Post on 17-Jan-2022
Comments Welcome
Syndicated Loan Risk: The Effects of Covenants and Collateral*
by
Jianglin Dennis Ding
Mario J. Gabelli School of Business Roger Williams University
Email: dding@rwu.edu
and
George G. Pennacchi
Department of Finance University of Illinois
Email: gpennacc@illinois.edu
This version: 16 April 2021
Abstract
This paper presents a new approach that quantifies how a credit rating agency and investors judge the effects of collateral and various covenants on syndicated loan risk. It addresses firms’ self-selection of these contract terms by analyzing how a loan’s collateral and covenants affect: 1) the difference between the loan’s credit rating and the senior, unsecured credit rating of the borrowing firm; and 2) the difference between the borrowing firm’s senior, unsecured CDS spread and its loan’s credit spread. The results show that the rating agency and investors agree that a collateral requirement, a capital expenditure covenant, and a dividend restriction covenant are most important for reducing a loan’s credit risk. However, equity issuance and excess cashflow sweeps increase a loan’s risk.
*Valuable comments were provided by Heitor Almeida, Tobias Berg, Iftekhar Hasan, Rustom Irani, Mathias Kronlund, Neil Pearson, Joshua Pollet, Gregory Udell, and Mao Ye; participants of seminars at Australian National University, California Polytechnic State University, Florida International University, HEC Paris, Lehigh University, the University of Delaware, and the University of Illinois; and participants of the Financial Intermediation Research Society Conference, the Marstrand Finance Conference, and the BAFFI CAREFIN Banking Conference.
1
Over the past decade, global syndicated loan issuance averaged $4.2 trillion per year,
with U.S. syndicated loans accounting for 53% of the volume.1 As syndicated lending has grown
and evolved, concerns over its effect on financial stability have arisen.2 Yet evaluating the risks
of these loans is complex because a loan’s default losses depend on the financial condition of its
borrowing firm and also the intricate terms of its loan agreement. Specifically, a loan contract’s
covenants and collateral may alter the loan’s default risk relative to that of debt that lack these
provisions, such as the typical corporate bond.3 Quantifying how default or “credit” risk is
influenced by collateral and various types of covenants is the focus of this study. It examines how
these loan contract terms affect credit risk from the perspectives of both a credit rating agency
and debt market investors.
Estimating the effects of covenants and collateral on a loan’s credit risk is challenging
because, like other corporate finance decisions, the choice of these loan contract terms may be
endogenous to a firm’s characteristics. Empirical analysis must account for a firm’s self-selection
of covenants and collateral. Naively comparing loan risk measures, such as credit spreads or
credit ratings, for loans of different firms that vary in these contract terms can lead to biased
inference. If collateral and covenants are chosen more often by high (low) risk firms, credit
spreads of collateralized and covenant-intensive loans will under- (over-) estimate the impact of
these contract terms if there is not proper account for the firms’ higher (lower) risk.
There is substantial theoretical research on the role of collateral and covenants in debt
contracts. This literature generally agrees that, ceteris paribus, collateral reduces a loan’s credit
risk, but it lacks consensus on whether riskier firms are more likely to collateralize their debt.4
1 See Refinitiv/Thomson Reuters Global Syndicated Loans Review for the years 2011-2020. Loan issuance peaked in 2018 at $5.1 trillion when, for comparison, global bond issuance was $6.6 trillion. 2 For example, see Regulators Voice Concerns over US Leveraged Loan Risk,” Reuters 8 May 2019. 3 Bradley and Roberts (2015) survey prior research showing that public bonds tend to have few, if any, covenants. 4 For example, in Bester (1985), Chan and Kanatas (1985), and Besanko and Thakor (1987) collateral creates incentives for a firm to select low-risk projects or it acts as a quality screening device. In these models, firms with collateralized debt have low risk, in equilibrium. In contrast, collateralized debt can
2
Similarly, most covenants tend to reduce loan risk, but theories differ on whether riskier firms
will have more covenant-intensive loans.5 Empirical evidence, however, suggests that collateral
requirements are more common in the debt of riskier firms since Berger and Udell (1990) find
that collateralized loans have higher credit spreads than uncollateralized ones and John, Lynch,
and Puri (2003) present evidence that the credit spreads of secured bonds exceed those of
unsecured bonds.6 Empirical research also finds that smaller and higher-leveraged firms are more
likely to have covenants in their bonds (Malitz 1986, Begley 1994, and Billett et al. 2007) and
syndicated loans (Bradley and Roberts 2015).
The first contribution of our paper is methodological. We present an approach to
estimating the effects of covenants and collateral on a loan’s credit risk that addresses a firm’s
self-selection of these contract terms. Implementing our approach requires that credit risk
measures for different debt securities of the same firm be observable at the same date.
Specifically, by comparing a credit risk measure for a firm’s loan that includes covenants and
collateral versus the credit risk measure of another debt security of the firm that lacks these
contract terms, we can quantify the effects of individual covenants and collateral on credit risk. It
is the within-firm variation in how covenants and collateral affect credit risk that allows us to
account for self-selection.
Our paper’s second contribution is show that this approach is, indeed, feasible for many
firms and leads to important empirical insights. Furthermore, our approach can be implemented
using two alternative measures of credit risk: credit ratings and credit spreads. While these two
solve a firm’s underinvestment problem, as in Stulz and Johnson (1985), or give the lender the power to force a debt renegotiation, as in Bester (1994). In these models, collateralized loans are more likely to be chosen by riskier firms. 5 Covenants will be more common in loans of risky firms if they give lenders the power to force early liquidation, as in Berlin and Mester (1992), or create incentives for lenders to monitor, as in Rajan and Winton (1995) and Park (2000). However, in Gârleanu and Zwiebel (2009) covenants are a screening device where lower-risk firms choose covenant-intensive loans. 6 Bao and Kolasinski (2017) also find that firms are more likely to issue debt that is secured, rather than unsecured, following weak performance or if their credit rating is poor.
3
measures reflect the beliefs of different sets of agents, namely rating agencies versus investors,
we show that they often agree on how particular covenants and collateral affect loan risk.
Our first credit risk measure takes advantage of the now-common practice of assigning
credit ratings to syndicated loans. Agencies began rating syndicated loans in the mid-1990s. By
1998 approximately 45% of syndicated loans were rated and around 80% are rated in recent years
(Standard & Poor’s (2013)). In most cases, a firm that receives an “issue” rating for its syndicated
loan will also be rated at the firm or “issuer” level. A firm rating is a general rating of the firm’s
senior unsecured debt and debt-like obligations that is not linked to a specific debt contract. In
contrast, the firm’s syndicated loan rating is an issue-specific rating that reflects the existing firm
rating but also incorporates the specific terms (including collateral and covenants) of the loan
being rated. Therefore, the difference between a given borrower’s loan rating and firm rating
reflects the collateral and covenant-specific effects on loan risk relative to the firm’s debt that
lack these terms.
The second measure used in this paper is based on the credit spread of a firm’s syndicated
loan and a maturity-matched credit default swap (CDS) spread of the same firm’s senior,
unsecured debt. Similar to a firm credit rating, a senior, unsecured CDS spread reflects the risk of
a generic class of the firm’s uncollateralized debt that typically lacks covenants. Moreover, in a
frictionless, arbitrage-free market, a CDS spread equals the credit spread on a floating-rate bond,
making it comparable to the spread on a syndicated loan which is also floating-rate debt.7
Therefore, the difference between a firm’s senior, unsecured CDS spread and its syndicated loan
spread reflects the reduction in risk associated with the loan’s covenants and collateral.
By differencing loan-specific versus firm-specific measures of credit ratings or credit
spreads at the time that a loan is originated, the paper employs a large number of observations on
7 See, for example, Bomfim (2016) page 73 for a derivation of this result. In principle, the yield spread on a syndicated loan could be compared to the yield spread on one of the firm’s bonds. However, most bonds have fixed coupons and are callable, requiring additional adjustments that would involve more approximation error.
4
a credit rating agency’s or investors’ ex-ante views of how covenants and collateral affect credit
risk. Moody’s (2003, 2004b) carried out a similar “ex-post” credit comparison based on a smaller
sample of firms that had both syndicated loans and bonds at the time the firms defaulted. It found
that the default rate on syndicated loans was 20% lower than that on bonds, and those firms that
defaulted on both had a median loss rate on loans that was 56% of that on senior unsecured
bonds.8 The current paper also uses within-firm but “ex-ante” credit comparisons using a sample
of over 3,500 loans. The larger sample size allows for more detailed analysis of the risk-reduction
effects of collateral and specific types of covenants on particular types of syndicated loans.
Our paper’s empirical work analyzes the risk reduction properties of both non-financial
covenants and financial covenants. Non-financial covenants include restrictions on a borrower’s
actions, such as limiting dividends and requiring a loan prepayment, known as a “sweep,” if the
borrower obtains proceeds from a debt or equity issue or from an asset sale. Financial or
“maintenance” covenants require borrowers to maintain stipulated levels of a financial ratio or
value, such as maintaining an interest coverage ratio or an amount of net worth above pre-
specified minimums. A borrower that violates a covenant is technically in default, though lenders
often waive violations in return for renegotiating the loan’s terms, such as raising the interest rate
or requiring more collateral. In this way, covenants can mitigate lenders’ default losses.
Previous research proposed various measures of covenant intensity. Covenant intensity is
often calculated simply as the total number of a loan’s covenants. Alternatively, Bradley and
Roberts (2015) introduced a covenant intensity index that is a weighted average of a subset of
covenants that primarily restrict borrower actions but put little weight on financial covenants. A
contribution of Matvos (2013), as well as our paper, is to jointly determine the relative
importance of different types of covenants based on how they impact credit risk. However while -
8 Loans and bonds have different (non-technical) default rates because some defaults occur outside of bankruptcy. For example, a firm may violate one of its loan covenants, and lenders may force a default on the firm’s bond payment that leads to a bond restructuring but no default on the loan.
5
Matvos (2013) examines the relative importance of financial covenants for revolving lines of
credit, our paper analyzes both financial and non-financial covenants for revolving lines of credit
and also term loans.9 Moreover, different from prior research that evaluates loan risk solely in
terms of credit spreads, we examine loan risk measured by both credit spreads and credit ratings.
The paper’s empirical work consists of two sets of regression exercises. The first uses the
difference between a loan’s credit rating and the borrower’s firm rating as the dependent variable,
while in the second the dependent variable is the “CDS – loan basis,” equal to the borrowing
firm’s senior, unsecured CDS spread minus its loan’s all-in-drawn spread. Each regression’s
explanatory variables are indicators for whether the loan is collateralized and whether it contains
various types of covenants, as well as other variables controlling for the loan’s characteristics.
Our results confirm that collateral and various covenant intensity measures are associated
with an improvement in loan ratings and a reduction in loan credit spreads. On average, secured
loans are rated around 1.4 notches higher, and have about a 25 basis point lower credit spread,
than unsecured loans. Furthermore, based on various measures of covenant intensity, the marginal
value of each additional covenant raises a loan’s credit rating by around 0.02 to 0.04 of a rating
notch and lowers a loan’s credit spread by around 7 to 11 basis points.10 Hence, the marginal
improvement from each additional covenant appears relatively small.
