In 1997, a financial contagion swept through the banking system in Southeast Asia, causing a credit crisis and initiating a worldwide economic slowdown.[2] The then-dubbed “Asian flu”[3] threw financial markets as far as the U.S. into a frenzy, even causing the New York Stock Exchange (NYSE) to activate its “circuit breaker” rules for the first time, and to temporarily close the trading floor in an effort to curb panic trading.[4]
For the first time since 1997, the NYSE circuit breaker protocols were again activated this March. They were activated four times.[5] The coronavirus pandemic has wreaked havoc through the world’s financial markets, and brought the global economy to a projected contraction for 2020, with the U.S. economy now expected to contract by 6% this year.[6]
Government-mandated shutdowns in response to this unprecedented public health emergency have put a strain on businesses, dramatically increasing the chance that large companies across many industries will default on their credit obligations, and be forced into financial restructuring or bankruptcy.[7]
In an attempt to avoid a liquidity crunch, the U.S. Congress approved the largest stimulus package in U.S. history, the Board of Governors of the Federal Reserve System (the “Fed”) slashed interest rates to between zero and 0.25%, regulators encouraged banks to “use their capital and liquidity buffers,”[8] and both the Secretary of the Treasury and the Fed started to quickly inject liquidity across the economy through direct cash grants and quantitative easing.[9] Aggressive monetary policy, however, is unlikely to prevent any pre-virus market bubbles from bursting, and, naturally, not meant to stop the riskiest investments from generating losses.
For a few years now, market analysts[10] and regulators[11]—including a sitting senator and former bankruptcy professor[12]—have been describing the $1.2 to 1.5 trillion leveraged loan market[13] as a ticking time bomb likely to play a role in causing the next financial crisis.[14] In a future economic slowdown, an increased rate of defaults and rating downgrades of leveraged bank loans and loan-backed securities (best known as Collateralized Loan Obligations or CLOs) would “induce fire sales and further depress prices” potentially bringing down overexposed large financial institutions, and freezing corporate credit markets.[15]
Eerily, as economic activity was intentionally grinded to a halt in an effort to mitigate the coronavirus spread, the voices forecasting the implosion of leveraged loans and loan-backed securities also grew louder in predicting possible contagion and cascade effects spreading across the financial institutions that hold them.[16] These voices are denouncing problematic fundamentals in the current leveraged loan market and comparing them to those that caused the 2007 subprime mortgage crisis to spread into a full-blown financial crisis towards the second half of 2008.[17] Two legal innovations in the world of corporate finance are the main anticipated culprits of a future financial meltdown: covenant-lite (or “cov-lite”) loans and CLOs.[18]
Below we explain why concerns of a systemic breakdown in the financial system as a result of cov-lite loans and CLOs may be overstated. In addition, we argue that a third innovation, borrower-friendly EBITDA add-backs, while more likely to distort credit monitoring, seems to be currently removed from the ratings analysis that goes into leveraged loans and CLOs.
II. The perceived problem: loose credit standards, dangerous financial engineering, and industry-friendly accounting
A. The apparent loose credit standards of cov-lite loans
Leveraged loans are large extensions of credit made to a corporation with below investment grade rating (BB+ or lower) or that leave the borrower with a level of debt-to-equity ratios that “significantly exceeds industry norms for leverage.”[19] In practice, these complex debt structures are mostly used to finance large corporate transactions such as a mergers and acquisitions (M&A), leveraged buyouts (LBOs), recapitalization (share repurchases) or refinancing of debt.[20]
Cov-lite loans are leveraged loans that contain weak or no financial covenants. They specifically lack maintenance covenants.[21] As a general rule, lenders prefer strong financial or maintenance covenants because they allow them “to take action earlier if an issuer experiences financial distress” or “to wrest some concessions from an issuer that is in violation of covenants (a fee, incremental spread, or additional collateral) in exchange for a waiver.”[22]
Currently, cov-lite loans account for over 85% of the total leveraged loan market, having grown dramatically from 1% of all leveraged loans in 2005, to 25% in 2007[23] to 64%[24] in 2015. Commentators argue that the prevalence of cov-lite loans makes the leveraged lending market financially unstable.[25] They assert that the reason for the proliferation of cov-lite agreements is the hot debt market spurred by low interest rates and investor yield cravings, and that this has caused the loosening of time-tested, responsible lending practices.[26]
One commentator described the perceived negative effect of cov-lite loans this way:[27]
Before the current debt binge, investor protection covenants used to be strict. A company, for instance, might be barred from taking on debt above a certain point, or perhaps be required to maintain a certain debt-to-cash-flow ratio. Once the issuer violated the covenants, then creditors could demand repayment of their principal immediately.
No more. Today, cov-lite, where the customary safeguards are minimal or nonexistent, is a bigger trend than plant-based meats, at least in the debt world. Currently, 80% of leveraged loans and junk bonds sport weak or no covenants, compared to 5% right before the financial crisis.
So why would anyone want to invest in a cov-lite issue, other than perhaps out of ignorance? The reason is money: cov-lite debts pay their investors higher interest rates, around 0.9 percentage points more than traditional covenant-bound debt, due to the higher risk. The problem is, without adequate covenant safeguards, troubled companies can grow even sicker and lurch along, zombie-like, until they collapse, delivering still more pain to debt holders.
B. The “dangerous” financial engineering behind CLOs evoke concerns of CDOs
Close to 60% of the totality of leveraged loans are then pooled, packaged into Collateralized Loan Obligations (CLOs) and sold as securities in the world’s capital markets.[28] The rise in popularity in CLOs has in fact been an important driver for the leveraged loan market as a whole to recently reach the $1.2 trillion mark, after a “post-crisis low of around $500 billion in 2010.”[29]
CLOs are “similar in concept to mortgage bonds that pool a large number of loans into a single security to provide diversification.”[30] CLOs are in fact a subcategory of Collateralized Debt Obligations (CDOs), with the main difference being that while CDOs are typically composed of bonds or other CDOs, CLOs contain corporate bank loans.[31]
In the leadup to the 2008 financial crisis, billions in CDOs securitizing subprime mortgages were sold throughout the financial system. Despite largely dealing with speculative grade—or “junk”—underlying assets, the higher tranches of these CDOs obtained investment-grade AAA ratings through the use of a financial technique known in the securitization industry as “credit enhancement.”[32] This “financial engineering” mechanism has also been widely used in the origination and packaging of CLOs and so this has raised the alarm that an outcome similar to that of the financial crisis may follow suit.[33]
Pre-2008, the below-investment grade tranches of CDOs—those that didn’t make the AAA rating of the senior tranches as a result of the credit enhancement process—were then repackaged into new CDOs (called “CDO-squared” or, if again resecuritized, “CDO-cubed”) and, through another round of credit enhancement, they were again resold with investor-grade ratings through the financial system.[34] In addition, “synthetic CDOs,” a derivative created for investors to bet around the performance of actual CDOs, were also traded widely through the financial system. Finally, Credit Default Swaps (CDSs), also a derivative, were created as a way to insure physical and synthetic CDO holders against the risk of default, thus increasing the complexity, the size of the exposure, and further spreading the risk of a subprime mortgage market implosion.[35]
As real estate prices began to fall and mortgage default rates to rise, ensuing collateral calls on rapidly deteriorating CDOs and payment demands on CDSs ultimately caused large asset write-downs to the balance sheets of highly interconnected large financial institutions, causing cascade failures and a liquidity drought among large financial institutions and initiating an unprecedented credit crisis. By the end of 2008, Bear Sterns had collapsed and been absorbed by JP Morgan, Lehman Brothers had filed for bankruptcy, Merrill Lynch had agreed to be acquired by Bank of America, and AIG had been bailed out by the Federal Reserve Bank of New York.[36]
As a former legal practitioner explained,[37]
…interests in questionable real estate loans were widely dispersed and had proliferated not only throughout the traditional banking sector in the U.S. and abroad but among non-traditional market participants—the so-called “shadow banking” sector—and were very difficult to trace. It was not possible to “contain” the problem—for example by facilitating the writedown and restructuring of bad loans by banks holding mortgages directly. Instead these loans had been securitized, in some cases through multiple steps and vehicles, so that indirect interests in a single loan might be held (and borrowed against) by a number of different institutions.
In the movie “The Big Short,” synthetic CDOs are labeled an “atomic bomb,”[38] while back in 2002 Warren Buffett said that derivatives—a category that includes synthetic CDOs and CDSs—were “financial weapons of mass destruction, carrying dangers that, while now latent, are potentially lethal.”[39]
The structural similarity between CLOs and CDOs[40] causes concern for a disastrous repeat:[41]
The big danger of lev loans lies in the eerie similarity the CLOs bear to the mortgage securities instruments that imploded during the financial crisis. Housing boom mortgages were collected into bonds, which in turn were placed into CLO-like instruments, known as collateralized debt obligations or CDOs. When the mortgages went belly up, a lot of CDOs imploded, leaving investors with nothing, other than regret.
In 2018, ten years after the implosion of CDOs triggered the great financial crisis, a record $128.2 billion CLOs were issued in the U.S.[42] By the end of the year, there were a total $600 billion outstanding CLOs in the country.
C. The industry-friendly EBITDA add-backs
Earnings before interest, tax, depreciation and amortization (EBITDA), is an accounting metric meant to “indicate a company’s ability to generate (gross) cash flow and, implicitly, service debt,”[43] while “eliminat[ing]the effects of financing and accounting decisions.”[44]
In the context of large M&A and LBO transactions, it is common practice for financial advisors and rating firms at the time of the financing to allow borrowers to carry out upward adjustments to future expected EBITDA.[45] These adjustments known as “EBITDA add-backs” are carried out under the notion that the synergies or operational improvements to be achieved as a result of these transactions will increase earnings.[46] The problem is, however, that borrowers have an incentive to overstate EBITDA, and thus understate the debt-to-EBITDA ratio in an attempt to increase their flexibility under the loan’s covenant vis-à-vis the monitoring lender[47] and to “improve the attractiveness of their debt issuance when they go to market.”[48]
And the adjustments are not small. The Financial Stability Board report estimates that “incurrence covenants in current deals are subject to EBITDA adjustments of 15-30%.”[49]
In 2018, researchers at S&P Global analyzed the actual performance of these forecasted EBITDA add-backs in major leveraged loan transactions, such as large LBOs and M&As.[50] Their report concluded that “companies and deal arrangers have become increasingly creative in presenting what qualifies as an add-back, resulting in an increase in both the number and types of adjustments.”[51] The report found that, (a) management-adjusted EBITDA including add-backs was not a good indicator for future EBITDA; (b) companies overestimate debt repayment; (c) combined, these effects understate future leverage and credit risk; and (d) add-backs also present incremental credit risk in the form of future event risk because covenants that rely on EBITDA are likely to provide additional flexibility under negative covenants and restricted payments of dividends, debt and lien allowances, etc.[52]
A 2019 follow-up report by the same rating firm, confirmed that EBITDA add-backs, which they also refer to as “marketing EBITDA,” continued to be substantial and overstated:[53]
Aggressive EBITDA adjustments have understated high leverage and purchase price multiples. Our study led us to several conclusions consistent with our previous study of the 2015 cohort: marketing EBITDA including add-backs is generally not a good indicator for future EBITDA and companies tend to overestimate debt repayment. These effects understate future leverage and credit risk, and add-backs also present incremental credit risk in the form of future event risk since covenants that rely on EBITDA may provide additional flexibility under negative covenants and restricted payments (e.g. dividends, debt, and lien allowances).
