Abstract

We use credit-arbitrage asset-backed commercial paper vehicles as a laboratory to empirically examine financial institutions’ motivations to take bad-tail systematic risk. By comparing the characteristics of global banks that sponsored credit-arbitrage vehicles prior to the global financial crisis to those that did not, we show that owner–manager agency problems, government safety nets, and government ownership of banks are associated with bad-tail systematic risk-taking. Although good governance is associated with less risk-taking on average, well-governed banks that also have a high ex ante expectation of being bailed out by the government take more risk. Lastly, we find mixed evidence that tougher bank capital regulation deters bad-tail risk-taking.

1. Introduction

Understanding what drives banks’ risk exposure is important for bank governance, regulatory policy design, and macroeconomic and financial stability analysis. Many studies have empirically examined bank risk-taking (see, e.g., Keeley, 1990; Laeven and Levine, 2009; De Jonghe, 2010; DeYoung and Torna, 2013), but these studies have largely focused on the average risk from all projects using measures such as bank loan losses or equity volatility. In contrast, we analyze the determinants of systematic bad-tail risk, meaning exposure to large losses in low-probability, high-risk premium states of the world. Data limitations typically prevent researchers from disentangling systematic bad-tail risk from other types of risk; as a result, we know very little about its determinants. Understanding these determinants is important because bad-tail risk is costly to hedge and poses particular threats to the stability of the financial system.

In this article, we use bank-sponsored credit-arbitrage asset-backed commercial paper (credit-arb ABCP) vehicles to offer evidence on the reasons why major banks exposed themselves to systematic bad-tail risk in the period leading up to the Global Financial Crisis (GFC) that began in 2007. We find that government safety nets and shareholder–manager agency problems are correlated with bad-tail systematic risk-taking decisions, but capital regulation is not robustly related to this type of risk-taking.1

In July 2007, just before a run on their liabilities began, credit-arb ABCP vehicles had about $700 billion in assets. The vast majority were sponsored by banks that provided their vehicles with committed backup lines of credit and other support, so sponsors bore the vehicles’ risks. When vehicles experiencing ABCP runoffs turned to their sponsors for help, the sponsors sought large amounts of new funding in interbank and other money markets. Most sponsors were European banks, but most vehicle assets and liabilities were denominated in US dollars; thus, sponsoring banks were forced to raise funds outside their home money markets and their national central banks were not immediately able to provide dollar liquidity support. These circumstances increased the cost to sponsors and also helped transmit the ABCP shock throughout the global financial system.

Credit-arb ABCP vehicles were a pure play on bad-tail systematic risk: sponsoring banks only suffered losses in extremely bad states of the world. The vehicles’ strategy was to go long the high-grade credit spread with mismatched funding and very high leverage.2 They borrowed at the short end of the credit spread term structure by issuing ABCP and lent at intermediate and long maturities, primarily by investing in AAA and AA-rated asset-backed securities (ABS). The vehicles were almost immune to all but two events: significant declines in the credit quality of high-grade ABS, or a prolonged loss of access to funding in the ABCP market. Such events were predictably more likely to occur in tandem with high-risk premiums or broad disruptions in financial markets, making the costs to sponsors of distressed vehicles especially large. The credit-arb vehicle strategies had some similarity to those of Long-Term Capital Management (LTCM), which have been likened to “picking up nickels in front of a steamroller” because they produce low per-unit returns with modest volatility in most states of the world, but large negative returns in some states of the world (Duarte, Longstaff, and Yu, 2007).

For our research design, bank-sponsored ABCP vehicles need to satisfy two requirements: first, the vehicle needs to primarily follow a credit spread arbitrage strategy, and second, the vehicle should not invest in assets originated by the sponsoring bank. The credit-arb vehicles that we study meet both conditions and allow us to observe a bad-tail risk decision that is isolated from the bank’s other business decisions. Moreover, we can identify the universe of potential credit-arb vehicle sponsors because only large banks have the capital and market access needed to convince rating agencies that the ABCP issued by these vehicles is nearly riskless (without the low funding costs associated with top-rated ABCP, the vehicles would not be profitable). Within this universe, both sponsor and non-sponsor banks existed in the USA and throughout most of Europe. By studying variations in the regulatory treatment, governance, safety net protection, and other characteristics of sponsor and non-sponsor banks, we can learn about the determinants of bad-tail risk-taking decisions.

Immediately before the GFC, there were three distinct regulatory regimes with respect to credit-arb ABCP vehicles. Throughout most of Europe, banks could completely avoid capital charges by placing assets in such off-balance sheet vehicles.3 While US-based banks faced substantially higher regulatory costs than the typical European-based bank, these banks still had lower capital charges on credit-arb vehicles relative to on-balance sheet assets. In contrast, banks in Spain and Portugal paid the same capital charges on sponsored vehicle assets as they did for on-balance sheet assets. To the extent that sponsorship decisions were motivated solely by the desire to avoid capital requirements, these differences in regulatory framework would predict that Spanish and Portuguese banks would not sponsor vehicles, while US banks would be less likely than other European banks to sponsor vehicles.

The evidence in support of these predictions is mixed. There were no Spanish or Portuguese bank sponsors of credit-arb vehicles, which is consistent with the idea that sponsorship was not feasible without some degree of regulatory capital relief. However, we find that US banks were actually more likely to become sponsors than European banks, other things equal, which suggests that avoiding capital charges was not the only determinant of bad-tail systematic risk decisions.

We next estimate cross-sectional regressions using other bank characteristics and find evidence that both implicit and explicit government support are associated with bad-tail systematic risk-taking. Banks that ex ante were expected to be bailed out by the government were significantly more likely to become sponsors. Similarly, the existence of a government-backed deposit insurance system and the dollar level of protection in that system are positively correlated with sponsorship. The effects of government safety nets are large; for example, the existence of a government-funded deposit insurance scheme is associated with an increase in the probability of sponsoring a credit-arb vehicle of about 14 percentage points, or nearly 60% relative to the baseline level of sponsorship.

We also find that shareholder–manager agency problems are correlated with decisions to sponsor credit-arb ABCP vehicles. Banks with managers that had better aligned incentives—for example, banks that used more incentive-compatible compensation practices—were on average less likely to become sponsors. Additionally, we find similar relations for banks with a large shareholder or with higher institutional ownership—both of which arguably reduce agency conflicts. These findings are consistent with Ellul and Yerramilli’s (2013) result showing that stronger internal risk controls reduce tail risk-taking, and suggest that good governance may reduce the propensity to take bad-tail systematic risks.

At first glance, this result is surprising given that several papers find evidence that better governance practices led to more risk-taking in the run-up to the GFC (Fahlenbrach and Stulz, 2011; Minton, Taillard, and Williamson, 2014; Cheng, Hong, and Scheinkman, 2015). We reconcile our results with this evidence by showing that the relationship between risk-taking and governance depends on expected government support. For the subset of banks with high implicit government support, better compensation practices, higher insider and institutional ownership, and the presence of large shareholders were associated with a significantly higher probability of sponsoring credit-arb vehicles. Together, these results suggest that good governance practices may inhibit bad-tail systematic risk-taking on average, but could encourage this type of risk-taking when shareholders expect to be bailed out by the government in the event of financial trouble.

German Landesbanks represent an extreme case of government support, as they are government controlled. These banks were particularly prone to sponsor credit-arb ABCP vehicles. Moreover, Landesbank-sponsored vehicles were more likely to be a relatively large fraction of the sponsor’s assets, so the chance of severe distress at the sponsoring Landesbank as a result of vehicle distress was larger. While it is possible that some of the Landesbank sponsorship was due to bad governance (Hau and Thum, 2009), we find that the Landesbanks with fewer potential agency conflicts were the most likely to become sponsors. Importantly, the results described earlier are robust to excluding Landesbanks from the sample.

Finally, we provide suggestive evidence that sponsorship was not simply a mistake by banks that did not understand the risks they were taking. At the time of the sponsorship decision, the most salient risks had been well-demonstrated by multiple historical examples of market meltdowns in commercial paper and ABS markets (Gatev and Strahan, 2006). Additionally, we show that late-adopters of sponsorship, who had several years to learn about the risks of sponsorship by observing the early-adopting banks, were no less likely to fail, be acquired due to financial duress, or receive a government bailout during the GFC. We also document that sponsorship expanded during the mid-2000s at the same time that vehicle profitability plummeted. Since bad-tail systematic risk was unlikely to be decreasing over this period, these sponsorship patterns are inconsistent with undistorted risk-taking behavior.

The empirical setting we use in this article allows us to rule out many alternative explanations for our results, including differences in time-invariant country-level characteristics as well as sample selection bias. However, the details surrounding the creation and sponsorship of credit-arb vehicles do not allow us to employ any of the usual methods for establishing causality. Consequently, this article provides evidence that government support and owner–manager agency problems are associated with bad-tail systematic risk-taking, but we do not claim that this relationship is causal.

The most closely related paper in the literature is Acharya, Schnabl, and Suarez (2013) (referred to as Acharya et al., hereafter), which provides evidence that sponsors bore the risks associated with ABCP vehicles and, in contrast to our article, suggests that avoidance of regulatory capital requirements was the main motivation for sponsorship. Our article differs from theirs in two important respects: the focus on tail risk and the consideration of additional motivations for sponsorship.

While we analyze the thirty-five sponsors of credit-arb vehicles, Acharya et al. analyze a pool of 127 sponsors of all types of ABCP vehicles. These other types of ABCP vehicles do not meet the requirements of our research design. In particular, multiseller vehicles, which dominate the Acharya et al. sample, did not primarily use a credit-arbitrage investment strategy; instead, they bought receivables or other short-term assets typically financed with bank lines of credit. Consequently, the maturity transformation typical of credit-arb vehicles was absent; additionally, sellers of receivables typically provided credit enhancements which further reduced the risk to the sponsoring bank. These vehicles also heavily invested in assets originated by the sponsoring bank, and many banks offered trade financing to their clients through multiseller vehicles. The other noncredit-arb vehicles in the Acharya et al. sample are described in the Online Appendix; though the details vary, all of these vehicles invested in assets that were connected to the sponsoring bank’s business. As a result, the risk profile of the typical vehicle studied in Acharya et al. is very similar to the average risk of the bank’s loan portfolio. We find it persuasive that banks’ primary motivation to sponsor such vehicles was reduction of required capital, and it is not surprising that they find evidence of this in a sample dominated by noncredit-arb vehicles.

Viewed in this light, we do not see our results as conflicting with Acharya et al. Instead, we view the combined papers as evidence that capital regulation matters for average bank risk-taking, but is not the only or even a material motivator for bad-tail systematic risk-taking. Moreover, we expand upon Acharya et al. by exploring the role of government guarantees and bank governance practices for sponsoring ABCP vehicles, neither of which were examined in their paper.

While our focus in this article is on understanding bank risk choices, our work is also related to the causes and dynamics of the GFC (e.g., Brunnermeier, 2009; Gorton, 2010; Covitz, Liang, and Suarez, 2013) and on the role of shadow banks (Adrian et al., 2010; Gorton et al., 2010). Our work also bears some similarity to recent studies on rare disasters in the presence of heterogeneity, such as Chen, Joslin, and Tran (2012), and the pricing of bad-tail risk, such as Bates (1991), though we analyze risk decisions, not asset valuations. Understanding the conditions that led banks to seek exposure to bad-tail systematic risk in the form of credit-arb ABCP vehicles helps further illuminate the role of the shadow-banking sector in the GFC. Our results might also help shareholders and regulators better understand hedge fund risks, since credit-arb vehicles are similar to hedge funds in some respects (e.g., Fung and Hsieh, 2001; Chan et al., 2006).

The remainder of the article is structured as follows. Section 2 describes hypotheses and identification strategies, while Section 3 describes ABCP vehicles and the risks they pose to sponsors, and also describes how credit-arb vehicles differ from others. Section 4 describes the data and Section 5 presents results. Section 6 provides concluding remarks.

2. Hypotheses and Identification

Prior literature has offered several potential motivations for imprudent bank risk-taking, including: financial regulation, government safety nets, agency problems and corporate governance failures, mistakes in the form of underestimation of risk (Calomiris, 2009; Coval, Jurek, and Stafford, 2009), and ex ante first-best decisions that simply turned out poorly ex post.4 Because it is difficult to explicitly test for bad luck, we focus our article on exploring the influence of regulation, safety nets, and governance failures.

Three types of government-related distortions might lead banks to sponsor ABCP vehicles. First, banks’ risk-reward tradeoffs may be altered by regulations (Acharya, Schnabl, and Suarez, 2013) or, second, by deposit insurance or other implicit or explicit government “safety nets” (Merton, 1977; Keeley, 1990; Gropp, Hakenes, and Schnabel, 2011). Third, government ownership or control of banks might also distort their decision-making, either because of agency problems introduced by government ownership or because of the implied government guarantee (Dinc, 2005).

Another hypothesis is that risk-taking decisions were distorted by agency problems. Gorton and Rosen (1995) suggest that low-skill bank managers may take negative-net present value bad-tail risks in order to boost earnings and avoid being fired (such managers accept a higher probability of being fired eventually in order to avoid being fired soon when low earnings reveal their type). Compensation-related problems may also be material.5 For example, a combination of opaque bank risk postures and incentive compensation may cause bank CEOs to take tail risk that shareholders cannot observe in an attempt to boost their compensation.

To better understand the role of government-induced distortions and agency problems in bad-tail systematic risk-taking, we exploit the institutional details of credit-arb vehicle sponsorship. Three features, discussed in detail in the next section, are particularly important for our study: credit-arb vehicles were designed to transfer systematic bad-tail risk to their sponsors, vehicle assets were not originated by the sponsor, and sponsors were predominately large banks.

Typically, it is difficult to disentangle bad-tail systematic risks from other types of risks and risk-taking motivations from other business decisions, but the first two features of credit-arb vehicle sponsorship allow us to do so in this setting. Furthermore, we observe in our sample every institution that has the option to take this form of systematic risk. This eliminates many concerns about sample selection that would exist if, for example, most credit-arb vehicles were sponsored by nonbank institutions. We use this sample to examine the characteristics that are associated with bank sponsorship of credit-arb vehicles.

One additional advantage of our setting is that the results are unlikely to be explained by the potential impact of the credit-arb vehicle on the sponsoring bank. Vehicle sizes and profits were usually small relative to the sponsoring bank. Consequently, it is unlikely that the sponsorship of these vehicles led banks to change their governance environment.6 However, we cannot definitively rule out the possibility that some third variable influences both sponsorship and corporate governance proxies.

3. Credit-arb Vehicles and Risks to Investors and Sponsors

This section presents information about credit-arb vehicle design and operation. Credit-arb vehicles include securities arbitrage vehicles (SAVs), structured investment vehicles (SIVs), and selected hybrid vehicles that have portfolios similar to SAVs.7 We provide additional details, including a brief summary of ABCP vehicles not studied in this article, in the Online Appendix.

Prior to the GFC, credit-arb vehicles invested in diversified portfolios of assets, mostly ABS, that appeared to pose low credit risk (about 90% of assets were rated AAA and AA before the crisis and almost all the remainder were rated A).8 These vehicles diversified both by investing in a variety of types of ABS and other bonds (see Figure 1), and also by virtue of the fact that each ABS was backed by a pool of loans that were diversified along some dimensions. Importantly, credit-arb vehicles generally did not invest in ABS originated by the sponsoring bank, so the risk-taking decision was clearly separated from the bank’s other lending businesses.

Figure 1.

Credit-arbitrage vehicle asset mix as of March 2007. This figure shows the percentage breakdown of each asset type held by the average credit-arbitrage vehicle in March 2007. Source: Moody’s Investors Service.

Figure 1.

Credit-arbitrage vehicle asset mix as of March 2007. This figure shows the percentage breakdown of each asset type held by the average credit-arbitrage vehicle in March 2007. Source: Moody’s Investors Service.

To fund their portfolio, credit-arb vehicles sold short-term ABCP, primarily to prime money market funds. In order to obtain this low-cost short-term funding, vehicles sought credit enhancements and committed backup lines of credit from a sponsor. In return, the sponsor received most of the net revenue (“excess spread”) from a vehicle after funding costs and operating expenses. While in principle any firm could sponsor a credit-arb vehicle, in practice only large banks had the financial capital necessary to issue credible guarantees. As shown in Table I, more than 80% of credit-arb ABCP outstanding as of June 2007 was sponsored by US and European banks. At this date, credit-arb vehicles accounted for $418 billion in ABCP outstanding ($346 billion sponsored by US and European banks), about one-quarter of total ABCP outstanding, which totaled $1,464 billion.

Table I.

ABCP outstanding by vehicle type

This table shows the total amount of ABCP outstanding by vehicle type at the end of the second quarter of 2007, measured in millions of US dollars. The top panel shows ABCP outstanding for the types of vehicles that we study in this article, while the bottom panel covers other types of vehicles. The middle column examines vehicles sponsored by US and European banks (the focus of this article), while the right column shows total ABCP outstanding.