However, when our empirical work accounts for the joint occurrence of different types of
covenants in loan contracts, we find that a covenant’s effect on risk reduction varies greatly by
covenant type. In terms of financial covenants, a maximum capital expenditure covenant is found
to significantly improve a loan’s credit rating and also reduce its credit spread, particularly if the
9 Matvos (2013) also analyzes the effects of the strictness of individual financial covenants, an issue that has been the focus of other research such as Drucker and Puri (2009), Demiroglu and James (2010), and Murfin (2012). Our analysis examines only the existence of covenants in loans and not their strictness. However, Matvos (2013) finds that once a covenant is chosen for a loan, differences in strictness have a small marginal effect. 10 Credit ratings can take 21 levels, known as notches. For example, using Moody’s scale, Aaa is the highest and C is the lowest. A one notch difference is the difference between neighboring notches, such as Aaa versus Aa1 or Ba1 versus Ba2.
6
loan is a revolving line of credit. A fixed cost coverage covenant also significantly reduces credit
spreads for lines of credit. Turning to nonfinancial covenants, our empirical findings show that a
dividend restriction covenant significantly improves a loan’s credit rating and reduces its credit
spread for both term loans and revolving lines of credit.
An intriguing and unexpected finding is that a credit rating agency and investors agree
that two nonfinancial covenants actually raise loan risk for both credit lines and term loans. Our
empirical tests show that inclusion of an equity issuance sweep or an excess cashflow sweep in a
loan contract worsens the loan’s credit rating and raises its credit spread. A logical explanation is
that new equity and excess cashflow tend to improve a loan’s value, and covenants requiring pre-
payment of loans after these events may reduce firms’ incentives to issue new equity or generate
additional cashflow.
The paper’s last set of findings examine how the effects of collateral and covenants differ
by the level of the borrowing firm’s risk. Here the credit rating agency and investors largely agree
that collateral has the biggest impact on improving loan risk for the least credit-worthy firms.
Interestingly, dividend restrictions improve a loan’s credit rating and lower its credit spread
mainly for firms with sub-investment grade firm ratings. Conversely, it is mainly loans issued by
investment grade firms that have worse loan ratings and higher spreads when they contain equity
issuance or excess cashflow sweeps.
The remainder of the paper is structured as follows. Section I presents our methodology
for analyzing the effects of covenants and collateral and shows how it addresses self-selection in
firms’ choice of these contract terms. Section II provides information about the data, variables,
and summary statistics for our analysis of credit ratings. Section III gives the credit rating results.
Section IV describes the credit spread data sample, and Section V provides the results from
analyzing these credit spreads. Section VI concludes.
7
I. Methodology for Addressing Self Selection of Covenants and Collateral
This section models a firm’s selection of covenants and collateral for its debt and the
outcome these decisions have on the debt’s credit risk.11 It also shows how the effects of
covenants and collateral can be estimated if credit risk measures are available for multiple debt
securities of the same firm.
A. The Selection Problem
Consider firm i’s selection of a set of k different covenants and collateral on a particular
debt security, d. Let ( )1 2 ... d d d di i i ikC c c c= be debt security d’s 1×k vector of covenants and
collateral where 1dijc = indicates that debt security d of firm i includes covenant/collateral j, j =
1,…,k. If the covenant/collateral of type j is not in debt security d ’s contract, then 0dijc = . Let
there be k latent variables Sij, j = 1,…,k that determine whether firm i selects covenant j for its
debt. The standard selection equations are:
, 1,...,
1 if 00 if 0
ij i j ij
ijdij
ij
S X u j k
Sc
S
γ= + =
>= ≤
(1)
where Xi is a vector of firm i’s observable characteristics that determine selection and uij are firm-
and covenant-specific variables that are unobserved by the econometrician and satisfy E[uij ]=0,
Var(uij )=1, and Cov(uij,uiv) = ρjv.
Next consider a credit risk measure for the firm’s debt, which can be the debt’s credit
rating, credit spread, or yield, and for concreteness we will refer to it as the debt’s yield. Thus, let
diy be the yield on firm i’s debt security d, which is assumed to be a function of
covenants/collateral, other debt contract terms, and firm i’s characteristics. In the typical
covenant/collateral selection problem, such as Bradley and Roberts (2005) or Matvos (2013), it is
11 See Li and Prabhala (2007) for a general review self-selection problems in the field of corporate finance.
8
assumed that only one debt security and its associated collateral/covenants is observed for each
firm at a given point in time. Referring to this single debt security as a loan and its vector of
collateral/covenants as liC , the standard outcome equation for the yield on firm i’s debt is:
1 2 3 for unobserved for
d dd i i i ii
C Z X d ly
d lβ β β ε + + + =
= ≠
(2)
where diZ is a vector of debt security d’s contract terms unrelated to covenants or collateral, Xi is
the vector of firm i’s observable characteristics that affect its yields, and εi denotes variables that
affect the yield on firm i’s debt that are unobserved by the econometrician and satisfy E[εi] = 0,
Var(εi)=σ 2, and Cov(uij,εi ) = ρj.12 Equation (2) says that we observe firm i’s yield on its loan that
includes the covenants/collateral chosen by the firm, liC . However, prior studies typically assume
that one does not observe a yield for the same firm that has other possible covenant/collateral
combinations li
diC C≠ .
Our goal is to estimate how covenants/collateral affect yields; that is, the coefficient
vector β1. However, the selection problem is that if Cov(uij,εi ) = ρj ≠ 0, then an ordinary least
squares regression of 1 2 3l l li i i i iy C Z Xβ β β ε= + + + using data on loans for a cross-section of
different firms would lead to biased estimates since
| | 0l li i i ij iE C E u Cε ε = ⇒ ≠ (3)
B. Potential Solutions to the Selection Problem
For the case of one covenant/collateral choice, i.e. k =1, a common method for estimating
equations (1) to (3) is to use a 2-stage Heckman selection adjustment. This method estimates the
inverse Mills ratio from a first-stage Probit model of equation (1) and includes it in a second stage
12 For example, Xi may include characteristics such as the firm’s operational (asset) risk and its overall leverage while d
iZ includes the debt security d’s contract terms such as its amount and maturity. εi might reflect hard-to-quantify characteristics such as the quality of the firm’s management.
9
regression of equation (2), where the inverse Mills ratio corrects for the bias reflected in equation
(3). However, this procedure becomes numerically challenging or infeasible when k is more than
2 or 3. For example, Bradley and Roberts (2015) use a Heckman adjustment that essentially
analyzes one covenant choice at a time (k=1). They do not formally model the joint selection of
multiple covenants but attempt to adjust for dependency between covenants by including a
measure of covenant intensity.
Matvos (2013) specifies an equation similar to that of (2) for each firm and identifies a
firm’s unobservable characteristics that affect yields, εi, by modeling how they change over
time.13 These unobserved characteristics are estimated using panel data on individual firms’
repeated borrowing in the syndicated loan market. Hence, his estimation requires a dynamic
model and data on within-firm changes in covenant choice over time. Implementing this approach
requires data where each firm issues multiple loans at different points in time.
Unlike the prior literature, our approach utilizes information on the yields of two different
debt securities of firm i that are observed at the same point in time. One is the yield on firm i’s
loan that contains covenants/collateral, and the second is the yield on firm i’s senior, unsecured
debt (bond) that is free of covenants and collateral. Thus, in our context, the yield outcome
equation for firm i at the time a loan is originated becomes:
1 2 3
1 2 3
for for
unobserved for or
d di i i i
d d di i i i i
C Z X d ly C Z X d b
d l b
β β β εβ β β ε
+ + + == + + + = ≠
(4)
where ( )0 0 ... 0biC = and b
iZ denote the bond’s vector of covenants/collateral and other contract
terms, respectively.
The advantage of simultaneously observing the yield on firm i’s loan and its bond is that
it allows one to substitute out for the firm’s unobserved (and observed) characteristics that affect
13 See Matvos (2013) equations (6) to (10).
10
yields, which is the root of the selection problem. Specifically, subtract the first line of equation
(4) from the second to obtain:
( ) ( )
( )1 2
1 2
b l b l b li i i i i i
l b li i i
y y C C Z Z
C Z Z
β β
β β
− = − + −
= − + − (5)
where the second line in equation (5) follows from the fact that ( )0 0 ... 0biC = when a bond is
free of covenants and collateral. Equation (5) says that the difference between firm i’s unsecured
bond yield and its loan yield depends on the covenants and collateral contract terms in firm i’s
loan as well as the difference in other contract terms for the bond and loan.
Importantly, both observed firm-specific characteristics, Xiβ3, and the unobserved firm-
specific characteristic that is correlated with covenant/collateral choice, εi, no longer appear in
equation (5). Thus, in principle for given values of β1 and β2, the reduction in yields from
collateral and particular covenants can be measured without the typical source of selection bias.
Equation (5) can be interpreted as a regression equation by assuming that the bond yield
or the loan yield is measured with error. For example, our empirical work will use a firm’s credit
default swap (CDS) spread as a proxy for the yield on the firm’s senior, unsecured bond or it will
use the generic long-term firm rating as a proxy for the rating on an actual bond without
covenants and collateral.14 These proxies are likely to have some approximation or measurement
error, say ηi, so that equation (5) can be modified as
( )1 2b l l l bi i i i i iy y C Z Zα α η− = + − + (6)
where α1= -β1 and α2= -β2. As long as the usual assumption holds that this approximation error is
uncorrelated with the selection of covenants/collateral, then the equation (6) can be estimated by
ordinary least squares using a cross-section, time series of firms that issued syndicated loans.
14 In the case of ratings, measurement error can also derive from interpreting ratings as a discrete approximation of a “true” continuous rating variable since ratings are set only at discrete notches.
11
C. Caveats
Equation (2), and thus equation (6), reflects the standard assumption that covenants and
collateral affect a debt’s yield in a linear manner that is separable from the firm’s characteristics
(c.f., Matvos (2013) equation 6). For example, it assumes the sensitivity of covenants and
collateral on a yield, β1, does not depend on the firm’s characteristics such as Xi. In reality, that
may not be the case. Therefore, our empirical work will explore whether α1=-β1 varies across
subsamples of firms sorted by the firm’s overall risk as proxied by the level of the firm’s
(issuer’s) rating.
We now turn to estimating the model in equation (6), first using differences in credit
ratings as the dependent variable and, second, using differences in credit spreads (yields) as the
dependent variable. The next section describes our data on credit ratings and loan contract terms.
II. Credit Rating Data, Variables and Summary Statistics
A. Data and Matching
The credit rating data used in this study is a 1995 to 2012 sample of Moody’s syndicated
loan ratings at the time of each loan’s origination, together with the contemporaneous firm issuer
rating of each borrowing firm. Moody’s credit ratings are ideal for the analysis because they
reflect expected loss rates, equal to the product of a debt instrument’s probability of default and
loss given default.15 Each rated loan has a Moody’s unique loan identifier and a firm-level
identifier which are used to match the loan rating with its firm’s rating.
To find the firm rating contemporaneous to the loan rating at origination, a search was
done for firm rating updates up to 365 days before the loan is rated. If an updated firm rating
within this range was not found, then the search range is extended up to 60 days after the loan
was issued, since a firm rating update immediately following the loan issue is likely a more
15 As discussed earlier, covenants and collateral can affect both the default probabilities and losses given default (recoveries). In contrast, Standard & Poor’s (2011) ratings reflect only default probabilities. In December 2003, S&P began issuing a separate rating reflecting loss given default.
12
accurate measure of the firm’s condition when the loan was rated. If a firm rating does not fall
within this time window, the loan rating observation is eliminated from the sample. This
procedure generates 7,355 unique loans with both loan ratings and firm ratings.16
Next, this sample of rated loans was matched to the LPC Dealscan database which
contains information on loan covenants, collateral, maturity, loan type (term loan or revolver),
loan purpose, loan size, and starting and ending date.17 Because there is no common identifier for
loans between Dealscan and Moody’s, loans were matched in two steps starting at the firm level.