III. The reality of cov-lite loans and CLOs
A. Underwriting standards: cov-lite loans aren’t really “lite”
A recent seminal report by a team of researchers at the Federal Reserve Bank of Philadelphia has found that the mainstream understanding that cov-lite loans are borrower-friendly, excessively risky, or lacking in investor protections, is erroneous.[54] Instead, the report found that 95% of these cov-lite loans were issued as part of a larger loan syndication deal where the borrower remained constrained by financial covenants, and where a leading bank remained in charge of monitoring compliance with such covenants on behalf of all lenders.[55]
Moreover, the October 2019 report rejects “supply-side explanations” that suggest that the rise in cov-lite loans is the result of competition among lenders and an ensuing race to the bottom in lending standards.[56] Instead, the report concludes that cov-lite loans emerged in the early 2000s to concentrate monitoring powers on the lender with most skin in the game, namely, the bank arranging the syndicate, and that the purpose for this new design was to reduce renegotiation costs for the benefit of all lenders, in a context where nonbank institutional investors—like mutual funds and CLOs—started participating consistently as ultimate investors in these transactions.[57]
According to the report, the fundamental mistake in the analysis by the majority of commentators is that they look at specific cov-lite loan contracts in isolation and fail to see the larger loan syndication structure under which they are issued.[58]
By looking at a large set of loans issued between 2005 and 2014 and focusing on the totality of leveraged loans issued by each publicly-traded corporation analyzed, the report discovered that virtually all cov-lite loans were issued as part of a larger syndication deal[59] that includes or is preceded by a “revolving credit facility” or “revolver.”[60] And the revolver loan invariably contains the typical financial covenants—including the maintenance covenant explained above[61]—and so it forces the borrower to remain disciplined by allowing the lender to continuously assess credit risk and intervene in case of covenant violations.[62]
The revolver is an open credit line—the equivalent of an individual’s credit card—granted by a group of lenders (a syndicate of banks), that is typically arranged by one leading “agent bank” who “leads the negotiation on behalf of the syndicate lenders, [and]typically has a large stake of the loan both absolutely and proportionally.”[63]
As part of the same deal, a second loan facility or set of facilities consist of a cov-lite “term loan” or series of term loans, which are meant to be sold directly or through CLOs to institutional investors such as hedge funds, pension funds or insurance companies.[64]
According to the Philadelphia Federal Reserve researchers, this two-tiered loan deal consisting of revolvers with covenants, followed by cov-lite term loans—an arrangement they call “split control”—evolved organically in the early 2000s and became prevalent after the financial crisis in order to reduce renegotiation fees[65] and to “mitigate bargaining frictions in loan deals with institutional lenders.”[66]
The practical result, then, is that the borrower of cov-lite term loans in reality remains subject to the same strictures included in the revolver loan, namely, to the same supervision by the agent bank who maintains the most skin in the game.[67] Not all lenders have the right to enforce, amend or waive compliance with the covenant—or the duty to monitor it—because only the revolver lender who is also the agent of the bank syndicate has this exclusive right.[68]
Under this arrangement, “[the]requirements placed upon borrowers [limiting]leverage or overall riskiness benefit both bank and non-bank lenders,”[69] as both revolver and term loan lenders maintain the same level of seniority and so will recover equally in the event of a default that prompt financial restructuring or, in the worst case scenario, liquidation in bankruptcy.
Ultimately, “pension funds and insurance companies that do not regularly engage in credit analysis ultimately benefit from the expertise of banks,” as they continue to exercise oversight and fulfill their credit monitoring role.[70] Moreover, the report concludes that banks have continued to actively and efficiently exercise these roles[71] and that the frequency of covenant violations as well as renegotiations with borrowers have remained similar under “split control” arrangements, as it was under older deals when cov-lite loans were not used extensively and institutional lenders were involved in the renegotiations.[72]
B. Cov-lite loans are largely structured as, and managed by, CLOs
Once a syndicate’s agent bank originates the leveraged loan deal, the chief vehicle by which these leveraged loans—62% of them worldwide today[73]—trade in the secondary market and reach institutional investors is a bankruptcy remote special purpose vehicle known as a CLO.[74] As the researchers of the Philadelphia Federal Reserve put it, “[i]nstitutional tranches of leveraged loans have a larger and more diverse set of lenders, […] and a secondary market that permits lenders to change during the life of the loan.”[75]
Researchers for the Congressional Research Service have explained the role of the originating bank with great clarity:[76]
The institution that originates a leveraged loan rarely, if ever, subsequently holds the loan entirely on its own balance sheet, because a lender often would be wary of taking on a large exposure to a single highly indebted company. Instead, the originating lender typically will either (1) partner with colenders, (2) sell pieces of a single loan to investors, or (3) bundle part or all of the loan into a pool of other leveraged loans in a process called securitization, then sell pieces of the pool to investors. The first two options [are]referred to as syndication and participation, respectively […]. The third option creates securities called collateralized loan obligations (CLOs), [which it then sells to]insurance companies, pension funds, mutual funds, hedge funds, and other private investment funds.
To avoid confusion, the “CLO” denomination applies both to the “fund”-like vehicle that structures and manages the diversified pool of assets and pays investors under the CLO’s priority terms or “tranches,” as well as to the “securities” or notes that a CLO issues to investors. Tranching is the process by which the assets—the different loans—purchased from the originator bank and pooled into the CLO are structured in order to achieve different risk profiles that are then converted into payment priorities and different credit ratings.
Tranching is the key behind the “credit enhancement” technique of “subordination,” under which the CLO manager won’t stop making any payments to upper tranches until the subordinated tranches have borne all the losses.[77] As explained above, credit enhancement fulfills the purpose of turning a bundle of similar speculative-grade loans into a waterfall of different credit ratings. A CLO will typically have a top tranch of senior notes rated AAA and AA, a middle tranch of mezzanine securities rated A/BBB/BB, and a bottom tranch of so-called “equity” securities—called that way because their high riskiness makes them closer to a stock, and even affords them a different tax treatment.[78]
As other structured finance products,[79] CLOs are popular with originators—in this case, the banks that originated the leveraged loans—for many reasons, including that securitization provides them with the benefits of (a) an efficient funding source, as they stand to turn the entire term loan amount into cash in only a few months;[80] (b) balance sheet relief, as the large relatively illiquid leveraged loans are removed off their books no longer conditioning a bank’s mandatory capital and liquidity ratios;[81] and (c) an extra source of cash, as the bank itself, which charges a fee for arranging the entire loan syndication, now stands to charge another fee for itself structuring, or helping the CLO to structure the deal, and for “servicing” or managing amortization and interest payments under the original loans for the benefit of the CLO.[82]
Also like other types of securitizations, CLOs are popular with investors because they provide them with an investment opportunity in: (a) a security that is highly diversified among many issuers and industries; (b) a security that is senior secured and so equal in priority to a lead agent bank’s revolver loan, as opposed to corporate bonds that are subordinated and unsecure; (c) a security of relative high yield compared to investment-grade bonds; and (d) a debt security that even if a wave of defaults occurs within the asset class it is unlikely to affect the investor’s recovery rate, particularly if the investment was made in the CLO’s highest tranches.[83]
C. Conservative financial engineering: Today’s CLOs are less risky than CDOs
The credit enhancement and tranching processes, by which a speculative-grade loan—some call it “junk loan”[84]—is effectively converted into a AAA-rated security, increases the concern that we may be looking at the same shaky fundamentals behind CDOs and CDSs that caused the great financial crisis. However, while CLOs are in fact very similar in nature to CDOs, there are a few substantial differences between today’s CLO structure and ecosystem, and the pre-financial crisis CDO market.
First, while equally benefitting from a diverse pool of underlying assets, CLOs today are not further broken down and dispersed through the financial system the way CDOs used to be. In a practice that proved to obscure ultimate ownership of its riskiest tranches, many CDO equity tranches went through a second round of credit enhancement and were turned into a “CDO-squared” (again having a top AAA tranch, followed by lower ones), and then many of these new CDOs’ lower tranches were again enhanced and turned into “CDO-cubed.” There are currently no CLO-squared or CLO-cubed in the market for CLOs, so their structures are simpler for rating firms and investors to understand and agree on, and the ownership rights—and opportunity for losses—represented by them are easier to identify.
Second, derivative equivalents to “synthetic CDOs,” which were used to track and bet on the performance of actual or “physical” CDOs, do not currently exist in the CLO market. Similarly, credit default swaps (CDSs), a much more common type of derivative, but which was widely used prior to the financial crisis to hedge the risk of default not only of “physical” CDOs but also of “synthetic” CDOs, do not perform an equivalent role today in the CLO market. This type of obscure and large risk dispersion is what Warren Buffett was referring to in the 2002 letter to his company’s shareholders, where he called them a weapon of mass destruction, after criticizing the “burgeoning quantities of long-term derivatives contracts and the massive amount of uncollateralized receivables that are growing alongside.”[85]
As researchers for the Bank of International Settlements have put it,[86]
CLOs are backed by simpler, more diversified pools of collateral than CDOs. CDOs issued in the run-up to the [great financial crisis]consisted mainly of subprime MBS, and CDOs backed by other CDOs (so-called CDO-squared) were common. In 2006, almost 70% of the collateral of newly issued CDOs corresponded to subprime MBS, and a further 15% was backed by other CDOs. Furthermore, more than 40% of the collateral gathered by the CDOs issued that year was not cash MBS, but CDS written on such securities. When conditions in the housing market turned, the complexity and opacity of CDOs amplified financial stress. In contrast, CLOs are much less complex. Their collateral is diversified across firms and sectors, and the known incidence of synthetic collateral or resecuritisations is minimal.
Finally, even though CLOs performed well during the 2008 crisis,[87] the structuring of CLO tranches today is more conservative than it was years prior, meaning that holders of higher tranches today are less likely to realize losses compared to a few years back:[88]
Unlike some securitized sectors that suffered from massive principal losses, CLO structures proved more resilient during the crisis, and they’ve benefited from structural improvements since then. Today’s CLOs, loosely referred to as CLO 2.0, have a more robust design than those of the 1.0 vintages. The latter commonly carried larger highly rated tranches, so the layers of lower-quality pieces ready to absorb losses in the event of loan defaults were meaningfully smaller than those of today’s CLOs. There are more nuanced features as well, including a tendency to hold fewer of the junkiest deals and nonloan holdings. Today’s loan market also isn’t as exposed to the multitude of related synthetic derivatives that dragged down the sector during the global financial crisis, and it’s extremely rare to find derivatives linked to loans in mutual funds today.
IV. Default rates and recovery expectations for cov-lite loans and CLOs
A. Strong historic performance
In 2019, and “[d]espite mounting concern from high-profile regulators concerning the asset class, the default rate on U.S. leveraged loans [80% of which are estimated to be cov-lite] dipped to 0.93%, its lowest level in almost seven years.”[89] That year, the U.S. leveraged loan default rate was “holding near multi-year lows, and [was]markedly below the 2.93% historical average, according to the S&P/LSTA Index.”[90] The U.S. leveraged loan default rate increased to its highest level in 15 months in January 2020, but “at 1.83%, the rate remain[ed]stubbornly low.”[91]
According to Bain Capital, default rates were lowering because of the prevalence of cov-lite loans, which were also likely to negatively affect recovery rates:[92]
Counterintuitively, it is our view that covenant-heavy lending agreements actually triggered a higher total default rate during the [great financial crisis]. The existence of covenants created multiple near-term triggers that could push a company into a restructuring.
…
We expect overall defaults will be lower given fewer companies will have covenant-based triggers to enter bankruptcy: The result will likely be a binary default and recovery pattern – some companies will survive that otherwise would not, while others will delay an inevitable outcome, possibly leading to lower recoveries.