Vehicle typeSponsored by major US and European banks Total
Hybrid 4,851 17,956 
Sec. Arbitrage 274,426 296,347 
SIV 66,831 104,094 

 
Subtotal 346,108 418,397 

 
Collateralized Debt Obligation (CDO) 2,642 68,833 
Loan-backed 3,539 3,539 
Multiseller 456,125 651,224 
Repo/Total return swap 4,979 83,251 
Single-seller 122,966 227,104 
Other 5,806 11,936 
Total 942,165 1,464,284 
Vehicle typeSponsored by major US and European banks Total
Hybrid 4,851 17,956 
Sec. Arbitrage 274,426 296,347 
SIV 66,831 104,094 

 
Subtotal 346,108 418,397 

 
Collateralized Debt Obligation (CDO) 2,642 68,833 
Loan-backed 3,539 3,539 
Multiseller 456,125 651,224 
Repo/Total return swap 4,979 83,251 
Single-seller 122,966 227,104 
Other 5,806 11,936 
Total 942,165 1,464,284 
Table I.

ABCP outstanding by vehicle type

This table shows the total amount of ABCP outstanding by vehicle type at the end of the second quarter of 2007, measured in millions of US dollars. The top panel shows ABCP outstanding for the types of vehicles that we study in this article, while the bottom panel covers other types of vehicles. The middle column examines vehicles sponsored by US and European banks (the focus of this article), while the right column shows total ABCP outstanding.

Vehicle typeSponsored by major US and European banks Total
Hybrid 4,851 17,956 
Sec. Arbitrage 274,426 296,347 
SIV 66,831 104,094 

 
Subtotal 346,108 418,397 

 
Collateralized Debt Obligation (CDO) 2,642 68,833 
Loan-backed 3,539 3,539 
Multiseller 456,125 651,224 
Repo/Total return swap 4,979 83,251 
Single-seller 122,966 227,104 
Other 5,806 11,936 
Total 942,165 1,464,284 
Vehicle typeSponsored by major US and European banks Total
Hybrid 4,851 17,956 
Sec. Arbitrage 274,426 296,347 
SIV 66,831 104,094 

 
Subtotal 346,108 418,397 

 
Collateralized Debt Obligation (CDO) 2,642 68,833 
Loan-backed 3,539 3,539 
Multiseller 456,125 651,224 
Repo/Total return swap 4,979 83,251 
Single-seller 122,966 227,104 
Other 5,806 11,936 
Total 942,165 1,464,284 

Details of credit enhancements varied across vehicles, but in most cases their purpose was to limit variation in vehicle net-asset value associated with changes in ratings of individual assets in the vehicle’s portfolio, and to achieve a low risk that individual assets would default.9 Liquidity backup lines of credit for the full amount of vehicle assets were also usually bought from the sponsor. If ABCP funding became unavailable for any reason, the provider of the line would pay off ABCP investors and assume the risks posed by portfolio assets.10

These contractual guarantees were weaker for SIVs than for SAVs (see the Online Appendix for details).11 However, almost all bank-sponsored SIVs were rescued by their sponsors during the GFC. Thus, as a practical matter, the sponsor of a credit-arb vehicle effectively bore all of the credit, market, and liquidity risks associated with the vehicle’s portfolio.

This risk was systematic bad-tail risk: ABCP investors were likely to withdraw funding in only three circumstances: (i) an exogenous disruption in money markets, in which case the sponsoring bank was likely to be forced to enter interbank markets to fund the vehicle at a time when funding liquidity was impaired; (ii) a sharp deterioration in the credit quality of a substantial fraction of vehicle assets, which was likely to occur only when market credit spreads were high and when buying vehicle assets at par would impose substantial mark-to-market losses on the sponsor; and (iii) when doubts arose about the ability of the sponsor to meet its obligations to the vehicle. In the latter case, the sponsor’s own distress was likely to be worsened by a need to attract interbank deposits to fund the vehicle. In August and September of 2007, both (i) and (ii) occurred, and (iii) occurred for some sponsors.

4. Data

We use information on the characteristics of ABCP vehicles and their sponsors, focusing on the period ending in mid-2007. Data on ABCP vehicles are from Moody’s Investors Service (Moody’s), primarily from quarterly “Program Index” spreadsheets, which include information on the characteristics of all ABCP vehicles rated by Moody’s.12 The spreadsheets capture well over 90% of global vehicles by assets. Figure 2 shows the amount of ABCP outstanding by vehicle type starting at year-end 1999. Growth was positive until 2007 and accelerated after 2004. ABCP outstanding increased by a factor of five between 2000 and 2007 for the set of credit-arb vehicles included in our sample, a much larger increase than at other types of vehicles. After the GFC began in the summer of 2007, issuance and outstandings decreased sharply for credit-arb vehicles. Multiseller ABCP held steady for several quarters, but began to fall as output and demand for financing from nonfinancial firms and consumers plummeted during the peak of the crisis in late 2008.

Figure 2.

ABCP vehicle global CP outstanding by vehicle type. This figure shows the total quarterly amount of ABCP outstanding by vehicle type from December 1999 to December 2011. Source: Moody’s Investors Service.

Figure 2.

ABCP vehicle global CP outstanding by vehicle type. This figure shows the total quarterly amount of ABCP outstanding by vehicle type from December 1999 to December 2011. Source: Moody’s Investors Service.

We include in our analysis only credit-arb vehicles sponsored by US and European banks.13Table I shows that as of the second quarter of 2007, this sample captures over 90% of outstanding ABCP for SAVs. Our sample only includes about 65% of outstanding ABCP for SIVs, primarily because a few large SIVs were sponsored by UK nonbanks. We are unable to determine the sponsor of one large hybrid vehicle; as a result, our sample represents just over a quarter of the outstanding hybrid ABCP.14 Throughout the article, our results are robust to eliminating SIV and hybrid vehicles.

We obtain balance sheet and income statement data about sponsor banks from Bankscope. We collect annual data on sponsors’ underwriting activity for high-yield bonds, ABS, and mortgage-backed securities (MBS) from Bloomberg and Dealogic’s DCM Analytics.

Figure 3 shows the amount of bank-sponsored credit-arbitrage vehicle ABCP outstanding grouped by the domicile of the vehicles’ sponsor. ABCP outstanding from European bank-sponsored vehicles grew rapidly from 2002.15 Vehicles sponsored by US banks grew moderately and ABCP outstanding was less than half the amount for European-sponsored vehicles by June 2007.

Figure 3.

Global ABCP outstanding by region. This figure plots the total quarterly amount of ABCP outstanding grouped by the domicile of the vehicles’ sponsor from December 1999 to December 2007. Only vehicles sponsored by banks are included. Europe is defined as all members in the EU-15 (the fifteen Western European member countries of the European Union before its expansion in 2004) plus Norway and Switzerland. Source: Moody’s Investors Service.

Figure 3.

Global ABCP outstanding by region. This figure plots the total quarterly amount of ABCP outstanding grouped by the domicile of the vehicles’ sponsor from December 1999 to December 2007. Only vehicles sponsored by banks are included. Europe is defined as all members in the EU-15 (the fifteen Western European member countries of the European Union before its expansion in 2004) plus Norway and Switzerland. Source: Moody’s Investors Service.

To evaluate the propensity of banks to sponsor vehicles, we must include both sponsor and non-sponsor banks in the sample. We include all sponsors as well as non-sponsor banks with $25 billion or more in total assets that were domiciled in the USA or in Europe. Small banks do not have the financial capacity to sponsor credit-arb vehicles; the smallest sponsor bank in our sample is Mellon Bank with assets of just over $25 billion. Our results are not sensitive to this cutoff and are robust to including banks with assets greater than $10 billion (untabulated). Table II presents the number of sponsor and non-sponsor banks in each country in mid-2007. On aggregate, we have thirty-five sponsoring banks out of a total sample of 145 banks. Germany has a relatively large number of sample banks and an especially large number of sponsor banks. Many German sponsors are government-owned or controlled, such as Landesbanken. As described below, our results are robust to excluding Landesbanken from the sample.

Table II.

Number of banks with assets greater than $25 billion by country

This table presents counts of the number of sponsor and non-sponsor banks in each country at the end of the second quarter of 2007. To be included as a sponsor, a bank must sponsor at least one SIV, securities arbitrage, or hybrid ABCP vehicle. Non-sponsors must have at least $25 billion in total assets to be part of our sample.

CountryBanks by country
Not sponsoring ABCP vehiclesSponsoring ABCP vehiclesTotal
Austria 
Belgium 
Denmark 
Finland 
France 12 14 
Germany 13 11 25 
Greece 
Ireland 
Italy 12 13 
Luxembourg 
The Netherlands 
Norway 
Portugal 
Spain 
Sweden 
Switzerland 
UK 11 
USA 25 35 
Total 110 35 145 
CountryBanks by country
Not sponsoring ABCP vehiclesSponsoring ABCP vehiclesTotal
Austria 
Belgium 
Denmark 
Finland 
France 12 14 
Germany 13 11 25 
Greece 
Ireland 
Italy 12 13 
Luxembourg 
The Netherlands 
Norway 
Portugal 
Spain 
Sweden 
Switzerland 
UK 11 
USA 25 35 
Total 110 35 145 
Table II.

Number of banks with assets greater than $25 billion by country

This table presents counts of the number of sponsor and non-sponsor banks in each country at the end of the second quarter of 2007. To be included as a sponsor, a bank must sponsor at least one SIV, securities arbitrage, or hybrid ABCP vehicle. Non-sponsors must have at least $25 billion in total assets to be part of our sample.

CountryBanks by country
Not sponsoring ABCP vehiclesSponsoring ABCP vehiclesTotal
Austria 
Belgium 
Denmark 
Finland 
France 12 14 
Germany 13 11 25 
Greece 
Ireland 
Italy 12 13 
Luxembourg 
The Netherlands 
Norway 
Portugal 
Spain 
Sweden 
Switzerland 
UK 11 
USA 25 35 
Total 110 35 145 
CountryBanks by country
Not sponsoring ABCP vehiclesSponsoring ABCP vehiclesTotal
Austria 
Belgium 
Denmark 
Finland 
France 12 14 
Germany 13 11 25 
Greece 
Ireland 
Italy 12 13 
Luxembourg 
The Netherlands 
Norway 
Portugal 
Spain 
Sweden 
Switzerland 
UK 11 
USA 25 35 
Total 110 35 145 

Table III lists the sponsor banks, the number of vehicles they sponsor, and the total ABCP outstanding in dollars and as a fraction of their total assets. About one-fifth of the banks in the list are from the USA (eight banks). The remainder are from Europe (twenty-seven banks, representing about 72% of sponsors). In most cases, the vehicles are modest in size relative to sponsor bank assets. (The biggest outliers in this regard, IKB Bank and Sachsen Landesbank, both government-controlled, failed early in the crisis; results presented below are robust to their elimination from the sample.) As shown in column (2), only $48 billion of a total of $347 billion of ABCP outstanding at vehicles of the type we analyze was in US-sponsored vehicles.

Table III.

List of banks sponsoring credit-arb ABCP vehicles

This table lists the sponsor banks in our sample at the end of the second quarter of 2007. A sponsor bank is required to sponsor at least one SIV, securities arbitrage, or hybrid ABCP vehicle. Banks that we perceive to be widely recognized global banks are shown in boldface.

Bank nameSecond quarter 2007
ABCP vehiclesTotal ABCP (US$ millions)Total ABCP to total assets (%)
European banks  

 
ABN Amro 9,263 0.7 
Barclays 3,840 0.2 
Bayerische Landesbank 12,687 2.9 
Commerzbank 1,007 0.1 
Danske Bank 2,500 0.5 
Deutsche Bank 6,391 0.4 
Deutsche Zentral 4,033 0.7 
Dresdner Bank 5,292 0.7 
Fortis 26,375 2.6 
HBOS 36,002 3.1 
HSBC Holdings 32,918 1.8 
HSH Nordbank 9,174 3.7 
IKB Deutsche Industriebank 18,577 22.0 
ING Groep 10,964 0.7 
KBC Group 4,266 1.0 
Landesbank Baden-Wuerttemberg 9,113 1.7 
Landesbank Berlin Holding 2,138 1.1 
Lloyds TSB Group 22,889 3.4 
Nationwide Building Society 2,936 1.4 
Natixis 2,820 0.5 
NIBC Holding 506 1.2 
Rabobank Group 15,181 2.1 
Sachsen LB 17,875 23.1 
Société Générale 724 0.1 
Standard Chartered 6,205 2.3 
UniCredito Italiano 19,289 1.8 
WestLB 16,096 4.3 

 
Subtotal 41 299,061   

 
US banks  

 
Bank of America 2,685 0.2 
Bank of New York 139 0.2 
Citigroup Inc 26,021 1.4 
JP Morgan Chase & Co. 3,352 0.2 
Mellon Bank 3,790 14.5 
State Street Corporation 4,188 3.9 
Wachovia Corporation 3,641 0.5 
Zions Bancorporation 3,736 8.0 

 
Subtotal 16 47,552   

 
Total 57 346,613   
Bank nameSecond quarter 2007
ABCP vehiclesTotal ABCP (US$ millions)Total ABCP to total assets (%)
European banks  

 
ABN Amro 9,263 0.7 
Barclays 3,840 0.2 
Bayerische Landesbank 12,687 2.9 
Commerzbank 1,007 0.1 
Danske Bank 2,500 0.5 
Deutsche Bank 6,391 0.4 
Deutsche Zentral 4,033 0.7 
Dresdner Bank 5,292 0.7 
Fortis 26,375 2.6 
HBOS 36,002 3.1 
HSBC Holdings 32,918 1.8 
HSH Nordbank 9,174 3.7 
IKB Deutsche Industriebank 18,577 22.0 
ING Groep 10,964 0.7 
KBC Group 4,266 1.0 
Landesbank Baden-Wuerttemberg 9,113 1.7 
Landesbank Berlin Holding 2,138 1.1 
Lloyds TSB Group 22,889 3.4 
Nationwide Building Society 2,936 1.4 
Natixis 2,820 0.5 
NIBC Holding 506 1.2 
Rabobank Group 15,181 2.1 
Sachsen LB 17,875 23.1 
Société Générale 724 0.1 
Standard Chartered 6,205 2.3 
UniCredito Italiano 19,289 1.8 
WestLB 16,096 4.3 

 
Subtotal 41 299,061   

 
US banks  

 
Bank of America 2,685 0.2 
Bank of New York 139 0.2 
Citigroup Inc 26,021 1.4 
JP Morgan Chase & Co. 3,352 0.2 
Mellon Bank 3,790 14.5 
State Street Corporation 4,188 3.9 
Wachovia Corporation 3,641 0.5 
Zions Bancorporation 3,736 8.0 

 
Subtotal 16 47,552   

 
Total 57 346,613   
Table III.

List of banks sponsoring credit-arb ABCP vehicles

This table lists the sponsor banks in our sample at the end of the second quarter of 2007. A sponsor bank is required to sponsor at least one SIV, securities arbitrage, or hybrid ABCP vehicle. Banks that we perceive to be widely recognized global banks are shown in boldface.