If a firm is publicly-traded and has a stock ticker symbol in both datasets, the firm is matched
using the ticker. When a firm does not have a ticker, it is manually matched by the firm name,
checking for name changes and names of subsidiaries. Once both datasets are matched at the firm
level, loan matches are identified by the loan’s starting date, ending date, and loan type. This
procedure yields a final sample of 3,558 loans with both ratings and loan characteristics. Of these,
1,875 (53%) are term loans and 1,683 (47%) are credit lines.
B. Variables and Summary Statistics
Almost 90% of the sample’s observations are loans of firms that have a sub-investment
grade firm rating, and only 2.6% of the observations are loans of firms that have a firm rating of
A3 or higher. Figure 1 shows the sample’s overall distribution of firm and loan ratings at the
notch level, except that all ratings of A3 to Aaa are grouped into a single category due to the few
observations in this range. The figure shows that loan ratings are more frequent than firm ratings
for ratings B1 and better while the opposite occurs for ratings B2 and worse. The average firm
rating is B1 while the average loan rating is Ba3.
A numeric score is assigned to each of the 21 possible Moody’s ratings, with the lowest
16 Different window sizes including 180 days before and 30 day after, 90 days before and 15 days after, and only ratings before the loan issue where tried. The results are robust to choosing these different windows. 17 Dealscan loans with missing information on loan characteristics were dropped from the sample. However, there were very few instances of missing information for those Dealscan loans that could be matched with Moody’s loan ratings.
13
rating, C, equal to 1 and the highest rating, Aaa, equal to 21. Since each loan observation has both
a loan rating and a firm rating, it now has two numeric variables, Loan score and Firm score.
Taking each loan observation and subtracting the firm score from the loan score gives the Rating
difference, which is just the difference in ratings measured in notches. Figure 2 shows the
distribution of rating differences. The average rating difference is 1.03; that is, on average a
loan’s rating is one notch higher than the borrowing firm’s rating.
Based on each observation’s Dealscan information, several indicator (dummy) variables
were created. Appendix A gives detailed definitions of all the variables used in the paper. The
indicator variable Secured equals 1 if the loan is collateralized and equals zero otherwise. The
variable Term equals 1 if the loan is a term loan, and zero if it is a revolving line of credit.
Dealscan also includes information on 15 different types of financial covenants, though
several are used infrequently. The six most common financial covenants were selected,
accounting for more than 95% of all financial covenants in the sample. They are: capital
expenditure, debt to EBITDA, fixed charge coverage, interest coverage, leverage ratio, and net
worth. Dealscan also reports six non-financial covenants, all of which are included in the
analysis: dividend restrictions, asset sales sweep, debt issuance sweep, equity issuance sweep,
excess cash flow sweep, and insurance proceeds sweep. Indicator variables are created for each of
the six financial covenants and six non-financial covenants, where each variable equals 1 if the
covenant exists in the contract and 0 otherwise. Appendix B shows the correlation matrix of the
presence of covenants in our sample of loans.
Another three variables are created that measure covenant intensity. Total financial is the
sum of all financial covenant indicators while Total non-financial is the sum of all non-financial
covenant indicators. Each of these variables ranges from 0 to 6. Total covenant is the sum of all
covenants, ranging from 0 to 12. The analysis also considers the Covenant Intensity Index
proposed by Bradley and Roberts (2015). This index ranges from 0 to 6 and gains 1 point if one
14
of following exists: asset sales sweep, debt issuance sweep, equity issuance sweep, 2 or more
financial covenants, dividend restrictions, and collateral.
Using Bloomberg data, one additional variable is constructed to measure the reputation of
each loan’s lead arranger, which could potentially affect the loan’s rating. The variable Top 10
underwriter equals 1 if the loan’s lead arranger is among the top 10 underwriters for U.S.
syndicated loans measured over the prior 3 years. Approximately three-quarters of our sample’s
loans had a lead arranger that was a top 10 underwriter.
Table I presents summary statistics on the sample’s ratings, collateralization, the
frequencies of different covenants, and other loan characteristics. As mentioned earlier, the
average difference between a loan’s rating and that of its firm is 1.03 notches. The table also
shows that 87% of loans are secured. The single most common financial covenant is debt to
EBITDA (47.8% of all loans) and the single most common non-financial covenant is dividend
restrictions (46.8% of all loans). Summing these average covenant frequencies indicates that the
average number of total covenants per loan is 3.5. In addition, the average loan amount is $365
million, and the average maturity is 5.4 years.
Figures 3 and 4 show significant differences in the main variables based on the risk of the
borrowing firm as proxied by the firm’s rating. Figure 3 shows that the average difference
between loan and firm ratings is negligible for investment grade firms, but this average difference
steadily increases as the firm’s rating worsens. Figure 4 shows that collateral and covenant
intensities also vary by firm rating. The figure stacks different colored rectangles reflecting
collateral and various types of covenants. The height of each rectangle is the proportion of loans
that are secured or the proportion of loans that contain the specific covenant in its contract. The
overall height of the bar reflects the extent of collateralization and covenant intensity.
The right-most bar in Figure 4 reflects the overall frequencies of collateral and covenants
given in Table I’s summary statistics. The left-most bar shows the same frequencies for the loans
of firms with an investment grade firm rating. Consistent with Figure 3 that showed no average
15
difference between loan and firm ratings for investment grade firms, Figure 4 shows that less than
6% of the loans of investment grade firms are collateralized and, on average, these loans have
relatively few covenants. Also consistent with Figure 3 that shows the difference between loan
and firm ratings rise as firm ratings worsen, Figure 4 shows that the likelihood of a loan being
secured tends to rise as the firm’s rating declines. 86% of loans given to Ba-rated firms are
secured but over 99% of loans given to firms rated B or worse are secured.
Interestingly, unlike collateral, covenant intensity is non-monotonic. The average number
of covenants in the loans of firms rated investment-grade, Ba, B, and Caa and worse equals 1.6,
4.8, 3.5, and 3.4, respectively. So, for sub-investment grade firms, covenant intensity declines as
ratings worsen. Some Ba-rated firm may substitute more covenants in their loans because they are
somewhat less likely to be collateralized compared to worse-rated firms.
III. Credit Rating Empirical Specification and Results
A. Regression Specification
Based on equation (6), our regression’s dependent variable is the difference in credit risk
measures for two debt securities of the same firm. One security is a loan that includes possible
covenants, collateral, and other loan-specific contract terms. The other security is the generic
unsecured debt of the firm that is independent of particular contract terms. Specifically, denote
,l
i tr as the (issue) rating of firm i’s loan that is originated at date t, and denote ,f
i tr as the firm
(issuer) rating of firm i at date t. Since credit ratings are measured as a numeric scale from 1 (C-
rated) to 21 (Aaa-rated) that increases with credit quality, we specify the dependent variable as
, ,l
i tf
i tr r− so that it is increasing in the loan’s relative credit quality.
Also consistent with equation (6), the regression’s explanatory variables include a vector
of the loan’s covenants and collateral, liC , and a vector of other loan-specific contract terms, l
iZ .
Since the firm rating reflects that of a generic senior, unsecured bond whose contract terms do not
16
differ across firms or over time, the variable ,
b
i tZ is set to zero so that the elements of α2 reflect the
marginal sensitivity of loan contract terms relative to this generic standard. In summary, our
credit ratings regression takes the form
, , , 1 , 2 ,l f l l
i t i t i t i t t i tr r C Zα α φ η− = + + + (7)
where φt accounts for possible time (year) fixed effects on rating differences and ηi,t is an
independently-distributed, mean zero error term.
It is noteworthy that while equation (7) is based on our theoretical model of Section I, it
is also consistent with how Moody’s describes its actual loan rating process:
Moody’s bank loan ratings are measures of expected credit loss rates and use the
same ratings scale as Moody’s bond ratings….Moody’s rates bank loans
individually, not according to any simple rule of thumb, but based on a careful
analysis of structure and collateral. Typically, loans are rated anywhere from on
par with an issuer’s senior-implied rating to up to three alpha-numeric rating
categories higher. (Moody’s (2004b, page 2)).
Since ,li tC in regression (7) contains a series of variables indicating the possible presence
of different covenants and also an indicator of collateral (Secured), positively-signed estimates of
the elements in vector α1 would indicate that covenants or collateral improve a loan’s rating. In
other regressions, indicator variables of individual covenants will be replaced with a covenant
intensity measure.
The vector of variables of loan characteristics, ,li tZ , in regression (7) include the natural
logs of the loan amount (Size) and of the loan’s number of months until maturity (Maturity). It
also includes the indicator variable Term. An additional series of variables are indicators of the
loan’s purpose, including a general corporate purpose, debt refinancing, recapitalization, and
LBO or merger and acquisition. Finally, these loan characteristics include Top 10 underwriter, a
variable which has been used by John, Lynch, and Puri (2003) and Narayanan, Rangan, and
17
Rangan (2007). A top lead arranger bank may perform better credit screening and monitoring of
borrowers, which could affect a loan’s rating and credit spread. Livingston and Miller (2000) find
that non-convertible debt issues have lower yields when they have more prestigious underwriters.
It is important to note that there can be different interpretations of our results from
regression (7). One is that we are taking the firm credit rating as given and examining whether a
loan’s covenants and collateral improve the loan’s rating relative to this fixed firm rating. This
interpretation is implicit in the Moody’s quote given above. But a second interpretation notes that
the origination of a loan potentially changes the firm’s capital structure and, due to covenants and
collateral, might restrict the firm’s activities.18 In principle, this change could worsen or improve
the firm (issuer) credit rating.19 Thus, while the results of equation (7) are valid in measuring the
effects of a loan’s covenants and collateral on the difference between the loan’s rating and the
firm rating, it is possible that the loan’s contract terms influence both ratings.
B. Collateral and Covenant Intensity Measures
The first set of tests based on equation (7) analyze four different summary measures of
covenant intensity, rather than individual covenants. Table II column 1 reports the results where
,li tC contains Total Covenants, which takes a value from 0 to 12. Its coefficient is statistically
significant and indicates that each additional covenant increases the syndicated loan rating by
2.2% of one notch. Since the average number of covenants per loan is around 3.5, this implies
that the contribution of covenants to increasing loan ratings is about 8% of a notch. Columns 2
and (3) report similar regressions but proxy covenant intensity by either Total Non-financial
Covenants or Total Financial Covenants. The impact of each covenant on the loan’s rating
appears greater for financial ones (4.1% of a notch) compared to non-financial ones (3.3% of a
18 The loan may not change the firm’s capital structure if it replaces an existing or maturing debt instrument. Moreover, if the firm has other loans with similar covenants, a new loan may not further restrict the firm’s activities. 19 For example, if the loan significantly increases the firm’s leverage or absorbs collateral, its effect could worsen the firm rating. Alternatively, if new loan covenants lead to greater monitoring of the firm’s management, a reduction in agency problems could improve the firm rating.
18
notch).20 Column 4 reports results using the Covenant Intensity Index of Bradley and Roberts
(2015) and shows that each unit increase leads to a loan rating improvement of 4.1% of a notch.
Since this index includes collateral, column 5 reports a regression without the variable Secured in
the regression, in which case a unit increase in the index leads to a larger loan rating improvement
of 11% of a notch.
In Table II’s regression specifications, the coefficients on Secured range from 1.39 to
1.46 and are highly significant, indicating that secured loans are rated about 1.4 notches higher
than unsecured loans. Recall that previous studies such as Berger and Udell (1990) and John,
Lynch, and Puri (2003) find, counter-intuitively, that collateralization is associated with higher
loan or bond risk. Importantly, Table II’s results show that once proper account of firm risk is
taken, the independent effect of collateral is to lower credit risk; that is, improve the loan’s rating.
