Journal of Banking & Finance 37 (2013) 5511–5525

Contents lists available at ScienceDirect

Journal of Banking & Finance

journal homepage: www.elsevier.com/locate/jbf

Credit and liquidity components of corporate CDS spreads q

Filippo Corò, Alfonso Dufour, Simone Varotto ⇑
ICMA Centre, Henley Business School, University of Reading, UK

a r t i c l e i n f o a b s t r a c t

Article history: This paper investigates the role of credit and liquidity factors in explaining corporate CDS price changes
Available online 17 July 2013 during normal and crisis periods. We find that liquidity risk is more important than firm-specific credit
risk regardless of market conditions. Moreover, in the period prior to the recent ‘‘Great Recession’’ credit
JEL classification: risk plays no role in explaining CDS price changes. The dominance of liquidity effects casts serious doubts
G01 on the relevance of CDS price changes as an indicator of default risk dynamics. Our results show how
G11 multiple liquidity factors including firm specific and aggregate liquidity proxies as well as an asymmetric
G15
information measure are critical determinants of CDS price variations. In particular, the impact of
G32
informed traders on the CDS price increases when markets are characterised by higher uncertainty,
Keywords: which supports concerns of insider trading during the crisis.
CDS Ó 2013 Elsevier B.V. All rights reserved.
Liquidity
Credit risk
Financial crisis
Informed trading
Trade impact

1. Introduction
Early work on credit default swaps (CDSs) often assumes that
the CDS price is driven only by the credit risk of the reference en-
tity (e.g. Longstaff et al., 2005; Blanco et al., 2005). However, the
2007–2009 financial crisis has dramatically highlighted the effect
of illiquidity on asset prices. Chen et al. (2005) are among the first
to address the interplay between credit and liquidity risk in the
CDS market. Tang and Yan (2007) use panel regressions to show
that a set of liquidity factors helps explain CDS spreads. They also
employ the liquidity adjusted CAPM of Acharya and Pedersen
(2005) and estimate a liquidity premium similar to that observed
in the Treasury and corporate bond markets. Bongaerts et al.
(2011) provide a general equilibrium pricing model for CDS con-
tracts that extends the liquidity adjusted CAPM of Acharya and
Pedersen (2005). They find that illiquidity can push CDS prices up-
wards, evidence that credit protection sellers command a liquidity
premium. Other relevant works are Buhler and Trapp (2008) and
Chen et al. (2007) who develop closed-form pricing equations for
q
An earlier version of this paper was circulated under the title ‘‘The time varying
properties of credit and liquidity components of CDS spreads’’.
⇑ Corresponding author. Address: ICMA Centre, Henley Business School, Univer-
sity of Reading, Whiteknights Park, Reading RG6 6UR, UK. Tel.: +44 (0)118 378
6655; fax: +44 (0)118 931 4741.
E-mail addresses: coro.filippo@gmail.com (F. Corò), a.dufour@icmacentre.ac.uk
(A. Dufour), s.varotto@icmacentre.ac.uk (S. Varotto).
CDSs that incorporate credit and liquidity risk components. Our
study contributes to the existing literature by providing answers
to the following questions:
 How does the impact of credit and liquidity risk on CDS price
movements vary during tranquil and turbulent periods? If the
role of the two risks in explaining CDS price dynamics changes
over time, then interpreting the CDS spread as an indicator of
default risk, which is common in the industry, may not be
meaningful. For example, if illiquidity increases substantially
during a crisis then a corresponding rise in the CDS prices
may not necessarily reflect higher default risk.
 Does the presence of informed traders have a significant impact
on price formation in the CDS market? If so, how did informed
trading impact on CDS prices in the course of the 2007–2009
financial crisis? This is particularly relevant given the criticism
that CDSs have attracted in the light of anecdotal evidence of
insider trading (see Acharya and Johnson, 2007) and wide-
spread concerns about the destabilising effect of speculation
in the CDS market, which was a prominent factor in the demise
of Lehman Brothers and AIG (Tang and Yan, 2010).
 To what extent are CDS price changes influenced by industry-
wide liquidity and firm-specific liquidity? For example, during
the recent crisis illiquidity appeared to be widespread across
the whole banking sector. This raises the question of the impor-
tance of commonality of liquidity effects on CDS spreads, and its
implication for CDS pricing.