The Philadelphia Federal Reserve researchers agreed that in split control structures—where cov-lite loans are sold to secondary market institutions—the controlling bank agent was less likely to force a distressed borrower into bankruptcy and more likely to renegotiate credit conditions.[93] However, regarding the forecast that cov-lite structures will possibly reduce recovery rates, they thought that the sample size of corporate defaults was still too small for meaningful comparison on recovery rates under the old system and the cov-lite system to be made.[94]
S&P Global, on the other hand, has been more assertive on the positive track record of cov-lite structures, stating that “[h]istorically, recoveries in cases of default on cov-lite loans have been on par with that of traditionally covenanted credits.”[95]
CLOs, on their part, have also performed very strongly, according to a 2014 report from S&P Global that looked at 20 years of CLO performance from 1994 to 2013.[96] The report documented few negative rating actions—or downgrades—on senior notes due to underlying collateral deterioration, few defaults, and minimal loss rates since the agency started rating the asset class in the mid-1990s.[97]
In 2018, global CLOs underwent more upgrades than downgrades and only four defaults. “This positive CLO credit performance echoed that of speculative-grade U.S. and European corporates. Globally, the CLO default rate fell to 0.12% in 2018 from 0.16% in 2017, and similarly the corporate default rate fell to 1.0% from 1.2%.”[98] Moreover, “no investment-grade CLOs (those rated BBB- or higher) were downgraded during the year.”[99]
Even before the coronavirus hit, however, analysts continued to project that “recoveries on [leveraged loans], including those held by CLOs, will be less than what has historically been the case.”[100] Moreover, like Bain Capital above, they attributed these projections to the prevalence of “cov-lite and otherwise less-restrictive documentation on today’s leveraged loans.”[101]
In a similarly pessimistic vein, in its 2019 report, the Financial Stability Board found that “[a]number of factors suggest that vulnerabilities in the leveraged loan and CLO markets” have grown since the global financial crisis:[102]
The degree of borrowers’ leverage has increased and, although loans tend to have lower credit ratings, there is some evidence that certain changes to loan documentation that weaken creditor protection are not fully priced in by market participants and investors. This has the potential to not only increase default rates and decrease recovery rates for leveraged loans, but also to exacerbate investor reactions to shocks.
B. Weathering the coronavirus “perfect storm”[103]
As the coronavirus pandemic has swept through the U.S. economy, rating agencies have been announcing downgrades or negative outlooks on specific leveraged loan classes, with a particular focus on industries that are most at risk as a result of the pandemic, notably airlines, gaming, lodging & leisure, and metals & mining.[104]
For example, on April 7, Fitch Ratings updated its sensitivity stress scenario to potentially review ratings on CLOs exposed to the automobile industry and corporations on negative outlook, and projected that default rates would continue to increase and recoveries to shrink:[105]
This results in average default projections within portfolios of 7% for the first year and 8% for the second year for EMEA CLOs, and 7.5% and 9% for US CLOs. Fitch believes this to be broadly in line with the base scenario as published by its EMEA and US leverage finance teams (see “Europe’s High-Yield Default Rates Rise as Credit Cycle Turns” and “Fitch U.S. Leveraged Loan Default Insight”).
In addition, preexisting pricing pressures on the oil & gas industry have only been compounded by the pandemic.[106] Despite this, the latest forecast by S&P Global is generally optimistic regarding CLO structured diversification and fundamentals:
The economic stress caused by COVID-19 and the related social distancing measures (not to mention the stress in the energy sector related to the precipitous decline in oil prices) can play out in many different ways. But in all of the wide range of stress scenarios outlined in this article, the fundamentals of the CLO structure work to protect the senior tranches. Projected rating changes at the “AAA” CLO tranche level show nearly 99% of the ratings either being affirmed or downgraded by one notch (to “AA+”) even under our most punitive hypothetical scenario of 20% loan defaults and 40% CLO “CCC” buckets. And no CLO tranche rated in the “A” category or higher experienced a default under any of our hypothetical scenarios. Further down the capital stack, the results become more negative (as expected under stresses as significant as these), with “B” and “BB” rated CLO tranches showing the most adverse outcomes and “BBB” and higher rated tranches showing successively better results.
While these projections appear optimistic, ultimate default and recovery rates are very much uncertain as a big portion of the world and the country remains in lockdown. Only when the uncertain weeks and months ahead are behind us, may we find out how the leveraged loan market, as captured in highly-diversified CLO structures, will ultimately weather this global storm. Default and recovery rates will be telling, and, at that point, comparisons may be drawn with the their historic performance as well as with the performance of other asset classes and markets during the crisis.
V. Systemic risk: G-SIBs are better prepared for loans and CLOs to go bust
Today’s $1.2 to $1.5 trillion leveraged loan market—which includes $600 billion in CLOs—is only a small portion of the total corporate loan market, which is estimated at $10.1[107] to $15.5[108] trillion. The pre-coronavirus U.S. stock market was two-to-four times as big with a $39.6 trillion total market capitalization before the coronavirus panic hit.[109]
Even with the March wild fluctuations,[110] as of April 27, 2020, the total U.S. stock market was sized at approximately $33.3 trillion[111]—still two-to-three times larger than the total corporate loan market and almost ten times as large as the leveraged loan market. Compare these proportions with the size of the CDS market before the great financial crisis. In 2007, that market alone stood at $60 trillion, which was almost three times the size of that year’s $22 trillion stock market.[112]
CDOs, on the other hand, were worth about $641 billion prior to the financial crisis, but the U.S.’s largest financial institutions were so exposed to them that they had to bear the brunt of the losses that ensued. This is what a team of researchers for the Federal Reserve Bank of Philadelphia discovered in their post-mortem analysis of the great financial crisis:[113]
As for the dealers at the center of SF ABS CDO issuance, our results support the hypothesis that most were not fully aware of the risks in the CDOs, since the dealers that underwrote the worst-performing CDOs (Morgan Stanley, Citicorp, Bear Stearns, and UBS) all suffered large and debilitating losses from the “super-senior” AAA bonds of the CDOs they underwrote and held (see FCIC (2011) and Lewis (2010) for a list of firms that held CDO risk). Goldman Sachs and Deutsche Bank, which we later learned were selling off their risk and shorting the subprime mortgage market, were 7th and 11th in terms of rank and not statistically different in terms of write-downs from the small issuers. This makes sense given the size of the market. To absorb $641 billion in SF ABS CDOs required the participation of the largest players in the financial system.
This is the backdrop against which current efforts to closely monitor systemic risk must be considered.
A. The Financial Stability Board’s 2019 report
In December 2019, the Financial Stability Board (FSB)—an organization hosted by the influential Bank of International Settlements, and which gathers key international central bank officials—found that, globally, “banks have the largest direct exposures to leveraged loans and CLOs.”[114] Direct exposure to leveraged loan and CLO markets is “highly concentrated in a limited number of [global systemically important banks or]G-SIBs.”[115] The United States has eight G-SIBs: Bank of America, Bank of New York Mellon, Citigroup, Goldman Sachs, JPMorgan Chase, Morgan Stanley, State Street, and Wells Fargo.[116]
A big difference with the pre-2008 environment, however, is that—unlike with synthetic CDOs and CDSs—the exposure of both bank and non-bank financial institutions is a bit more transparent today. Using supervisory and market data, the FSB report was able to identify “the direct holders of roughly 79% of leveraged loans and 86% of CLOs.”[117] Also, according to an S&P March 2020 report, U.S. banks today own roughly $99 billion in CLO securities.[118]
Banks also have indirect exposure as they extend credit to intermediaries that invest in these markets.[119] According to the FSB, the resilience of banks is believed to have significantly improved since the financial crisis, in part, as a result of the Volcker Rule.[120] However, “concentration and indirect exposures could still make adverse developments in the leveraged loan and CLO impactful:”[121]
Liquidity risks [for example,]are three-fold. First, banks that provide liquidity to borrowers under revolvers can face sudden withdrawals, and may lack liquidity to meet payment obligations. Second, banks expect to be able to dispose of leveraged loans they originate in the open market. Should the demand for such assets dry up, banks would face the liquidity consequences of holding such loans for an extended period. [Something the Fed calls “pipeline risk”[122]] Third, in stress, warehouse facilities could also generate delays in repayment and reduce bank liquidity. The financial crisis showed that credit line drawdowns can increase significantly in the presence of liquidity shocks, since corporates often rely on these pre-arranged credit lines for their investment spending.
Despite these risks, the FSB concludes that, based on stress tests carried out in the U.S. and the U.K., “the credit, market and liquidity risks arising from their involvement in the leveraged loan market are unlikely to have an outsized impact on their ability to lend in a market downturn.”[123] Moreover, stress test results “suggest that these banks’ credit risk exposures are manageable and that banks would be able to withstand material losses on their loan portfolios, even when loans include fewer creditor protections.”[124]
B. The Fed’s 2019 stress test results and financial stability report
The FSB was referring to the Fed’s June 2019 stress tests report. As part of the ongoing supervision on large national banks and bank holding companies in the U.S., the Fed ran stress tests to evaluate the potential impact of sudden writedowns of the banks’ leveraged loan and CLO holdings, in a context of a wider financial and economic crisis—a “severely adverse scenario.”[125] In the Fed’s stress testing jargon, banking exposure to CLOs was included under the category of “available-for-sale” or “held-to-maturity” securities that suffer “credit-related” and “other-than-temporary-impairment” losses.[126]
The results were encouraging:[127]
In the U.S., the results of this year’s stress test cycle (DFAST 2019) suggest that, in the aggregate, the 18 firms subject to the supervisory stress test would experience substantial losses under both the adverse and severely adverse scenarios but could continue lending to businesses and households, due to the substantial build of capital since the financial crisis.
A similar sentiment had been echoed in the month prior, May 2019, when the Fed published its Financial Stability Report, where it also identified the particular risks arising from the leveraged loan market, which were similar to those outlined by the FSB.[128] According to the Fed, the larger banks typically involved in the origination of leveraged loans and the securitization of CLOs seemed well prepared to weather a severe crisis that could plummet leveraged loan prices and dry up the market:[129]
On the whole, banks appear well positioned to deal with these exposures. The annual stress-test exercises stress a range of participating banks’ direct and indirect exposures to shocks from the business sector. The tests require that participating banks have sufficient capital to withstand material losses on these exposures and continue lending. In addition, banks’ internal liquidity stress tests and the Liquidity Coverage Ratio requirement incorporate protections against draws on credit lines. With regard to leveraged lending, banks have improved their management of the associated risks—reflecting, in part, the 2013 interagency guidance on leveraged lending—even as underwriting standards have deteriorated over the past decade.[[130]] Moreover, large banks have improved their management of syndication pipelines.
Regarding institutional investors on CLOs, the Fed warned that “[i]t is hard to know with certainty how today’s CLO structures and investors would fare in a prolonged period of stress,” but underlined the positive aspect that “[c]ompared with the investment vehicles associated with subprime mortgages in the financial crisis, CLOs are structured in a way that avoids run risk.”[131]
C. The 2020 stress test scenario and coronavirus reality
As if anticipating the coronavirus catastrophe, in February 2020, the Fed published its annual instructions for banks to run stress tests based on a new set of dire hypothetical facts—a new “severely adverse scenario.” The Fed’s apocalyptic projection included the unemployment rate reaching 10%, and a severe shock to the leveraged loan market:[132]
This year’s global market shock for the severely adverse scenario emphasizes a heightened stress to highly leveraged markets that causes CLOs and private equity investments to experience larger market value declines relative to 2019. … [T]he leveraged loan market comes under considerable pressure. Open-ended mutual funds and exchange-traded funds (ETFs) that hold leveraged loans and high yield bonds face heavy redemptions. Due to liquidity mismatches, mutual fund and ETF managers sell their most liquid holdings, leading to more extensive declines in the prices of fixed income securities and other related assets. Price declines on leveraged loans flow through to the prices for collateralized loan obligations (CLOs). CLO prices suffer severe corrections associated with the devaluation of the underlying collateral and selling by concentrated holders desiring to reduce risk.