Table II also shows that, all else equal, term loans are on average rated around 0.21 to
0.29 notches lower than credit lines. Even when term loans and credit lines are of the same loan
package so that they are covered by the same covenants, they have different characteristics. In
particular, most term loan contracts give borrowers the option of cancelling the loan before
maturity by paying a cancellation fee (Berg, Saunders, and Steffen (2015)). Further, lenders may
cancel credit lines if a borrower’s credit condition worsens, due to the common “materially
adverse change” clause in credit line contracts.
The regressions also consistently show that loan size has a positive effect on ratings,
perhaps because lenders have a greater incentive to screen and monitor larger loans. The negative
sign on maturity indicates that, all else equal, longer maturity loans are rated worse than shorter
maturity ones: a doubling of maturity reduces the credit rating by about ¼ of a notch. Finally,
lead arranger reputation, as proxied by being a top-10 underwriter, is associated with a better
20 On average, loans have 1.4 financial covenants and 2.1 non-financial covenants.
19
rating. On average, loans made by a top lead arranger are rated 0.14 to 0.19 of a notch higher,
suggesting that rating agencies take a positive view of underwriter reputation.
C. Individual Covenant Effects
The next set of tests examines individual covenants as a preliminary step to determining
their relative importance. Twelve different regressions in the form of equation (7) were run where
the variable ,li tC contained only one of the 12 covenants. Five covenants were not statistically
significant in individually improving the loan rating. Three of the five were the three least
common financial covenants: fixed charge coverage, net worth, and leverage ratio covenants that
were used in only 19%, 6%, and 5% of loans, respectively. The other two were the non-financial
covenants; excess cashflow sweep and equity issuance sweep covenants used in 37% and 24% of
loans, respectively. The results for the seven covenants that did significantly improve the loan’s
rating are reported in Table III.
All seven coefficients have similar magnitudes. On average, the presence of one of these
covenants improves the rating of a loan by 0.172 of a notch, with the capital expenditure and
dividend restriction covenants improving the rating the most. The effects on ratings of collateral,
term loan status, loan size, loan maturity, and underwriter reputation are remarkably similar to
those in Table II that used covenant intensity measures rather than individual covenants.
D. Joint Covenant Effects
Now let us consider jointly including all 12 covenants, as well as the collateral indicator,
in the vector ,li tC of equation (7). Column 1 of Table IV reports the results and shows that, when
conditioning on the choice of all covenants, only two covenants significantly improve a loan’s
rating. One covenant is a limit on the firm’s capital expenditures, which raises a loan’s rating by
0.18 of a notch. This finding is consistent with Nini, Smith, and Sufi (2009) who show that
capital expenditure covenants are effective in restricting the investment of a borrower with
declining credit quality. Moreover, they show that when a loan incorporates such a covenant, the
20
borrowing firm subsequently experiences improved operating and market performance.
A second covenant that restricts dividends significantly improves a loan’s rating by 0.26
of a notch. The efficacy of dividend restriction covenants is supported by the findings of Healy
and Palepu (1990). They note that such covenants often limit dividends to a maximum proportion
of accounting earnings, which raises the possibility that firms may manipulate accounting to
inflate earnings and avoid a dividend reduction. However, their findings show no evidence of
such manipulation but, rather, firms employing dividend cuts. Hence, a covenant restricting
dividends appears to be effective in reducing shareholder – debtholder conflicts.
What, at first, appears puzzling is that Column 1 of Table IV shows that three covenants
actually reduce a loan’s rating relative to its issuing firm’s rating. The first is a minimum net
worth requirement whose presence decreases a loan’s rating by 0.23 of a notch. Dichev and
Skinner (2002) provide evidence that firms make accounting choices to reduce the likelihood of
violating net worth covenants and, moreover, Franz et al. (2014) find that firms that are close to
violating a net worth covenant are more likely to engage in both accounting earnings management
and real earnings management. As real earnings management involves changes in real operational
and investment activities that depart from optimal business decisions, they are counter-
productive.21 In addition, a banker has highlighted other problems with a net worth covenant:
The net worth category of covenants is ripe for misinterpretation and lender problems.
First these covenants are maintenance covenants and are measured on a balance sheet.
That means that the covenant applies at a specific date, and a month later the covenant
ratio can be substantially different. Second, problems arise when market values do not
match GAAP values. Most covenants are measured according to GAAP, however, the
discrepancy between GAAP and market value can be substantial. Tangible assets may
have substantial market value which is not captured in GAAP – for example real estate
improvements. Conversely intangible assets (such as patents, copyrights, distribution
21 They give as examples: increasing inventory production; reducing discretionary expenses; and providing price discounts or credit incentives to accelerate revenue.
21
agreements and marketing rights) can have tremendous value but may be subtracted by
the lender for covenant purposes. (Nichols (2016)).
The other two covenants that appear detrimental to a loan’s rating are an equity issuance
sweep and an excess cashflow sweep, which reduce its rating by 0.21 of a notch and 0.13 of a
notch, respectively. How can this result be logical? Recall that sweeps require a firm to prepay its
loan upon a particular event. However, unlike other types of sweeps equity issuance and excess
cashflow are events that are highly likely to reduce a loan’s default risk.22 Consequently,
requiring that a loan be repaid following these favorable situations is likely to unprofitable for
lenders and unproductive for borrowers. Indeed, the presence of these covenants could increase
default risk by incentivizing firms to avoid issuing equity or generating excess cashflow.
Finally, Column 1 of Table IV shows that when the possible presence of all covenants is
taken into account, the impact of collateral, term loan status, size, loan maturity, and underwriter
reputation remain similar to earlier regressions.
E. Effects of Covenants and Collateral on Term Loans versus Credit Lines
The previous results suggest that when collateral, covenants, and other loan controls are
assumed to have equal effects on term loans and credit lines, term loans tend to be rated about ¼
of a notch lower than revolving lines of credit. Columns 2 and 3 of Table IV loosen the restriction
that contract terms have equal effects on different loan types by reporting regressions run on
separate subsamples of term loans versus revolving credit lines.
The results for the term loan and revolver subsamples are qualitatively similar to that of
the full sample that combines the two loans types. But possibly due to the smaller subsample
sizes, some coefficients lose significance. In particular, the effect of a capital expenditure
covenant on a loan’s rating is significant only for revolvers. An explanation may be that revolvers
22 In contrast, other sweeps are more likely to protect lenders since debt issuance increases leverage and asset sales may reduce productive capacity, both of which could increase default risk. Similarly, insurance proceeds are likely to be a result of a loss in firm capital.
22
are more susceptible to excess capital spending since they allow for future drawdowns of credit.
The negative effects on loan ratings from a net worth covenant and excess cashflow and equity
issuance sweeps also seem more important for revolving lines of credit. However, the effect of a
dividend restriction covenant on improving loan ratings maintains its importance both for term
loans and revolvers.
The effect of collateral is also similar for term loans and revolvers by raising both of their
ratings by over 1.3 notches. Interestingly, a longer maturity is only a detriment to a term loan’s
rating, perhaps because most credit lines have a relatively short maturity.
F. Effects of Covenants and Collateral by Firm Credit Rating
Figures 3 and 4 indicate that the difference between loan and firm ratings, as well as
covenant intensity, varies by a firm’s rating. This fact motivates our exploration of whether the
sensitivities of a loan’s relative rating to the presence of covenants and collateral, which are the
elements of the vector α1 in equation (7), might differ by the level of a firm’s rating. Toward this
end we run regression (7) on the following four subsamples: Investment grade-rated firms (10.4%
of the full sample); Ba-rated firms (18.5% of the full sample); B-rated firms (61.5% of the full
sample); and C-rated firms (9.6% of the full sample). The results are reported in Table V where,
for comparison purposes, column 1 repeats the results for the full sample that were reported
earlier in Table IV.
Comparing Columns 2 to 5 of Table V reveals several differences across these firm-
rating subsamples. First, a capital expenditure covenant significantly improves a loan’s rating
only for investment-grade firms, and by a sizable 0.43 of a notch. Interestingly, a covenant
limiting leverage appears to improve loan ratings for the polar cases of investment-grade firms
and C-rated firms. In contrast, a dividend restriction covenant significantly improves a loan’s
rating only for the middle categories of Ba-rated and B-rated firms.
Other differences are revealed with respect to sweep covenants. A debt issuance sweep
seems detrimental to the rating of loans made to investment grade firms. An equity issuance
23
sweep worsens loan ratings for B-rated firms but actually improves loan ratings for investment-
grade and C-rated firms. Excess cashflow sweeps significantly reduce a loan’s rating by large
amounts of 0.82 of a notch and 0.65 of a notch for investment grade and Ba-rated firms.
Insurance proceeds sweeps reduce a loan’s rating for investment grade firms but raise it for B-
rated firms.
While the effect of collateral on loan ratings is consistently positive across all firm rating
subsamples, it has an extraordinarily large effect for C-rated firms where it improves loan ratings
by 3.8 notches. This result seems sensible since lending to relatively safe firms tends to be based
on “cashflow” but lending to relatively risky firms is based on “assets” or collateral (Berger and
Udell (1990)). Table V also reveals that the negative effects of loan maturity and the positive
effects of loan size and top 10 underwriter status are most important for the lower-quality B-rated
and C-rated firms.
G. Robustness: Financial Firms
Loans to financial firms make up 6% of all loans in our credit rating data sample.
Previous research often excludes financial firms due to their unique firm-level characteristics. But
since our tests use differences in loan versus firm ratings, our methodology in Section I argues for
keeping these firms in the sample. Nonetheless, regression tests on a sample that excludes
financial firms find similar results.
H. Robustness: Time Window for Matching Firm Ratings to Loan Ratings
Firm (issuer) ratings are not updated frequently, so it is sometimes difficult to find a fresh
firm rating that is close enough, and prior to, the loan rating date. The paper’s procedure is to start
by searching for firm ratings 365 days before a loan’s rating date. If a firm rating is not found
during that time window, then a search is performed that includes up to 60 days after the loan is
rated. For robustness, several alternative time windows are tried. New samples are generated
using windows such as 180 days and 90 days before loan rating dates, and 30 days, 15 days, and 0
days after loan rating dates. Using the smallest window of 90 days before and 0 days after reduces
24
the original 3,558 observation sample down to 2,917, a reduction of only 18%. The paper’s main
results are quite similar across the samples that use different windows.
Having examined the effects of covenants and collateral on a loan’s credit rating, let us
now consider how these loan contract terms affect a loan’s credit spread. The next section
describes the data sample used to analyze a loan’s credit spread and its borrowing firm’s senior
unsecured credit default swap spread.
IV. Credit Spread Data, Variables and Summary Statistics
A. Data and Matching
Data on individual loans in Dealscan, including each loan’s all-in-drawn spread, were
matched to Markit data on the borrowing firms’ credit default swap (CDS) spreads. When
available, the borrowing firm’s stock ticker in Dealscan was used to search Markit for the firm’s
senior, unsecured CDS spread corresponding to the day the loan was originated and to a maturity
closest to that of the loan.23 Also obtained was an “implied” credit rating of each borrowing firm
as constructed by Markit.24 Matching on stock tickers resulted in 3,748 loans with complete
Dealscan and Markit data. The sample covers the period from 2002 to 2015.
B. Variables and Summary Statistics
The summary statistics in Table VI show that the borrowing firms’ average senior,
unsecured CDS spread is 141 basis points while their all-in-drawn loan spreads averaged 139
basis points. The relatively small 2 basis point difference in spreads might be partly due to the
tendency of CDS contracts to be more liquid than a firm’s debt.25 Figure 5 shows the sample
distribution of individual firms’ differences in CDS – loan spreads, i.e., their CDS – loan basis.
