The Front Men of Wall Street: The Role of CDO Collateral Managers in the CDO Boom and Bust

• DOI: http://doi.org/10.1111/jofi.12520
• Date: 01 October 2017
• Published date: 01 October 2017

THE JOURNAL OF FINANCE •VOL. LXXII, NO. 5 •OCTOBER 2017

The Front Men of Wall Street: The Role of CDO

Collateral Managers in the CDO Boom and Bust

SERGEY CHERNENKO∗

ABSTRACT

I study the incentives of the collateral managers who selected securities for ABS

CDOs—securitizations that figured prominently in the financial crisis. Specialized

managers without other businesses that could suffer negative reputational conse-

quences invested in low-quality securities underwritten by the CDO’s arranger.These

securities performed significantly worse than observationally similar securities. Man-

agers investing in these securities were rewarded with additional collateral manage-

ment assignments. Diversified managers who did assemble CDOs suffered negative

reputational consequences during the crisis: institutional investors withdrew from

their mutual funds. Overall, the results are consistent with a quid pro quo between

collateral managers and CDO underwriters.

STRUCTURED PRODUCTS SUCH AS PRIVATE-LABEL residential mortgage-backed se-

curities (RMBS) and collateralized debt obligations (CDOs) were at the heart

of the recent credit boom and the 2007 to 2009 financial crisis. Given the role

of these securities in the crisis, researchers have analyzed the incentives of the

loan originators,1credit ratings agencies,2and underwriters3bringing these

securities to the market. The literature has also started to explore the demand

for securitizations, pointing to the reliance on credit ratings and regulatory

arbitrage on the one hand4and the neglect of risk on the other hand.5

Most of the demand for the non-AAA-rated tranches of private-label RMBS,

however, came not from end investors but from CDOs that invested in asset-

backed securities (ABS CDOs). During the 2002 to 2007 period, issuance of

∗Sergey Chernenko is with the Ohio State University. This paper was previously circulated

under the title “The Dark Side of Specialization: Evidence from Risk Taking by CDO Collateral

Managers.” I would like to thank Zahi Ben-David, Justin Birru, Jennifer Dlugosz, Isil Erel, Robin

Greenwood, John Griffin, Sam Hanson, Kenneth Singleton (Editor), Ren´

e Stulz, Adi Sunderam,

two anonymous referees, and seminar participants at the Ohio State University and Washington

University in St. Louis for helpful comments. I have read the Journal of Finance’s disclosure policy

and have no conflicts of interest to disclose.

1See Keys et al. (2009,2010) and Keys, Seru, and Vig (2012).

2See Griffin and Tang (2012) and Griffin, Nickerson, and Tang (2013).

3See Griffin, Lowery, and Saretto (2014).

4See Merrill, Nadauld, and Strahan (2014).

5See Gennaioli, Shleifer, and Vishny (2012) and Chernenko, Hanson, and Sunderam (2016).

DOI: 10.1111/jofi.12520

1893

1894 The Journal of Finance R

non-AAA-rated private-label RMBS amounted to about $661 billion.6Over

the same period, $860 billion of ABS CDOs, which invested primarily in non-

AAA-rated RMBS, were issued.7For the vast majority of these CDOs, the

ultimate responsibility for the selection of the underlying collateral rested with

independent asset management firms called CDO collateral managers. Several

years after the crisis, the academic literature still offers almost no systematic

evidence on the identity,incentives, and role played by these asset management

firms.

In the meantime, a popular view that has emerged in the wake of the crisis

and that has been popularized by Michael Lewis in his book The Big Short

is that collateral managers acted as front men for the investment banks un-

derwriting the CDOs: “investors felt better buying a Merrill Lynch CDO if it

didn’t appear to be run by Merrill Lynch” (Lewis (2010, p. 143)). According

to this front men view, collateral managers engaged in a quid pro quo with

investment banks: in return for being handed CDO management assignments,

collateral managers had their CDOs invest in low-quality RMBS collateral that

the investment banks were unable to sell to anyone else. The front men view

has been at the core of a number of recent SEC administrative proceedings and

private lawsuits against collateral managers, claiming breach of contract and

misrepresentation of who would make collateral selection decisions.8Perhaps

the most prominent of these lawsuits are those involving deals sponsored by

Magnetar Capital.