The broad selloff of corporate bonds and leveraged loans spills over to prices for other risky credit and private equity instruments. Credit spreads for emerging market corporate credit and sovereign bonds widen due to flight-to-safety considerations. Asset values for private equity experience sizable declines as leveraged firms face lower earnings and a weak economic outlook.
In the Fed’s doom scenario, the unemployment rate would be growing steadily before reaching 10% by the third quarter of 2021.[133] In real life, today, economists estimate that the unemployment rate in the U.S. stands at an unprecedented 13% due to the coronavirus lockdowns.[134]
Shocks to the leveraged loan market have already played out with a scary similarity to many of the 2019 stress projections by the FSB and the Fed, and even in the doomsday 2020 stress test scenario. For example, the “pipeline risk” anticipated by the FSB and the Fed materialized itself when a number of leveraged loans that were in the process of being pooled and sold to CLOs plummeted in price.[135]
To shore up prices and inject liquidity to the market, on April 9, 2020 the Fed responded by promising to invest in the highest tranches of these “newly-issued CLOs,”[136] in an aggressive use of its lender of last resort powers. The Fed’s quantitative easing efforts and lending promises have even included buying so-called “junk bonds”[137]—a move that was so unthinkable just a month ago that a hedge fund manager specializing in distressed debt took issue with the Fed intervening his niche market.[138]
In March, a month of wild fluctuations,[139] the leveraged loan market experienced a sell-off[140] as a high percentage of loans were categorized as “distressed,” which is when an asset is priced at 80 cents on the dollar. However, as a result of the Fed’s aggressive action, leveraged loan prices surged significantly through April, although not yet to pre-coronavirus levels.[141]
VI. EBITDA add-backs are tricky, but rating agencies do have an eye on them
As seen above, EBITDA-debt ratios are the most common metric lenders use to monitor borrower performance and trigger incurrence covenant violations.[142] Borrowers seem to have the incentive to overstate or “adjust” EBITDA in order to reduce the leverage ratios on their books, and, as a result, (a) diminish the likelihood of violating an incurrence covenant, and (b) increase their ability to borrow more, which, in the leveraged loan market, most likely means the ability to raise additional debt, including to refinance older debt.[143]
Compared to the amount of academic work on the economics of cov-lite loans and CLOs, the economics of EBITDA add-backs and negotiation dynamics in the leveraged loan market have been understudied. We do know, however, that in large M&A and LBO deals, EBITDA add-backs are vibrantly negotiated between bank lenders, on one side, and the borrowing company typically supported by a private equity sponsor, on the other.[144] And that in a typical LBO deal, the “private equity firm borrows as much as they can from a variety of lenders (up to 70 or 80 percent of the purchase price) and funds the balance with their own equity.”[145]
While private equity firms are expected to push for higher leverage and covenant flexibility, lending institutions are also expected to proceed with caution given the size of the credit exposures. Interestingly, the recent pressure for add-backs that can lower EBITDA-to-debt ratio appears to have increased as a result of a 2013 regulatory recommendation by the Fed, the Office of the Comptroller of the Currency (OCC), and the Federal Deposit Insurance Corporation (FDIC), which required banks to underwrite transactions that, “at a minimum,” do not exceed a “6x Debt/EBITDA” ratio.[146] For perspective, in 2010, the Blackstone bid to acquire Fidelity obtained a commitment of $10 billion dollars from a syndicate of banks, which amounted to a 7x Debt/EBITDA ratio.[147] That $1.4 billion dollar gap between what the regulator today thinks is prudent versus what the banks in 2010 thought was doable, may provide a clue.
For the time being, in the absence of more insight, we cannot determine if there is a more nuanced logic—beyond short-sighted appetite for a lucrative closing fee—in current add-back flexibility arrangements among lenders and borrowers in large leveraged transactions. In the meantime, however, the public can find comfort in the fact that, this time around (or perhaps once again[148]), a leading rating firm has not only been very active exposing the problems behind EBITDA add-backs, but it has also been emphatic assuring the public that it does not consider management-adjusted EBITDA in their ratings of borrowers, loans and CLOs:[149]
Regardless of the findings [that management overstates EBITDA], we want to emphasize that our ratings are based on our projections of a company’s growth and earnings and our view of issues like expected synergies or cost efficiencies. As this report illustrates, most speculative-grade issuers have initially been rated in the ‘B’ category by S&P Global Ratings over the past four years (83% in our sample, as reflected in chart 4b). Our rating is based on our calculation of leverage and not what is presented to us. If we took the marketing leverage presented to us and bought into pro forma add-backs, projected earnings, and debt reduction, our initial issuer credit ratings would likely be higher and most likely lowered as actual results are reported. Marketing leverage and the language around add-backs as defined in debt agreements do not determine our view of credit risk (other than in assessing covenant headroom when reviewing debt instruments containing financial maintenance covenants). We often do give some credit to add-backs or synergies that we view as achievable, especially when a company has demonstrated its ability to realize on similar items in past comparable transactions. However, we are almost always considerably less optimistic than management when it comes to certain elements pertaining to future growth, for example realizable revenue and/or cost synergies, as reflected in our projections. Additionally, we exclude adjustments for items that are ultimately cash operating costs like management fees and restructuring costs. In fact, our analysis goes much deeper than EBITDA and examines the true cash flow characteristics of issuers.
VII. Conclusion
Will the leveraged loan market trigger a financial pandemic? We don’t know, but a month and a half into the coronavirus crisis, it appears unlikely—either because any financial meltdown at this point will be blamed first and foremost on the pandemic, or because the systemic risk associated with the leveraged loan market isn’t as high as many commentators would have us believe. What we do know, now, is that cov-lite loans aren’t really “lite,” that CLO structures are much more conservative today than CDOs and CDSs were 13 years ago, and that we need to keep a closer eye on EBITDA add-backs.
Endnotes:
* JD Candidate ’21, Fordham Law School; LL.M. ’09, Columbia Law School.
± See https://news.law.fordham.edu/jcfl/blog/
[1] Note of the author: I finalized the first draft of this article days before the coronavirus pandemic brought the world economy to a halt. I have since updated it in a limited way to discuss relevant developments, like CLO downgrades from the leading rating agencies. Out of intellectual honesty, however, I have maintained the article’s main structure and its main thesis—that in the spectrum of things to be worried about in the financial markets, cov-lite loans and CLOs deserved perhaps less attention than it was being given to them. In hindsight, I would modify my argument to say, first, that awareness of the financial risks posed by a possible global pandemic is what deserved more attention than anything. Second, I would include the caveat that I did not expect the article’s validity to be tested so quickly, perhaps in the next few days or even hours. If I turn out getting it completely wrong, I apologize in advance.
[2] Masahiro Kawai et al., Crisis and Contagion in East Asia: Nine Lessons, (World Bank Policy Research Working Paper, WPS 2610, 2001). See also Richard S. Carnell et al., The Law of Financial Institutions 429 (6th ed., 2017) (describing “cascades, contagion, and asset implosions” as the main sources of systemic risk in the financial system).
[3] Galina Hale, Could We Have Learned from the Asian Financial Crisis of 1997-98?, Federal Reserve Bank of San Francisco Economic Letter, Feb. 2011 (“Financial contagion spread through the region so fast that it was nicknamed the ‘Asian flu’”), https://www.frbsf.org/economic-research/files/el2011-06.pdf.
[4] Division of Market Regulation, SEC., Trading Analysis of October 27 and 28, 1997, (Sept. 1998), https://www.sec.gov/news/studies/tradrep.htm.
[5] Akane Otani, Circuit Breaker Halts Stock Trading for First Time Since 1997, Wall St. J. (Mar. 9, 2020), https://www.wsj.com/articles/traders-closely-watching-circuit-breakers-thresholds-11583761223; and Minami Funakoshi & Travis Hartman, March Madness, Reuters (Mar. 18, 2020), https://graphics.reuters.com/usa-markets/0100B5L144C/index.html.
[6] Alan Rappeport & Jeanna Smialek, I.M.F. Predicts Worst Downturn Since the Great Depression, N.Y. Times (Apr. 14, 2020) (last updated Apr. 29, 2020), https://www.nytimes.com/2020/04/14/us/politics/coronavirus-economy-recession-depression.html.
[7] See Oaktree, Memos from Howard Marks: Which Way Now? (Mar. 31, 2020) (“I believe we’re likely to see defaults on the part of leveraged entities, based on price markdowns, ratings downgrades and perhaps defaults on their portfolio assets”), https://www.oaktreecapital.com/insights/howard-marks-memos.
[8] Joint Release, Board of Governors of the Federal Reserve System, Federal Deposit Insurance Corporation & Office of the Comptroller of the Currency, Statement on the Use of Capital and Liquidity Buffers (Mar. 17, 2020), https://www.fdic.gov/news/news/press/2020/pr20030b.pdf.
[9] Brian Cheung, A Glossary of the Federal Reserve’s Full Arsenal of ‘Bazookas’, Yahoo! Finance (Apr. 9, 2020), https://finance.yahoo.com/news/glossary-federal-reserves-emergency-measures-coronavirus-bazookas-120337473.html.
[10] See Mayra Rodriguez, U.S. Corporate Debt Continues To Rise As Do Problem Leveraged Loans, Forbes (July 25, 2019), https://www.forbes.com/sites/mayrarodriguezvalladares/2019/07/25/u-s-corporate-debt-continues-to-rise-as-do-problem-leveraged-loans/#1215441f3596; Colby Smith, Warnings Mount for Leveraged-Loan Market, Financial Times (Oct. 19, 2018), https://ftalphaville.ft.com/2018/10/19/1539934976000/Warnings-mount-for-leveraged-loan-market; and Tom Metcalf et. al., Wall Street’s Billionaire Machine, Where Almost Everyone Gets Rich, Bloomberg (Dec. 20, 2018), https://www.bloomberg.com/graphics/2018-collateralized-loan-obligations.
[11] See Financial Stability Board, Vulnerabilities associated with leveraged loans and collateralised loan obligations, 16 (2019), https://www.fsb.org/wp-content/uploads/P191219.pdf. See also Kristen Haunss, Regulators Voice Concerns Over US Leveraged Loan Risk, Reuters (May 8, 2019), https://www.reuters.com/article/regulators-levloans/regulators-voice-concerns-over-us-leveraged-loan-risk-idUSL2N22K0OC; John Glover, Loose Leveraged Lending Is Storing Up Economic Trouble, BIS Says, Bloomberg (Sept. 23, 2018), https://www.bloomberg.com/news/articles/2018-09-23/loose-leveraged-lending-is-storing-up-economic-trouble-bis-says; Colby Smith, Warnings Mount for Leveraged-Loan Market, Financial Times (Oct. 19, 2018), https://ftalphaville.ft.com/2018/10/19/1539934976000/Warnings-mount-for-leveraged-loan-market.
[12] Press Release, Sen. Elizabeth Warren [D-MA], Warren Presses Regulators on Risks in Leveraged Lending Market (Nov. 15, 2018). Warren, a former presidential candidate, is also a Bankruptcy professor emeritus at Harvard Law School and author of one of the seminal textbooks in the field, https://hls.harvard.edu/faculty/directory/10935/Warren/publications.