The distribution is positively skewed, indicating that there are relatively more firms with large
23 CDS spreads for contracts with the standard document clause “ex-restructuring” (XR) were obtained. 24 Markit reports an implied credit rating for each firm based on a comparison of its 5-year CDS spreads to the average CDS spreads of rated firms in the same industry. 25 The CDS – bond basis, equal to a firm’s CDS spread – the spread on the firm’s bond, tends to be negative. Oehmke and Zawadowski (2017) link the magnitude of this negative basis to measures of bond illiquidity and CDS liquidity as proxied by the volume of CDS trading.
25
positive spreads than large negative spreads.
Other summary statistics in Table VI reveal differences in this CDS – loan spread sample
compared to credit rating sample summarized in Table I. Fewer loans are secured (62% versus
87%), and there are relatively less term loans (25% versus 53%). The average number of
covenants per loan is also less (1.7 versus 3.5), and the loans are larger ($1,182 million versus
$365 million) and have a shorter maturity (4.0 years versus 5.4 years). Moreover, a higher
proportion of loans are underwritten by a top-10 lead arranger (93% versus 74%). Apparently, the
requirement that the borrowing firm have traded CDS contracts results in a sample of larger and
less-covenant intensive loans.
V. Credit Spread Empirical Design and Results
A. Methodology
As with the previous analysis of credit ratings, this section’s analysis of credit spreads is
based on the model regression equation (6) in Section I. Analogous to a yield or credit spread on
a firm’s bond that lacks covenants and collateral, the spread on a firm’s senior, unsecured credit
default swap contract is used. Specifically, the dependent variable is the difference between a
firm’s CDS spread and the firm’s syndicated loan spread:
, , , 1 , 2 ,cds l l li t i t i t i t t i ts s C Zα α φ η− = + + + (8)
where ,li ts is the all-in-drawn spread in basis points on firm i’s syndicated loan originated at date t
and ,cdsi ts is the date t equivalent-maturity CDS spread in basis points on firm i’s senior unsecured
debt. Thus, ,cdsi ts - ,
li ts is the firm’s “CDS – loan basis.” If covenants and collateral reduce loan risk
by decreasing the relative spread on loans, one expects that the elements of α1 would be
positive.26
26 We note that the results of estimating equation (8) are subject to the interpretations discussed at the end of Section III.A. While our tests are valid in measuring the effects of a loan’s covenants and collateral on
26
B. Collateral and Covenant Intensity Measures
Similar to Table II, Table VII examines the effects of covenant intensity and collateral on
credit spreads. Consistent with the results using credit ratings, each of the covenant intensity
measures is associated with a decline in loan credit spreads. Also similar to the results using
ratings, financial covenants tend to be slightly more important for reducing a loan’s credit spread
(11.0bp per covenant) compared to non-financial covenants (10.1bp per covenant).
Table VII also confirms the risk-reducing property of collateral: ceteris paribus, secured
loans have around a 30bp lower spread than unsecured loans. Also consistent with the credit
rating results, larger loans are associated with lower loan credit spreads. In contrast, the table
shows that term loans and longer-maturity loans have lower relative spreads, a reversal of the
previous finding that term loans and longer maturities worsened a loan’s credit rating. A further
difference from the credit rating results is that investors are not impressed by an underwriter’s
reputation. Being one of the top-10 lead arrangers does not have a statistically significant effect
on a loan’s spread relative to the firm’s CDS spread.27
C. Individual Covenant Effects
Next, a series of regressions are run with ,li tC in equation (8) containing only one of the 12
individual covenants. Table VIII reports the results of the 7 covenants that have a significantly
positive relationship to the CDS-loan basis, ,cdsi ts - ,
li ts . The same non-financial covenants that were
significant for improving loan credit ratings (dividend restrictions, asset sales sweep, debt
issuance sweep, insurance proceeds sweep) are also found to lower loan credit spreads, as well as
one additional non-financial covenant, excess cashflow sweep. In terms of financial covenants,
the maximum capital expenditures covenant also both improves loan ratings and reduces loan
the difference between the firm’s CDS spread and its loan’s spread, these covenants and collateral may influence both spreads. 27 This result comes with the caveat that only 6.6% of the sample’s loans are underwritten by a bank that is not a top-10 lead arranger.
27
spreads. Yet, while interest coverage and debt to EBITDA covenants improved loan ratings, they
do not significantly lower credit spreads. Rather, the fixed cost coverage covenant, which is
similar to the interest coverage covenant, significantly lowers loan spreads. In each regression,
collateral continues to significantly lower loan spreads, and the average reduction is 26.5bp.
D. Joint Covenant Effects
Table IX column 1 reports regression results where the vector ,li tC includes all 12
covenants. Similar to the findings using credit ratings, the results show that a capital expenditure
covenant and a dividend restriction covenant significantly improve a loan’s credit quality by
reducing its credit spread by 58bp and 16bp, respectively. Also very similar to the results using
credit ratings, we see that an equity issuance sweep and an excess cashflow sweep reduce a loan’s
credit quality via an increase in its credit spread by 70bp and 33bp, respectively. These
similarities are striking given that the loan sample using credit ratings versus the loan sample
using credit spreads are quite different based on a comparison of the summary statistics in Tables
I and VI.
However, the results in Table IX column 1 show that several additional covenants
improve a loan’s credit quality. A fixed cost coverage covenant reduces a loan’s credit spread by
26bp and an asset sales sweep and a debt issuance sweep reduce a loan’s credit spread by 22bp
and 50bp, respectively. Unlike the results for credit ratings, a leverage ratio covenant is shown to
actually worsen loan quality by 11bp. Yet since a leverage ratio covenant bears some similarity to
a net worth covenant which was detrimental to credit ratings, this result might not differ much.
Consistent with the results when using covenant intensity measures or individual
covenants as the dependent variables, we find that collateral, term loans, larger loans, and longer-
maturity loans are associated with lower loan spreads. Being a top 10 underwriter has no effects.
E. Effects of Covenants and Collateral on Term Loans versus Credit Lines
28
The next investigation compares the effects of covenants and collateral on the credit
spreads of term loans versus the credit spreads of revolving lines of credit by running regression
(8) on separate loan-type subsamples. The results are reported in Table IX’s column 2 (term
loans) and column 3 (revolvers). Almost all of the covenants that were significant for the full
sample continue to be significant and of the same sign and magnitude for the revolving line of
credit subsample. The only exception is that an asset sales sweep no longer reduces a revolver’s
credit spread but an insurance proceeds sweep does.
Perhaps because term loans represent less than 26% of the full sample, many of the
coefficients on covenants have similar signs to those of the full sample results but are no longer
statistically significant. An exception is the equity issuance sweep that continues to be detrimental
to a term loan’s credit quality. However, the effects of collateral, loan size, and maturity continue
to be significant and indicate that they are even more important for term loans compared to
revolvers.
It is noteworthy that, in general, a comparison of the results in Table IX to those in Table
IV show that the number of statistically significant covenants is greater for credit spreads
compared to credit ratings. This is especially true for revolving lines of credit. A possible
explanation might be linked to theory and empirical evidence showing that credit ratings reflect
actual (physical) expected default losses while credit spreads reflect risk-neutral expected default
losses. In other words, credit spreads contain possible systematic risk premiums while credit
ratings do not (Iannotta et al. (2019)). The greater overall sensitivity of credit spreads might be
due to covenants and collateral affecting not only expected default losses but also systematic
default risk premiums. This is consistent with credit line spreads having higher sensitivity to
covenants since credit lines are more likely to be drawn during economic downturns, making
them systematically risky (Pennacchi (2006)).
F. Effects of Covenants and Collateral by CDS Implied Firm Rating
29
Evidence discussed in Section III and reported in Table V showed that the impact of
covenants and collateral on loan ratings differed by a firm’s risk as proxied by its firm credit
rating. This section undertakes a similar examination of credit spreads based on subsamples of
loans categorized their issuing firm’s Markit “implied” credit rating. In order to categorize firms
in approximately equal subsamples, we divided the full sample of 3,748 observations into
subsamples of firms with implied ratings of A and Above (1,347 observations), BBB (1,070
observations), and BB and Below (1,331 observations). It is noteworthy that based on these
implied ratings, 64% of all observations are for investment-grade firms whereas investment-grade
firms made up only 10% of our credit rating sample observations.
Table X reports the results from running regression (8) where column 1 repeats the full
sample results and columns 2, 3, and 4 are results for the subsamples A and Above, BBB, and BB
and Below, respectively. The capital expenditure covenant reliably decreases loan spreads for the
full sample and all three implied firm rating subsamples by economically large amounts
exceeding 53bps. Also, an excess cashflow sweep significantly reduces loan credit quality,
especially for investment grade firms. But other than these two covenants, the sensitivity of credit
spreads to the presence of covenants differs by the category of implied firm rating.
Similar to the results found for credit ratings, a dividend restriction covenant improves
loan quality mainly for sub-investment grade firms, equal to a 23bp reduction in the loan’s
spread. Also, a debt issuance sweep and insurance proceeds sweep improves loan credit quality
the most for firms with an implied BBB rating, equal to 58bp and 52bp, respectively. In contrast,
an equity issuance sweep reduces loan quality the most for firms with an implied A rating or
better, equal to a rise in credit spread by 64bp. It seems very plausible that requiring a loan
repayment upon equity issuance would make little sense for high credit quality firms.
Regarding the effects of collateral, the results show that it is effective only for the sub-
investment grade category of BB and Below where it reduces loan spreads by 41bp. Indeed, it
appears that investors view collateral for loans whose firms have an implied rating of A and
30
Above with suspicion since these loans have spreads 10bp higher than without collateral. The
finding that collateral improves loan quality the most for the lowest quality firms is consistent the
earlier findings using credit ratings. Table X also shows that loan size and maturity are associated
with lower loan credit spreads across all implied firm rating categories.
G. Robustness: Financial Firms
Financial firms make up 12% of the firms in our sample, and the previous tests using
differences in these firms’ CDS spreads versus their loan spreads may already control for any
special characteristics and argue for keeping them in our sample. However, eliminating financial
firms reduces the number of observations from 3748 to 3288, but regression tests on this
subsample produces results similar to those of the full sample.
VI. Conclusion
This paper develops a new approach for quantifying the impact that collateral and
covenants have in reducing the risk of syndicated loans. It examined how these contract terms are
viewed both from a credit rating agency’s (Moody’s) and investors’ perspectives. Its
methodology addresses a firm’s self-selection of collateral and covenants by taking advantage of
the existence of risk measures for both the firm’s unsecured, covenant-free debt as well as its
syndicated loan that may be secured and contain various covenants. First, the paper analyzed how
a loan’s covenants and collateral affect the difference between a loan’s credit rating and the
generic senior, unsecured credit rating of its borrowing firm. Second, it examined how a loan’s
covenants and collateral affect the difference between the borrowing firm’s senior, unsecured
credit default swap (CDS) spread and its loan’s credit spread.
The credit rating agency and investors agree that collateral and some types of covenants
are important for reducing loan risk. In particular, both view collateral and covenants limiting
capital expenditures and dividends as particularly important for reducing the risk of loans.
Perhaps more surprising is that they also agree that other covenants, namely excess cashflow and
31
equity issuance sweeps, are detrimental to a loan’s credit quality, possibly because these
covenants require loan prepayment following loan-favorable events. Some disagreement exists on
other covenants. A fixed cost coverage covenant, an asset sales sweep, and a dividend issuance
sweep reduce a loan’s credit spread but not its credit rating. A leverage ratio covenant actually
increases a loan’s credit spread while a net worth covenant worsens a loan’s credit rating.