Bank nameSecond quarter 2007
ABCP vehiclesTotal ABCP (US$ millions)Total ABCP to total assets (%)
European banks  

 
ABN Amro 9,263 0.7 
Barclays 3,840 0.2 
Bayerische Landesbank 12,687 2.9 
Commerzbank 1,007 0.1 
Danske Bank 2,500 0.5 
Deutsche Bank 6,391 0.4 
Deutsche Zentral 4,033 0.7 
Dresdner Bank 5,292 0.7 
Fortis 26,375 2.6 
HBOS 36,002 3.1 
HSBC Holdings 32,918 1.8 
HSH Nordbank 9,174 3.7 
IKB Deutsche Industriebank 18,577 22.0 
ING Groep 10,964 0.7 
KBC Group 4,266 1.0 
Landesbank Baden-Wuerttemberg 9,113 1.7 
Landesbank Berlin Holding 2,138 1.1 
Lloyds TSB Group 22,889 3.4 
Nationwide Building Society 2,936 1.4 
Natixis 2,820 0.5 
NIBC Holding 506 1.2 
Rabobank Group 15,181 2.1 
Sachsen LB 17,875 23.1 
Société Générale 724 0.1 
Standard Chartered 6,205 2.3 
UniCredito Italiano 19,289 1.8 
WestLB 16,096 4.3 

 
Subtotal 41 299,061   

 
US banks  

 
Bank of America 2,685 0.2 
Bank of New York 139 0.2 
Citigroup Inc 26,021 1.4 
JP Morgan Chase & Co. 3,352 0.2 
Mellon Bank 3,790 14.5 
State Street Corporation 4,188 3.9 
Wachovia Corporation 3,641 0.5 
Zions Bancorporation 3,736 8.0 

 
Subtotal 16 47,552   

 
Total 57 346,613   
Bank nameSecond quarter 2007
ABCP vehiclesTotal ABCP (US$ millions)Total ABCP to total assets (%)
European banks  

 
ABN Amro 9,263 0.7 
Barclays 3,840 0.2 
Bayerische Landesbank 12,687 2.9 
Commerzbank 1,007 0.1 
Danske Bank 2,500 0.5 
Deutsche Bank 6,391 0.4 
Deutsche Zentral 4,033 0.7 
Dresdner Bank 5,292 0.7 
Fortis 26,375 2.6 
HBOS 36,002 3.1 
HSBC Holdings 32,918 1.8 
HSH Nordbank 9,174 3.7 
IKB Deutsche Industriebank 18,577 22.0 
ING Groep 10,964 0.7 
KBC Group 4,266 1.0 
Landesbank Baden-Wuerttemberg 9,113 1.7 
Landesbank Berlin Holding 2,138 1.1 
Lloyds TSB Group 22,889 3.4 
Nationwide Building Society 2,936 1.4 
Natixis 2,820 0.5 
NIBC Holding 506 1.2 
Rabobank Group 15,181 2.1 
Sachsen LB 17,875 23.1 
Société Générale 724 0.1 
Standard Chartered 6,205 2.3 
UniCredito Italiano 19,289 1.8 
WestLB 16,096 4.3 

 
Subtotal 41 299,061   

 
US banks  

 
Bank of America 2,685 0.2 
Bank of New York 139 0.2 
Citigroup Inc 26,021 1.4 
JP Morgan Chase & Co. 3,352 0.2 
Mellon Bank 3,790 14.5 
State Street Corporation 4,188 3.9 
Wachovia Corporation 3,641 0.5 
Zions Bancorporation 3,736 8.0 

 
Subtotal 16 47,552   

 
Total 57 346,613   

We use proxies for managerial incentives and bank governance to test the importance of agency problems. The main executive compensation measure is from RiskMetrics Group’s Corporate Governance Quotient (CGQ). We follow Aggarwal et al. (2010) and use the questions included in the CGQ index related to compensation and ownership. For each bank, we measure the percentage of compensation and ownership attributes that satisfy a threshold defined by the RiskMetrics Group. A higher value of Compensation Index means that the bank has established compensation practices that are more shareholder friendly. We also use the percent of shares held by individual insiders, collected from FactSet/Lionshares, as another measure of incentive alignment.

We follow Erkens, Hung, and Matos (2012) in creating two corporate governance indicators. One is the share of institutional ownership at each bank, from FactSet/Lionshares. A second is concentration of ownership, denoted Blockholder and measured by an indicator variable equal to 1 if a bank has at least one single owner with voting shares larger than 10%, constructed using data from FactSet/Lionshares, Bankscope, and banks’ annual reports.

A portion of our sample of banks is not covered by these databases, typically those that are not publicly traded or do not publish standardized financial disclosures. We hand collect information about the compensation, governance, and ownership structure of such banks by using annual reports, banks’ websites, and other publicly-available sources. This allows us to expand our sample by about twenty banks. We were not able to collect such information for remaining banks, as governance and compensation disclosures were not common in some countries prior to the GFC.

When examining the role that safety nets played in vehicle sponsorship, we use a bank-specific measure of government support similar to that in Brandao-Marques, Correa, and Sapriza (2018) and Correa et al. (2014). We exploit the fact that Moody’s assigns two separate ratings to each bank that it covers. The bank financial strength rating (BFSR) is intended to measure the bank’s intrinsic safety and is constructed to ignore external support that the bank might receive from any other entity (including the government). The bank deposit rating measures the bank’s ability to repay its deposit obligations and incorporates both the BFSR and Moody’s opinion of the likelihood and amount of external support. Our government support measure is defined as the difference (in rating notches) between a bank’s BFSR and its long-term foreign currency deposit rating.

Ratings-based measures of expected government support have two advantages. First, they capture the rating agencies’ assessment of the likelihood of bank bailouts by the government conditional on banks entering a state of distress. Second, these support measures are bank-specific, which facilitates the identification of the impact of expected government support on different outcome variables across banks within a country. Ratings based measures of expected government support have been extensively used in the literature. Some studies have used them to assess the informativeness of market signals, such as bond and equity prices, and to infer bank default probabilities conditional on expected government support (Gropp, Vesala, and Vulpes, 2006). Others have focused on the impact of expected government support on competition and risk-taking (Gropp, Hakenes, and Schnabel, 2011). Recently, ratings-based measures of government support have been used to assess the impact of too-big-to-fail implicit guarantees on bond (Acharya, Anginer, Warburton, 2016) and equity prices (Correa et al., 2014).

We use two additional measures of explicit government support: the existence of a government-backed deposit insurance system and the natural logarithm of the government-mandated minimum required amount of coverage, measured in US dollars, in that system. These measures are obtained from Demirgüç-Kunt, Kane, and Laeven (2015). We exclude voluntary privately backed insurance amounts, because we believe that the implicit government support of this insurance is weaker than for the government-mandated portion.

Table IV reports summary statistics for our sample and Table V examines how bank and country characteristics vary across sponsor and non-sponsor banks. Definitions and sources for each of the variables are in the Appendix. Table V reveals some differences between sponsor and non-sponsor banks: As of June 2007, relative to non-sponsors, sponsor banks were larger, were more complex in their assets and activities, but less profitable, had smaller loans to assets and equity to assets ratios, and underwrote larger volumes of ABS and MBS.

Table IV.

Sample summary statistics

This table presents summary statistics for the variables used in our analysis. The sample consists of all banks domiciled in the USA or in Europe with at least $25 billion in total assets. Sponsor is a dummy variable equal to one if the bank sponsored at least one SIV, securities arbitrage, or hybrid ABCP vehicle as of June 2007. The remaining variables are measured as of December 2006; definitions for each variable are found in the Appendix.

VariableNMeanMedianStd DevMinMax
Sponsor 145 0.24 0.00 0.43 0.00 1.00 
Total Assets (billion dollars) 145 339 104 488 25 1957 
ln(Total Assets) 145 11.89 11.56 1.28 10.13 14.49 
Return on Assets 145 0.89 0.80 0.54 0.06 3.07 
Equity to Assets 145 6.33 6.00 3.38 1.44 19.07 
Loans to Assets 145 52.45 57.73 20.44 0.26 93.07 
Deposits to Assets 145 54.04 57.92 18.14 0.18 91.35 
Non-Interest Operating Income to Assets 145 1.54 1.25 1.44 –0.13 8.17 
High Yield Underwriter 145 0.16 0.00 0.37 0.00 1.00 
Securitization Underwriter 145 0.33 0.00 0.47 0.00 1.00 
Dummy Landesbank 145 0.06 0.00 0.24 0.00 1.00 
Government Support 145 0.49 0.00 1.40 –1.00 8.25 
Probability of Support 145 0.10 0.00 0.26 0.00 1.00 
Deposit Insurance 145 0.66 1.00 0.48 0.00 1.00 
Level of Deposit Insurance 145 2.44 3.00 1.87 0.00 4.61 
Compensation Index 111 0.58 0.57 0.27 0.00 1.00 
Insider Ownership 115 0.16 0.03 0.24 0.00 1.00 
Institutional Ownership 113 0.31 0.29 0.27 0.00 1.00 
Blockholder 145 0.51 1.00 0.50 0.00 1.00 
First Mover 145 0.13 0.00 0.34 0.00 1.00 
VariableNMeanMedianStd DevMinMax
Sponsor 145 0.24 0.00 0.43 0.00 1.00 
Total Assets (billion dollars) 145 339 104 488 25 1957 
ln(Total Assets) 145 11.89 11.56 1.28 10.13 14.49 
Return on Assets 145 0.89 0.80 0.54 0.06 3.07 
Equity to Assets 145 6.33 6.00 3.38 1.44 19.07 
Loans to Assets 145 52.45 57.73 20.44 0.26 93.07 
Deposits to Assets 145 54.04 57.92 18.14 0.18 91.35 
Non-Interest Operating Income to Assets 145 1.54 1.25 1.44 –0.13 8.17 
High Yield Underwriter 145 0.16 0.00 0.37 0.00 1.00 
Securitization Underwriter 145 0.33 0.00 0.47 0.00 1.00 
Dummy Landesbank 145 0.06 0.00 0.24 0.00 1.00 
Government Support 145 0.49 0.00 1.40 –1.00 8.25 
Probability of Support 145 0.10 0.00 0.26 0.00 1.00 
Deposit Insurance 145 0.66 1.00 0.48 0.00 1.00 
Level of Deposit Insurance 145 2.44 3.00 1.87 0.00 4.61 
Compensation Index 111 0.58 0.57 0.27 0.00 1.00 
Insider Ownership 115 0.16 0.03 0.24 0.00 1.00 
Institutional Ownership 113 0.31 0.29 0.27 0.00 1.00 
Blockholder 145 0.51 1.00 0.50 0.00 1.00 
First Mover 145 0.13 0.00 0.34 0.00 1.00 
Table IV.

Sample summary statistics

This table presents summary statistics for the variables used in our analysis. The sample consists of all banks domiciled in the USA or in Europe with at least $25 billion in total assets. Sponsor is a dummy variable equal to one if the bank sponsored at least one SIV, securities arbitrage, or hybrid ABCP vehicle as of June 2007. The remaining variables are measured as of December 2006; definitions for each variable are found in the Appendix.

VariableNMeanMedianStd DevMinMax
Sponsor 145 0.24 0.00 0.43 0.00 1.00 
Total Assets (billion dollars) 145 339 104 488 25 1957 
ln(Total Assets) 145 11.89 11.56 1.28 10.13 14.49 
Return on Assets 145 0.89 0.80 0.54 0.06 3.07 
Equity to Assets 145 6.33 6.00 3.38 1.44 19.07 
Loans to Assets 145 52.45 57.73 20.44 0.26 93.07 
Deposits to Assets 145 54.04 57.92 18.14 0.18 91.35 
Non-Interest Operating Income to Assets 145 1.54 1.25 1.44 –0.13 8.17 
High Yield Underwriter 145 0.16 0.00 0.37 0.00 1.00 
Securitization Underwriter 145 0.33 0.00 0.47 0.00 1.00 
Dummy Landesbank 145 0.06 0.00 0.24 0.00 1.00 
Government Support 145 0.49 0.00 1.40 –1.00 8.25 
Probability of Support 145 0.10 0.00 0.26 0.00 1.00 
Deposit Insurance 145 0.66 1.00 0.48 0.00 1.00 
Level of Deposit Insurance 145 2.44 3.00 1.87 0.00 4.61 
Compensation Index 111 0.58 0.57 0.27 0.00 1.00 
Insider Ownership 115 0.16 0.03 0.24 0.00 1.00 
Institutional Ownership 113 0.31 0.29 0.27 0.00 1.00 
Blockholder 145 0.51 1.00 0.50 0.00 1.00 
First Mover 145 0.13 0.00 0.34 0.00 1.00 
VariableNMeanMedianStd DevMinMax
Sponsor 145 0.24 0.00 0.43 0.00 1.00 
Total Assets (billion dollars) 145 339 104 488 25 1957 
ln(Total Assets) 145 11.89 11.56 1.28 10.13 14.49 
Return on Assets 145 0.89 0.80 0.54 0.06 3.07 
Equity to Assets 145 6.33 6.00 3.38 1.44 19.07 
Loans to Assets 145 52.45 57.73 20.44 0.26 93.07 
Deposits to Assets 145 54.04 57.92 18.14 0.18 91.35 
Non-Interest Operating Income to Assets 145 1.54 1.25 1.44 –0.13 8.17 
High Yield Underwriter 145 0.16 0.00 0.37 0.00 1.00 
Securitization Underwriter 145 0.33 0.00 0.47 0.00 1.00 
Dummy Landesbank 145 0.06 0.00 0.24 0.00 1.00 
Government Support 145 0.49 0.00 1.40 –1.00 8.25 
Probability of Support 145 0.10 0.00 0.26 0.00 1.00 
Deposit Insurance 145 0.66 1.00 0.48 0.00 1.00 
Level of Deposit Insurance 145 2.44 3.00 1.87 0.00 4.61 
Compensation Index 111 0.58 0.57 0.27 0.00 1.00 
Insider Ownership 115 0.16 0.03 0.24 0.00 1.00 
Institutional Ownership 113 0.31 0.29 0.27 0.00 1.00 
Blockholder 145 0.51 1.00 0.50 0.00 1.00 
First Mover 145 0.13 0.00 0.34 0.00 1.00 
Table V

Bank characteristics by sponsorship decision

This table reports the mean value as of December 2006 of various bank-level characteristics for banks that sponsored at least one SIV, securities arbitrage, or hybrid ABCP vehicle as of June 2007 and banks that did not sponsor any of these vehicle types. The sample consists of all banks domiciled in the USA or in Europe with at least $25 billion in total assets. We also report the difference in means between sponsor and non-sponsor banks and perform a two-sample t test to determine if this difference is statistically significant; standard errors are shown in parenthesis. *** and * indicate significance at the 1% and 10% levels, respectively. Definitions for the variables are found in the Appendix.

CharacteristicSponsorNon-sponsorDifference
Total Assets 739.627 211.837 527.790*** 
(106.169) 
Return on Assets 0.745 0.934 –0.189* 
(0.107) 
Equity to Assets 4.829 6.806 –1.976*** 
(0.552) 
Loans to Assets 41.701 55.867 –14.167*** 
(3.501) 
Deposits to Assets 54.263 53.971 0.292 
(2.841) 
Non-Interest Operating Income to Assets 1.601 1.515 0.086 
(0.279) 
High Yield Underwriter 0.457 0.064 0.394*** 
(0.089) 
Securitization Underwriter 0.743 0.200 0.543*** 
 (0.084) 

 
Observations 35 110 145 
CharacteristicSponsorNon-sponsorDifference
Total Assets 739.627 211.837 527.790*** 
(106.169) 
Return on Assets 0.745 0.934 –0.189* 
(0.107) 
Equity to Assets 4.829 6.806 –1.976*** 
(0.552) 
Loans to Assets 41.701 55.867 –14.167*** 
(3.501) 
Deposits to Assets 54.263 53.971 0.292 
(2.841) 
Non-Interest Operating Income to Assets 1.601 1.515 0.086 
(0.279) 
High Yield Underwriter 0.457 0.064 0.394*** 
(0.089) 
Securitization Underwriter 0.743 0.200 0.543*** 
 (0.084) 

 
Observations 35 110 145 
Table V

Bank characteristics by sponsorship decision

This table reports the mean value as of December 2006 of various bank-level characteristics for banks that sponsored at least one SIV, securities arbitrage, or hybrid ABCP vehicle as of June 2007 and banks that did not sponsor any of these vehicle types. The sample consists of all banks domiciled in the USA or in Europe with at least $25 billion in total assets. We also report the difference in means between sponsor and non-sponsor banks and perform a two-sample t test to determine if this difference is statistically significant; standard errors are shown in parenthesis. *** and * indicate significance at the 1% and 10% levels, respectively. Definitions for the variables are found in the Appendix.

CharacteristicSponsorNon-sponsorDifference
Total Assets 739.627 211.837 527.790*** 
(106.169) 
Return on Assets 0.745 0.934 –0.189* 
(0.107) 
Equity to Assets 4.829 6.806 –1.976*** 
(0.552) 
Loans to Assets 41.701 55.867 –14.167*** 
(3.501) 
Deposits to Assets 54.263 53.971 0.292 
(2.841) 
Non-Interest Operating Income to Assets 1.601 1.515 0.086 
(0.279) 
High Yield Underwriter 0.457 0.064 0.394*** 
(0.089) 
Securitization Underwriter 0.743 0.200 0.543*** 
 (0.084) 

 
Observations 35 110 145 
CharacteristicSponsorNon-sponsorDifference
Total Assets 739.627 211.837 527.790*** 
(106.169) 
Return on Assets 0.745 0.934 –0.189* 
(0.107) 
Equity to Assets 4.829 6.806 –1.976*** 
(0.552) 
Loans to Assets 41.701 55.867 –14.167*** 
(3.501) 
Deposits to Assets 54.263 53.971 0.292 
(2.841) 
Non-Interest Operating Income to Assets 1.601 1.515 0.086 
(0.279) 
High Yield Underwriter 0.457 0.064 0.394*** 
(0.089) 
Securitization Underwriter 0.743 0.200 0.543*** 
 (0.084) 

 
Observations 35 110 145 

5. Results for Determinants of Vehicle Sponsorship

In this section, we test hypotheses that distortions from government regulations, safety nets, and agency problems motivated banks’ sponsorship of ABCP vehicles. We end with informal evidence about the possibility that sponsorship decisions resulted from undistorted risk-return tradeoffs.

We use a linear probability model to examine the effect of government and agency distortions on ABCP sponsorship. The dependent variable is an indicator equal to 1 for banks that sponsored at least one credit-arb vehicle and 0 otherwise. Specifically, we estimate
Pr(Sponsori=1)= α+βBank Traitsi+ρj
(1)
for each bank i in country j as of June 2007. Independent variables are measured, in most cases, as of the previous December.16 The primary advantage of the linear probability model is that it allows us to control for country fixed effects (ρj), which helps alleviate concerns that our results are driven by unobservable cross-country differences.17 Standard errors (reported in parentheses) are clustered at the country level.