[13] Compare Sean Campbell, Taking It All In: Understanding Leveraged Lending in the U.S. Economy, Financial Services Forum, April 9, 2019 (“The Loan Syndications and Trading Association (LSTA), for example, estimates the total amount of outstanding leveraged loans to be $1.5 trillion. This is a sizable number by any definition, but this amount must be compared to the amount of other forms of debt and risk in the U.S. economy”), https://www.fsforum.com/types/press/blog/understanding-leveraged-lending-in-the-u-s-economy; with Financial Stability Board, supra note 11 at 6 (“In 2018 the U.S. institutional leveraged loans outstanding and high-yield bond markets were comparable in size at around $1.2 trillion”).
[14] Some have been predicting this doom scenario since before the 2008 financial crisis, and then again in 2011 and 2014. See Bethany McLean, Corporate Subprime: The Default Crisis that Never Happened, Slate (Feb. 22, 2011), https://slate.com/business/2011/02/covenant-lite-the-default-crisis-that-never-happened-will-it-now.html; and Kate Marino, Cov-lite Loan Controversy Fades To Whisper While Portfolio Managers Chase Yield, Forbes (June 3, 2014) (“While highlighting the data is helpful for headlines, one issue remains: market participants chasing yield don’t seem to care much, despite signs that the quality of today’s covenant-lite loan has deteriorated significantly from the last cycle. Just four short years after the syndicated corporate loan market emerged from the slumber of the global financial meltdown, cov-lite term loans have skyrocketed to account for nearly two-thirds of issuance”), https://www.forbes.com/sites/mergermarket/2014/06/03/cov-lite-loan-controversy-fades-to-whisper-while-portfolio-managers-chase-yield/#7fad0045c887.
[15] Tirupan Goel, The Rise of Leveraged Loans: A Risky Resurgence?, Bank of Int’l Settlements (BIS) Q. Rev. (extract from pages 10-11, Sep. 2018), https://www.bis.org/publ/qtrpdf/r_qt1809u.htm.
[16] Letter from Sen. Elizabeth Warren [D-MA] to Secretary of the Treasury Steven T. Mnuchin (Mar. 19, 2020), https://www.warren.senate.gov/imo/media/doc/2020.03.19%20Letter%20to%20FSOC%20from%20Senator%20Warren%20Re%20Leveraged%20Lending.pdf. See also Mark Carey, Financial Innovation May Cause the Next Recession as Protection-Lite Loans Falter, Barron’s (Apr. 8, 2020), https://www.barrons.com/articles/innovative-loans-face-high-risk-of-default-as-economy-nears-recession-51586349000.
[17] Id.
[18] See Andrew Thomson & Andrew Hedlund, Covenant-lite: The Slow-Burning Fuse that Could Blow Up Loans, Private Debt Investor, Feb. 2020, https://reorg.com/wp-content/uploads/2020/01/PDI_Feb2020_V2-private-debt-investor-magazine.pdf; Larry Light, A Decade After the Financial Crisis, Corporate Finance Must Contend with These New Ticking Debt Time Bombs, Fortune (Dec. 29, 2019) (“…Corporate America’s overweening debt and weak covenants are an accident waiting to happen, according to fixed-income impresario Jeffrey Gundlach, CEO of asset manager DoubleLine. He warns that the problem’s size means that cleaning up the eventual imbroglio, once everything goes wrong, will be daunting. ‘When you have the recession, there won’t be an ability to fix it,’ he said during a London speech in September. How bad could the next mess be? Right now, the load of U.S. junk bonds outstanding is valued at $1.5 trillion. Leveraged loans are almost as high, around $1.2 trillion. Adding them together, you near the size of the mortgage-backed securities that imploded during the 2008 crisis—and crippled the world economy. And that’s not all. Some 80% of junk issues and lev loans are cov-lite, giving issuers too much leeway, to the detriment of investors. … In light of dubious leveraged loans and threadbare covenant protection, in addition to teetering junk bonds, the credit markets and a lot of fixed-income investors should brace themselves for rough seas ahead”), https://fortune.com/2019/12/29/financial-crisis-corporate-debt-lev-loans-covenants; and Cleary Gottlieb, Has the European Leveraged Loan Market Reached Boiling Point? (May 3, 2019) (raising similar alarm bells on cov-lite loans in Europe), https://www.clearygottlieb.com/news-and-insights/publication-listing/has-the-european-leveraged-loan-market-reached-boiling-point.
[19] See Office of the Comptroller of the Currency, Leveraged Lending: Comptroller’s Handbook, 2 (2008) (“The OCC broadly considers a leveraged loan to be a transaction where the borrower’s post-financing leverage, when measured by debt-to-assets, debt-to-equity, cash flow-to-total debt, or other such standards unique to particular industries, significantly exceeds industry norms for leverage”), https://www.occ.treas.gov/publications-and-resources/publications/comptrollers-handbook/files/leveraged-lending/index-leveraged-lending.html. See also Financial Stability Board, supra note 11 at 5 (“There is no commonly agreed definition for leveraged loans. However, criteria used by regulators and data providers to classify a loan as ‘leveraged’ typically include: (i) high indebtedness of the borrowing corporate (e.g. gross debt to earnings before interest, tax, depreciation and amortisation (EBITDA) ratio 4x or higher); (ii) below investment grade credit rating for the loan (or borrower) (i.e. below BBB); (iii) loan purpose is to finance an acquisition (e.g. management buy-out (MBO) or leveraged buy-out (LBO)); (iv) presence of a private equity sponsor in the transaction (e.g. financing of borrowers owned by financial sponsors); or (v) high loan spread at issuance (e.g. +125 basis points)”).
[20] See Financial Stability Board, supra note 11 at 16 (“Leveraged loans are mainly used for leveraged buy-outs (LBOs), mergers and acquisitions (M&A), recapitalisation or refinancing of debt”). See also Office of the Comptroller of the Currency, supra note 19 at 1 (“Leveraged lending is a type of corporate finance used for mergers and acquisitions, business recapitalization and refinancing, equity buyouts, and business or product line build-outs and expansions”).
[21] Incurrence covenants in loan agreements restrict the borrower from engaging specific debt-increasing actions—dividend payments, acquisitions, divestitures, etc.—that may affect certain credit ratios. These covenants do not consider it to be a violation of the covenant if those same limits are exceeded as a result of the company’s financial results. Maintenance covenants, on the other hand, call for quarterly financial tests and consider any failure to comply with the pre-established test conditions to be a violation of the covenant (triggering technical default and the remedy of acceleration, which makes the maturity date current), regardless of the actions taken by the company. See S&P Global Market Intelligence, LCD Loan Primer, 15 (2017), https://www.spglobal.com/marketintelligence/en/pages/toc-primer/lcd-primer#sec15. See also Office of the Comptroller of the Currency, supra note 19 at 62-64.
[22] S&P Global Market Intelligence, supra note 21.
[23] Bo Becker & Victoria Ivashina, Covenant-Light Contracts and Creditor Coordination 3 (Swedish House of Finance Research Paper, No . 16-09, 2016), https://www.hbs.edu/faculty/pages/item.aspx?num=50952.
[24] S&P Global Market Intelligence, Covenant-Lite Share of US Leveraged Loan Market Hits Record 79% (Feb. 14, 2020) (“The cov-light share of market has increased steadily from roughly 64% in August 2015”), https://www.spglobal.com/marketintelligence/en/news-insights/latest-news-headlines/leveraged-loan-news/covenant-lite-share-us-leveraged-loan-market-hits-record-79.
[25] See Andrew Thomson & Andrew Hedlund and Larry Light, supra note 18. See also Lisa Lee, How Leveraged Loans Are (and Aren’t) Like Junk Bonds, Bloomberg (Sept. 30, 2018) (“Leveraged loans were once normally protected by contract clauses called maintenance covenants that allowed lenders to monitor the borrowing company’s performance and take action, like forcing the sale of assets, if earnings deteriorated. During the financial crisis of the last decade, many companies with leveraged loans saw earnings plummet. Some were able to make the case to investors that this wasn’t their fault and they needed these covenants eliminated. Investors agreed. These loans with looser rules, known as covenant-lite, gradually spread and now represent 80 percent of new issuance. They’ve become so common that many investors shy away from making loans with strict covenants, figuring that if a borrower needs to agree to that many restrictions, the loan could be dicey. Since high-yield bonds don’t have maintenance covenants that give investors oversight over performance, investors and borrowers began to view covenant-lite loans similarly”), https://www.bloomberg.com/news/articles/2018-09-30/how-leveraged-loans-are-and-aren-t-like-junk-bonds-quicktake.
[26] Id. See also William D. Cohan, The Big, Dangerous Bubble in Corporate Debt, N.Y. Times (Aug. 9, 2018), https://www.nytimes.com/2018/08/09/opinion/corporate-debt-bubble-next-recession.html.
[27] Larry Light, supra note 18.
[28] See S&P Global Market Intelligence, Those $700B in US CLOs: Who holds them, what risk they pose (June 21, 2019) (“To be sure, terms on leveraged loans have been getting more aggressive, and many of those loans—about 60% by LCD’s estimate—are being packaged into these CLOs. These are then purchased, in large part, by overseas investors, reminding many of the pre-2008 CDO machine, the end of which saw some tens of billions of dollars in losses across the financial system”), https://www.spglobal.com/marketintelligence/en/news-insights/latest-news-headlines/leveraged-loan-news/those-700b-in-us-clos-who-holds-them-what-risk-they-pose. See also Bain Capital, Implications of the Growth in Covenant-Lite Loans, Credit Market Insights Q2 2019 (stating that CLOs make up “approximately 50%” of total leveraged loans), https://www.baincapitalcredit.com/sites/baincapitalcredit.com/files/Credit_Market_Insights-Implications_of_Growth_in_Cov-Lite_Loans_060419.pdf.
[29] Eric Jacobson, Should You Worry About CLOs in Your Funds?, Morningstar (Sep. 24, 2019), https://www.morningstar.com/articles/946877/should-you-worry-about-clos-in-your-funds.
[30] Sean Campbell, supra note 13.
[31] Carol M. Kopp, Asset-Backed Security vs. Collateralized Debt Obligation, Investopedia (Apr. 20, 2019) (“Collateralized loan obligations (CLOs) are CDOs made up of bank loans”), https://www.investopedia.com/ask/answers/040715/what-difference-between-collateralized-debt-obligation-cdo-and-asset-backed-security-abs.asp.
[32] Frank J. Fabozzi & Vinod Kothari, Introduction to Securitization 97-99 (Wiley, 2008) (“The amount of credit enhancement needed to obtain a specific credit rating is specified by the rating agencies from which a rating is sought and is referred to as the sizing of the transaction. The mechanisms of credit enhancement are classified into three categories: (1) originator-provided [e.g. excess spread, cash collateral], (2) structural [e.g. overcollateralization and stratification or subordination], and (3) third-party provided [e.g. monoline insurance, letters of credit and pool insurance policies]”).
[33] Cezary Podkul & Paul J. Davis, Financial Engineering Made Risky Loans Seem Safe. Now They Face a Huge Test, Wall St. J. (Mar. 20, 2020), https://www.wsj.com/articles/financial-engineering-made-risky-loans-seem-safe-now-they-face-a-huge-test-11584702000.
[34] Russ Ray, CDOs Cubed: The First-Ever Triple Derivative, 12 (3) Derivatives Use, Trading & Regulation 183 (2005), https://link.springer.com/content/pdf/10.1057/palgrave.dutr.1850038.pdf.