New to this literature, the paper also explored how the effects of covenants and collateral
differ by the type of loan and also the level of risk of the borrowing firm. In general, covenants
have a greater influence on the loan ratings and spreads of revolving lines of credit compared to
term loans. It also uncovers evidence that the sensitivity of loan ratings and credit spreads to
covenants and collateral differ by a firm’s risk as proxied by its firm rating. In particular,
collateral is most important in improving a loan’s credit rating and reducing its credit spread
when the borrowing firm has a high level of risk.
Overall, these findings quantify the importance of collateral and particular types of
covenants in improving syndicated loan ratings. They might be applied to construct a covenant
intensity index that best measures covenants’ improvement on credit quality. The results also can
guide loan contract design for the purpose of targeting a given credit rating or a given credit
spread.
32
Appendix A Variable Definition Variable Source Description Firm ratings Moody's Aaa to C, 21 notches. Moody's long term issuer rating. Long term
unsecured rating and long term corporate family ratings are used when not available
Loan rating Moody's Aaa to C, 21 notches. Moody's syndicated loan rating at origination Firm score Calculated Number of notches of firm rating, ranging from 21(Aaa) to 1 (C) Loan score Calculated Number of notches of loan rating, ranging from 21(Aaa) to 1 (C) Rating difference Calculated Dependent variable equal to loan score minus firm score Size Dealscan ln(Facility amount) Maturity Dealscan ln(months until maturity) Term Dealscan Indicator variable equals 1 if term loan Secured Dealscan Indicator variable equals 1 if loan is secured (has collateral) Asset sales sweep Dealscan Indicator variable equals 1 if asset sales sweep covenant exists Debt issuance sweep Dealscan Indicator variable equals 1 if debt issuance sweep covenant exists Equity issuance sweep Dealscan Indicator variable equals 1 if equity issuance sweep covenant exists Excess CF sweep Dealscan Indicator variable equals 1 if excess cash flow sweep covenant exists Insurance sweep Dealscan Indicator variable equals 1 if insurance proceeds sweep covenant exists Dividend restrictions Dealscan Indicator variable equals 1 if dividend restrictions covenant exists Debt to EBITDA Dealscan Indicator variable equals 1 if debt to EBITDA covenant exists Interest coverage Dealscan Indicator variable equals 1 if interest coverage covenant exists Capital Expenditure Dealscan Indicator variable equals 1 if capital expenditure covenant exists Fixed charge coverage Dealscan Indicator variable equals 1 if fixed charge coverage covenant exists Leverage ratio Dealscan Indicator variable equals 1 if leverage ratio covenant exists Net worth Dealscan Indicator variable equals 1 if net worth covenant exists Year Dealscan The year the loan is issued Purpose Dealscan Primary purpose field in the Facility table of Dealscan database Loan spread Dealscan All-in-drawn spread, in basis points CDS spread Markit CDS spread for contract maturity closest to loan maturity Spread difference Calculated Dependent variable equal to CDS spread minus Loan spread Top 10 underwriter Bloomberg Indicator variable equals 1 if the bank is among top 10 lead arrangers of
US loans over the past 3 years
33
Appendix B Correlations Between Covenants
Debt to EBITDA
Interest Cov. CapEx
Fixed Charge
Cov. Leverage Net
Worth
Asset Sales Sweep
Debt Issue Sweep
Equity Issue Sweep
Excess Cashflow
Sweep Insurance
Sweep Dividend Restrict
Debt to EBITDA 1.000 0.649 0.534 0.377 -0.140 0.130 0.684 0.657 0.500 0.571 0.617 0.668 Interest Cov. 0.538 1.000 0.476 0.082 0.049 0.069 0.505 0.478 0.369 0.436 0.444 0.506 CapEx 0.248 0.211 1.000 0.269 -0.074 0.064 0.638 0.633 0.603 0.568 0.582 0.571 Fixed Charge Cov. 0.121 -0.058 0.069 1.000 0.048 0.201 0.372 0.338 0.417 0.304 0.319 0.385 Leverage -0.280 -0.024 -0.099 0.011 1.000 0.081 -0.102 -0.094 -0.041 -0.112 -0.082 0.030 Net Worth -0.075 0.019 0.033 0.116 0.290 1.000 0.089 0.052 0.094 -0.020 0.033 0.180 Asset Sales Swp 0.344 0.257 0.341 0.155 -0.108 -0.030 1.000 0.878 0.663 0.728 0.832 0.796 Debt Issue Swp 0.307 0.209 0.347 0.138 -0.078 -0.009 0.770 1.000 0.675 0.727 0.800 0.724 Equity Issue Swp 0.177 0.106 0.203 0.153 -0.018 0.055 0.501 0.610 1.000 0.565 0.593 0.560 Excess Cash Swp 0.277 0.161 0.366 0.086 -0.124 -0.050 0.545 0.565 0.217 1.000 0.675 0.618 Insurance Swp 0.362 0.269 0.330 0.146 -0.149 -0.024 0.778 0.685 0.394 0.606 1.000 0.691 Dividend Restrict 0.402 0.332 0.257 0.217 0.060 0.084 0.410 0.348 0.197 0.292 0.433 1.000
The correlations in the upper triangle of the table are covenant correlations for the 3,558 sample of loans used in the analysis of credit ratings. The correlations in the lower triangle of the table are the covenant correlations for the 3,748 sample of loans used in the analysis of credit spreads.
34
References
Bao, J. and A. Kolasinski, 2017, Why Do Firms Issue Secured Debt?, University of Delaware and Texas A&M
University working paper.
Begley, J., 1994, Restrictive Covenants Included in Public Debt Agreements: An Empirical Investigation,
University of British Columbia working paper.
Berg, T., A. Saunders, and S. Steffen, 2016, The Total Cost of Corporate Borrowing in the Loan Market: Don't
Ignore the Fees, Journal of Finance 71, 1357-1392.
Berger, A. N. and G. F. Udell, 1990, Collateral, Loan Quality, and Bank Risk, Journal of Monetary Economics
25, 21-42.
Berger, A. N., and Udell, G. F. 1995, Relationship Lending and Lines of Credit in Small Firm Finance, Journal of
Business 68, 351-381.
Berlin, M., and L. Mester, 1992, Debt Covenants and Renegotiation, Journal of Financial Intermediation 2, 95-
133.
Besanko, D. and A. Thakor, 1987, Collateral and Rationing: Sorting Equilibria in Monopolistic and Competitive
Credit Markets, International Economic Review 28, 671-689.
Bester, H., 1985, Screening versus Rationing in Credit Markets with Imperfect Information, American Economic
Review 75:850–55.
Bester, H., 1994, The Role of Collateral in a Model of Debt Renegotiation, Journal of Money, Credit and Banking
26, 72-86.
Billett, M., T-H. King, and D. Mauer, 2007, Growth Opportunities and the Choice of Leverage, Debt Maturity,
and Covenants, Journal of Finance 62, 697-730.
Bomfim, A., 2016, Understanding Credit Derivatives and Related Instruments, 2nd Ed., Academic Press,
Waltham, MA.
Boot, A, A. Thakor, and G. Udell, 1991, Secured Lending and Default Risk: Equilibrium Analysis, Policy
Implications, and Empirical Results, Economic Journal 101:458–72.
35
Booth, J. R. 1992, Contract Costs, Bank Loans, and the Cross-monitoring Hypothesis, Journal of Financial
Economics 31:25–41.
Bradley, M., and M. R. Roberts, 2015, The Structure and Pricing of Corporate Debt Covenants, Quarterly Journal
of Finance 5, 1-37.
Chan, Y., and Kanatas, G. 1985, Asymmetric Valuation and the Role of Collateral in Loan Agreements, Journal
of Money, Credit and Banking 17:85–95.
Coffey, M., 2005, Opportunity Knocks... and Borrowers Answer, LSTA Loan Market Chronicle, 39-44.
Demiroglu, C., and C. James, 2010, The Information Content of Bank Loan Covenants, Review of Financial
Studies 23, 3700–3737.
Dichev, I. and D. Skinner, 2002, Large-Sample Evidence on the Debt Covenant Hypothesis, Journal of
Accounting Research 40, 1091-1123.
Drucker, S. and M. Puri, 2009, On Loan Sales, Loan Contracting, and Lending Relationships, Review of Financial
Studies 22, 2835-2872.
Franz, D., H. HassabElnaby, and G. Lobo, 2014, Impact of Proximity to Debt Covenant Violation on Earnings
Management, Review of Accounting Studies 19, 473-505.
Gârleanu, N. and J. Zwiebel, 2009, Design and Renegotiation of Debt Covenants, Review of Financial Studies 22,
749-781.
Healy, P. and K. Palepu, 1990, Effectiveness of Accounting-Based Dividend Covenants, Journal of Accounting
and Economics 12, 97-123.
Heckman, J., 1979, Sample Selection Bias as a Specification Error, Econometrica 47, 153-161.
Iannotta, G., G. Pennacchi, and J. Santos, 2019, Ratings-Based Regulation and Systematic Risk Incentives,
Review of Financial Studies 32, 1374-1415.
John, K., A. Lynch, and M. Puri, 2003, Credit Ratings, Collateral and Loan Characteristics: Implications for
Yield, Journal of Business 76, 371-410.
Khieu, H., D. Mullineaux, and H-C. Yi, 2012, The Determinants of Bank Loan Recovery Rates, Journal of
Banking & Finance 36, 923-933.
36
Li, K. and N. Prabhala, 2007, Self-Selection Models in Corporate Finance, Chapter 2 in Handbook of Empirical
Corporate Finance 1, Elsevier, Amsterdam.
Livingston, M. and R. E. Miller, 2000, Investment Bank Reputation and the Underwriting of Nonconvertible
Debt, Financial Management 29(2), 21-34.
Malitz, I., 1986, On Financial Contracting: The Determinants of Bond Covenants, Financial Management 15(2),
18-25.
Matvos, G., 2013, Estimating the Benefits of Contractual Completeness, Review of Financial Studies 26, 2798-
2844.
Moody’s Investor Service, 2003, Relative Default Rates on Corporate Loans and Bonds, Special Comment
(September).
Moody’s Investor Service, 2004a, Characteristics and Performance of Moody’s-Rated U.S. Syndicated Bank
Loans, Special Comment (March).
Moody’s Investor Service, 2004b, Recent Bank Loan Research: Implications for Moody’s Bank Loan Rating
Practices, Special Comment (December).
Murfin, J. ,2012, The Supply-Side Determinants of Loan Contract Strictness, Journal of Finance 67, 1565–1601.
Narayanan, R. P., K. P. Rangan, and N. K. Rangan, 2007, The Effect of Private-Debt-Underwriting Reputation on
Bank Public-Debt Underwriting, Review of Financial Studies 20, 597-618.
Nichols, C., 2016, Financial Covenants All Commercial Lenders Must Understand, CenterState Bank
Correspondent Division, available at https://csbcorrespondent.com/blog/financial-covenants-all-
commercial-lenders-must-understand .
Nini, G., D. Smith, and A. Sufi, 2009, Creditor Control Rights and Firm Investment Policy, Journal of Financial
Economics 92, 400-420.
Oehmke, M. and A. Zawadowski, 2017, The Anatomy of the CDS Market, Review of Financial Studies 30, 80-
119.
Park, C., 2000, Monitoring and Debt Seniority Structure, Journal of Finance 55, 2157–2195.
37
Pennacchi, G., 2006, Deposit Insurance, Bank Regulation, and Financial System Risks, Journal of Monetary
Economics 53, 1-30.
Rajan, R. and A. Winton, 1995, Covenants and Collateral as Incentives to Monitor, Journal of Finance 50, 1113–
1146.