Even though sponsoring a credit-arb vehicle is a bad-tail systematic risk that is mostly isolated from a bank’s other business decisions, banks that focus on traditional activities might find it more costly to sponsor a vehicle. The univariate statistics reported in Table V are suggestive of this, because sponsor banks have much lower loans-to-assets ratios and are much more likely to underwrite high-yield debt and securitized assets. To control for this, we include deposit-to-assets and loans-to-assets ratios as proxies for the degree to which a bank engages in traditional activities, and noninterest operating income-to-assets ratios and indicator variables for high-yield debt and securitized asset underwriting as proxies for the degree to which a bank is involved in more exotic activities. The inclusion of these proxies alleviates concerns that differences in business models across the two groups bias our results.

We do not analyze in detail the determinants of credit-arb vehicle size. A barrier to empirical analysis is that the proper normalization is not obvious.18 We run Tobit regressions of credit-arb vehicle assets as a fraction of the sponsor’s total assets and of the sponsor’s total equity on the predictors shown in Table VI, VII, and VIII and obtain qualitatively similar results (see the Online Appendix). However, because vehicle size was modest relative to bank assets in most cases, its value as a measure of systematic bad-tail risk is not clear. In contrast, the existence of a sponsored vehicle was a signal to counterparties early in the crisis that a bank had taken systematic bad-tail risk, in addition to increasing the bank’s contingent funding needs and credit losses in bad states of the world.

Table VI.

Capital regulation and sponsorship of credit-arb vehicles

This table shows the results from estimating a linear probability model in which the dependent variable is a binary indicator equal to 1 for banks that sponsored at least one SIV, securities arbitrage, or hybrid ABCP vehicle. We examine sponsorship as of June 2007. Most independent variables are measured as of December of the previous year. Dummy US indicates banks that are headquartered in the USA; these banks face higher regulatory capital charges when sponsoring ABCP vehicles than most European-based banks, but sponsorship still offers some regulatory relief as compared to on-balance sheet financing. Dummy Spain/Portugal indicates banks that are headquartered in these countries; these banks receive no regulatory capital relief for sponsoring vehicles. Definitions for the other independent variables are found in the Appendix. Region fixed effects are defined as indicator variables for banks headquartered in the U.S., the British Isles, Northern Europe, Western Europe, and Southern Europe. The intercept is suppressed to conserve space. Standard errors (shown in parenthesis) are clustered at the country level. ***, **, * indicates significance at the 1%, 5%, and 10% level, respectively.


Dependent variable: Sponsoring a credit-arbitrage ABCP Vehicle = 1
Independent Variable(1)(2)(3)(4)
Dummy US 0.082*  0.055 — 
(0.047) (0.045) 
Dummy Spain/Portugal  –0.199** –0.185* –0.139* 
(0.082) (0.090) (0.071) 
ln(Total Assets) 0.050 0.047 0.048 0.040 
(0.037) (0.036) (0.035) (0.031) 
Return on Assets –0.139** –0.135*** –0.136*** –0.125*** 
(0.051) (0.045) (0.046) (0.043) 
Equity to Assets –0.019** –0.019* –0.021** –0.019* 
(0.008) (0.011) (0.010) (0.009) 
Loans to Assets –0.001 –0.000 –0.000 –0.001 
(0.003) (0.003) (0.003) (0.003) 
Deposits to Assets 0.001 0.002 0.001 0.001 
(0.001) (0.001) (0.001) (0.001) 
Non-Interest Operating Income to Assets 0.051*** 0.059*** 0.055** 0.053** 
(0.017) (0.020) (0.020) (0.018) 
High Yield Underwriter 0.197 0.189 0.185 0.199 
(0.178) (0.186) (0.184) (0.186) 
Securitization Underwriter 0.282** 0.303** 0.299** 0.300** 
(0.114) (0.119) (0.121) (0.122) 

 
Region fixed effects No No No Yes 
Observations 145 145 145 145 
Countries 18 18 18 18 
Adjusted R-squared 0.349 0.358 0.355 0.353 

Dependent variable: Sponsoring a credit-arbitrage ABCP Vehicle = 1
Independent Variable(1)(2)(3)(4)
Dummy US 0.082*  0.055 — 
(0.047) (0.045) 
Dummy Spain/Portugal  –0.199** –0.185* –0.139* 
(0.082) (0.090) (0.071) 
ln(Total Assets) 0.050 0.047 0.048 0.040 
(0.037) (0.036) (0.035) (0.031) 
Return on Assets –0.139** –0.135*** –0.136*** –0.125*** 
(0.051) (0.045) (0.046) (0.043) 
Equity to Assets –0.019** –0.019* –0.021** –0.019* 
(0.008) (0.011) (0.010) (0.009) 
Loans to Assets –0.001 –0.000 –0.000 –0.001 
(0.003) (0.003) (0.003) (0.003) 
Deposits to Assets 0.001 0.002 0.001 0.001 
(0.001) (0.001) (0.001) (0.001) 
Non-Interest Operating Income to Assets 0.051*** 0.059*** 0.055** 0.053** 
(0.017) (0.020) (0.020) (0.018) 
High Yield Underwriter 0.197 0.189 0.185 0.199 
(0.178) (0.186) (0.184) (0.186) 
Securitization Underwriter 0.282** 0.303** 0.299** 0.300** 
(0.114) (0.119) (0.121) (0.122) 

 
Region fixed effects No No No Yes 
Observations 145 145 145 145 
Countries 18 18 18 18 
Adjusted R-squared 0.349 0.358 0.355 0.353 
Table VI.

Capital regulation and sponsorship of credit-arb vehicles

This table shows the results from estimating a linear probability model in which the dependent variable is a binary indicator equal to 1 for banks that sponsored at least one SIV, securities arbitrage, or hybrid ABCP vehicle. We examine sponsorship as of June 2007. Most independent variables are measured as of December of the previous year. Dummy US indicates banks that are headquartered in the USA; these banks face higher regulatory capital charges when sponsoring ABCP vehicles than most European-based banks, but sponsorship still offers some regulatory relief as compared to on-balance sheet financing. Dummy Spain/Portugal indicates banks that are headquartered in these countries; these banks receive no regulatory capital relief for sponsoring vehicles. Definitions for the other independent variables are found in the Appendix. Region fixed effects are defined as indicator variables for banks headquartered in the U.S., the British Isles, Northern Europe, Western Europe, and Southern Europe. The intercept is suppressed to conserve space. Standard errors (shown in parenthesis) are clustered at the country level. ***, **, * indicates significance at the 1%, 5%, and 10% level, respectively.


Dependent variable: Sponsoring a credit-arbitrage ABCP Vehicle = 1
Independent Variable(1)(2)(3)(4)
Dummy US 0.082*  0.055 — 
(0.047) (0.045) 
Dummy Spain/Portugal  –0.199** –0.185* –0.139* 
(0.082) (0.090) (0.071) 
ln(Total Assets) 0.050 0.047 0.048 0.040 
(0.037) (0.036) (0.035) (0.031) 
Return on Assets –0.139** –0.135*** –0.136*** –0.125*** 
(0.051) (0.045) (0.046) (0.043) 
Equity to Assets –0.019** –0.019* –0.021** –0.019* 
(0.008) (0.011) (0.010) (0.009) 
Loans to Assets –0.001 –0.000 –0.000 –0.001 
(0.003) (0.003) (0.003) (0.003) 
Deposits to Assets 0.001 0.002 0.001 0.001 
(0.001) (0.001) (0.001) (0.001) 
Non-Interest Operating Income to Assets 0.051*** 0.059*** 0.055** 0.053** 
(0.017) (0.020) (0.020) (0.018) 
High Yield Underwriter 0.197 0.189 0.185 0.199 
(0.178) (0.186) (0.184) (0.186) 
Securitization Underwriter 0.282** 0.303** 0.299** 0.300** 
(0.114) (0.119) (0.121) (0.122) 

 
Region fixed effects No No No Yes 
Observations 145 145 145 145 
Countries 18 18 18 18 
Adjusted R-squared 0.349 0.358 0.355 0.353 

Dependent variable: Sponsoring a credit-arbitrage ABCP Vehicle = 1
Independent Variable(1)(2)(3)(4)
Dummy US 0.082*  0.055 — 
(0.047) (0.045) 
Dummy Spain/Portugal  –0.199** –0.185* –0.139* 
(0.082) (0.090) (0.071) 
ln(Total Assets) 0.050 0.047 0.048 0.040 
(0.037) (0.036) (0.035) (0.031) 
Return on Assets –0.139** –0.135*** –0.136*** –0.125*** 
(0.051) (0.045) (0.046) (0.043) 
Equity to Assets –0.019** –0.019* –0.021** –0.019* 
(0.008) (0.011) (0.010) (0.009) 
Loans to Assets –0.001 –0.000 –0.000 –0.001 
(0.003) (0.003) (0.003) (0.003) 
Deposits to Assets 0.001 0.002 0.001 0.001 
(0.001) (0.001) (0.001) (0.001) 
Non-Interest Operating Income to Assets 0.051*** 0.059*** 0.055** 0.053** 
(0.017) (0.020) (0.020) (0.018) 
High Yield Underwriter 0.197 0.189 0.185 0.199 
(0.178) (0.186) (0.184) (0.186) 
Securitization Underwriter 0.282** 0.303** 0.299** 0.300** 
(0.114) (0.119) (0.121) (0.122) 

 
Region fixed effects No No No Yes 
Observations 145 145 145 145 
Countries 18 18 18 18 
Adjusted R-squared 0.349 0.358 0.355 0.353 
Table VII.

Government support and sponsorship of credit-arb vehicles

This table shows the results from estimating a linear probability model in which the dependent variable is a binary indicator equal to 1 for banks that sponsored at least one SIV, securities arbitrage, or hybrid ABCP vehicle. We examine sponsorship as of June 2007. Most independent variables are measured as of December of the previous year. Government Support is a measure of expected government support, defined for each bank as Moody’s long-term foreign currency deposit rating minus Moody’s bank financial strength rating (see Brandao-Marques, Correa, and Sapriza, 2018). Probability of Support is an alternative rating-based measure proposed by Gropp, Hakenes, and Schnabel (2011), which adds the rating agencies’ information on default frequencies across different rating buckets to infer the probability of bank support, Deposit Insurance is an indicator variable equal to 1 if the country has government managed deposit insurance. Level of Deposit Insurance is the natural logarithm of the US dollar value of the deposit insurance limit for government-backed deposit insurance systems. Definitions for the other independent variables are found in the Appendix. Region fixed effects are defined as indicator variables for banks headquartered in the USA, the British Isles, Northern Europe, Western Europe, and Southern Europe. The intercept is suppressed to conserve space. Standard errors (shown in parenthesis) are clustered at the country level. ***, **, * indicates significance at the 1%, 5%, and 10% level, respectively.


Dependent variable: Sponsoring a credit-arbitrage ABCP vehicle = 1
Independent Variable(1)(2)(3)(4)(5)
Government Support 0.061***     
(0.019) 
Probability of Support  0.133    
(0.174) 
Dummy Landesbank   0.318***   
(0.034) 
Deposit Insurance    0.160**  
(0.075) 
Level of Deposit Insurance     0.015** 
(0.006) 
ln(Total Assets) 0.039 0.040 0.037 0.045 0.045 
(0.037) (0.035) (0.035) (0.029) (0.029) 
Return on Assets –0.091** –0.092** –0.105*** –0.132** –0.126** 
(0.036) (0.038) (0.035) (0.048) (0.045) 
Equity to Assets –0.006 –0.007 –0.007 –0.015 –0.014 
(0.010) (0.010) (0.011) (0.009) (0.009) 
Loans to Assets –0.001 –0.002 –0.001 –0.000 –0.000 
(0.003) (0.003) (0.003) (0.003) (0.003) 
Deposits to Assets 0.001 0.002 0.001 0.001 0.000 
(0.002) (0.002) (0.002) (0.001) (0.001) 
Non-Interest Operating Income to Assets 0.056*** 0.054** 0.062** 0.065** 0.064** 
(0.019) (0.020) (0.024) (0.025) (0.025) 
High Yield Underwriter 0.288 0.254 0.272 0.194 0.195 
(0.193) (0.198) (0.200) (0.188) (0.188) 
Securitization Underwriter 0.199* 0.221* 0.221* 0.306** 0.307** 
(0.107) (0.123) (0.125) (0.123) (0.124) 

 
Country fixed effects Yes Yes Yes No No 
Region fixed effects No No No Yes Yes 
Observations 145 145 145 145 145 
Countries 18 18 18 18 18 
Adjusted R-squared 0.415 0.389 0.408 0.366 0.370 

Dependent variable: Sponsoring a credit-arbitrage ABCP vehicle = 1
Independent Variable(1)(2)(3)(4)(5)
Government Support 0.061***     
(0.019) 
Probability of Support  0.133    
(0.174) 
Dummy Landesbank   0.318***   
(0.034) 
Deposit Insurance    0.160**  
(0.075) 
Level of Deposit Insurance     0.015** 
(0.006) 
ln(Total Assets) 0.039 0.040 0.037 0.045 0.045 
(0.037) (0.035) (0.035) (0.029) (0.029) 
Return on Assets –0.091** –0.092** –0.105*** –0.132** –0.126** 
(0.036) (0.038) (0.035) (0.048) (0.045) 
Equity to Assets –0.006 –0.007 –0.007 –0.015 –0.014 
(0.010) (0.010) (0.011) (0.009) (0.009) 
Loans to Assets –0.001 –0.002 –0.001 –0.000 –0.000 
(0.003) (0.003) (0.003) (0.003) (0.003) 
Deposits to Assets 0.001 0.002 0.001 0.001 0.000 
(0.002) (0.002) (0.002) (0.001) (0.001) 
Non-Interest Operating Income to Assets 0.056*** 0.054** 0.062** 0.065** 0.064** 
(0.019) (0.020) (0.024) (0.025) (0.025) 
High Yield Underwriter 0.288 0.254 0.272 0.194 0.195 
(0.193) (0.198) (0.200) (0.188) (0.188) 
Securitization Underwriter 0.199* 0.221* 0.221* 0.306** 0.307** 
(0.107) (0.123) (0.125) (0.123) (0.124) 

 
Country fixed effects Yes Yes Yes No No 
Region fixed effects No No No Yes Yes 
Observations 145 145 145 145 145 
Countries 18 18 18 18 18 
Adjusted R-squared 0.415 0.389 0.408 0.366 0.370 
Table VII.

Government support and sponsorship of credit-arb vehicles

This table shows the results from estimating a linear probability model in which the dependent variable is a binary indicator equal to 1 for banks that sponsored at least one SIV, securities arbitrage, or hybrid ABCP vehicle. We examine sponsorship as of June 2007. Most independent variables are measured as of December of the previous year. Government Support is a measure of expected government support, defined for each bank as Moody’s long-term foreign currency deposit rating minus Moody’s bank financial strength rating (see Brandao-Marques, Correa, and Sapriza, 2018). Probability of Support is an alternative rating-based measure proposed by Gropp, Hakenes, and Schnabel (2011), which adds the rating agencies’ information on default frequencies across different rating buckets to infer the probability of bank support, Deposit Insurance is an indicator variable equal to 1 if the country has government managed deposit insurance. Level of Deposit Insurance is the natural logarithm of the US dollar value of the deposit insurance limit for government-backed deposit insurance systems. Definitions for the other independent variables are found in the Appendix. Region fixed effects are defined as indicator variables for banks headquartered in the USA, the British Isles, Northern Europe, Western Europe, and Southern Europe. The intercept is suppressed to conserve space. Standard errors (shown in parenthesis) are clustered at the country level. ***, **, * indicates significance at the 1%, 5%, and 10% level, respectively.