As described in a popular NPR report,9the “Magnetar Trade” involved the

hedge fund sponsoring a number of CDOs by investing in their equity tranches

and simultaneously betting against them by entering into credit default swaps

(CDS) on the CDOs’ collateral securities and higher-rated tranches. At the

same time, Magnetar pressured the CDO collateral managers into purchasing

risky bonds for their portfolios, thereby increasing the probability that the

CDOs would fail and that Magnetar’s short bets would pay off.

Not all collateral managers were willing to abdicate their responsibility for

the selection of collateral, however. Many worried about the possibility of neg-

ative reputational effects for their non-CDO businesses. One example is Ischus

6According to the Securities Industry and Financial Markets Association (SIFMA), $5,328

billion of nonagency RMBS was issued during the 2002 to 2007 period. Given that the

mean “fraction” rated AAA was around 87.6% (Ashcraft, Goldsmith-Pinkham, and Vick-

ery (2011)), issuance of non-AAA-rated RMBS is estimated at 0.124 ×$5,328 billion =

$661 billion.

7This number includes cash flow, hybrid, and synthetic CDOs.

8A typical example is Aozora Bank’s lawsuit against Credit Suisse as the underwriter and

Harding Advisory as the collateral manager. The lawsuit alleges that the collateral manager

breached its contractual obligations to CDO investors and “allowed the CDO’s arranger: (1) to

control a substantial portion of the CDO’s considerable purchasing power, and (2) to direct [the

collateral manager] to purchase collateral products from the CDO arranger’s own inventory.”

(Aozora Bank, Ltd. v. Credit Suisse Group, Docket No. 652274/2013 (N.Y.Sup Ct. June 26, 2013)).

9The episode of “This American Life” was based on ProPublica’s investigation of the financial cri-

sis. The full report is available at https://www.propublica.org/article/all-the-magnetar-trade-how-

one-hedge-fund-helped-keep-the-housing-bubble.

The Front Men of Wall Street 1895

Capital Management, the CDO management subsidiary of Resource America,

a publicly traded asset manager. In September 2006, Ischus and Magnetar

sought to work together on a deal called Hydrus 2006-1. In an email exchange

with Magnetar, Ischus appeared to be uncomfortable with the portfolio of secu-

rities suggested by Magnetar, saying “we will not assemble a portfolio we are

not proud of.”10 As a result of Ischus’s unwillingness to invest in collateral it

considered too risky, negotiations between Ischus and Magnetar fell apart and

the CDO was never issued.

Based on the above discussion, the front men narrative leads to four key

predictions. First, underwriters rewarded managers willing to invest in low-

quality collateral with future collateral management assignments. Second,

because the franchise value of their other businesses was at stake, diver-

sified managers were less likely to engage in bad deal-making in an ef-

fort to attract deal flow. This implies, in turn, that diversified managers

lost market share over time, and that the CDOs of diversified managers

performed better than deals of specialized managers. Third, a key driver

of specialized managers’ underperformance was their willingness to cater

to CDO underwriters by purchasing the banks’ unsold inventory of sub-

prime RMBS and CDOs. And fourth, diversified managers who did as-

semble ABS CDOs suffered negative reputational consequences during the

crisis.

This paper studies the identity, incentives, and performance of the collateral

managers of ABS CDOs issued during the 2002 to 2007 period and provides

evidence that supports the main predictions of the front men narrative. I start

by examining the relationship between the characteristics and performance

of a manager’s past CDOs and his future deal volume. I find that managers

whose past CDOs perform poorly get more deal assignments. While CDO per-

formance was not observable in real time, the results are consistent with the

idea that investment banks knew which collateral managers had invested in

lower quality collateral, which then performed worse ex post. Consistent with

this interpretation, the negative effect of past performance on future deal ac-

tivity works largely through the collateral manager’s investment in securities

underwritten by the CDO’s arranger and securities that Magnetar Capital bet

against.

Given these incentives, which managers were more likely to engage in bad

deal-making to attract deal flow? The front men narrative predicts that man-

agers with low franchise value should be more willing to engage in such behav-

ior. It is worth noting that collateral managers generally have small stakes, if

any, in the CDOs that they manage and that a large fraction of their manage-

ment fees is in the form of upfront and senior fees that do not depend much on

the performance of the CDO. Therefore, from the collateral manager’s perspec-

tive, the major risk in this setting is not so much financial risk as reputational

10 Parts of the email exchange between Ischus and Magnetar were released in the

course of ProPublica’s investigation and are available at http://www.propublica.org/documents/

item/e-mails-between-magnetar-and-ischus.