.
Although EBITDA adjustments are not new, recent new issue loans have been marketed based on heavily adjusted EBITDA numbers. This has two impacts. Firstly, leverage in the market is often misreported. Secondly, weaker loan managers are likely to make bad investment choices because they do not properly determine which adjustments are reasonable (e.g. one-off transaction costs) and which should be disregarded (e.g. expected future cost savings).
More importantly, loan documentation now uses adjusted EBITDA as the basis for incurrence covenants (e.g. dividends only allowed if Debt/Adjusted EBITDA is below 4.5x). Typically, adjustments are limited (e.g. to 15% of EBITDA) so the EBITDA adjustment gives only limited additional flexibility. We have also seen examples where there is no limitation on adjustments, effectively rendering ratio-based maintenance covenants ineffective.
We believe that these developments in loan documentation make manager selection more important than ever. A good loan manager will not be influenced by the numbers an arranging bank uses to market a loan, will assess incurrence covenants taking into account any allowable EBITDA adjustments and will avoid loans with the loosest documentation.