Dependent variable: Sponsoring a credit-arbitrage ABCP vehicle = 1
Independent Variable(1)(2)(3)(4)(5)
Government Support 0.061***     
(0.019) 
Probability of Support  0.133    
(0.174) 
Dummy Landesbank   0.318***   
(0.034) 
Deposit Insurance    0.160**  
(0.075) 
Level of Deposit Insurance     0.015** 
(0.006) 
ln(Total Assets) 0.039 0.040 0.037 0.045 0.045 
(0.037) (0.035) (0.035) (0.029) (0.029) 
Return on Assets –0.091** –0.092** –0.105*** –0.132** –0.126** 
(0.036) (0.038) (0.035) (0.048) (0.045) 
Equity to Assets –0.006 –0.007 –0.007 –0.015 –0.014 
(0.010) (0.010) (0.011) (0.009) (0.009) 
Loans to Assets –0.001 –0.002 –0.001 –0.000 –0.000 
(0.003) (0.003) (0.003) (0.003) (0.003) 
Deposits to Assets 0.001 0.002 0.001 0.001 0.000 
(0.002) (0.002) (0.002) (0.001) (0.001) 
Non-Interest Operating Income to Assets 0.056*** 0.054** 0.062** 0.065** 0.064** 
(0.019) (0.020) (0.024) (0.025) (0.025) 
High Yield Underwriter 0.288 0.254 0.272 0.194 0.195 
(0.193) (0.198) (0.200) (0.188) (0.188) 
Securitization Underwriter 0.199* 0.221* 0.221* 0.306** 0.307** 
(0.107) (0.123) (0.125) (0.123) (0.124) 

 
Country fixed effects Yes Yes Yes No No 
Region fixed effects No No No Yes Yes 
Observations 145 145 145 145 145 
Countries 18 18 18 18 18 
Adjusted R-squared 0.415 0.389 0.408 0.366 0.370 

Dependent variable: Sponsoring a credit-arbitrage ABCP vehicle = 1
Independent Variable(1)(2)(3)(4)(5)
Government Support 0.061***     
(0.019) 
Probability of Support  0.133    
(0.174) 
Dummy Landesbank   0.318***   
(0.034) 
Deposit Insurance    0.160**  
(0.075) 
Level of Deposit Insurance     0.015** 
(0.006) 
ln(Total Assets) 0.039 0.040 0.037 0.045 0.045 
(0.037) (0.035) (0.035) (0.029) (0.029) 
Return on Assets –0.091** –0.092** –0.105*** –0.132** –0.126** 
(0.036) (0.038) (0.035) (0.048) (0.045) 
Equity to Assets –0.006 –0.007 –0.007 –0.015 –0.014 
(0.010) (0.010) (0.011) (0.009) (0.009) 
Loans to Assets –0.001 –0.002 –0.001 –0.000 –0.000 
(0.003) (0.003) (0.003) (0.003) (0.003) 
Deposits to Assets 0.001 0.002 0.001 0.001 0.000 
(0.002) (0.002) (0.002) (0.001) (0.001) 
Non-Interest Operating Income to Assets 0.056*** 0.054** 0.062** 0.065** 0.064** 
(0.019) (0.020) (0.024) (0.025) (0.025) 
High Yield Underwriter 0.288 0.254 0.272 0.194 0.195 
(0.193) (0.198) (0.200) (0.188) (0.188) 
Securitization Underwriter 0.199* 0.221* 0.221* 0.306** 0.307** 
(0.107) (0.123) (0.125) (0.123) (0.124) 

 
Country fixed effects Yes Yes Yes No No 
Region fixed effects No No No Yes Yes 
Observations 145 145 145 145 145 
Countries 18 18 18 18 18 
Adjusted R-squared 0.415 0.389 0.408 0.366 0.370 
Table VIII.

Agency conflicts and sponsorship of credit-arb vehicles

This table shows the results from estimating a linear probability model in which the dependent variable is a binary indicator equal to 1 for banks that sponsored at least one SIV, securities arbitrage, or hybrid ABCP vehicle. We examine sponsorship as of June 2007. Most independent variables are measured as of December of the previous year. We use several measures of potential agency conflicts: Compensation Index is calculated from RiskMetrics Group’s CGQ as in Aggarwal et al. (2010); higher values represent compensation practices that are better aligned with shareholder value. Insider Ownership is the percent of shares owned by individual insiders, while Institutional Ownership is the percent of shares owned by institutions. Blockholder is an indicator variable that is equal to 1 if the bank has at least one single owner with voting shares larger than 10%. In Panels A, B, and C, we interact these measures of agency conflicts with various proxies for government support. In Panel A, we define government support using High Probability of Support, which is an indicator equal to 1 for banks which have a ratings-based probability of support (defined as in Gropp, Hakenes, and Schnabel, 2011) greater than 0. In Panel B, we define government support using High Government Support, which is an indicator for banks that expect to receive a significant amount of government support in the event of trouble. It is equal to one for banks which have a Moody’s long-term foreign currency deposit rating greater than Moody’s bank financial strength rating (see Brandao-Marques, Correa, and Sapriza, 2018). In Panel C, government support is defined as a dummy variable for German Landesbanks which are government owned. While not shown to conserve space, the model includes an intercept and the other independent variables shown in Table VII. Standard errors (shown in parenthesis) are clustered at the country level. ***, **, * indicates significance at the 1%, 5%, and 10% level, respectively.

Panel A: Interacting governance with probability of support

Dependent variable: Sponsoring a credit-arbitrage ABCP vehicle = 1
Independent Variable(1)(2)(3)(4)
Compensation Index –0.523***    
(0.124) 
Insider Ownership  –0.245   
(0.178) 
Institutional Ownership   –0.398**  
(0.164) 
Blockholder    –0.159*** 
(0.052) 
High Probability of Support –0.803*** –0.330* –0.079 –0.150 
(0.229) (0.164) (0.190) (0.101) 
High Probability of Support × Compensation Index 1.825***    
(0.284) 
High Probability of Support × Insider Ownership  1.336**   
(0.531) 
High Probability of Support × Institutional Ownership   0.368  
(0.525) 
High Probability of Support × Blockholder    0.319** 
(0.139) 

 
Country fixed effects Yes Yes Yes Yes 
Observations 111 115 113 145 
Countries 18 18 18 18 
Adjusted R-squared 0.478 0.464 0.395 0.422 
Panel A: Interacting governance with probability of support

Dependent variable: Sponsoring a credit-arbitrage ABCP vehicle = 1
Independent Variable(1)(2)(3)(4)
Compensation Index –0.523***    
(0.124) 
Insider Ownership  –0.245   
(0.178) 
Institutional Ownership   –0.398**  
(0.164) 
Blockholder    –0.159*** 
(0.052) 
High Probability of Support –0.803*** –0.330* –0.079 –0.150 
(0.229) (0.164) (0.190) (0.101) 
High Probability of Support × Compensation Index 1.825***    
(0.284) 
High Probability of Support × Insider Ownership  1.336**   
(0.531) 
High Probability of Support × Institutional Ownership   0.368  
(0.525) 
High Probability of Support × Blockholder    0.319** 
(0.139) 

 
Country fixed effects Yes Yes Yes Yes 
Observations 111 115 113 145 
Countries 18 18 18 18 
Adjusted R-squared 0.478 0.464 0.395 0.422 
Panel B: Interacting governance with government support

Dependent variable: Sponsoring a credit-arbitrage ABCP vehicle = 1
Independent Variable(1)(2)(3)(4)
Compensation Index –0.508**    
(0.194) 
Insider Ownership  –0.199   
(0.163) 
Institutional Ownership   –0.437*  
(0.226) 
Blockholder    –0.160** 
(0.056) 
High Government Support –0.765* –0.143 –0.063 –0.181** 
(0.377) (0.198) (0.259) (0.084) 
High Government Support × Compensation Index 1.574***    
(0.490) 
High Government Support × Insider Ownership  0.806*   
(0.428) 
High Government Support × Institutional Ownership   0.444  
(0.793) 
High Government Support × Blockholder    0.277** 
(0.105) 
 
Country fixed effects Yes Yes Yes Yes 
Observations 111 115 113 145 
Countries 18 18 18 18 
Adjusted R-squared 0.447 0.418 0.398 0.418 

 
Panel B: Interacting governance with government support

Dependent variable: Sponsoring a credit-arbitrage ABCP vehicle = 1
Independent Variable(1)(2)(3)(4)
Compensation Index –0.508**    
(0.194) 
Insider Ownership  –0.199   
(0.163) 
Institutional Ownership   –0.437*  
(0.226) 
Blockholder    –0.160** 
(0.056) 
High Government Support –0.765* –0.143 –0.063 –0.181** 
(0.377) (0.198) (0.259) (0.084) 
High Government Support × Compensation Index 1.574***    
(0.490) 
High Government Support × Insider Ownership  0.806*   
(0.428) 
High Government Support × Institutional Ownership   0.444  
(0.793) 
High Government Support × Blockholder    0.277** 
(0.105) 
 
Country fixed effects Yes Yes Yes Yes 
Observations 111 115 113 145 
Countries 18 18 18 18 
Adjusted R-squared 0.447 0.418 0.398 0.418 

 
Panel C: Interacting governance with Landesbanks


Dependent variable: Sponsoring a credit-arbitrage ABCP vehicle = 1
Independent Variable(1)(2)
Compensation Index –0.205  
(0.243) 
Insider Ownership  –0.053 
(0.094) 
Dummy Landesbank –0.053 0.013 
(0.140) (0.044) 
Dummy Landesbank × Compensation Index 0.947***  
(0.219) 
Dummy Landesbank × Insider Ownership  1.224*** 
(0.176) 

 
Country fixed effects Yes Yes 
Observations 111 115 
Countries 18 18 
Adjusted R-squared 0.406 0.409 

 
Panel C: Interacting governance with Landesbanks


Dependent variable: Sponsoring a credit-arbitrage ABCP vehicle = 1
Independent Variable(1)(2)
Compensation Index –0.205  
(0.243) 
Insider Ownership  –0.053 
(0.094) 
Dummy Landesbank –0.053 0.013 
(0.140) (0.044) 
Dummy Landesbank × Compensation Index 0.947***  
(0.219) 
Dummy Landesbank × Insider Ownership  1.224*** 
(0.176) 

 
Country fixed effects Yes Yes 
Observations 111 115 
Countries 18 18 
Adjusted R-squared 0.406 0.409 

 
Table VIII.

Agency conflicts and sponsorship of credit-arb vehicles

This table shows the results from estimating a linear probability model in which the dependent variable is a binary indicator equal to 1 for banks that sponsored at least one SIV, securities arbitrage, or hybrid ABCP vehicle. We examine sponsorship as of June 2007. Most independent variables are measured as of December of the previous year. We use several measures of potential agency conflicts: Compensation Index is calculated from RiskMetrics Group’s CGQ as in Aggarwal et al. (2010); higher values represent compensation practices that are better aligned with shareholder value. Insider Ownership is the percent of shares owned by individual insiders, while Institutional Ownership is the percent of shares owned by institutions. Blockholder is an indicator variable that is equal to 1 if the bank has at least one single owner with voting shares larger than 10%. In Panels A, B, and C, we interact these measures of agency conflicts with various proxies for government support. In Panel A, we define government support using High Probability of Support, which is an indicator equal to 1 for banks which have a ratings-based probability of support (defined as in Gropp, Hakenes, and Schnabel, 2011) greater than 0. In Panel B, we define government support using High Government Support, which is an indicator for banks that expect to receive a significant amount of government support in the event of trouble. It is equal to one for banks which have a Moody’s long-term foreign currency deposit rating greater than Moody’s bank financial strength rating (see Brandao-Marques, Correa, and Sapriza, 2018). In Panel C, government support is defined as a dummy variable for German Landesbanks which are government owned. While not shown to conserve space, the model includes an intercept and the other independent variables shown in Table VII. Standard errors (shown in parenthesis) are clustered at the country level. ***, **, * indicates significance at the 1%, 5%, and 10% level, respectively.

Panel A: Interacting governance with probability of support

Dependent variable: Sponsoring a credit-arbitrage ABCP vehicle = 1
Independent Variable(1)(2)(3)(4)
Compensation Index –0.523***    
(0.124) 
Insider Ownership  –0.245   
(0.178) 
Institutional Ownership   –0.398**  
(0.164) 
Blockholder    –0.159*** 
(0.052) 
High Probability of Support –0.803*** –0.330* –0.079 –0.150 
(0.229) (0.164) (0.190) (0.101) 
High Probability of Support × Compensation Index 1.825***    
(0.284) 
High Probability of Support × Insider Ownership  1.336**   
(0.531) 
High Probability of Support × Institutional Ownership   0.368  
(0.525) 
High Probability of Support × Blockholder    0.319** 
(0.139) 

 
Country fixed effects Yes Yes Yes Yes 
Observations 111 115 113 145 
Countries 18 18 18 18 
Adjusted R-squared 0.478 0.464 0.395 0.422 
Panel A: Interacting governance with probability of support

Dependent variable: Sponsoring a credit-arbitrage ABCP vehicle = 1
Independent Variable(1)(2)(3)(4)
Compensation Index –0.523***    
(0.124) 
Insider Ownership  –0.245   
(0.178) 
Institutional Ownership   –0.398**  
(0.164) 
Blockholder    –0.159*** 
(0.052) 
High Probability of Support –0.803*** –0.330* –0.079 –0.150 
(0.229) (0.164) (0.190) (0.101) 
High Probability of Support × Compensation Index 1.825***    
(0.284) 
High Probability of Support × Insider Ownership  1.336**   
(0.531) 
High Probability of Support × Institutional Ownership   0.368  
(0.525) 
High Probability of Support × Blockholder    0.319** 
(0.139) 

 
Country fixed effects Yes Yes Yes Yes 
Observations 111 115 113 145 
Countries 18 18 18 18 
Adjusted R-squared 0.478 0.464 0.395 0.422 
Panel B: Interacting governance with government support

Dependent variable: Sponsoring a credit-arbitrage ABCP vehicle = 1
Independent Variable(1)(2)(3)(4)
Compensation Index –0.508**    
(0.194) 
Insider Ownership  –0.199   
(0.163) 
Institutional Ownership   –0.437*  
(0.226) 
Blockholder    –0.160** 
(0.056) 
High Government Support –0.765* –0.143 –0.063 –0.181** 
(0.377) (0.198) (0.259) (0.084) 
High Government Support × Compensation Index 1.574***    
(0.490) 
High Government Support × Insider Ownership  0.806*   
(0.428) 
High Government Support × Institutional Ownership   0.444  
(0.793) 
High Government Support × Blockholder    0.277** 
(0.105) 
 
Country fixed effects Yes Yes Yes Yes 
Observations 111 115 113 145 
Countries 18 18 18 18 
Adjusted R-squared 0.447 0.418 0.398 0.418 

 
Panel B: Interacting governance with government support

Dependent variable: Sponsoring a credit-arbitrage ABCP vehicle = 1
Independent Variable(1)(2)(3)(4)
Compensation Index –0.508**    
(0.194) 
Insider Ownership  –0.199   
(0.163) 
Institutional Ownership   –0.437*  
(0.226) 
Blockholder    –0.160** 
(0.056) 
High Government Support –0.765* –0.143 –0.063 –0.181** 
(0.377) (0.198) (0.259) (0.084) 
High Government Support × Compensation Index 1.574***    
(0.490) 
High Government Support × Insider Ownership  0.806*   
(0.428) 
High Government Support × Institutional Ownership   0.444  
(0.793) 
High Government Support × Blockholder    0.277** 
(0.105) 
 
Country fixed effects Yes Yes Yes Yes 
Observations 111 115 113 145 
Countries 18 18 18 18 
Adjusted R-squared 0.447 0.418 0.398 0.418 

 
Panel C: Interacting governance with Landesbanks


Dependent variable: Sponsoring a credit-arbitrage ABCP vehicle = 1
Independent Variable(1)(2)
Compensation Index –0.205  
(0.243) 
Insider Ownership  –0.053 
(0.094) 
Dummy Landesbank –0.053 0.013 
(0.140) (0.044) 
Dummy Landesbank × Compensation Index 0.947***  
(0.219) 
Dummy Landesbank × Insider Ownership  1.224*** 
(0.176) 

 
Country fixed effects Yes Yes 
Observations 111 115 
Countries 18 18 
Adjusted R-squared 0.406 0.409 

 
Panel C: Interacting governance with Landesbanks


Dependent variable: Sponsoring a credit-arbitrage ABCP vehicle = 1
Independent Variable(1)(2)
Compensation Index –0.205  
(0.243) 
Insider Ownership  –0.053 
(0.094) 
Dummy Landesbank –0.053 0.013 
(0.140) (0.044) 
Dummy Landesbank × Compensation Index 0.947***  
(0.219) 
Dummy Landesbank × Insider Ownership  1.224*** 
(0.176) 

 
Country fixed effects Yes Yes 
Observations 111 115 
Countries 18 18 
Adjusted R-squared 0.406 0.409 

 

5.1 Capital Regulations

One potential explanation for credit-arb vehicle sponsorship is that banks created these vehicles to avoid capital charges, as discussed in Acharya et al. (2013). We first examine the impact of regulation using a specific difference on the regulatory capital treatment of credit-arb vehicles. At the time of our sample, most European banks were not required to have extra capital for the funding backstops and credit enhancements that bank sponsors provided for their SAVs, but US banks did face such requirements.19 Though the precise cost of these requirements is unknown, back-of-the-envelope calculations suggest that capital charges cut the profit for a US bank relative to the profit that a European bank could earn by 40–70% (see Online Appendix for details).

Given this cost differential, a hypothesis that favorable regulatory capital treatment incentivized sponsorship implies that US banks should have been less likely to sponsor at the margin, after controlling for other characteristics. We explore this possibility by including an indicator variable that is 1 for US banks and 0 for European banks in our prediction model while controlling for bank and country characteristics. The result is reported in column (1) of Table VI. The coefficient on US banks is positive rather than negative, suggesting that moderately tougher capital regulation did not hinder bad-tail risk-taking.