0378-4266/$ - see front matter Ó 2013 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.jbankfin.2013.07.010
5512 F. Corò et al. / Journal of Banking & Finance 37 (2013) 5511–5525
Our analysis is three-pronged. First, we study the presence and
relevance of asymmetric information in the CDS market. Acharya
and Johnson (2007) consider this issue but, surprisingly, conclude
that the degree of asymmetric information, which they proxy with
the number of a firm’s bank relationships, does not appear to influ-
ence CDS prices. They suggest ‘‘using intraday data on actual trans-
actions in the CDS market’’ to derive a more accurate proxy of
information asymmetries. We follow their suggestion and reach
opposite conclusions. With intraday data, we estimate asymmetric
information through the persistent price impact of order flow on
CDS prices. This is a measure of the magnitude of informed trading
based on the analysis of Hasbrouck (1991) in the stock market.1
Our results show that order flow has a significant, positive and time-
varying impact on CDS prices, which is consistent with the
implications of theoretical asymmetric information models. To our
knowledge, ours is the first empirical work that investigates the
influence of informed trading on CDS prices using a microstructure
approach. Tang and Yan (2007) use the ‘‘probability of informed trad-
ing’’ or PIN developed by Easley et al. (1996) to look at the sensitivity of
liquidity variables to changes in the level of asymmetric information
in the CDS market. However, they do not attempt to measure the
direct contribution of informed trading to CDS price variations.
In the second part of our study we provide an investigation of
the determinants of CDS price changes. While controlling for
firm-specific credit risk and macro-economic factors, we look at
whether CDS price dynamics are related to firm-specific and indus-
try-wide illiquidity. Several authors have investigated firm-specific
illiquidity effects in CDS prices (Acharya and Johnson, 2007; Tang
and Yan, 2007; Pires et al., 2010). However, an analysis of indus-
try-wide illiquidity in the CDS market is new to this study. The
motivation for considering industry-wide liquidity stems from
the work of Chordia et al. (2000) and Hasbrouck and Seppi
(2001) who find common liquidity effects in the equity market.
We use two industry-wide liquidity proxies: the average bid-ask
spread and an asymmetric information measure. In addition to the
widely used bid-ask spread we consider the latter liquidity proxy,
estimated as the average permanent price impact of CDS trades in a
given industry, as we exploit the original intuition in Chordia et al.
(2000) who suggest that asymmetric information may exhibit
commonality across firms in the same industry. Our results show
that there is a significant liquidity effect on CDS price movements
and a large portion of the CDS variability is explained by both firm-
specific and industry-wide liquidity variables. Interestingly, it ap-
pears that asymmetric information at the industry level plays an
important role in the CDS price formation during crisis periods
but is not significant when markets are stable.
Finally, we look at how liquidity and credit effects vary in tran-
quil and turbulent markets. We find that, before the ‘‘Great Reces-
sion’’, credit effects on CDS price changes are unimportant, which
confirms the conjecture that the perception of credit risk was arti-
ficially low during the credit boom preceding the crisis (Acharya
et al., 2009, p. 13). Credit effects, however, become highly statisti-
cally significant during the crisis. Strikingly, the explanatory power
of the liquidity measures is greater than that of the firm-specific
credit measures in both tranquil and crisis periods.
The paper is organised as follows. In Section 2 we describe the
data. In Section 3 we present our methodology to estimate (i) the
permanent trade impact induced by information asymmetries in
the CDS market and (ii) the determinants of CDS price changes.
In Section 4 we describe our findings. Robustness tests are pre-
sented in Section 5. Section 6 concludes the paper.
1
Several other authors have looked at the price impact of informed traders in the
stock market including Kyle (1985), Glosten and Harris (1988), Brennan and
Subrahmanyam (1996), Easley (1996) and Dufour and Engle (2000).
Table 1
CDS trades and quotes.
Industry sector No. of No. of No. of No. of
firms trades bids offers
Consumer goods 32 5956 7495 7512
Consumer services 29 7316 9382 9416
Financials 41 2560 3384 3377
Telecommunications 33 8881 10,902 10,901
Total 135 24,713 31,163 31,206
This table shows summary statistics for our credit default swap dataset. For the
purpose of our study we focus on European companies with CDS contracts of 5-
years maturity, unsecured underlying debt, from January 1, 2006 to July 31, 2009.
The ‘‘No. of Trades’’ column indicates the number of executed trades for which
quotes are available on the same day; Bid and Offer denotes the number of bid and
offer price updates in the sample period.
2. Data
Our dataset combines data from two sources, GFI Group and
Bloomberg. The former provides intraday prices on credit default
swaps as well as descriptive information on individual CDS con-
tracts. The latter is used to retrieve market capitalisation and bal-
ance sheet information for the companies on which the CDS
contracts are written, the so-called ‘‘reference entities.’’ The com-
bined dataset covers the period from January 1, 2006 to July 31,
2009. In the empirical analysis the sample is split into two parts:
a pre-crisis sub-sample from January 2006 to March 2007 and a cri-