[35] Carolyn E.C. Paris, Drafting for Corporate Finance: Concepts, Deals, and Documents loc. 760 (2nd ed. 2014) (ebook).
[36] Id.
[37] Id.
[38] The Big Short (Paramount Pictures 2015). See also Christopher Whittall & Mike Bird, In a Blast From a Financial Crisis Past, Synthetic CDOs Are Back, Wall St. J. (Apr. 28, 2017), https://www.wsj.com/articles/in-a-blast-from-a-financial-crisis-past-synthetic-cdos-are-back-1503912601.
[39] Berkshire Hathaway Inc., Warren Buffett’s Letters to Berkshire Shareholders 2002, 15 (Feb. 21, 2003), https://www.berkshirehathaway.com/letters/2002pdf.pdf.
[40] Yalman Onaran, Can We Survive the Next Financial Crisis?, Bloomberg (Sept. 10, 2018), https://www.bloomberg.com/graphics/2018-lehman-anniversary.
[41] Larry Light, supra note 18.
[42] Sean Campbell, supra note 13.
[43] S&P Global Market Intelligence, When The Credit Cycle Turns: The EBITDA Add-Back Fallacy (Sept. 24, 2018) (“Thus, for example, a company with a debt-to-EBITDA ratio of 5x could repay the debt within a reasonable period since, it was argued, that working capital requirements were negligible and capital expenditures could be immediately scaled back to service debt. The banks (and other investors) seeking lending opportunities bought into the concept, and in hindsight did objectively well as the default rate among cable companies was relatively benign”), https://www.capitaliq.com/CIQDotNet/CreditResearch/RenderArticle.aspx?articleId=2104625&SctArtId=460373&from=CM&nsl_code=LIME.
[44] Office of the Comptroller of the Currency, supra note 19 at 60.
[45] Financial Stability Board, supra note 11 at 9 (“The rationale for M&A and private equity buyouts often includes synergies or operational improvements, and it is standard practice to recognise these as adjustments (add-backs) to EBITDA, which is used to measure compliance with incurrence covenants. These add-backs, however, are uncertain, both in magnitude and timing, and may overstate EBITDA and thus understate debt-to-EBITDA”).
[46] S&P Global Market Intelligence, When The Credit Cycle Turns: The EBITDA Add-Back Fallacy (Sept. 24, 2018), https://www.spglobal.com/en/research-insights/articles/when-the-credit-cycle-turns-the-ebitda-add-back-fallacy.
[47] Id.
[48] Moody’s Investors Service, EMEA Spec-Grade Firms Are Making Higher Earnings Adjustments To Attract Investors (June 27, 2018), https://www.moodys.com/research/Moodys-EMEA-spec-grade-firms-are-making-higher-earnings-adjustments–PR_385895.
[49] Id.
[50] S&P Global Market Intelligence, supra note 46 (“Over time, management-adjusted EBITDA as a concept arose as a proxy for a company’s future run-rate cash flow generation, post-acquisition or -divestiture or -merger. In these cases, transaction costs, a current expense, or expected cost savings (synergies) were added back to the reported EBITDA to create the management-adjusted EBITDA, which would serve as a better proxy of future (gross) cash flow”).
[51] Id.
[52] Id.
[53] S&P Global Market Intelligence, When The Cycle Turns: The Continued Attack Of The EBITDA Add-Back (Sept. 19, 2019), https://www.spglobal.com/ratings/en/research/articles/190919-when-the-cycle-turns-the-continued-attack-of-the-ebitda-add-back-11156255#ID292.
[54] Mitchell Berlin et al., Concentration of Control Rights in Leveraged Loan Syndicates 4 (Federal Reserve Bank of Philadelphia Working Paper, WP 19-41, 2019) (concluding that cov-lite loan borrowers, in practice, remain equally subject to the discipline of strict financial covenants, and that “bank lenders maintain the incentive and ability to monitor their borrowers even in the presence of a covenant-lite term loan, which [should]alleviate[] concerns that covenant-lite loans create a large number of unmonitored borrowers”).
[55] Id. at 2 (“In our sample, 95% of leveraged loan borrowers with a term loan also have a line of credit. In some cases, the revolver was issued prior to the term loan but is still outstanding at the time of term loan issuance”).
[56] Id. at 4, 26-29 (“We rule out several alternative explanations for the rise in split control rights for institutional deals. We find no evidence that the nature of leveraged loan borrowers has changed over time, which suggests that changes in the contract structure are not due to changes in loan demand or borrower composition. We also show that deals with institutional tranches have experienced no differential changes in other contract terms, including interest rate, amount, or maturity, that would suggest alternative supply-side explanations for the rise in split control rights. In the Online Appendix, we explain how to replicate our results using only data from Thomson-Reuters’ LPC DealScan and show that split control rights comprise the majority of institutional loans through the middle of 2018, so our results are not due to the extraordinarily low level of interest rates that existed for our main sample period”).
[57] See id. at 18 (“Third, we offer evidence that institutional deals experienced more acute bargaining frictions during the financial crisis, a period prior to the widespread adoption of split control rights. Finally, we show that split control rights are part of a menu of contractual terms that have been explicitly designed to mitigate bargaining frictions following the experience of the financial crisis”); and id. at 7 (“A unique feature of the leveraged loan market since the early 2000s has been the increase in participation of nonbank institutional investors such as CLOs and mutual funds. According to the Shared National Credit (SNC) Program administered by U.S. regulatory agencies, the fraction of all syndicated loans held by nonbanks increased from less than 10% in 2001 to nearly 20% by 2008 and almost 25% by 2016”).
[58] Id. at 3.
[59] Id. at 8 (“…we define a deal as the set of loan facilities that are current on the same day to the same borrower. We use the concept of a deal under the assumption that loan contracts are written with an understanding of all of the firm’s borrowings. In most cases, separate facilities are governed by the same loan contract, and in cases of separate contracts we observe language that references other facilities”).
[60] Id. at 2.
[61] Id. at 11 (“In some cases, the monitoring of the maintenance covenants depends on the utilization of the borrower’s line of credit, so-called springing covenants”).
[62] Id. See also Sean Campbell, supra note 13 (“The researchers show […] that when both the revolving and term loans are considered together, less than 2 percent of leveraged loans are truly ‘covenant-lite’ and this proportion has remained stable since 2005”).
[63] Id. Compare with Financial Stability Board, supra note 11 at 22 (“Contract terms in leveraged loan agreements and placement documents typically limit the direct risks to originators that resell loans. And […] while banks originate leveraged loans, institutional investors now play a larger role in purchasing syndicated loan packages. These factors might weaken incentives for banks to conduct due diligence and apply strict standards in good times to some extent, contributing to procyclicality in the provision of credit”).
[64] See Mitchell Berlin et al., supra note 54 at 7 (“In response to the emergence of institutional investors, the arrangers of leveraged loans began to design a tranche of the deal intended specifically for institutional investors. As described in Taylor et al. (2006), an ‘institutional term loan (term loan B, C, or D) is a term loan facility carved out for nonbank institutional investors.’ This tranche is different from the so-called pro-rata tranches that are traditionally funded by banks, which include a revolving credit facility and a ‘term loan A’ that are funded in identical proportions by lenders. An institutional term loan typically does not amortize and often has a longer maturity than the pro-rata tranches”); and Office of the Comptroller of the Currency (OCC), supra note 19 at 48 (providing the example of a four-tiered leveraged loan deal). In the OCC example, the first two tiers—comprising a Revolving Credit (RC or “revolver”) and a Term Loan A (called “TLA”)—are presumably taken in equal parts by the bank leading the syndicate and by the syndicate members, with the lead bank or syndicate agent exercising oversight over the borrower. The bottom two tiers—called TLB and TLC at origination (also referred to as “institutional loans”)—are absorbed by “institutional investors.” In the example, the revolver is a one-year open line of credit of $50 million. The TLA is a larger loan and has a longer maturity—5 years—while providing for regular amortization payments as well as interest at set intervals. The TLB and TLC loans, on the other hand, provide for a steady stream of interest payments and a similarly long maturity as the TLA, but instead lack amortization payments and call for repayment at maturity or close to maturity. This structure seems to reflect the fact that institutional lenders or investors are more interested in the fixed interest-based income and only expect to recoup their principal towards the end of the maturity—with “ballooning” payments—or at maturity through a one time “bullet” payment. Drawing from Mitchell Berlin et al and from the OCC example, the reason for this complex structure seems to be that the cov-lite loans in the bottom two tiers are meant to be absorbed by private equity firms whose main interest is to sponsor a leveraged deal—an acquisition or an LBO—and by CLO funds or insurance companies whose interest is to structure and invest in long-term fixed-income CLOs, while the top two tiers are meant to remain closely monitored by the bank.
[65] Mitchell Berlin et al., supra note 54 at 4-5, 23 (“Similar to other research on nonbank lenders, we show that renegotiating an institutional loan was particularly costly during the financial crisis, which we conjecture prompted a change in the optimal contract to facilitate renegotiation. For example, during 2006-2010, the average fee paid to amend a loan following a covenant violation was more than three times larger for borrowers with an institutional loan”).
[66] Id. at 2, 4 (“Instead, as our evidence shows, contracts have evolved to facilitate bank monitoring while mitigating the bargaining frictions that have arisen with the entry of nonbank lenders. […] Similar to other research on nonbank lenders, we show that renegotiating an institutional loan was particularly costly during the financial crisis, which we conjecture prompted a change in the optimal contract to facilitate renegotiation”).
[67] Id. at 5-6 (“Our results are consistent with canonical models of corporate debt structure that justify giving bank lenders unilateral control rights in certain states of the world. These models were designed to explain the commonly observed debt structure of a senior bank loan with strong control rights and subordinated bonds with much weaker rights. We show that a similar structure has developed in the leveraged loan market, with the notable difference that the covenant-lite term loan typically has equal seniority to the revolving loan”).
[68] Id. at 11 (“… the [split control]arrangement means that any financial covenants can be waived or modified by only a subset of the lenders. In the event of a covenant violation, these lenders have the unilateral right to negotiate with the borrower about how to cure the default”).
[69] Sean Campbell, supra note 13.
[70] Mitchell Berlin et al., supra note 54 at 13 (“Across the entire sample, the frequency of deals with no maintenance covenant is only 1% and has not risen in recent years. This striking fact is at odds with the typical interpretation of covenant-lite loans and can only be known by properly accounting for all of the loans that a borrower has. Instead of removing financial covenants from the standard loan contract, covenant-lite loans concentrate control rights with the revolving lenders”).
[71] Id. at 13-17 (“…efficient monitoring does not require that the interests of revolving lenders be aligned with those of the term lenders ex-post. Indeed, canonical multi-creditor models permit the monitoring lender to expropriate other lenders when it monitors efficiently. In these models, expropriated lenders are compensated ex-ante in the pricing of the loan, but the benefit of concentrating control rights is more efficient monitoring and renegotiation outcomes”).
[72] Id. at 4 (“… our evidence suggests that split control rights have little effect on the frequency of renegotiation but rather change the set of lenders who are involved in the renegotiation”).
[73] Eva Su et al., Leveraged Lending and Collateralized Loan Obligations: Frequently Asked Questions, 5 (Congressional Research Service Report, R46096, 2019) (“Investors can hold leveraged loans by either (1) investing directly in individual leveraged loans, typically through syndications and participations or (2) investing in CLOs. Institutions that directly hold large shares of outstanding leveraged loans include mutual funds (19%), banks (8%), and insurance companies (6%), as shown in Figure 2.21 According to one study, mutual fund holdings are split fairly evenly between funds offered to institutional investors and funds offered to retail investors. Nearly all of the remainder of leveraged loans (62%) are held by CLOs”), https://crsreports.congress.gov/product/pdf/R/R46096.