While most European banks enjoyed a clear cost advantage with regards to sponsoring credit-arb vehicles, banks headquartered in Spain and Portugal were an important exception. Regulators in these countries required assets held in off-balance sheet vehicles to be treated as if they were on-balance sheet assets of the vehicle’s sponsor for regulatory capital purposes. Table II shows that no bank in Spain or Portugal sponsored a credit-arb vehicle. In column (2) of Table VI, we include an indicator variable for banks located in Spain and Portugal and confirm that even after controlling for bank characteristics, banks located in Spain and Portugal were approximately 20 percentage points less likely to sponsor a vehicle. This result is robust to controlling for region fixed effects.20 It may be that extremely tough capital regulation made it financially unattractive to become a sponsor, or regulators in Spain and Portugal may have discouraged sponsorship in other ways that are not observable.

To ensure that the results in column (1) are not biased by comparing US banks to Spanish/Portuguese banks, in column (3), we simultaneously include indicators for both US banks and Spanish/Portuguese banks. In this specification, the coefficient on US banks becomes insignificant.

The above analysis could be misleading because SIVs, which are a subset of our credit-arb sample, actually had similar regulatory treatment across both Europe and the USA (see Online Appendix for details). In unreported results, we estimate the probability that a bank sponsors at least one SAV and find that the coefficient on the US dummy variable has a similar magnitude, but becomes significant at the 5% level (even after controlling for the Spain/Portugal dummy in column (3)). Since US banks had an especially pronounced regulatory disadvantage when sponsoring SAVs, this result suggests that regulatory arbitrage was not the only driver of the decision to sponsor.

An additional bit of circumstantial evidence suggesting that capital regulation per se was not the primary determinant of sponsorship is that European banks’ vehicles grew rapidly during 2004–07 even though it was clear that the impending implementation of Basel II would soon impose capital requirements on SAV activities similar to capital requirements in the USA. ABCP vehicles are difficult to wind up quickly, so the growth amounted to a pre-commitment by European banks to continue operations at scale even after Basel II implementation. Such a pre-commitment is inconsistent with low capital charges being the primary motivation for European banks to sponsor ABCP vehicles relative to US banks.21

Together, the evidence presented above offers a nuanced view of the effects of capital regulation. Unlike the Acharya et al. (2013) result on average risk-taking, avoiding capital requirements does not seem to be the primary motivation for taking bad-tail systematic risk. However, at least with respect to the specific form of bad-tail risk we study (i.e., credit-arb vehicle sponsorship), some degree of regulatory capital relief appears to have been a necessary prerequisite.

5.2 Government Safety Nets

We next examine government safety nets—that is, the loss-shifting ability of bank equity holders due to the existence of some combination of deposit insurance, regulatory forbearance, too-big-to-fail government bailouts, and other safety net features. Shareholders who expect to be rescued by the government might be more willing to take bad-tail risks.

To explore the correlation between safety nets and risk-taking, we use several bank-level measures of the strength of government safety nets: the difference between a bank’s financial strength rating and its all-in rating (Government Support), a ratings-based measure of the probability of receiving a government bailout (Probability of Support) from Gropp, Hakenes, and Schnabel (2011), and a dummy for Landesbanks (which are government-controlled, and thus highly likely to receive government support). In these specifications, we include country fixed effects, so the effects are identified off of bank-specific differences in expected government support within a country. We also examine two measures of deposit insurance: a dummy for whether the deposit insurance system in a bank’s home country is government-controlled (Deposit Insurance), and the natural logarithm of the dollar value of government-provided deposit insurance (Level of Deposit Insurance). Because these are country-level variables, we include region fixed effects in these specifications. Across all five measures, higher values imply a stronger safety net.

Table VII reports the results. Consistent with Brandao-Marques, Correa, and Sapriza (2018), we find that Government Support is associated with bank risk-taking, as shown by the positive and significant relation between this variable and credit-arb vehicle sponsorship. The coefficient in column (1) implies that a one standard deviation increase in Government Support leads to a nearly 7 percentage point increase in the probability of sponsorship (the unconditional probability of sponsorship is 24%). Column (2) shows that a higher Probability of Support is also positively correlated with bad-tail risk-taking, but the relationship is not statistically significant.

We next examine the effect of government ownership by looking at the correlation between Landesbanks and sponsorship. The coefficient in column (3) implies that Landesbanks are about 32 percentage points more likely to sponsor vehicles after controlling for bank characteristics and country fixed effects; this relationship is highly statistically significant. Because Landesbanks are state-owned, it is plausible that this result reflects expectations of receiving a government bailout. However, Landesbanks have a history of getting into financial trouble and so it is also possible that they sponsored vehicles in order to conceal continued risk-taking from stakeholders (including the government).22 While we cannot rule out agency conflicts, in Section 5.3, we show that among the subset of Landesbanks, those banks that had better governance were more likely to become sponsors, which is consistent with the Landesbank indicator at least partially reflecting expected government support.

While the Landesbanks were clear culprits in the sponsorship of credit-arb vehicles, it is important to note that our results are not driven by these banks. In Section 5.3, we show that all of our main regressions are robust to excluding Landesbanks from the sample.

Finally, in columns (4) and (5) of Table VII, we explore the role of government-provided deposit insurance. In column (4), we find that the existence of a government-backed deposit insurance system increases the probability of sponsorship by about 16 percentage points. We also find that higher levels of government deposit insurance are associated with higher probabilities of credit-arb vehicle sponsorship. The coefficient reported in column (5) implies that a one standard deviation increase in the level of deposit insurance raises the probability of sponsorship by just under 3 percentage points, which is roughly a 12% increase from the baseline probability of sponsorship.

In summary, government-induced incentive distortions played an important role in banks’ decision to sponsor ABCP vehicles. While we find mixed evidence that differences in US and European bank sponsorship of vehicles were due to bank regulation, we find consistent evidence that banks with more explicit and implicit government support were more likely to take bad-tail systematic risks.

5.3 Agency Problems

Owner–manager agency problems might reduce or increase risk-taking relative to shareholders’ desires. The literature has implicitly focused on average risk, but the intuition works for bad-tail risk as well. On the one hand, senior executive wealth may be very exposed to bank performance due to lack of diversification. If executives are risk averse, they may therefore take too little risk in the absence of corrective actions by shareholders, such as tailored compensation contracts (Smith and Stulz, 1985). On the other hand, compensation arrangements might provide incentives to take too much risk by rewarding earnings without regard to risk (e.g., John and John, 1993; Diamond and Rajan, 2009; Bannier, Feess, and Packham, 2013), or low-quality executives might increase risk in hopes that associated increases in short-term earnings will help them retain their jobs (Saunders, Strock, and Travlos, 1990; Gorton and Rosen, 1995). Thus, differences in compensation arrangements and other governance mechanisms by which shareholders influence risk-taking may influence both managers’ incentives and their ability to sponsor vehicles.23 Moreover, other things equal, the more difficult it is for shareholders to monitor and exert control, the more the risk-taking may deviate from shareholders’ desired level.

To examine the extent to which agency problems played a role in banks’ decisions to sponsor ABCP vehicles, we use four proxies for the quality of bank governance and report the results in Table VIII. This table uses a smaller sample than the rest of the article because the governance measures are not available for all banks.24 Across three of the four measures, we find statistically significant evidence that shareholder–manager agency problems are associated with systematic bad-tail risk-taking. In the first column of Panel A of Table VIII, we show that more shareholder-friendly pay practices are correlated with less risk-taking. The negative coefficient implies that a move from the 25th percentile to the 75th percentile in the sample distribution of Compensation Index is associated with a decrease of 20 percentage points in the probability of sponsoring a credit-arb vehicle, which is economically large (the unconditional probability of sponsorship in our sample is 24%). Columns (2) through (4) provide similar evidence using various measures of ownership. More insider ownership, higher institutional ownership, and the presence of a blockholder all reduce the probability of sponsorship, though the effect of insider ownership is not statistically significant.

This evidence suggests that, on average, managers took too much risk from the shareholders’ perspective. However, the results in the previous section suggest that shareholders’ desired level of risk-taking is a function of expected government bailouts. If the probability of a bailout is high enough, shareholders might prefer that the bank take even more risk. In Panel A, we examine this by interacting our proxies for governance with an indicator for banks that have a non-zero Probability of Support. With the exception of institutional ownership, the coefficient on the interaction terms of governance proxies is large, positive, and statistically significant across all three panels. Moreover, these estimates imply that for banks that ex ante expect to receive a government bailout in times of trouble, better governance (in the form of more incentive compatible pay, higher insider ownership, or the presence of a blockholder) is always positively associated with higher bad tail risk-taking. For example, a bank with a high probability of support is actually 50 percentage points more likely to sponsor a vehicle when moving from the 25th to 75th percentile of Compensation Index.25

In Panels B and C, we show that results are robust to alternative ways of identifying banks that have high levels of implicit government support. Specifically, in Panel B, we define high expected support as banks with a non-zero value of Government Support, while in Panel C we define Landesbanks as expecting a high level of support.26 Regardless of which measure of expected support we use, Panels A, B, and C of Table VIII show that the relationship between sponsorship and governance flips signs for banks that expect to be bailed out.

Together, the results in Table VIII are consistent with shareholder–manager agency problems playing an important role in the decision to take bad-tail systematic risk. For banks that expected to be bailed out, which make up about 15–20% of our sample depending on the measure we use, ownership, and compensation structures that better aligned managers and shareholders’ interests were positively associated with bad-tail systematic risk-taking. In our sample, this includes banks such as Credit Suisse, Natixis, IKB, and Standard Chartered. In contrast, for banks with little expected government support, good governance was negatively correlated with bad-tail risk-taking.27

Since vehicles were small relative to assets for almost all sponsors (see Table III), it is unlikely that the decision to sponsor a credit-arb vehicle has any meaningful feedback effect on governance. In untabulated tests, we find that our results are similar when we measure ownership characteristics as of the year 2000. Since very few banks sponsored vehicles at that time, these measures are pre-determined and unlikely to be driven by omitted variables. Importantly, these results are also not driven by Landesbanks’ participation in the ABCP market. We repeat the estimations in Table VIII after excluding all Landesbanks from the sample and find results that are nearly identical in both magnitude and statistical significance (untabulated).

5.4 Just Bad Luck

Data are not available to support formal testing of a hypothesis that banks made entirely undistorted risk-reward decisions to sponsor credit-arb vehicles. However, by examining the sponsorship behavior of first-movers as compared to followers, we can gain some insight into the likelihood that banks made a mistake. On the one hand, the first-movers had more time to put risk mitigants in place and to learn the risks associated with credit-arb vehicles, but on the other hand, the followers may have learned from the experience of the first-movers. We do not know which group was more likely to take imprudent risk, but it is plausible that the risk-taking behavior differed across the two groups. Consequently, evidence that the banks invested in the same way would suggest that it is less likely that the risk-taking was simply a mistake.

We split the sponsor banks into first-movers and followers based on the date that the bank first sponsored a credit-arb vehicle. Banks with a duration of credit-arb sponsorship above the sample median are defined as first-movers, while the remaining sponsoring banks are defined as followers.28 The first column of Table IX shows that there was no difference between first-movers and followers in the intensive margin of sponsorship (measured by the amount of sponsored credit-arb vehicle commercial paper to bank assets) as of June 2007, so it appears that both groups were similarly exposed to credit-arb vehicles.

Table IX.

First-movers versus followers

This table compares bank outcomes for banks that were early vs late adopters of vehicle sponsorship. For each sponsor bank, we calculate sponsorship age as the number of years from the time that the bank first sponsored a credit-arb vehicle. First-Mover is defined as banks above the median duration of sponsorship. Sponsor x First Mover is in bold face to identify the variable of most interest. Definitions for the other independent variables are found in the Appendix. In column (1), we examine the extent of sponsorship exposure as of June 2007 (measured as credit-arbitrage vehicle commercial paper scaled by bank assets). In columns (2)–(4), we examine bank outcomes during the GFC (August 2007 to July 2009). Standard errors (shown in parenthesis) are clustered at the country level. ***, **, * indicates significance at the 1%, 5%, and 10% level, respectively.

Independent variable(1)(2)(3)(4)
Risky CP to assetsReceived capital injectionBank failedBank acquired
Sponsor 0.056* 0.162 –0.084* 0.183*** 
(0.029) (0.148) (0.044) (0.037) 
Sponsor × First Mover –0.014 0.014 0.010 –0.133 
(0.025) (0.160) (0.029) (0.120) 
ln(Total Assets) –0.004 0.154*** –0.013** 0.001 
(0.003) (0.042) (0.006) (0.039) 
Return on Assets –0.006 0.173* –0.014 –0.037 
(0.005) (0.090) (0.015) (0.077) 
Equity to Assets –0.000 –0.042 –0.012** –0.003 
(0.001) (0.029) (0.005) (0.010) 
Loans to Assets –0.000 0.007 0.000 –0.004** 
(0.000) (0.004) (0.001) (0.001) 
Deposits to Assets –0.000 –0.004 0.000 0.003 
(0.000) (0.003) (0.001) (0.002) 
Non-Interest Operating Income to Assets 0.002 0.081 0.004 –0.031 
(0.002) (0.060) (0.005) (0.020) 
High Yield Underwriter –0.014 0.053 –0.039 –0.097 
(0.009) (0.159) (0.026) (0.175) 
Securitization Underwriter –0.012* –0.089 0.091 –0.024 
(0.006) (0.108) (0.054) (0.054) 

 
Country Fixed Effects Yes Yes Yes Yes 
Observations 145 145 145 145 
Countries 18 18 18 18 
Adjusted R-squared 0.256 0.265 –0.048 –0.009 
Independent variable(1)(2)(3)(4)
Risky CP to assetsReceived capital injectionBank failedBank acquired
Sponsor 0.056* 0.162 –0.084* 0.183*** 
(0.029) (0.148) (0.044) (0.037) 
Sponsor × First Mover –0.014 0.014 0.010 –0.133 
(0.025) (0.160) (0.029) (0.120) 
ln(Total Assets) –0.004 0.154*** –0.013** 0.001 
(0.003) (0.042) (0.006) (0.039) 
Return on Assets –0.006 0.173* –0.014 –0.037 
(0.005) (0.090) (0.015) (0.077) 
Equity to Assets –0.000 –0.042 –0.012** –0.003 
(0.001) (0.029) (0.005) (0.010) 
Loans to Assets –0.000 0.007 0.000 –0.004** 
(0.000) (0.004) (0.001) (0.001) 
Deposits to Assets –0.000 –0.004 0.000 0.003 
(0.000) (0.003) (0.001) (0.002) 
Non-Interest Operating Income to Assets 0.002 0.081 0.004 –0.031 
(0.002) (0.060) (0.005) (0.020) 
High Yield Underwriter –0.014 0.053 –0.039 –0.097 
(0.009) (0.159) (0.026) (0.175) 
Securitization Underwriter –0.012* –0.089 0.091 –0.024 
(0.006) (0.108) (0.054) (0.054) 

 
Country Fixed Effects Yes Yes Yes Yes 
Observations 145 145 145 145 
Countries 18 18 18 18 
Adjusted R-squared 0.256 0.265 –0.048 –0.009 
Table IX.

First-movers versus followers

This table compares bank outcomes for banks that were early vs late adopters of vehicle sponsorship. For each sponsor bank, we calculate sponsorship age as the number of years from the time that the bank first sponsored a credit-arb vehicle. First-Mover is defined as banks above the median duration of sponsorship. Sponsor x First Mover is in bold face to identify the variable of most interest. Definitions for the other independent variables are found in the Appendix. In column (1), we examine the extent of sponsorship exposure as of June 2007 (measured as credit-arbitrage vehicle commercial paper scaled by bank assets). In columns (2)–(4), we examine bank outcomes during the GFC (August 2007 to July 2009). Standard errors (shown in parenthesis) are clustered at the country level. ***, **, * indicates significance at the 1%, 5%, and 10% level, respectively.

