The CDO market is slowly picking up steam, with 15 transactions issued since 2015, according to LCD. This year alone has featured in excess of $3.5 billion of collateralized debt obligation activity, more than double that of each of the past three years.
For many, the term CDO immediately evokes the financial crisis, with banks and investors ultimately having to write down tens of billions in losses over the ensuing years.
Consequently, since the crisis, CLOs—which share with CDOs two letters and several structural features—have been working hard to distance themselves from their pre-crisis cousins. But now, GSO, the credit arm of private equity firm Blackstone, recently became the latest high-profile fund to issue its first post-crisis CDO, out of a platform it has named Cirrus Funding.
Other CDO issuers include Anchorage Capital Group, Brigade Capital Management, Fortress Investment Group, and Credit Suisse Asset Management.
Anchorage launched the first post-crisis CDO in Europe in August, and others in that region are expected to follow suit.
Not your pre-crisis CDOs
There are notable differences between the CDOs being issued now versus those from over 10 years ago.
The most noteworthy entails underlying collateral. Many pre-crisis CDOs were filled with subprime mortgage bonds. Subsequent litigation charged that the bulk of the mortgages were fraudulently underwritten, forcing many of the originating banks to eventually repurchase them, or pay large settlements to both investors and regulators.
Today’s iteration of CDOs instead invests strictly in the credit of corporate issuers, whose financials are audited. And in many cases those issuers have a running history with investors in either the high yield or leveraged loan markets.
Part of the growing appeal of the new CDOs is, in fact, due to the regulatory regime. Following the enactment of the Volcker Rule in 2014, CLO managers could no longer purchase a single high-yield bond if they wanted to sell to banks, one of the largest buyers of the senior debt tranches on a CLO. As a result, these new CDOs, which are issued without the intent to comply with the Volcker Rule, allow managers to have the flexibility to switch between investing in high yield bonds and leveraged loans. (The LSTA, incidentally, recently made its case to regulators that CLOs be allowed bond buckets of up to 10%.)
The primary buyers of these new CDOs are insurance companies in the U.S. and overseas. This investor base is particularly interested in these assets given their fixed-rate coupons, which offer an investment grade rating and a long duration matching their liabilities, and exceeding most yields elsewhere in the fixed-income universe.
On GSO’s Cirrus Funding 2018-1, which matures in 18 years, the Aaa tranche (Moody’s) offers a coupon of 4.80%, while the junior-most Baa3 debt tranche pays a coupon of 7.05%.
This time is different?
At first glance, the notion of CDOs receiving investment grade ratings might raise eyebrows, considering this debt’s prominent role in the financial crisis.
These new CDOs certainly are more risky than the CLOs currently in market. However, Moody’s analysts have required more safeguards with these vehicles, compared to CLOs.
Particularly noteworthy in this iteration of CDOs is the ability of managers to buy up to 70% of the portfolio in second-lien loans and/or unsecured or subordinated bonds, according to analysts at Moody’s, who’ve rated all of the latest CDOs.
CDOs are allowed to hold up to 17.5% of their portfolio in Caa rated assets and below, compared to the 7.5% in CLOs, according to Moody’s. A number of the CDOs already have CCC rated assets of roughly 10% in their portfolios, according to trustee reports.
But these CDOs are leveraged at 2–3x, versus the 10–13x leverage of most CLOs, sources say.
On the more conservative end of recent CDOs is the $327 million Brigade Debt Funding II CDO, which Moody’s assigned a weighted average rating factor (WARF)—a numerical estimate of a portfolio’s credit risk, with a higher WARF indicating more risk—of 2748, roughly equivalent to a B2 rating. It is important to note here that the portfolio was only 19% ramped at the time Moody’s looked at the portfolio. For comparison, the median WARF on CLOs issued in 2018 was 2760.
The most recent trustee report on the debut $408 million Brigade Debt Funding I issued in February meanwhile showed that its WARF was 3331.
In order to account for the higher allowance of higher-risk second-liens and unsecured bonds, the analysts at Moody’s have required managers to hold more credit protection to achieve a rating comparable to what they would normally assign a CLO.
For example, on the CDO tranches that have been rated Aaa by the Moody’s analysts so far, those tranches have another 52% of the debt stack below them can absorb any principal losses, compared to those that have 36% of debt below them to get a Aaa rating in CLOs. Lower in the capital stack, this amount is 29% on a Baa3 CDO tranche, compared to 13% for CLOs.
But perhaps forgotten over the years—again, the negatives associated with collateralized debt obligations have been stubborn—are the benefits of how CDOs are structurally similar to CLOs: Namely, they are able to obtain term funding, just on a riskier set of underlying assets. This assumes, however, that defaults are contained so that the CDO doesn’t fail a number of tests, which would cut off payments to the equity holders, many of whom are the managers themselves.
A key test that these CDOs have, compared to CLOs: at least one of them is based on the portfolio’s market value, including a metric known as market value overcollateralization (MVOC). This is important because the CDO can fail those tests if markets become more volatile, and the underlying loans and bonds start selling off, even if they manage to avoid default later. The thresholds for those tests range anywhere from 113–118% in excess of the total outstanding debt tranches.
History may not repeat, but does it rhyme?
CLO investors often point to how well the 2006 and 2007 vintages performed, even as the volatility thereafter was at times frightening, at one point shutting off the equity to nearly half of the CLOs, in 2009. The median cumulative equity distributions on 2006- and 2007-vintage CLOs ended up at roughly 220% and 239%, respectively, according to Morgan Stanley. But those returns did not come easy, as nearly half of CLOs were not receiving equity distributions, as they were failing key tests in early 2009.
A number of investors have pointed to some important differences this time around, however, namely that the credit deterioration then was focused primarily on household debt, instead of corporate borrowers. Today, of course, the amount of corporate borrowing and leverage levels has risen in recent years, while the widespread lack of covenant protections has yet to be tested.
About that corporate borrowing: The leveraged loan market, for one, had grown to about $1.1 trillion at the end of September from $600 billion in 2008, according to the S&P/LSTA Leveraged Loan Index. Today’s larger pool of loans contains a greater share that are lower-rated: 59% are rated B+ or lower, compared to 37% in 2008. There is as much paper outstanding rated B+ and below now as there was total paper in 2008.
The percentage of lower-rated debt continues to be watched because CLOs, and the new CDOs, which are the natural buyers of this debt, have limits on how much they can hold, as mentioned earlier. – Andrew Park
LCD comps is an offering of S&P Global Market Intelligence. LCD’s subscription site offers complete news, analysis and data covering the global leveraged loan and high yield bond markets. You can learn more about LCD here.