Risk has become the talk of the leveraged finance markets.
Over the past year, market observers of many stripes have been increasingly vocal about the rising leverage in the credit markets, and the prevalence of covenant-lite loan structures has only further raised the volume of chatter.
But it is important to remember that in the rush to assess the health of this long-standing bull market, cov-lite in and of itself is not a sign of credit risk. Cov-lite simply means that maintenance covenants—a lender’s early warning system—are no longer present, and that the loan market has adopted a bond market approach to covenants.
If the underlying credit is healthy and the business model supports the issuer’s ability to manage debt, the loss of maintenance covenants is moot. However, if credit quality is poor and market conditions weaken the performance of borrowers, cov-lite makes it more difficult for lenders to identify—much less intervene with and influence—weakening credits.
Cov-lite credits are not new. They have been in the loan market for the past 20 years and represented nearly a quarter of the pre-crisis market. Cov-lite was initially a feature for healthier issuers—those with stronger credit profiles who had limited risk of tripping covenants. The removal of maintenance covenants has been a feature of revolvers and term loans historically, and, to this day, both first-lien and second-lien credits can be cov-lite.
In the years leading up to the pre-crunch bull market, LCD tracked a handful of cov-lite transactions, with their numbers growing as the market ramped up. By 2006 they represented 6% of transactions. A year later, just before the market imploded, 22% (198) of new-issue transactions were cov-lite. As a result, that same year, 17% of the S&P/LSTA Leveraged Loan Index was cov-lite, triple the share in 2006.
To be sure, cov-lite does run today’s market. Some 85% of institutional new issuance is cov-lite, as is 80% of the par outstandings underlying the S&P/LSTA Leveraged Loan Index. Of the nearly 1,100 tranches in the Index, 983 are cov-lite. Investors have little option but to accept this convergence with the bond market, with regards to credit terms.
Knowing that the loss of maintenance covenants has become, basically, a foregone conclusion, it is important to dig deeper into cov-lite metrics. Just how big, bad, and risky are these credits, versus the minority that include a full covenant package?
It should come as no surprise that cov-lite deals today are larger than those with full covenants. For all transactions in the year to date, the average deal size for cov-lite is $680 million, 56% greater than the $430 million average for full package credits. The disparity increases significantly if we exclude refinancings. In that case, the average deal size for cov-lite declines to $580 million, but the average for full-package credits drops to roughly half that, at $300 million.
Throughout its history, cov-lite has outsized full-covenant deals—even in pre-crisis years. Over the past decade, however, the gap between the two has grown steadily, to its widest point today.
At the same time, cov-lite’s credit stats, in the form of leverage and interest coverage, are weaker than those for full-covenant deals. Leverage is higher and coverage ratios are thinner for cov-lite deals—though the gap is narrowing.
Currently, leverage on cov-lite deals is at post-crunch high. The average pro forma total debt/EBITDA for cov-lite deals (excluding refinancings) is 5.7x, or 33 bps higher than the 5.4x for full-covenant credits. This gap is slightly higher than 2017’s 23 bps, but much lower than the gaps seen in the years before then, at 50–90 bps. The narrowing gap is mostly due to the rise in leverage in the full-covenant deals. These have risen nearly a full turn over the past eight years, from 4.4x. The average for cov-lite has risen about half a turn over that time, from 5.3x.
Coverage ratios have behaved similarly over time, converging as ratios decline to post-crunch lows. Currently, excluding refinancings, the average coverage ratio for cov-lite loans is 3.1x, versus 3.3x for full-covenant credits. This is a wider gap than last year, when they were roughly similar. But in 2017 the average coverage ratio was also about 20 bps higher in both segments. Over the last four years, as LIBOR has steadily risen, the average coverage ratio for cov-lite has fallen from about a third of a turn, from 3.5x. The average coverage ratio for full-covenant deals has fallen twice that amount, from 4.1x.

On the reward side of the equation, the pricing on cov-lite is also lower, and since 2006, on average, cov-lite has rarely carried a premium to full-covenant deals. In the year to date, the average TLB spread for cov-lite is 337 bps, 75 bps lower than the 412 bps average for full-covenant TLBs. This gap is about the same as the 78 bps in 2017, when spreads were slightly wider. And over the past few years, as LIBOR has risen, the gap has narrowed from about 101 bps in 2015, when average TLB spreads were 60–80 bps higher. The narrowing of the gap is due to spreads tightening in both segments, but more so on the full-covenant package side.

The gap in pricing exists in both the higher- and lower-rated credit segments, with the better-rated BB/BB– segment having a narrower gap than the B+/B segment. The average TLB spread gap for BB/BB– tranches was 62 bps, based upon 268 bps for cov-lite and 330 bps for full-covenant tranches. This is slightly wider than the 56 bps gap in 2017 and the 45 bps gap in the years before. One would assume that the prevalence of cov-lite deals in the past five years would give more fuel to the refinancings that have driven down institutional spreads over that time. However, excluding refinancings from the analysis drives the gap out to 114 bps, based upon 278 bps for cov-lite and 392 bps for full covenants. This is more than 20 bps wider than the 91 bps gap in 2017 and more than triple the 34 bps in 2016.
For B+/B credits, the behavior of the spread gap is more consistently wide, and the exclusion of refinancings has less impact. Year-to-date, the gap between cov-lite and full-package TLB tranches is about 90 bps, based upon relative averages of 368 and 458 bps. This is slightly higher than the 82 bps in 2017 but lower than the 106 bps in 2016. Excluding refinancings widens the gap slightly, but not as significantly as with the better-rated credits. Without refinancings, the average gap in B+/ TLB spreads year to date widens to 117 bps, versus 97 bps in 2017 and 89 bps in 2016.

As a result, when we look at how spread per leverage has declined in the market, cov-lite credits are leading the way. The average spread per leverage, excluding refis and based upon TLB spreads, is 64 bps for cov-lite credits, 24 bps tighter than the 88 bps average for full-covenant deals. The richest SPL figure in either segment was in 2009, when it was 135 bps for cov-lite and 175 bps for full-covenant credits. However, at the time, the market was just regaining its footing, and SPL has steadily dropped since then. Over the past few years, the gap between the two seems to have settled at roughly 25 bps.

However, it is important to keep in mind that the average spread per leverage in the market today—cov-lite or full-covenant—is generally 50% higher than it was in the pre-crisis era (and that is before factoring in a rising LIBOR). Risk metrics are higher and spreads are thinner today than they were five years ago, and cov-lite may lead the way in those trends, but the market has moved in that direction across the board.
Further, in assessing the additional “risk” that cov-lite may incur due to the loss of the canary in the coal mine, one other key factor about cov-lite is critical: It is almost exclusively a rated universe. Only 1.3% of cov-lite deals lack a bank loan rating, compared with 23% on the full-covenant credits. This factor brings the cov-lite segment of the loan market even closer to the bond market in that, while it currently has set aside the requirement of maintenance covenants, it has ensured that the general health is still being monitored through the rating process.
Source：LCD