sis sub-sample from April 2007 to July 2009. The reference entities
are European companies and all contracts are denominated in
Euros. The CDS price data consists of time stamped best bid and of-
fer quotations and transaction prices. Bid and offer quotations are
binding for the dealer until they are executed, revised or with-
drawn. There are trading days where, for some CDS contracts, we
do not have any intra-day CDS quotes. This is a common feature
in the CDS electronic platforms as, unlike equity markets, there is
no obligation on dealers to continuously post prices on either side
of the market. Unfortunately, no information is provided on the vol-
ume of the transactions, the identity of buyers and sellers, and the
depth of the market. The representativeness of this dataset is guar-
anteed by the primary role of the GFI Group in the credit derivative
market (Tang and Yan, 2010). The GFI Group supplies a hybrid trad-
ing system that combines voice and electronic trading. Both types
of trades are recorded in the GFI database. For our empirical analy-
sis we keep only CDS data which satisfy the following conditions:
the maturity of the CDS contract is 5 years; the restructuring clause
is modified–modified2 (standard for the European market); the
underlying debt is senior unsecured; the quoted bid-ask spread is
greater than 0; and transactions are executed during days when bid
and offer quotations are recorded. In addition, we select only refer-
ence entities for which we are able to source market capitalisation
and short-term and long-term liabilities over the whole sample per-
iod. Finally, we consider only those industry sectors well represented
throughout the observation period, namely, financials, consumer
goods, consumer services and telecommunications.
In Table 1 we report summary statistics of trades and quotes for
our sample, which includes a total of 135 reference entities. On
average, there is approximately one trade execution per week
per reference entity. This is not representative of the actual trading
activity in the sample, which varies substantially from company to
company. The paucity of our data requires us to proceed with cau-
tion when estimating trade impact. This is because the estimation
procedure is based on the proximity of traded prices and quotes for
the same reference entity. It is to ensure that such proximity is
2
For a complete description of alternative restructuring clauses see ISDA defini-
tions (ISDA, 2003).
 F. Corò et al. / Journal of Banking & Finance 37 (2013) 5511–5525
80 Bid price
Offer price
Trade price
70
Basis points
60
50
40
08:53:20 09:26:40
preserved that our sample includes only those traded prices for
which active quotes are available at the same time. In addition,
in order to compensate for the relatively low frequency of the data
for some reference entities, we depart from the firm level analysis
in Hasbrouck (1991) and perform our investigation of the price im-
pact of trades using the whole sample of reference entities through
panel regressions. Table 1 indicates that the sectors with highest
and lowest trading activity and number of quote revisions are tele-
communications and financials respectively. Note that the number
of bid quotes and the number of offer quotes may differ because
market makers are allowed to post one-side quote revisions. Inter-
estingly, the number of quote revisions increases with the number
of trades which suggests, as one should expect, a more efficient
price discovery as trade execution intensifies. This is supported
by Fig. 1 which is a snapshot of the trading and price dynamics
of Telecom Italia 5-year CDS contract on June 27, 2007. It is evident
from the Figure that quote updating activity increases with trade
arrival. The price discovery process and the effect of CDS trades
on prices is examined next.
3. The model
In this section we present the econometric framework for our
empirical analysis and we derive a series of testable assumptions.
3.1. Evidence of informed trading in the CDS market
We conduct our investigation of informed trading in the CDS
market by employing a model inspired by the work of Hasbrouck
(1991). The rationale behind Hasbrouck’s model is that, in a market
with informed traders, order flow conveys information and has a
permanent impact on prices. To minimise the losses that may de-
rive from trading with better informed traders, liquidity providers
tend to raise their quotes after buyer-initiated trades and lower
them after seller-initiated trades. As theoretically formalised in
Easley and O’Hara (1987), the magnitude of the quote revision is
positively related to the likelihood of informed trading and the le-
vel of informational asymmetry. Furthermore, Hasbrouck (1991)
shows that the trade impact may also depend on the size of the
bid-ask spread at the time of the trade, the wider the spread the
larger the trade impact. With these considerations in mind we cap-
ture the price impact of informed trading in the CDS market with
the following panel regression model:
r it ¼ ai þ k0 Q it þ k1 Q i;t1 þ k2 Q i;t2 þ gQ it St þ uit
5513
10:00:00 10:33:20 11:06:40 11:40:00 12:13:20 12:46:40
Time
Fig. 1. Intra-day quotes of Telecom Italia.
where i is the reference entity, t is the trade time, rit is the difference
between the mid-quotes prevailing at t + 60 s and at t, respectively.
When two consecutive trades occur within a 60-s time-frame, the
t + 60 mid-quote is taken to be the one at the time of the second
trade.
As a result, rit is positive when the midpoint is revised upwards
and negative when it is revised downwards. ai denotes firm spe-
cific fixed effects. Qt denotes the ‘‘sign’’ of the trade at time t, that