[74] See id. at n. 20 (“U.S. CLOs are often structured in offshore SPVs to benefit from legal isolation and favorable tax treatments. Around 74% of all U.S. CLO securities held both domestically and abroad were issued out of the Cayman Islands”). See also Financial Stability Board, supra note 11 at 3 (“CLOs are asset-backed securities issued by a special purpose vehicle (SPV). The SPV acquires a portfolio of leveraged loans (typically syndicated loans originated by banks), which it finances through the issuance of securities in the form of bonds (senior and mezzanine tranches) and equity. As codified in the CLO agreements (‘indentures’), these securities are collateralised by the underlying loan portfolio. Payments received from portfolio assets are pooled and flow, in order, to senior, mezzanine, and equity tranches; this is known as the ‘waterfall’. Except for the most junior security representing the residual interest (‘equity’), almost all CLO securities are structured as floating rate notes (‘notes’) and receive ratings. CLOs differ from many other securitisation structures, in that they are actively managed by a CLO manager”); and Frank J. Fabozzi & Vinod Kothari, supra note 32 at 6, 15 (on the role of bankruptcy remoteness or bankruptcy protection, which is a legal structuring mechanism meant to shield the SPV from a possible bankruptcy of the debt originator).
[75] Mitchell Berlin et al., supra note 54 at 4.
[76] Eva Su et. al., supra note 73 at 2.
[77] Fabozzi & Vinod Kothari, supra note 32 at 22.
[78] Id. See also Eva Su et. al., supra note 73 at 7 (“[Tranches] give the holder the right to the payment of cash flow on the underlying loans. The different tranches are assigned different payment priorities, so some will incur losses before others. […] The tranches are often known as senior, mezzanine, and equity tranches, in order from highest to lowest payment priority, credit quality, and credit rating. Through this process […], the loan portfolio’s risks are redistributed to the lower tranches first, and tranches with higher credit ratings are formed”).
[79] Fabozzi & Vinod Kothari, supra note 32 at 5 (citing definition by Fabozzi, Davis and Choudhry: “[Structured finance consist of] techniques employed whenever the requirements of the originator or owner of an asset, be they concerned with funding, liquidity, risk transfer, or other need, cannot be met by an existing, off-the-shelf product or instrument. Hence, to meet this requirement, existing products and techniques must be engineered into a tailor-made product or process. Thus, structured finance is a flexible financial engineering tool”).
[80] Id. at 9.
[81] Id. at 13, 216 (explaining that this type of structure is meant to achieve “regulatory and/or economic capital relief”).
[82] Id. at 13, 160 (discussing origination fees and servicing fees, respectively).
[83] Id. at 275-89.
[84] Bradley Keoun, Junk Loans, Now at $1.2 Trillion, Could Haunt U.S. Banks, Top Regulator Warns, The Street (May 16, 2019), https://www.thestreet.com/markets/top-us-banking-regulator-warns-of-growing-risks-of-junk-loan-defaults-14961419.
[85] Berkshire Hathaway Inc., supra note 39 at 15.
[86] Sirio Aramonte & Fernando Avalos, Structured Finance Then and Now: A Comparison of CDOs and CLOs, BIS Q. Rev. (extract from pages 11-14, Sept. 2019), https://www.bis.org/publ/qtrpdf/r_qt1909w.htm.
[87] S&P Global Market Intelligence, Those $700B in US CLOs: Who holds them, what risk they pose (June 21, 2019) (“Yet part of the reason why a number of banks, insurance companies and other investors have been comfortable building their holdings in CLOs is that these vehicles performed astoundingly well during the financial crisis, compared to the similar-sounding CDOs, which were often filled with fraudulently underwritten subprime mortgages”), https://www.spglobal.com/marketintelligence/en/news-insights/latest-news-headlines/leveraged-loan-news/those-700b-in-us-clos-who-holds-them-what-risk-they-pose.
[88] Eric Jacobson, supra note 29.
[89] Tim Cross, U.S. Leveraged Loan Default Rate Dips To 7-Year Low, Forbes (Mar. 12, 2019), https://www.forbes.com/sites/spleverage/2019/03/12/us-leveraged-loan-default-rate-dips-to-7-year-low/#31f706662d18.
[90] S&P Global Market Intelligence, Despite Aging Credit Cycle, Near-Term Spike In Leveraged Loan Defaults Unlikely (June 10, 2019), https://www.spglobal.com/marketintelligence/en/news-insights/latest-news-headlines/leveraged-loan-news/despite-aging-credit-cycle-near-term-spike-in-leveraged-loan-defaults-unlikely. Compare with Larry Light, supra note 9 (“Right now, the default rate on leveraged loans is very low: a measly 1.42%. And the average historical default rate for these loans is 3.1%, while the rate for junk bonds (also called high-yield bonds) is a bit higher at 3.3%”).
[91] S&P Global Market Intelligence, Ranks Of Leveraged Loan ‘Weakest Links’ Hit Record As Credit Downgrades Mount (Feb. 14, 2020), https://www.spglobal.com/marketintelligence/en/news-insights/latest-news-headlines/leveraged-loan-news/ranks-of-leveraged-loan-weakest-links-hit-record-as-credit-downgrades-mount.
[92] Bain Capital, supra note 28. (“As a manager of loan portfolios and CLOs, we are well aware of the growth in cov-lite issuance and its potential impact on defaults and recoveries. Cov-lite has become standard practice in the market, which we believe will reduce recoveries in the future, but may also lead to fewer companies defaulting. Importantly, we have long considered document analysis a key component of our investment process alongside fundamental research. We believe combining these two elements is the best strategy to minimize defaults and maximize recoveries”). See also Kate Marino, supra note 14 (“[t]he low default expectation is in part because so few covenants are out there to trip up borrowers”).
[93] Mitchell Berlin et al., supra note 54 at 20-23.
[94] Id. at 31.
[95] S&P Global Market Intelligence, supra note 21 (“Historically, recoveries in cases of default on cov-lite loans have been on par with that of traditionally covenanted credits, though there is consensus that recent-vintage deals will recover somewhat less than their predecessors, due to a larger share of lesser-quality issues being cov-lite, along with other types of credit deterioration”).
[96] S&P Global Market Intelligence, Twenty Years Strong: A Look Back At U.S. CLO Ratings Performance From 1994 Through 2013 (June 19, 2019) (“Since the mid-1990s, Standard & Poor’s Ratings Services has rated more than 1,100 U.S. cash-flow CLOs with an aggregate note balance of more than $645 billion (including equity). At year-end 2013, 705 of these CLOs were still outstanding, with an aggregate note balance of approximately $293 billion, versus 654 CLOs and $279 billion at the start of the year. […] Analysis of the full universe of rated CLOs shows that at year-end 2013 only eight investment-grade CLO tranches (or 0.15% of the notes originally rated BBB- or higher) have defaulted, while 17 speculative-grade CLO tranches (or 1.78% of the notes originally rated BB+ or lower) have defaulted. This compares very favorably with the percentage of rated speculative corporate loans that have defaulted”), https://www.spglobal.com/_media/documents/2018-annual-global-leveraged-loan-clo-default-and-rating-transition.pdf.
[97] Id.
[98] Id.
[99] Id.
[100] Tim Cross, supra note 89.
[101] Id.
[102] Financial Stability Board, supra note 11 at 1.
[103] The Editorial Board, A Perfect Storm Wreaks Havoc on Global Markets, Financial Times (Mar. 13, 2020), https://www.ft.com/content/69af7eae-6527-11ea-b3f3-fe4680ea68b5.
[104] Fitch Ratings: Coronavirus Risk to US Leveraged Finance Issuers, CLOs, Loan Syndications and Trading Association (LSTA), https://www.lsta.org/content/fitch-ratings-coronavirus-risk-to-us-leveraged-finance-issuers-clos (last visited May 1, 2020).
[105] See Fitch Ratings, Fitch Ratings Updates CLO Sensitivity Stress for Coronavirus Vulnerabilities (Apr. 7, 2020) (“As Fitch projects default rates to increase it also expects recoveries in the eight vulnerable industries to decrease. The agency therefore applies a haircut to recovery assumptions in its updated stress scenario. The agency applies a multiplier of 0.85 to recoveries at all rating scenarios to issuers in these eight industries. This recovery haircut reflects anticipated movement in going-concern EBITDAs for the recovery analysis, increased use of the liquidation approach instead of the going-concern approach and otherwise weaker valuations in the affected sectors”) https://www.fitchratings.com/research/structured-finance/fitch-ratings-updates-clo-sensitivity-stress-for-coronavirus-vulnerabilities-07-04-2020. See also Fitch Ratings, CLO Stress Testing for Industries Vulnerable to Coronavirus (Mar. 27, 2020), https://www.fitchratings.com/research/structured-finance/clo-stress-testing-for-industries-vulnerable-to-coronavirus-27-03-2020.
[106] LSTA, supra note 104.
[107] Board of Governors of the Federal Reserve System (US), Nonfinancial Corporate Business; Debt Securities and Loans; Liability, Level [BCNSDODNS], retrieved from FRED, Federal Reserve Bank of St. Louis; https://fred.stlouisfed.org/series/BCNSDODNS, April 15, 2020.
[108] Mayra Rodriguez, supra note 10.
[109] Wilshire Associates, Wilshire 5000 Full Cap Price Index [WILL5000PRFC], retrieved from FRED, Federal Reserve Bank of St. Louis; https://fred.stlouisfed.org/series/WILL5000PRFC, April 28, 2020. On February 14, 2020, the Wilshire 5000 Full Cap Price Index closed at 34,438.39 points, which would indicate a $39.6 trillion total value for the U.S. stock market on that date. This calculation was made using the latest Wilshire update, from 2013, which explains that their index points (measured in billions) should be multiplied by 1.15 in order to achieve an approximate total stock market value. Even though they disclaim that their index does not quite represent the full market size, the Wilshire 5000 Full Cap Price Index is seen as the best proxy for that calculation in the industry. See Wilshire 5000 Family: Wilshire 5000 Total Market Index, Wilshire https://wilshire.com/indexes/wilshire-5000-family/wilshire-5000-total-market-index (last visited May 1, 2020). See also William T. Ziemba et al., Stock Market Crashes: Predictable and Unpredictable and What to Do About Them 135 (World Scientific, 2018).
[110] On March 23, 2020, for example, the Wilshire 5000 Full Cap Price Index closed at 22,463.62 points, which would mean a $25.83 trillion total size for the U.S. stock market on that date. See id.
[111] The number is obtained by multiplying 28,990.17 billion (the Wilshire 5000 Full Cap Price Index’s points on April 28, 2020) by 1.15, for a total of 33,338.69 billion. See id.
[112] Compare Chris White, The Rise and Fall of the Hottest Financial Product in the World, Business Insider (Aug. 15, 2016) (estimating a 60 trillion value), https://www.businessinsider.com/rise-and-fall-of-cds-market; with Janet Morrissey, Credit Default Swaps: The Next Crisis?, TIME (Mar. 17, 2008) (estimating a 47 trillion value), http://content.time.com/time/business/article/0,8599,1723152,00.html.
[113] Larry Cordell et al., Collateral Damage: Sizing and Assessing the Subprime CDO Crisis 21 (Federal Reserve Bank of Philadelphia Working Paper, WP 11-30, 2012), https://philadelphiafed.org/-/media/research-and-data/publications/working-papers/2011/wp11-30R.pdf.