Standard & Poor’s, 2011, A Guide to the Loan Market, available at
https://www.lcdcomps.com/d/pdf/LoanMarketguide.pdf.
Standard & Poor’s Financial Services, 2013, Primer, available at www.leveragedloan.com/primer.
Stulz, R. and H. Johnson, 1985, An Analysis of Secured Debt, Journal of Financial Economics 14, 501-52
38
Figure 1 Sample Distribution of Firm and Loan Ratings
Figure 2 Distribution of Rating Differences (Loan Rating - Firm Rating, in Notches)
39
Figure 3 Difference in Loan – Firm Ratings by Firm Rating
Figure 4 Proportions of Loans with Collateral and Various Covenants By Firm Rating
40
Figure 5 Distribution of Spread Differences (CDS Spread – Loan Spread, in Basis Points)
0
2
4
6
8
10
12
-300
-260
-220
-190
-170
-150
-130
-110 -9
0-7
0-5
0-3
0-1
0 10 30 50 70 90 110
130
150
170
190
220
260
300
340
380
420
460
500
Perc
ent o
f Obs
erva
tions
CDS Spread - Loan Spread (Basis Points)
41
Table I Summary Statistics of Credit Rating Sample This table provides summary statistics for ratings scores, collateral, covenants, and other loan characteristics. For indicator variables, mean values in column 2 show the prevalence of this characteristic in the dataset. The sample is based on syndicated loans originated from 1995 to 2012. Variables are defined in Appendix A. (1) (2) (3) (4) (5) Obs. Mean Std. Dev Min Max Firm Score 3,558 7.978 2.502 2 21 Loan Score 3,558 9.010 2.216 3 21 Rating Difference 3,558 1.031 1.318 -3 8 Term Loan 3,558 0.527 0.499 0 1 Loan Amount ($ millions) 3,558 365.3 586.8 1 10700 Maturity (years) 3,558 5.40 1.50 0.25 12 Secured 3,558 0.871 0.335 0 1 Capital Expenditure 3,558 0.251 0.434 0 1 Debt to EBITDA 3,558 0.478 0.500 0 1 Fixed Charge Coverage 3,558 0.194 0.395 0 1 Interest Coverage 3,558 0.363 0.481 0 1 Leverage Ratio 3,558 0.0514 0.221 0 1 Net Worth 3,558 0.0618 0.241 0 1 Dividend Restrictions 3,558 0.468 0.499 0 1 Asset Sales Sweep 3,558 0.403 0.491 0 1 Debt Issuance Sweep 3,558 0.356 0.479 0 1 Equity Issuance Sweep 3,558 0.243 0.429 0 1 Excess Cash Flow Sweep 3,558 0.367 0.482 0 1 Insurance Proceeds Sweep 3,558 0.328 0.470 0 1 Top 10 Underwriter 3,558 0.741 0.438 0 1
42
Table II Effects of Covenant Intensity and Collateral on Credit Ratings This table provides results of a linear regression of the rating difference on four covenant strictness measures, collateral status, and other control variables. Columns 1 to 3 use total covenants, total non-financial covenants, and total financial covenants as measures. Columns 4 and 5 provide results using the Bradley and Roberts (2015) covenant intensity index as the measure, with and without collateral. Variables are defined in Appendix A. t-statistics are in parentheses. ***, **, and * denote significance at the 1%, 5%, and 10% level, respectively.
(1) (2) (3) (4) (5)
Total Covenants
Total Non-financial
Covenants
Total Financial
Covenants
Covenant Intensity
Index
Covenant Intensity
Index
Measure 0.0220*** 0.0329*** 0.0411*** 0.0411*** 0.107***
(3.477) (3.383) (2.872) (3.478) (9.235) Secured 1.427*** 1.421*** 1.458*** 1.386***
(17.46) (17.26) (18.08) (16.30) Term -0.292*** -0.294*** -0.290*** -0.292*** -0.214***
(-6.668) (-6.722) (-6.620) (-6.686) (-4.749) Size 0.215*** 0.216*** 0.218*** 0.215*** 0.159***
(10.06) (10.08) (10.22) (10.04) (7.230) Maturity -0.135** -0.135** -0.130** -0.133** 0.0884
(-2.080) (-2.078) (-2.013) (-2.047) (1.344) Top 10 Underwriter 0.193*** 0.193*** 0.191*** 0.193*** 0.143***
(3.719) (3.718) (3.689) (3.718) (2.655)
Loan Purpose FE YES YES YES YES YES Year FE YES YES YES YES YES Observations 3,558 3,558 3,558 3,558 3,558 Adjusted R-squared 0.217 0.217 0.216 0.217 0.155
43
Table III Effects of Individual Covenants on Credit Ratings
This table provides results of a linear regression of the rating difference on covenants individually, collateral, loan type and control variables. Columns 1 to 7 report the 7 out of 12 covenants that show significant effects. Variables are defined in Appendix A. t-statistics are in parentheses. ***, **, and * denote significance at the 1%, 5%, and 10% level, respectively.
(1) (2) (3) (4) (5) (6) (7)
Min. Interest Coverage
Max. Debt to EBITDA
Max. Capital Expenditure
Insurance Proceeds Sweep
Asset Sales Sweep
Debt Issuance Sweep
Dividend Restrictions
Covenant 0.126*** 0.157*** 0.205*** 0.151*** 0.181*** 0.143*** 0.242***
(2.689) (3.401) (3.790) (3.085) (3.733) (2.917) (5.194) Secured 1.468*** 1.447*** 1.427*** 1.431*** 1.413*** 1.434*** 1.429***
(18.27) (17.90) (17.50) (17.44) (17.16) (17.49) (17.73) Term -0.291*** -0.289*** -0.292*** -0.294*** -0.294*** -0.294*** -0.287***
(-6.651) (-6.598) (-6.682) (-6.711) (-6.734) (-6.717) (-6.577) Size 0.215*** 0.211*** 0.226*** 0.215*** 0.213*** 0.217*** 0.211***
(10.01) (9.785) (10.64) (10.06) (9.908) (10.11) (9.871) Maturity -0.131** -0.139** -0.131** -0.133** -0.135** -0.135** -0.132**
(-2.020) (-2.139) (-2.021) (-2.055) (-2.084) (-2.077) (-2.036) Top 10 Underwriter 0.185*** 0.188*** 0.185*** 0.194*** 0.192*** 0.190*** 0.190***
(3.573) (3.618) (3.565) (3.747) (3.708) (3.660) (3.674)
Loan Purpose FE YES YES YES YES YES YES YES Year FE YES YES YES YES YES YES YES Observations 3,558 3,558 3,558 3,558 3,558 3,558 3,558 Adjusted R-squared 0.216 0.217 0.218 0.216 0.217 0.216 0.221
44
Table IV Joint Effects of Covenants and Collateral on Credit Ratings
This table provides results of a linear regression of the rating difference on all covenants, collateral, and control variables. Column 1 provides results of the full sample of loans with and a term loan indicator. Column 2 provides results of the subsample of only term loans. Column 3 provides results of the subsample of only revolving lines of credit. Variables are defined in Appendix A. t-statistics are in parentheses. ***, **, and * denote significance at the 1%, 5%, and 10% level, respectively.
(1) (2) (3) Full Sample Term Loan Only Revolver Only Capital Expenditure 0.178** 0.113 0.260***
(2.509) (1.032) (2.806) Debt to EBITDA 0.0321 0.00603 -0.0130
(0.444) (0.0514) (-0.147) Fixed Cost Coverage 0.0485 0.0469 0.0660
(0.736) (0.447) (0.791) Interest Coverage 0.0112 0.0429 -0.00810
(0.178) (0.421) (-0.104) Leverage Ratio -0.000268 -0.0257 -0.0359
(-0.00252) (-0.115) (-0.317) Net Worth -0.229** -0.280 -0.179*
(-2.429) (-1.585) (-1.698) Dividend Restriction 0.264*** 0.295** 0.242***
(3.452) (2.093) (2.749) Asset Sales Sweep 0.0769 -0.0765 0.216
(0.684) (-0.402) (1.573) Debt Issuance Sweep -0.0256 0.0124 -0.0485
(-0.256) (0.0782) (-0.378) Equity Issuance Sweep -0.214*** -0.0996 -0.374***
(-2.876) (-0.877) (-3.786) Excess Cashflow Sweep -0.133* -0.0935 -0.171*
(-1.915) (-0.904) (-1.824) Insurance Proceeds Sweep -0.00344 0.0468 -0.0391
(-0.0407) (0.358) (-0.357) Secured 1.430*** 1.314*** 1.351***
(16.93) (6.641) (15.04) Term -0.280***
(-6.404) Size 0.208*** 0.284*** 0.115***
(9.449) (8.375) (4.028) Maturity -0.136** -0.326*** 0.0318
(-2.104) (-2.667) (0.431) Top 10 Underwriter 0.181*** 0.215*** 0.146**
(3.476) (2.634) (2.211)
Loan Purpose FE YES YES YES Year FE YES YES YES Observations 3,558 1,875 1,683 Adjusted R-squared 0.280 0.135 0.325
45
Table V Credit Ratings: Differences in Covenants and Collateral by Firm Ratings
This table provides results of a linear regression of the rating difference on all covenants, collateral, and control variables. Column 1 provides results of the full sample of loans. Columns 2, 3, 4, and 5 provide results for the subsamples of loans where the issuing firms’ rating is investment grade, Ba, B, and C, respectfully. Variables are defined in Appendix A. t-statistics are in parentheses. ***, **, and * denote significance at the 1%, 5%, and 10% level, respectively. (1) (2) (3) (4) (5)
Full Sample Invest Grade Ba Rated B Rated C Rated Capital Expenditure 0.178** 0.433** 0.109 -0.0450 -0.228
(2.509) (2.105) (1.074) (-0.502) (-0.992) Debt to EBITDA 0.0321 0.0950 0.0372 0.0582 -0.101
(0.444) (1.291) (0.297) (0.598) (-0.571) Fixed Cost Coverage 0.0485 -0.00643 -0.00216 0.0739 0.404*
(0.736) (-0.0740) (-0.0206) (0.861) (1.889) Interest Coverage 0.0112 -0.111* 0.0967 0.160* 0.303*
(0.178) (-1.783) (1.002) (1.825) (1.660) Leverage Ratio -0.000268 0.129** 0.233 0.0145 1.076**
(-0.00252) (2.155) (1.326) (0.0688) (2.318) Net Worth -0.229** 0.108 0.00192 -0.374** 1.138*
(-2.429) (1.303) (0.0171) (-2.348) (1.912) Dividend Restriction 0.264*** -0.0953 0.251** 0.408*** 0.155
(3.452) (-1.506) (2.047) (3.608) (0.672) Asset Sales Sweep 0.0769 -0.00419 0.200 -0.302* 0.0919
(0.684) (-0.0267) (1.236) (-1.848) (0.291) Debt Issuance Sweep -0.0256 -0.682*** -0.253* 0.0698 -0.360
(-0.256) (-4.047) (-1.947) (0.509) (-0.954) Equity Issuance Sweep -0.214*** 0.346** -0.0515 -0.209** 0.538**
(-2.876) (2.181) (-0.474) (-2.202) (2.239) Excess Cashflow Sweep -0.133* -0.817*** -0.652*** -0.136 0.0954
(-1.915) (-3.686) (-6.101) (-1.579) (0.392) Insurance Proceeds Sweep -0.00344 -0.465** 0.163 0.228** 0.118
(-0.0407) (-2.399) (1.252) (2.178) (0.353) Secured 1.430*** 0.576*** 0.674*** 0.573* 3.844***