Independent variable(1)(2)(3)(4)
Risky CP to assetsReceived capital injectionBank failedBank acquired
Sponsor 0.056* 0.162 –0.084* 0.183*** 
(0.029) (0.148) (0.044) (0.037) 
Sponsor × First Mover –0.014 0.014 0.010 –0.133 
(0.025) (0.160) (0.029) (0.120) 
ln(Total Assets) –0.004 0.154*** –0.013** 0.001 
(0.003) (0.042) (0.006) (0.039) 
Return on Assets –0.006 0.173* –0.014 –0.037 
(0.005) (0.090) (0.015) (0.077) 
Equity to Assets –0.000 –0.042 –0.012** –0.003 
(0.001) (0.029) (0.005) (0.010) 
Loans to Assets –0.000 0.007 0.000 –0.004** 
(0.000) (0.004) (0.001) (0.001) 
Deposits to Assets –0.000 –0.004 0.000 0.003 
(0.000) (0.003) (0.001) (0.002) 
Non-Interest Operating Income to Assets 0.002 0.081 0.004 –0.031 
(0.002) (0.060) (0.005) (0.020) 
High Yield Underwriter –0.014 0.053 –0.039 –0.097 
(0.009) (0.159) (0.026) (0.175) 
Securitization Underwriter –0.012* –0.089 0.091 –0.024 
(0.006) (0.108) (0.054) (0.054) 

 
Country Fixed Effects Yes Yes Yes Yes 
Observations 145 145 145 145 
Countries 18 18 18 18 
Adjusted R-squared 0.256 0.265 –0.048 –0.009 
Independent variable(1)(2)(3)(4)
Risky CP to assetsReceived capital injectionBank failedBank acquired
Sponsor 0.056* 0.162 –0.084* 0.183*** 
(0.029) (0.148) (0.044) (0.037) 
Sponsor × First Mover –0.014 0.014 0.010 –0.133 
(0.025) (0.160) (0.029) (0.120) 
ln(Total Assets) –0.004 0.154*** –0.013** 0.001 
(0.003) (0.042) (0.006) (0.039) 
Return on Assets –0.006 0.173* –0.014 –0.037 
(0.005) (0.090) (0.015) (0.077) 
Equity to Assets –0.000 –0.042 –0.012** –0.003 
(0.001) (0.029) (0.005) (0.010) 
Loans to Assets –0.000 0.007 0.000 –0.004** 
(0.000) (0.004) (0.001) (0.001) 
Deposits to Assets –0.000 –0.004 0.000 0.003 
(0.000) (0.003) (0.001) (0.002) 
Non-Interest Operating Income to Assets 0.002 0.081 0.004 –0.031 
(0.002) (0.060) (0.005) (0.020) 
High Yield Underwriter –0.014 0.053 –0.039 –0.097 
(0.009) (0.159) (0.026) (0.175) 
Securitization Underwriter –0.012* –0.089 0.091 –0.024 
(0.006) (0.108) (0.054) (0.054) 

 
Country Fixed Effects Yes Yes Yes Yes 
Observations 145 145 145 145 
Countries 18 18 18 18 
Adjusted R-squared 0.256 0.265 –0.048 –0.009 

We next look for evidence that the banks invested in different types of assets by examining bank outcomes during the GFC based upon sponsorship decisions as of June 2007. If first-movers were savvier and invested in higher quality assets, we would expect to find that they were less likely to receive a capital injection, be acquired due to duress, or fail during the crisis. Alternatively, if followers were able to learn from observing first-movers and thus invest in better vehicles, they should fare better during the GFC. Instead, the data confirms that there are no differences in outcomes during the GFC between first-movers and followers (see columns (2)–(4)), which suggests that the two groups invested in similar quality assets. Overall, the lack of differences between first-movers and followers is inconsistent with sponsorship decisions being driven by underestimating the risk.

Informal evidence is also inconsistent with undistorted risk decisions: vehicle profitability was low, and grew lower at the same time as vehicles expanded. Profits are generally not disclosed, but a couple of anecdotes give a sense of vehicle profitability. At year-end 2006, Mellon Bank’s vehicle had $3.2 billion in assets, which if consolidated would have increased Mellon’s total assets by 7.7% (a large fraction). But the vehicle provided only $3 million of gross revenue to the bank (10 basis points of vehicle assets). Somewhat unrealistically assuming that Mellon itself had no expenses associated with the vehicle (all revenue was net profit), the vehicle contributed only 0.3% of Mellon’s 2006 net income.29 Similarly, Acharya et al. (2013) note that Deutsche Bank’s 2007 annual report reveals that its vehicles generated €6 million of fees relative to commitments of €6.3 billion, again about 10 basis points before expenses.

In the aggregate, vehicles grew as profitability fell. Figure 4 plots a proxy for the gross spread earned on vehicle investments (the yield on ABS less LIBOR for two types of underlying assets; ABCP rates were generally close to LIBOR until the crisis began). As shown in Figure 2, outstanding ABCP at credit-arb vehicles grew from 2004 on, while Figure 4 shows gross spreads fell to low levels during the same period.30 Unless the risk associated with the vehicles fell even more rapidly than did gross spreads, or unless the returns to all other strategies fell more, increasing exposure to credit-arb vehicles as expected profits fall appears to be a violation of portfolio theory. As we argued previously, risks posed by the vehicles were almost entirely bad-tail risk borne by the sponsors, and such bad-tail risk does not vary much during booms (or perhaps is increasing).31 Furthermore, the events of the 1990s and 2000s clearly demonstrated the risks involved in sponsoring these vehicles, making it less likely that banks underestimated the risks.32

Figure 4.

ABCP interest rate spreads. This figure plots a proxy for the gross spread earned on ABCP vehicle investments from December 2009 to June 2007. The credit card spread represents the yield on 5-year AAA credit card ABS less LIBOR, while the auto spread is the yield on 3-year AAA automobile loan ABS less LIBOR. Source: Salomon Smith Barney and Citigroup Global Markets.

Figure 4.

ABCP interest rate spreads. This figure plots a proxy for the gross spread earned on ABCP vehicle investments from December 2009 to June 2007. The credit card spread represents the yield on 5-year AAA credit card ABS less LIBOR, while the auto spread is the yield on 3-year AAA automobile loan ABS less LIBOR. Source: Salomon Smith Barney and Citigroup Global Markets.

Overall, the volume and profitability data, combined with first-movers and followers experiencing similar outcomes during the GFC, suggest that sponsorship decisions were not simply due to bad luck. However, these vehicles did not add substantially to most sponsors’ profits, which raises the question of why bank managers would sponsor them. Motivations are not observable, but we speculate that managers in the competitive banking industry were under constant pressure to improve earnings by even small amounts, and bank managers did so partly by taking bad-tail systematic risks.

6. Concluding Remarks

Overall, the evidence from our article suggests that differences in regulatory capital treatment of credit-arbitrage ABCP vehicles were not the only factor in banks’ decisions to sponsor such vehicles and to take the systematic bad-tail risk that was associated with them. Instead, we find that agency problems and government safety nets (and their interactions) are also associated with bad-tail systematic risk-taking. Such risk-taking is of particular interest to designers of financial regulations. Regulatory reforms implemented after the GFC have both increased capital ratios and attempted to reduce some agency problems.33 But our results imply that bad-tail risk-taking is unlikely to be solved by increasing capital ratios. Furthermore, reducing agency problems might actually increase the incentives to take bad-tail risks if banks expect to be bailed out by the government. As a result, further attention to simultaneously reducing both agency problems and expected government support is warranted.

Footnotes

*

This paper represents the authors’ opinions and not necessarily those of the Board of Governors, the Federal Reserve System, the World Bank, or other members of its staff. We thank Thorsten Beck (the editor), two anonymous referees, an anonymous editor, Adam Ashcraft, Norah Barger, Tom Boemio, Sally Davies, Bob DeYoung, Toni Gravelle, Gary Gorton, David Jones, Steve Kamin, Michal Kowalik, George Pennacchi, Nathan Sheets, Philipp Schnabl, Gustavo Suarez and many others for useful conversations and suggestions.

1

Our benchmark for capital regulation is the established treatment of ABCP vehicles as of the mid-2000s. However, our sample period does not encompass periods with higher capital requirements such as those set after Basel III.

2

The vehicles’ liabilities included no equity or subordinated debt, or small amounts of subordinated debt.

3

Though Basel II would impose a capital cost on sponsoring vehicles similar to that faced by US banks, at the time of our sample European banks had not yet implemented Basel II standards.

4

With regard to the latter, banks may simply have struck an ex ante reasonable risk-return balance. Even if profits were low, risks may have been commensurately low. Problems at ABCP vehicles may simply have been bad luck, and systemic consequences were an externality that did not affect bank decisions.

6

The results are robust to eliminating the few outlier banks with total ABCP/assets greater than 4%. Of the remaining sample, the average ABCP/assets conditional on sponsorship is about 1%.

7

Generally, hybrid vehicles have portfolios that mix SAV and multiseller characteristics.

8

For brevity, we use ABS to refer to the full menu of ABS, including MBS, CDO, ABS, etc.

9

For example, if an originally AAA asset is downgraded a little bit, the provider of the enhancement (typically the sponsoring bank) adds assets to the vehicle to restore net asset value. If the asset is downgraded further, say below single-A, the terms of the enhancement give the provider an incentive to buy the asset out of the vehicle at par. In principle, ABCP investors bear the risk of a jump to default without an intervening downgrade, but such jumps are very rare for assets rated A or better, and liquidity backstops offer additional protection.

10

For ABCP investors to suffer a credit loss, a variant of a double-default event must occur: the sponsoring bank must fail to meet its contractual obligations to the vehicle, and the value of vehicle assets must be insufficient to pay off investors as ABCP matures.

11

Our results are robust to controlling for or eliminating SIVs.

12

The spreadsheets include the average amount of ABCP outstanding—issued in the US and European commercial paper markets—each quarter. For commercial paper denominated in currencies other than the US dollar, we convert the amount outstanding to nominal US dollars using the historical exchange rates published by the Federal Reserve in Table H.10.

13

We omit a small number of vehicles sponsored by nonbanks or Asian banks because we are unable to determine regulatory treatment of them or we are unable to find data on sponsor characteristics. We also omit a small number of vehicles sponsored by Canadian banks because liquidity backup lines of credit were structured differently in Canada than elsewhere.

14

“Hybrid” vehicles, as classified by Moody’s, combine characteristics of SAVs and other vehicle types. We include in our analysis only hybrids that were effectively SAVs. We exclude some vehicles classified as SAVs by Moody’s where the vehicle invested mainly in securities originated by the sponsor, which for our purposes makes it more like a multiseller vehicle.

15

For the purpose of this paper, Europe includes all members in the EU-15 (the fifteen Western European member countries of the European Union before its expansion in 2004) plus Norway and Switzerland.

16

Bankscope variables are measured as of the previous fiscal year-end. Approximately 95% of our sample has a December fiscal year end date. For the remaining banks, we associate fiscal years ending in January through March with the prior year and fiscal years ending in April through November with the current year. Consequently, financial variables for a bank with a March fiscal year are measured as of the current March, while financial variables for a bank with a September fiscal year are measured as of the previous September.

17

For specifications where the main independent variable is defined at the country-level, we use region fixed effects instead of country fixed effects. In unreported results, we further confirm that the relationships we report are robust to a nonlinear probit model.

18

Total assets of a bank’s credit-arb vehicles relative to consolidated total assets of the sponsoring bank does not fully reflect both the capacity of the bank to obtain liquidity in the event of a run on the vehicles and its capacity to absorb credit losses in the event that vehicle assets have to be taken onto the bank’s balance sheet. The literature has so far not put forward a convincing way to simultaneously measure both credit loss-bearing capacity and funding capacity in a way that is useful for studying credit-arb vehicle size.

19

The US regulatory requirements enacted between 2001 and 2004 imposed capital charges on assets held in vehicles. See the Online Appendix for details.

20

We define geographic regions based on banks headquartered in the USA, the British Isles, Northern Europe, Western Europe, and Southern Europe.

21

Had they been in effect, high incremental capital requirements associated with bank sponsorship of credit-arb vehicles might have greatly reduced the extent of sponsorship. However, existence of low capital requirements alone is not necessarily the marginal motivation for sponsorship. We do not argue that capital requirements were irrelevant, only that the evidence is more consistent with other factors driving decisions at the margin. Relatedly, Iannotta, Pennacchi, and Santos (2018) argue that ratings-based regulations incentivize financial firms to choose bonds that have more systematic risk. However, such a mechanism could not have driven systematic risk of credit-arb vehicles because European sponsors’ vehicles were not subject to such regulations.

22

Perhaps consistent with this, we note that there are a number of other types of government entities acting as owners in our sample (including in Greece, Finland, Italy, The Netherlands, France, and Germany), but government ownership outside of the Landesbanks does not predict sponsorship (untabulated).

23

Berger, Imbierowicz, and Rauch (2016) find that higher CEO ownership reduces the probability of bank default during the financial crisis. In contrast, Beltratti and Stulz (2012) show that banks with more shareholder-friendly boards have significantly worse stock performance during the financial crisis, and Fahlenbrach and Stulz (2011) find that CEOs with compensation packages that better align shareholder-CEO interests perform at best no better than CEOs with worse incentives during the crisis.

24

As noted in Section 4, many of the banks in our sample disclose very little information aside from financial statements. For banks missing this governance information, we hand-searched all available bank statements and annual reports to verify that the information was not publicly obtainable.

25

The implied magnitudes are also large for the other measures in Table VIII. For example, using Panel A: a one standard deviation increase in insider ownership reduces the probability of sponsorship by about 6 percentage points for a firm with low government support, but increases the probability of sponsorship by 26 percentage points for a firm with high government support. Similarly, the presence of a blockholder reduces sponsorship by about 16 percentage points if the bank has low government support, but increases sponsorship by about 16 percentage points if the bank has high government support.

26

Note that Landesbanks do not have any institutional owners or non-government blockholders, so we only use compensation and insider ownership data to measure governance for Landesbanks.

27

Examples of such banks include many smaller US banks (e.g., US Bancorp and BB&T), UBS, and Santander. These banks had low expectations of government support, but were well governed.

28

Our results are not sensitive to this particular definition. We continue to find no difference between first-mover and follower banks when we define first-movers based on the upper quartile of sponsorship age.

29

By early 2008, Mellon had used “excess liquidity” to payoff ABCP investors and wind up the vehicle, taking the assets onto the bank’s balance sheet.

30

The pattern raises the possibility that bank managers tried to compensate for reduced per-unit revenue by increasing the size of vehicles, but we cannot verify such a conjecture.

31

The vehicles are difficult to wind up, so a sponsor that increased vehicle size was committing to bear increased tail risk for many years.

32

There was recent historical experience of the relevant kinds of meltdowns. A fixed-income hedge fund blowup (LTCM) occurred less than 10 years before sponsorship decisions were made, an asset-backed securities meltdown occurred in 1994 (collateralized mortgage obligations), and a commercial paper market meltdown occurred in 2001.

33

Federal banking and housing agencies adopted a final rule implementing credit risk-retention requirements mandated by the Dodd–Frank Act on October 21, 2014. This rule applies to the ABCP vehicles discussed in the article.

Appendix

This appendix defines the variables used throughout the article.

Table AI.

Variable definitions

VariableDefinition
Sponsor Dummy variable equal to 1 if the bank sponsored at least one SIV, securities arbitrage, or hybrid ABCP vehicle during a given year. Source: Moody's quarterly program index. 
ln(Total Assets) Natural logarithm of the total assets of the bank (US$). Source: Bankscope. 
Return on Assets Net income divided by average assets. Source: Bankscope. 
Equity to Assets Ratio of book equity to total assets. Source: Bankscope. 
Loans to Assets Ratio of net loans to total assets. Source: Bankscope. 
Deposits to Assets Ratio of deposits to total assets. Source: Bankscope. 
Non-Interest Operating Income to Assets Ratio of non-interest operating income to total assets. Source: Bankscope. 
High Yield Underwriter Indicator variable equal to one if there is a positive amount of high yield debt underwritten by the bank during 2006. Source: Bloomberg, Dealogic DCM Analytics. 
Securitization Underwriter Indicator variable equal to one if there is a positive amount of asset-backed securities and MBS underwritten by the bank during 2006. Source: Bloomberg and Dealogic DCM Analytics. 
Dummy Landesbank Dummy variable equal to 1 if the bank is a German Landesbank. Source: Bankscope. 
Dummy US Dummy variable equal to 1 if the bank is domiciled in the United States. Source: Bankscope. 
Dummy Spain/Portugal Dummy variable equal to 1 if the bank is domiciled in Spain or Portugal. Source: Bankscope. 
Government Support For each bank, Moody’s long-term foreign currency deposit rating minus Moody’s bank financial strength rating (see Brandao-Marques, Correa and Sapriza, 2018). Source: Moody’s Investors Services and author’s calculations. 
Probability of Support Measure of government support calculated as in Gropp, Hakenes, and Schnabel (2011). This measure uses Fitch’s Individual Rating and Issuer Rating, maps these ratings to default probabilities, and then infers the probability of being bailed out by the government. Source: Fitch Ratings Global CorporateFinance 2006 Transition and Default Study and author’s calculations. 
Government Blockholder Dummy variable equal to 1 if the bank has at least one government shareholder that owns voting shares greater than or equal to 10%. Source: Bankscope, FactSet/Lionshares, Annual Reports. 
Deposit Insurance Dummy variable equal to one if the government is involved in managing the deposit insurance system. Source:Demirgüç-Kunt, Kane, and Laeven, 2015
Level of Deposit Insurance The natural logarithm of the U.S. dollar value of minimum required amount of deposit insurance mandated by the government. Source:Demirgüç-Kunt, Kane, and Laeven (2015)
Compensation Index The percentage of the bank's compensation and ownership attributes that satisfy thresholds specified by RiskMetrics Group's CGQ, calculated as in Aggarwal et al. (2010). Source: RiskMetrics. 
Insider Ownership Percentage of shares owned by individual insiders. Source: FactSet/Lionshares. 
Institutional Ownership Percentage of shares owned by institutions. Source: FactSet/Lionshares. 
Blockholder Dummy variable equal to 1 if the bank has at least one single owner with voting shares larger than 10%. Source: FactSet/Lionshares, Bankscope, Annual Reports. 
VariableDefinition
Sponsor Dummy variable equal to 1 if the bank sponsored at least one SIV, securities arbitrage, or hybrid ABCP vehicle during a given year. Source: Moody's quarterly program index. 
ln(Total Assets) Natural logarithm of the total assets of the bank (US$). Source: Bankscope. 
Return on Assets Net income divided by average assets. Source: Bankscope. 
Equity to Assets Ratio of book equity to total assets. Source: Bankscope. 
Loans to Assets Ratio of net loans to total assets. Source: Bankscope. 
Deposits to Assets Ratio of deposits to total assets. Source: Bankscope. 
Non-Interest Operating Income to Assets Ratio of non-interest operating income to total assets. Source: Bankscope. 
High Yield Underwriter Indicator variable equal to one if there is a positive amount of high yield debt underwritten by the bank during 2006. Source: Bloomberg, Dealogic DCM Analytics. 
Securitization Underwriter Indicator variable equal to one if there is a positive amount of asset-backed securities and MBS underwritten by the bank during 2006. Source: Bloomberg and Dealogic DCM Analytics. 
Dummy Landesbank Dummy variable equal to 1 if the bank is a German Landesbank. Source: Bankscope. 
Dummy US Dummy variable equal to 1 if the bank is domiciled in the United States. Source: Bankscope. 
Dummy Spain/Portugal Dummy variable equal to 1 if the bank is domiciled in Spain or Portugal. Source: Bankscope. 
Government Support For each bank, Moody’s long-term foreign currency deposit rating minus Moody’s bank financial strength rating (see Brandao-Marques, Correa and Sapriza, 2018). Source: Moody’s Investors Services and author’s calculations. 
Probability of Support Measure of government support calculated as in Gropp, Hakenes, and Schnabel (2011). This measure uses Fitch’s Individual Rating and Issuer Rating, maps these ratings to default probabilities, and then infers the probability of being bailed out by the government. Source: Fitch Ratings Global CorporateFinance 2006 Transition and Default Study and author’s calculations. 
Government Blockholder Dummy variable equal to 1 if the bank has at least one government shareholder that owns voting shares greater than or equal to 10%. Source: Bankscope, FactSet/Lionshares, Annual Reports. 
Deposit Insurance Dummy variable equal to one if the government is involved in managing the deposit insurance system. Source:Demirgüç-Kunt, Kane, and Laeven, 2015
Level of Deposit Insurance The natural logarithm of the U.S. dollar value of minimum required amount of deposit insurance mandated by the government. Source:Demirgüç-Kunt, Kane, and Laeven (2015)
Compensation Index The percentage of the bank's compensation and ownership attributes that satisfy thresholds specified by RiskMetrics Group's CGQ, calculated as in Aggarwal et al. (2010). Source: RiskMetrics. 
Insider Ownership Percentage of shares owned by individual insiders. Source: FactSet/Lionshares. 
Institutional Ownership Percentage of shares owned by institutions. Source: FactSet/Lionshares. 
Blockholder Dummy variable equal to 1 if the bank has at least one single owner with voting shares larger than 10%. Source: FactSet/Lionshares, Bankscope, Annual Reports. 
Table AI.