[114] Compare with participation by investment funds, Financial Stability Board, supra note 11 at 23 (“US registered investment funds held approximately US$216 billion in leveraged loans and US$63 billion in CLOs as at December 2018. These holdings represent approximately 12% and 8% of outstanding institutional leveraged loans and CLOs, respectively, but represent less than 1% of U.S. registered investment fund total net assets. U.S. registered investment funds’ holdings of junior CLO tranches are limited and concentrated in a small number of closed-end funds; since mid-2017, these holdings have been declining”). The comment on BIS being influential is a reference to its Basel rules being followed in most of the world including largely in the United States. See Basel Regulatory Framework, Board of Governors of the Federal Reserve System, https://www.federalreserve.gov/supervisionreg/basel/basel-default.htm (last visited May 1, 2020).
[115] Id.
[116] John C. Hull, Risk Management and Financial Institutions 389 (Wiley, 5th ed. 2018). On systemic risk, see Richard S. Carnell et al., supra note 2.
[117] Financial Stability Board, supra note 11 at 2.
[118] S&P Global Market Intelligence, US banks’ CLO security holdings near $100B after 12% jump in 2019 (Mar. 3, 2020), https://www.spglobal.com/marketintelligence/en/news-insights/latest-news-headlines/leveraged-loan-news/us-banks-clo-security-holdings-near-100b-after-12-jump-in-2019.
[119] Financial Stability Board, supra note 11 at 6 (“Adverse shocks to the leveraged loan and CLO markets could impact lenders directly — through their holdings of leveraged loans and CLOs — and indirectly — through exposures to entities investing or participating in those markets”).
[120] See Oaktree, supra note 7 (“In the Global Financial Crisis, leveraged investment vehicles like Collateralized Mortgage Obligations and Collateralized Debt Obligations melted down, bringing losses to the banks that held their junior debt and equity. […] This time, leveraged securitizations are less pervasive in the financial system, and their risk capital wasn’t supplied by banks (thanks to the Volcker Rule), but mostly by non-bank lenders and funds”).
[121] Financial Stability Board, supra note 11 at 21-22. See also id. at 2 (“A comprehensive assessment of the system-wide implications of the exposures of financial institutions to leveraged loans and CLOs is challenging. First, there remain important data gaps. Using supervisory and market data, the direct holders of roughly 79% of leveraged loans and 86% of CLOs were identified in this report. Little is known, in particular, about the direct exposures of certain non-bank investors to these markets, including their holdings of lower-rated CLO tranches. Moreover, limited information on indirect linkages between banks and non-banks makes it difficult to assess possible risks from spillovers and interconnectedness, and their systemic implications. Second, the propagation of adverse developments across the financial sector would depend on the behaviour of the holders of leveraged loans and CLOs in stressed scenarios. This behaviour, in turn, depends on factors such as sources of funding, investment horizons and risk management practices, some of which may encourage shedding of exposures under stressed market conditions that potentially amplify strains. At the same time, more resilient CLO structures might act as a mitigating factor. The magnitude of the stress itself would depend on macroeconomic conditions, and could be compounded during an economic downturn…”).
[122] Board of Governors of the Federal Reserve System, Financial Stability Report 24 (May, 2019) (“…pipeline risk arises between the time when a bank commits to underwrite a leveraged loan and the time when it sells portions of the loan to nonbank investors. If market sentiment shifts abruptly, resulting in lower investor demand for these loans, the bank may need to hold a larger portion of the loan on its balance sheet than it expected”), https://www.federalreserve.gov/publications/files/financial-stability-report-201905.pdf.
[123] Id.
[124] Id.
[125] Board of Governors of the Federal Reserve System, Dodd-Frank Act Stress Test 2019: Supervisory Stress Test Results 13 (June, 2019), https://www.federalreserve.gov/publications/files/2019-dfast-results-20190621.pdf.
[126] Id. at 1, 13-15.
[127] Id. at 22. According to the FSB, the tests were consistent with the “Basel III liquidity framework—particularly the Liquidity Coverage Ratio (LCR) and the Net Stable Funding Ratio (NSFR)”. See Financial Stability Board, supra note 11 at 22.
[128] Board of Governors of the Federal Reserve System, supra note 122 at 24 (“Banks are important sources of funding for businesses and, as a result, face exposure to credit risk through [commercial and industrial]C&I loans, leveraged loans, and CLOs held on their balance sheets. Large banks, in particular, face additional risk given their central role in the leveraged loan market. For example, pipeline risk arises between the time when a bank commits to underwrite a leveraged loan and the time when it sells portions of the loan to nonbank investors. If market sentiment shifts abruptly, resulting in lower investor demand for these loans, the bank may need to hold a larger portion of the loan on its balance sheet than it expected. Finally, banks may also face indirect exposures because they act as lenders to nonbank financial intermediaries that may, in turn, be directly exposed to elevated losses on leveraged loans”).
[129] Id. at 24.
[130] Compare with Mitchell Berlin et al., supra note 54.
[131] Board of Governors of the Federal Reserve System, supra note 122 at 24 (“In particular, they do not rely on funding that must be rolled over before the underlying assets mature, in contrast to other securitization vehicles such as asset-backed commercial paper. Moreover, the investor base for CLOs has become more stable than in the past. CLOs are now predominantly held by investors with relatively stable funding. In contrast, before the financial crisis CLO tranches were commonly held by leveraged structured investment vehicles that relied heavily on short-term wholesale funding”).
[132] Board of Governors of the Federal Reserve System, 2020 Supervisory Scenarios for Annual Stress Tests Required under the Dodd-Frank Act Stress Testing Rules and the Capital Plan Rule 7 (Feb., 2020), https://www.federalreserve.gov/newsevents/pressreleases/files/bcreg20200206a1.pdf.
[133] Id. at 4.
[134] Harriet Torry & Anthony DeBarros, WSJ Survey: Coronavirus to Cause Deep U.S. Contraction, 13% Unemployment, Wall St. J. (Apr. 8, 2020), https://www.wsj.com/articles/wsj-survey-coronavirus-to-cause-deep-u-s-contraction-13-unemployment-11586354400.
[135] Robert Smith & Joe Rennison, Big Banks Left Hanging After ‘Disaster’ In Risky Loan Market, Financial Times (Mar. 30, 2020) https://www.ft.com/content/49ee0c64-cd97-4342-9f03-fec019963fef.
[136] See Deborah Festa et al., Milbank, Client Alert: TALF Expanded to Include AAA Tranches of Static CLOs 3 (Apr. 9, 2020), https://www.milbank.com/images/content/1/3/v2/130259/Client-Alert-TALF-Expanded.pdf. See also Lisa Lee, CLOs, Leveraged Loans Will Miss Most of New Fed Plan Bounty, Bloomberg (Apr. 9, 2020), https://www.bloomberg.com/news/articles/2020-04-09/fed-s-easter-present-won-t-do-much-for-clos-or-leveraged-loans.
[137] Patti Domm, Fed Fires an Even Bigger Bazooka, Expands Its Shopping List to Include Junk Bonds, CNBC (Apr. 9, 2020), https://www.cnbc.com/2020/04/09/fed-fires-an-even-bigger-bazooka-expands-its-shopping-list-to-include-junk-bonds.html.
[138] Rick Green, Howard Marks Bemoans Fed Help for Junk Bonds, Leveraged Debt, Bloomberg, Bloomberg (Apr. 14, 2020), https://www.bloomberg.com/news/articles/2020-04-14/howard-marks-bemoans-fed-help-for-junk-bonds-leveraged-debt.
[139] S&P Global Market Intelligence, US Leveraged Loan Market Claws Back Some Losses After Brutal March (Apr. 13, 2020), https://www.spglobal.com/marketintelligence/en/news-insights/latest-news-headlines/leveraged-loan-news/us-leveraged-loan-market-claws-back-some-losses-after-brutal-march-58011234.
[140] Akin Oyedele, Coronavirus Shock Is Pushing Highly Indebted ‘Zombie’ Companies Toward Financial Ruin, Business Insider (Mar. 12, 2020), https://www.businessinsider.com/coronavirus-sell-off-recession-risks-crisis-zombie-companies-leveraged-loans-2020-3.
[141] S&P Global Market Intelligence, Fed Rally & Default Fears Bring Bifurcation Back To Leveraged Loans (Apr. 30, 2020), https://www.spglobal.com/marketintelligence/en/news-insights/latest-news-headlines/leveraged-loan-news/fed-rally-default-fears-bring-bifurcation-back-to-leveraged-loans.
[142] Mitchell Berlin et al., supra note 54 at 10 (“The most common covenants are tied to an agreed-upon definition of the borrower’s cash flow available for debt service, typically defined as [EBITDA]”).
[143] See Moody’s Investors Service, supra note 48; and Lisa Lee, Leverage Downplayed in Loan Market as Borrowers Get Aggressive, Bloomberg (June 15, 2017) https://www.bloomberg.com/professional/blog/leverage-downplayed-loan-market-borrowers-get-aggressive.
[144] See Michael E. Mariani & Eric S. Goodwin, Cravath, Market Trends 2019: LBO Leveraged Finance 5 https://www.cravath.com/files/Uploads/Documents/Publications/5268285_1.pdf (last visited May 1, 2020); and Marco P. Rodrigues, Miller Thomson, A Canadian Banking Lawyer’s Perspective on EBITDA Add-Backs (April 21, 2016), https://www.millerthomson.com/en/publications/communiques-and-updates/financial-services-restructuring-communique/april-21-2016-financial-services-insolvency-communique/a-canadian-banking-lawyers-perspective-on.
[145] What Is A Leveraged Buyout (LBO)?, Corporate Finance Institute, https://corporatefinanceinstitute.com/resources/knowledge/finance/leveraged-buyout-lbo (last visited May 1, 2020).
[146] Board of Governors of the Federal Reserve System, Federal Deposit Insurance Corporation, Office of the Comptroller of the Currency, Interagency Guideline on Leveraged Lending 1 (Mar. 21, 2013) (“[D]ebt agreements have frequently included features that provided relatively limited lender protection including, but not limited to, the absence of meaningful maintenance covenants in loan agreements […], which lessened lenders’ recourse in the event of a borrower’s subpar performance”), https://www.federalreserve.gov/supervisionreg/srletters/sr1303a1.pdf. See also Michael E. Mariani & Eric S. Goodwin, supra note 144 at 5 (“With deal valuations high, borrowers and creditors have faced pressure to justify significant leverage in marketing (regulators have cautioned about 6x or worse), and borrowers have demanded achievable EBITDA-based tests to make restricted payments, incur indebtedness, and satisfy other covenants. As a result, sponsor documents have incorporated liberal add-backs for calculating EBITDA. In particular, synergy add-backs have become entrenched in the market, despite estimates being widely recognized as aspirational”).
[147] How to Profit from Frontrunning LBOs, Nasdaq (May 19, 2010) (“Blackstone talked to a group of banks, namely Bank of America, Barclays, Citigroup, Credit Suisse, Deutsche Bank, and JP Morgan Chase, to see how much they’d be willing to lend based on $1.4 billion in annual EBITDA. The banks decided they would lend up to $10 billion, or 7x EBITDA”), https://www.nasdaq.com/articles/how-profit-frontrunning-lbos-2010-05-19.
[148] Claire Hill, Why Did Rating Agencies Do Such a Bad Job Rating Subprime Securities?, 71 U. Pitt. L. Rev. 585 (2010), https://scholarship.law.umn.edu/faculty_articles/80.
[149] S&P Global Market Intelligence, When The Cycle Turns: The Continued Attack Of The EBITDA Add-Back (Sept. 19, 2019), https://www.spglobal.com/ratings/en/research/articles/190919-when-the-cycle-turns-the-continued-attack-of-the-ebitda-add-back-11156255.
