In April, covenant-lite loans reached a post-credit-crisis milestone by capturing 40% of total institutional new-issue volume, the most since November 2007.
Viewed through a wider lens, however, the trend is less dramatic. Covenant-lite loans represent a more modest 22% of 2012 volume (as of May 16), down from 27% during the comparable period in 2011, when market conditions were even more robust.
For this reason, the percent of S&P/LSTA Index loans that are covenant-lite has increased just slightly, to 24.3% as of May 11, from 24.1% at year-end.
Of course, covenant-lite activity, like clearing yields, generally tracks the market. Therefore, arrangers expect incurrence-test-only loan activity to continue apace in the near term as paper forged in the hothouse conditions of February and March rolls into the market. Indeed, the forward calendar is dominated by such loans.
Further out, participants speculate that the recent cooling in the primary market may put a damper on covenant-lite activity, for obvious reasons.
On an apples-to-apples basis, arrangers say issuers pay a small premium to avoid maintenance tests – after all, financial covenants are effectively a repricing option that lenders can exercise if an issuer’s financial heath declines. Participants estimate the premium at 25-50 bps. The only evidence, though, is circumstantial: in late April, Schrader International’s $235 million, six-year first-lien term loan cleared at a roughly 50 bps premium to talk – L+500/1.25%/98, versus L+450-475/1.25%/98.5 – after the B/B2 issuer stripped covenants. However, sources speculate that had Schrader approached investors with a covenant-lite deal initially, it might have seen a tighter execution.
The truth, though, is that no one really knows how much more issuers pay to avoid maintenance tests. Here’s why:
In time-honored fashion, the recent batch of covenant-lite loans skews toward the more desirable issuers. That is apparent in clearing yields, what with incurrence-test-only loans printing, on average, inside loans with maintenance tests.
Even controlling for rating, this pattern holds true, suggesting that within a credit-quality bracket, issuers that are more creditworthy command both better spreads and looser structures.
- Add-on paper represents a disproportionate amount of covenant-lite loans, at 22%, versus 7.7% for covenant-heavy loans.
- Covenant-lite structures remain less prominent in the middle market, accounting for 11% of 2012 volume for issuers with $50 million of EBITDA or less, versus 22% for loans to larger borrowers.
Perhaps because stronger issuers have the muscle to avoid maintenance tests, covenant-lite loans have generated superior returns. Since January 2006 – the first date for which there is a statistically significant sample – covenant-lite S&P/LSTA Index loans outperformed covenant-heavy loans, at 39.28% to 36.36%. The pattern has held in 2012, with incurrence-test only loans ahead 5.18% to 4.86% as of May 14.
These statistics are even more remarkable considering that among loans outstanding as of Dec. 31, 2007, 23% have since has covenant amendments that pushed spreads up an average of 198 bps. In comparison, only 9% of covenant-lite loans have been repriced higher since that date via a waiver or amend-to-extend.
The reason for the return gap is simple: covenant-lite loans have enjoyed a lower default experience over time – another sign of positive bias in the sample. Of the S&P/LSTA Index loans that were outstanding at year-end 2007, just before the default rate spiked, 9.3% of covenant-lite loans have defaulted, versus 14.7% of covenant-heavy loans. And the gap may grow further still. After all, TXU, the biggest potential pending defaulter by a wide margin, has a covenant-heavy loan.
As for recovery levels, there doesn’t seem to be any daylight between covenant-lite and covenant-heavy loans. Take the average price of loans at default. Since 2007, the average for covenant-lite loans is 52.3 cents on the dollar, one cent inside the 53.3 average for covenant-heavy loans. – Steve Miller