Variable definitions

VariableDefinition
Sponsor Dummy variable equal to 1 if the bank sponsored at least one SIV, securities arbitrage, or hybrid ABCP vehicle during a given year. Source: Moody's quarterly program index. 
ln(Total Assets) Natural logarithm of the total assets of the bank (US$). Source: Bankscope. 
Return on Assets Net income divided by average assets. Source: Bankscope. 
Equity to Assets Ratio of book equity to total assets. Source: Bankscope. 
Loans to Assets Ratio of net loans to total assets. Source: Bankscope. 
Deposits to Assets Ratio of deposits to total assets. Source: Bankscope. 
Non-Interest Operating Income to Assets Ratio of non-interest operating income to total assets. Source: Bankscope. 
High Yield Underwriter Indicator variable equal to one if there is a positive amount of high yield debt underwritten by the bank during 2006. Source: Bloomberg, Dealogic DCM Analytics. 
Securitization Underwriter Indicator variable equal to one if there is a positive amount of asset-backed securities and MBS underwritten by the bank during 2006. Source: Bloomberg and Dealogic DCM Analytics. 
Dummy Landesbank Dummy variable equal to 1 if the bank is a German Landesbank. Source: Bankscope. 
Dummy US Dummy variable equal to 1 if the bank is domiciled in the United States. Source: Bankscope. 
Dummy Spain/Portugal Dummy variable equal to 1 if the bank is domiciled in Spain or Portugal. Source: Bankscope. 
Government Support For each bank, Moody’s long-term foreign currency deposit rating minus Moody’s bank financial strength rating (see Brandao-Marques, Correa and Sapriza, 2018). Source: Moody’s Investors Services and author’s calculations. 
Probability of Support Measure of government support calculated as in Gropp, Hakenes, and Schnabel (2011). This measure uses Fitch’s Individual Rating and Issuer Rating, maps these ratings to default probabilities, and then infers the probability of being bailed out by the government. Source: Fitch Ratings Global CorporateFinance 2006 Transition and Default Study and author’s calculations. 
Government Blockholder Dummy variable equal to 1 if the bank has at least one government shareholder that owns voting shares greater than or equal to 10%. Source: Bankscope, FactSet/Lionshares, Annual Reports. 
Deposit Insurance Dummy variable equal to one if the government is involved in managing the deposit insurance system. Source:Demirgüç-Kunt, Kane, and Laeven, 2015
Level of Deposit Insurance The natural logarithm of the U.S. dollar value of minimum required amount of deposit insurance mandated by the government. Source:Demirgüç-Kunt, Kane, and Laeven (2015)
Compensation Index The percentage of the bank's compensation and ownership attributes that satisfy thresholds specified by RiskMetrics Group's CGQ, calculated as in Aggarwal et al. (2010). Source: RiskMetrics. 
Insider Ownership Percentage of shares owned by individual insiders. Source: FactSet/Lionshares. 
Institutional Ownership Percentage of shares owned by institutions. Source: FactSet/Lionshares. 
Blockholder Dummy variable equal to 1 if the bank has at least one single owner with voting shares larger than 10%. Source: FactSet/Lionshares, Bankscope, Annual Reports. 
VariableDefinition
Sponsor Dummy variable equal to 1 if the bank sponsored at least one SIV, securities arbitrage, or hybrid ABCP vehicle during a given year. Source: Moody's quarterly program index. 
ln(Total Assets) Natural logarithm of the total assets of the bank (US$). Source: Bankscope. 
Return on Assets Net income divided by average assets. Source: Bankscope. 
Equity to Assets Ratio of book equity to total assets. Source: Bankscope. 
Loans to Assets Ratio of net loans to total assets. Source: Bankscope. 
Deposits to Assets Ratio of deposits to total assets. Source: Bankscope. 
Non-Interest Operating Income to Assets Ratio of non-interest operating income to total assets. Source: Bankscope. 
High Yield Underwriter Indicator variable equal to one if there is a positive amount of high yield debt underwritten by the bank during 2006. Source: Bloomberg, Dealogic DCM Analytics. 
Securitization Underwriter Indicator variable equal to one if there is a positive amount of asset-backed securities and MBS underwritten by the bank during 2006. Source: Bloomberg and Dealogic DCM Analytics. 
Dummy Landesbank Dummy variable equal to 1 if the bank is a German Landesbank. Source: Bankscope. 
Dummy US Dummy variable equal to 1 if the bank is domiciled in the United States. Source: Bankscope. 
Dummy Spain/Portugal Dummy variable equal to 1 if the bank is domiciled in Spain or Portugal. Source: Bankscope. 
Government Support For each bank, Moody’s long-term foreign currency deposit rating minus Moody’s bank financial strength rating (see Brandao-Marques, Correa and Sapriza, 2018). Source: Moody’s Investors Services and author’s calculations. 
Probability of Support Measure of government support calculated as in Gropp, Hakenes, and Schnabel (2011). This measure uses Fitch’s Individual Rating and Issuer Rating, maps these ratings to default probabilities, and then infers the probability of being bailed out by the government. Source: Fitch Ratings Global CorporateFinance 2006 Transition and Default Study and author’s calculations. 
Government Blockholder Dummy variable equal to 1 if the bank has at least one government shareholder that owns voting shares greater than or equal to 10%. Source: Bankscope, FactSet/Lionshares, Annual Reports. 
Deposit Insurance Dummy variable equal to one if the government is involved in managing the deposit insurance system. Source:Demirgüç-Kunt, Kane, and Laeven, 2015
Level of Deposit Insurance The natural logarithm of the U.S. dollar value of minimum required amount of deposit insurance mandated by the government. Source:Demirgüç-Kunt, Kane, and Laeven (2015)
Compensation Index The percentage of the bank's compensation and ownership attributes that satisfy thresholds specified by RiskMetrics Group's CGQ, calculated as in Aggarwal et al. (2010). Source: RiskMetrics. 
Insider Ownership Percentage of shares owned by individual insiders. Source: FactSet/Lionshares. 
Institutional Ownership Percentage of shares owned by institutions. Source: FactSet/Lionshares. 
Blockholder Dummy variable equal to 1 if the bank has at least one single owner with voting shares larger than 10%. Source: FactSet/Lionshares, Bankscope, Annual Reports. 

References

Acharya
V.
,
Anginer
D.
,
Warburton
J.
(
2016
): The end of market discipline? Investor expectations of implicit government guarantees. Unpublished working paper, New York University.

Acharya
V.
,
Schnabl
P.
,
Suarez
G.
(
2013
):
Securitization without risk transfer
,
Journal of Financial Economics
107
,
515
536
.

Adrian
T.
,
Ashcraft
A.
,
Boesky
H.
,
Pozsar
Z.
(
2010
) Shadow banking, Staff Report No. 458, Federal Reserve Bank of New York.

Aggarwal
R.
,
Erel
I.
,
Stulz
R.
,
Williamson
R.
(
2010
):
Differences in governance practices between U.S. and foreign firms: measurement, causes, and consequences
,
Review of Financial Studies
23
,
3131
3169
.

Bannier
C. E.
,
Feess
E.
,
Packham
N.
(
2013
):
Competition, bonuses, and risk-taking in the banking industry
,
Review of Finance
17
,
653
690
.

Bates
D. S.
(
1991
):
The crash of ’87: was it expected? The evidence from options markets
,
Journal of Finance
46
,
1009
1044
.

Beltratti
A.
,
Stulz
R.
(
2012
):
The credit crisis around the globe: why did some banks perform better?
,
Journal of Financial Economics
105
,
1
17
.

Berger
A. N.
,
Imbierowicz
B.
,
Rauch
C.
(
2016
):
The role of corporate governance in bank failures during the recent financial crisis
,
Journal of Money, Credit, and Banking
48
,
729
770
.

Brandao-Marques
L.
,
Correa
R.
,
Sapriza
H.
(
2018
):
Government support, regulation, and risk taking in the banking sector
,
Journal of Banking and Finance
, forthcoming.

Brunnermeier
M.
(
2009
):
Deciphering the liquidity and credit crunch of 2007–2008
,
Journal of Economic Perspectives
23
,
77
100
.

Calomiris
C.
(
2009
):
Financial innovation, regulation, and reform
,
Cato Journal
29
,
65
91
.

Chan
N.
,
Getmansky
M.
,
Haas
S. M.
,
Lo
A. W.
(
2006
) Systemic risk and hedge funds, in:
Carey
M.
,
Stulz
R. M.
(eds.),
The Risks of Financial Institutions
,
University of Chicago Press
, Chicago, pp.
235
340
.

Chen
H.
,
Joslin
S.
,
Tran
N. K.
(
2012
):
Rare disasters and risk sharing with heterogeneous beliefs
,
Review of Financial Studies
25
,
2189
2224
.

Cheng
I. H.
,
Hong
H.
,
Scheinkman
J. A.
(
2015
):
Yesterday's heroes: compensation and risk at financial firms
,
Journal of Finance
70
,
839
879
.

Correa
R.
,
Lee
K. H.
,
Sapriza
H.
,
Suarez
G.
(
2014
):
Sovereign credit risk, banks' government support, and bank stock returns around the world
,
Journal of Money, Credit and Banking
46
,
93
121
.

Coval
J.
,
Jurek
J.
,
Stafford
E.
(
2009
):
The economics of structured finance
,
Journal of Economic Perspectives
23
,
3
25
.

Covitz
D.
,
Liang
N.
,
Suarez
G.
(
2013
):
The evolution of a financial crisis: collapse of the asset-backed commercial paper market
,
Journal of Finance
68
,
815
848
.

De Jonghe
O.
(
2010
):
Back to the basics in banking: a micro analysis of banking system stability
,
Journal of Financial Intermediation
19
,
387
417
.

Demirgüç-Kunt
A.
,
Kane
E.
,
Laeven
L.
(
2015
):
Deposit insurance around the world: a comprehensive analysis and database
,
Journal of Financial Stability
20
,
155
183
.

DeYoung
R.
,
Torna
G.
(
2013
):
Nontraditional banking activities and bank failures during the financial crisis
,
Journal of Financial Intermediation
22
,
397
421
.

Diamond
D. W.
,
Rajan
R. G.
(
2009
):
The credit crisis: conjectures about causes and remedies,
American Economic Review
99
,
606
610
.

Dinc
S.
(
2005
):
Politicians and banks: political influences on government-owned banks in emerging markets
,
Journal of Financial Economics
77
,
453
479
.

Duarte
J.
,
Longstaff
F. A.
,
Yu
F.
(
2007
):
Risk and return in fixed income arbitrage: nickels in front of a steamroller
,
Review of Financial Studies
20
,
769
812
.

Ellul
A.
,
Yerramilli
V.
(
2013
):
Stronger risk controls, lower risk: evidence from US bank holding companies
,
Journal of Finance
68
,
1757
1803
.

Erkens
D.
,
Hung
M.
,
Matos
P.
(
2012
):
Corporate governance in the 2007–2008 financial crisis: evidence from financial institutions worldwide
,
Journal of Corporate Finance
18
,
389
411
.

Fahlenbrach
R.
,
Stulz
R. M.
(
2011
):
Bank CEO incentives and the credit crisis
,
Journal of Financial Economics
99
,
11
26
.

Fung
W.
,
Hsieh
D. A.
(
2001
):
The risk in hedge fund strategies: theory and evidence from trend followers
,
Review of Financial Studies
14
,
313
341
.

Gatev
E.
,
Strahan
P.
(
2006
):
Banks’ advantage in hedging liquidity risk: theory and evidence from the commercial paper market
,
Journal of Finance
61
,
867
892
.

Gorton
G.
(
2010
):
Slapped by the Invisible Hand: The Panic of 2007
,
Oxford University Press
, Oxford.

Gorton
G.
,
Metrick
A.
,
Shleifer
A.
,
Tarullo
D.
(
2010
):
Regulating the shadow banking system
,
Brookings Papers on Economic Activity
2
,
261
312
.

Gorton
G.
,
Rosen
R.
(
1995
):
Corporate control, portfolio choice, and the decline of banking
,
Journal of Finance
50
,
1377
1420
.

Gropp
R.
,
Hakenes
H.
,
Schnabel
I.
(
2011
):
Competition, risk-shifting, and public bail-out policies
,
Review of Financial Studies
24
,
2084
2120
.

Gropp
R.
,
Vesala
J.
,
Vulpes
G.
(
2006
):
Equity and bond market signals as leading indicators of bank fragility
,
Journal of Money, Credit and Banking
38
,
399
428
.

Hau
H.
,
Thum
M.
(
2009
):
Subprime crisis and board (in-) competence: private vs. public banks in Germany
,
Economic Policy
24
,
701
752
.

Iannotta
G.
,
Pennacchi
G.
,
Santos
J. A.
(
2018
):
Ratings-based regulation and systematic risk incentives
,
Review of Financial Studies
32
,
1374
1415
.

John
T.
,
John
K.
(
1993
):
Top management compensation and capital structure
,
Journal of Finance
48
,
949
974
.

Keeley
M. C.
(
1990
):
Deposit insurance, risk, and market power in banking
,
American Economic Review
80
,
1183
1200
.

Laeven
L.
,
Levine
R.
(
2009
):
Bank governance, regulation and risk taking
,
Journal of Financial Economics
93
,
259
275
.

Merton
R. C.
(
1977
):
An analytic derivation of the cost of deposit insurance and loan guarantees
,
Journal of Banking and Finance
1
,
3
11
.

Minton
B. A.
,
Taillard
J. P.
,
Williamson
R.
(
2014
):
Financial expertise of the board, risk taking, and performance: evidence from bank holding companies
,
Journal of Financial and Quantitative Analysis
49
,
351
380
.

Saunders
A.
,
Strock
E.
,
Travlos
N. G.
(
1990
):
Ownership structure, deregulation, and bank risk taking
,
Journal of Finance
45
,
643
654
.

Smith
C. W.
,
Stulz
R. M.
(
1985
):
The determinants of firms’ hedging policies
,
Journal of Financial and Quantitative Analysis
20
,
391
405
.

This work is written by (a) US Government employee(s) and is in the public domain in the US.

Supplementary data