(16.93) (5.285) (5.350) (1.761) (6.504) Term -0.280*** -0.0282 -0.149** -0.297*** -0.195*
(-6.404) (-0.397) (-2.163) (-5.595) (-1.717) Size 0.208*** 0.00822 0.0500 0.310*** 0.175***
(9.449) (0.318) (1.239) (10.98) (3.131) Maturity -0.136** 0.00642 0.0454 -0.454*** -0.533*
(-2.104) (0.175) (0.451) (-4.225) (-1.907) Top 10 Underwriter 0.181*** 0.0443 0.00290 0.133** 0.627***
(3.476) (0.688) (0.0309) (2.186) (3.442)
Loan Purpose FE YES YES YES YES YES Year FE YES YES YES YES YES Observations 3,558 369 660 2,189 340 Adjusted R-squared 0.223 0.208 0.381 0.226 0.483
46
Table VI Summary Statistics of Credit Spread Sample
This table provides summary statistics for spreads (in basis points), collateral, covenants, and other loan characteristics. For indicator variables, mean values in column 2 show the prevalence of this characteristic in the dataset. The sample is based on syndicated loan market from 2002 to 2015. Variables are defined in Appendix A. (1) (2) (3) (4) (5) Obs. Mean Std. Dev Min Max CDS Spread 3,748 141 164 1.53 1160 Loan Spread (all-in-drawn) 3,748 138.5 104.9 1.750 850 Spread Difference 3,748 2.095 112.1 -487.6 491.5 Term Loan 3,748 0.246 0.431 0 1 Loan Amount ($ Millions) 3,746 1182 1689 8.9 30000 Maturity (years) 3,748 4.011 1.791 0.083 18 Secured 3,748 0.617 0.486 0 1 Capital Expenditure 3,748 0.0395 0.195 0 1 Debt to EBITDA 3,748 0.283 0.450 0 1 Fixed Charge Coverage 3,748 0.101 0.302 0 1 Interest Coverage 3,748 0.244 0.429 0 1 Leverage Ratio 3,748 0.228 0.419 0 1 Net Worth 3,748 0.0894 0.285 0 1 Dividend Restrictions 3,748 0.294 0.456 0 1 Asset Sales Sweep 3,748 0.135 0.342 0 1 Debt Issuance Sweep 3,748 0.118 0.322 0 1 Equity Issuance Sweep 3,748 0.0648 0.246 0 1 Excess Cash Flow Sweep 3,748 0.0547 0.227 0 1 Insurance Proceeds Sweep 3,748 0.0912 0.288 0 1 Top 10 Underwriter 3,748 0.934 0.248 0 1
47
Table VII Effects of Covenant Intensity and Collateral on Credit Spreads This table provides results of a linear regression of the difference in the CDS spread - loan spread on four covenant strictness measures, collateral status, and other control variables. Columns 1 to 3 use total covenants, total non-financial covenants, and total financial covenants as measures. Columns 4 and 5 provide results using Bradley and Roberts (2015) covenant intensity index as the measure, with and without collateral. Variables are defined in Appendix A. t-statistics are in parentheses. ***, **, and * denote significance at the 1%, 5%, and 10% level, respectively. (1) (2) (3) (4) (5)
Total
Covenants
Total Non-financial
Covenants
Total Financial
Covenants
Covenant Intensity
Index
Covenant Intensity
Index Measure 7.356*** 10.13*** 11.02*** 12.59*** 14.83*** (8.271) (5.895) (8.016) (8.577) (10.62) Secured 31.72*** 34.62*** 26.39*** 17.90*** (8.902) (9.301) (7.449) (4.804) Term 23.70*** 27.87*** 23.16*** 23.08*** 24.68*** (5.289) (6.266) (5.144) (5.144) (5.499) Size 12.07*** 12.85*** 11.37*** 11.93*** 12.02*** (7.033) (7.444) (6.609) (6.954) (6.984) Maturity 38.74*** 39.67*** 40.13*** 39.64*** 39.74*** (13.66) (13.89) (14.26) (14.08) (14.08) Top 10 Underwriter 4.813 4.704 5.463 4.820 3.889 (0.664) (0.646) (0.754) (0.666) (0.536) Loan Purpose FE YES YES YES YES YES Year FE YES YES YES YES YES Observations 3,748 3,748 3,748 3,748 3,748 Adjusted R2 0.217 0.210 0.217 0.219 0.214
48
Table VIII Effects of Individual Covenants and Collateral on Credit Spreads
This table provides results of a linear regression of the difference in the CDS spread - loan spread on covenants individually, collateral, loan type and control variables. Columns 1 to 7 report the 7 out of 12 covenants that show significant effects. Variables are defined in Appendix A. t-statistics are in parentheses. ***, **, and * denote significance at the 1%, 5%, and 10% level, respectively. (1) (2) (3) (4) (5) (6) (7)
Max. Capital
Expenditure
Min. Fixed Charge
Coverage
Dividend Restrictions
Asset Sales
Sweep
Debt Issuance Sweep
Excess Cash Flow
Sweep
Insurance Proceeds Sweep
Covenant 75.22*** 30.59*** 30.32*** 43.54*** 46.05*** 36.21*** 43.22*** (8.231) (5.378) (7.486) (7.753) (7.578) (4.295) (6.556) Secured 24.53*** 28.52*** 30.66*** 24.76*** 26.63*** 25.93*** 24.42*** (6.887) (8.022) (8.616) (6.947) (7.513) (7.218) (6.801) Term 28.91*** 29.51*** 27.47*** 23.53*** 24.03*** 27.65*** 25.91*** (6.555) (6.655) (6.198) (5.228) (5.346) (6.166) (5.788) Size 12.60*** 12.70*** 12.66*** 11.31*** 10.79*** 11.70*** 11.76*** (7.344) (7.356) (7.363) (6.565) (6.234) (6.749) (6.821) Maturity 40.20*** 41.80*** 39.27*** 40.76*** 42.33*** 41.04*** 40.47*** (14.30) (14.85) (13.84) (14.52) (15.13) (14.44) (14.32) Top 10 Underwriter 6.598 5.188 5.409 5.348 5.473 5.594 6.599 (0.911) (0.713) (0.746) (0.738) (0.755) (0.767) (0.908) Loan Purpose FE YES YES YES YES YES YES YES Year FE YES YES YES YES YES YES YES Observations 3,748 3,748 3,748 3,748 3,748 3,748 3,748 Adjusted R2 0.217 0.209 0.215 0.216 0.215 0.207 0.212
49
Table IX Joint Effects of Covenants and Collateral on Credit Spreads This table provides results of a linear regression of the difference in the CDS spread - loan spread on covenants jointly, collateral, loan type and control variables. Column 1 provides results using the full sample. Column 2 provides results of the subsample of only term loans. Column 3 provides results of the subsample of only revolving lines of credit. Variables are defined in Appendix A. t-statistics are in parentheses. ***, **, and * denote significance at the 1%, 5%, and 10% level, respectively. (1) (2) (3)
Full Sample Term Loan Only Revolver Only Capital Expenditure 58.11*** 32.34 63.65***
(5.899) (1.531) (5.694) Debt to EBITDA -2.570 6.247 -8.169
(-0.503) (0.424) (-1.587) Fixed Cost Coverage 25.89*** 6.420 28.82***
(4.452) (0.456) (4.704) Interest Coverage 2.482 -17.26 6.704
(0.517) (-1.241) (1.390) Leverage Ratio -11.36** 11.74 -12.96***
(-2.385) (0.698) (-2.836) Net Worth 1.753 9.658 -3.586
(0.285) (0.405) (-0.618) Dividend Restriction 15.74*** 20.92 14.45***
(3.396) (1.579) (3.134) Asset Sales Sweep 21.86** 28.48 13.30
(2.331) (1.630) (1.136) Debt Issuance Sweep 50.44*** 29.41* 60.51***
(5.078) (1.647) (4.653) Equity Issuance Sweep -69.54*** -48.66*** -75.97***
(-7.621) (-2.907) (-6.359) Excess Cashflow Sweep -33.40*** -17.51 -46.56***
(-3.118) (-0.952) (-2.974) Insurance Proceeds Sweep 1.563 -27.66 29.18**
(0.151) (-1.534) (2.010) Secured 21.42*** 25.59** 16.85***
(5.281) (2.051) (4.129) Term 23.90***
(5.373) Size 12.09*** 25.88*** 6.694***
(7.045) (6.078) (3.721) Maturity 35.63*** 64.25*** 23.78***
(12.46) (9.037) (7.788) Top 10 Underwriter 5.899 20.58 3.599
(0.826) (1.295) (0.453)
Loan Purpose FE YES YES YES Year FE YES YES YES Observations 3,748 921 2,827 Adjusted R-squared 0.244 0.354 0.184
50
Table X Effects of Covenants and Collateral on Credit Spreads By Implied Credit Rating
This table provides results of a linear regression of the rating difference on all covenants, collateral, and control variables. Column 1 provides results of the full sample of loans. Columns 2, 3, and 4 provide results for the subsamples of loans where the issuing firms’ Markit-implied rating is A and above, BBB, or BB and below, respectfully. Variables are defined in Appendix A. t-statistics are in parentheses. ***, **, and * denote significance at the 1%, 5%, and 10% level, respectively. (1) (2) (3) (4)
Full Sample A and Above BBB BB and Below Capital Expenditure 58.11*** 53.25*** 62.48*** 68.25***
(5.899) (2.585) (3.458) (4.635) Debt to EBITDA -2.570 -15.63*** 1.391 -8.992
(-0.503) (-2.965) (0.219) (-0.820) Fixed Cost Coverage 25.89*** 7.230 -7.669 11.39
(4.452) (1.144) (-0.787) (1.107) Interest Coverage 2.482 5.490 -11.68* -8.397
(0.517) (1.209) (-1.915) (-0.780) Leverage Ratio -11.36** -3.041 -12.74** -0.674
(-2.385) (-0.756) (-2.281) (-0.0522) Net Worth 1.753 3.631 18.73** -32.82**
(0.285) (0.628) (2.569) (-2.232) Dividend Restriction 15.74*** -9.622** 4.944 22.53**
(3.396) (-2.164) (0.807) (2.356) Asset Sales Sweep 21.86** -24.06* -54.06*** 14.63
(2.331) (-1.857) (-2.785) (0.967) Debt Issuance Sweep 50.44*** 20.58 58.40*** 20.98
(5.078) (1.325) (2.912) (1.345) Equity Issuance Sweep -69.54*** -63.74*** -0.412 -29.15*
(-7.621) (-4.250) (-0.0258) (-1.838) Excess Cashflow Sweep -33.40*** -93.24*** -135.8*** -41.13***
(-3.118) (-3.269) (-6.045) (-2.630) Insurance Proceeds Sweep 1.563 41.18** 51.94*** -11.70
(0.151) (2.086) (3.027) (-0.691) Secured 21.42*** -9.958*** -8.658* 41.17***
(5.281) (-2.737) (-1.685) (4.300) Term 23.90*** 5.915 -4.024 -0.228
(5.373) (1.104) (-0.611) (-0.0292) Size 12.09*** 14.21*** 9.426*** 12.36***
(7.045) (8.992) (3.950) (3.307) Maturity 35.63*** 9.788*** 17.98*** 65.27***
(12.46) (4.002) (4.865) (9.274) Top 10 Underwriter 5.899 -10.93 14.37 4.811
(0.826) (-1.501) (1.551) (0.331)
Loan Purpose FE YES YES YES YES Year FE YES YES YES YES Observations 3,748 1,347 1,070 1,331 Adjusted R-squared 0.244 0.397 0.253 0.313