A Closer Look at CLOs

New research into collateralized loan obligations (CLOs), which are helping reshape the investment policies of the life/annuity industry, explores whether they have the same vulnerabilities as the 'CDOs' in 'The Big Short,' the book and 2015 film starring actor Steve Carell (in photo), Brad Pitt and others.

Collateralized loan obligations, or CLOs, are a type of asset-backed security (ABS) whose returns, relative to their risks, are providing some extra yield to an annuity industry worn down by a dozen years of Federal Reserve low-interest rate policy. But important questions remain about CLOs.

Those questions resemble the ones that surrounded collateralized debt obligations, or CDOs, during the 2008 financial crisis. Both CLOs and CDOs are bundles of below-investment grade loans or mortgages that have been turned into a single security, segments of which (the “senior tranches”) are rated as investment-grade—and thus permissible for life insurers to buy and hold.

CDOs crashed back in 2008, and the debacle inspired the bestselling book and movie, “The Big Short.” In the movie, investors (played by Christian Bane, Steve Carell and others) reaped fortunes by identifying the most flawed mortgage-backed CDOs and betting against them. Now academic researchers are asking if CLOs might be as problematic as CDOs.

This is important, because asset managers who can select the best high-risk loans and bundle them into out-performing investment-grade fixed income instruments are helping life insurers battle the long interest rate drought. By redeploying a portion of fixed indexed annuity assets into CLO tranches, they are making old FIA books better-funded and new FIA issues more competitive. (See Part I of RIJ’s series on the “Bermuda Triangle” strategy.)

Two scholarly analyses of CLOs appeared in October. In one, credit ratings experts at the University of Texas and MIT ask why the ratings of CLOs didn’t drop as much as the ratings of the debt inside them during the financial crisis that started last March. The second study was written by economists at the Federal Reserve Bank Board and Yale School of Management.

The economists ask if traders (like those played by Bane and Carell) are hacking (my word, not theirs) into CLOs to find out which ones might be mis-priced. These two papers are highly academic, but if you want to understand the CLO market better, they’re worth your time.

Why do CLOs get higher ratings than their components?

In their paper, “Are CLO Collateral and Tranche Ratings Disconnected?” John Griffin (Texas) and Jordan Nickerson (MIT) assert, “CLOs appear considerably riskier than current ratings suggest,” adding that their findings “have current relevance for policymakers.”

“Beginning in March and accelerating in April, both S&P and Moody’s began downgrading large amounts of collateral,” the paper said. “Collateral downgrades stabilized by mid-June, with S&P downgrades standing at 30% of the par-weighted CLO assets while Moody’s downgraded 26% of their rated collateral at the end of August.

One of the figures in the Griffin Nickerson paper. Another is below.

“In contrast, the par-weighted percent of tranches downgraded is 2% for S&P and 1.75% for Moody’s. When considering tranches placed on negative credit watch these magnitudes increase slightly to approximately 3.5% and 5.5%, respectively. It is difficult to reconcile the current lack of tranche rating actions with either rating agency’s methodology, based on prior disclosures or current model outputs,” Griffin and Nickerson wrote.

“We are not able to explain the potential reliance on non-model considerations by rating agencies. Additionally, roughly twice as many CLO tranches would be considered failing by S&P standards had managers not engaged in trading that made the portfolios appear to be safer,” the paper said. Banks, insurance companies, and mutual funds “which hold such assets could contribute to systemic risk if assets are downgraded or experience correlated losses, as often happens with structured products during distress states.” 

‘Privately-produced debt is a double-edged sword’

In “Adverse Selection Dynamics in Privately-Produced Safe Debt Markets,” a new paper by Nathan Foley-Fisher and Stephane´ Verani of the Federal Reserve Board and Gary B. Gorton of Yale, the authors explore an issue closely related to the events in “The Big Short.” It’s the issue of adverse selection. In good times, the authors write, investors take the AAA-rated upper tranches of CLOs literally at face value, even though their underlying contents, structures, and trading strategies are relatively opaque. It just costs too much for investors to evaluate them, and there’s no reason to doubt them. In the authors’ words, this renders them “information-insensitive.”

“CLOs have complicated and opaque internal structures,” the authors explained. “All these attributes make it very expensive for agents to produce private information about the value of the AAA tranche, allowing buyers of this debt to avoid adverse selection because it is very expensive to produce private information. But, in bad times this is exactly the problem!”

In early 2020, as in 2008, investors start looking at CLOs and other ABS more closely. Like the curious investors in “The Big Short,” they sensed that it would be worthwhile to invest considerable time and money to identify mis-pricing that they might capitalize on. As the authors put it, “When the pandemic hit some agents became informed while others did not.”

CLOs suddenly moved from “information-insensitive” to “information-sensitive.” People on one side of a CLO trade knew more than the people on the other side, which the authors described as “a unique risk.” That’s why the trade prices in the CLO market changed dramatically in March 2020, they believe. Like income annuity issuers who know less about the health and longevity risks of their customers than do the clients themselves, traders of CLOs were vulnerable to adverse selection, and weren’t sure how to price the instruments.

(c) 2020 RIJ Publishing LLC.