is, whether the trade was initiated by a buyer or a seller. Unfortu-
nately, our dataset does not contain information about the initiator
of the trade. Hence, the sign of the trade is inferred using the algo-
rithm suggested by Hasbrouck (1991) where Qt takes the value +1
if the trade price is above the prevailing quote midpoint; 1 if the
trade price is below the prevailing quote midpoint and 0 if the
trade price is equal to the quote midpoint. Accordingly, trades
are classified as buyer-initiated, seller-initiated or undetermined,
respectively. The lagged signed trade variables Qt1, Qt2 are intro-
duced to control for potential delayed effects of trades on mid-
quotes. Finally, St is the prevailing bid-ask spread before the trade
at time t. Regression (1) is estimated with a standard ‘‘within
transformation’’ whereby all terms in the panel are expressed as
differences from the respective firm level means. As the demeaned
model will have a zero intercept we recover the regression con-
stant by adding to all the terms in (1) the respective panel means,
as suggested by Cameron and Trivedi (2005).3 We also derive t-sta-
tistics with panel-robust standard errors (with clustering at the firm
level) to control for autocorrelation and heteroskedasticity.
The focal point of the analysis is the persistent component of
the trade impact which we define as k ¼ k0 þ k1 þ k2 . A positive
and statistically significant estimate of k would reveal that liquid-
ity suppliers tend to revise strategically their quotations condition-
ally on the order flow. This would confirm that, as for equity and
bond markets, the CDS market is characterised by the presence
of informed traders, and market makers use order flow to revise
their estimates of CDS prices. On the contrary, a non-significant
estimate of k would reveal that trading activity is not relevant to
CDS price formation. Hasbrouck (1991) suggests using impulse re-
sponse functions to capture the informational (persistent) impact
of a trade on prices. He indicates that this measure could be
approximated by the sum of the trade coefficients in his return
equation. We follow the same approach and use k to measure in-
formed trading. This is achieved by estimating k using model (1)
with mid-quote adjustments rit defined over an extended period
ð1Þ
3
See Cameron and Trivedi (2005), pp. 697–706.
5514 F. Corò et al. / Journal of Banking & Finance 37 (2013) 5511–5525
of 60 s which we assume to be long enough to eliminate temporary
microstructure effects.
3.2. Determinants of CDS price changes
The second step of our empirical analysis is to provide a new
perspective on the determinants of CDS price changes. We make
the following contributions. First, using real market prices from
an electronic trading platform, we examine the liquidity effects
in the CDS market for European companies during the 2007–
2009 financial crisis. Second, contrary to Tang and Yan (2007),
Acharya and Johnson (2007) and Pires et al. (2011), our analysis
focuses on CDS price changes rather than CDS price levels. We do
so because we detect non-stationarity in the series of CDS prices
and most explanatory variables. This is consistent with Blanco
et al. (2005) who, in their study of price discovery in the bond
and CDS markets, find CDS prices and bond yields to be integrated
of order one for most of the companies in their sample. Our anal-
ysis is also in line with the work of Collin-Dufresne et al. (2001)
who investigate the determinants of credit spreads in the bond
market by looking at yield changes rather than levels. Thus far,
only Ericsson et al. (2009) have explored the components of CDS
price changes, but they limit their analysis to credit risk factors
only. Our work, instead, encompasses both credit and liquidity
proxies. Third, differently from Acharya and Johnson (2007) and
Tang and Yan (2007) who employ exclusively firm-specific liquid-
ity variables, we look at commonality in liquidity effects at the
industry level and explore the impact of informed trading on CDS
price changes with a microstructure proxy. Forth, we examine
whether changes in CDS prices are related to the demand of credit
protection. While for other markets (equity, bond and option mar-
kets)4 there is evidence that asset prices depend on demand pres-
sure, in the CDS market this is still an open question.
In the following, we first provide a summary list of the credit
and liquidity proxies as well as the macro-economic variables we
employ in our analysis, and then we describe each variable in de-
tail, justify its use and develop testable hypotheses:
 firm-specific credit variables: distance to default, leverage and
credit rating;
 firm-specific liquidity variables: time-weighted absolute bid-
ask spread, demand pressure, trading intensity and quote
imbalance;
 industry-wide liquidity variables: industry average bid-ask
spread and industry average trade impact; and
 macro-economic variables: return on the Euro Stoxx 50, implied
volatility of the Euro Stoxx 50, and 10-year Euro government
bond yield.
The analysis of the determinants of CDS price changes is per-
formed with the above explanatory variables sampled on a
monthly basis. We use changes in the monthly averages of the
explanatory variables rather than changes in their end-of-the
month values because end-of-the month values may not be avail-
able consistently for all our reference entities.
The distance to default (DTD) is calculated as in Vassalou and
Xing (2004). We source data on equity prices and book values of
short term and long term liabilities from Bloomberg. The distance
to default is first computed for every week of the sample and then
averaged over each calendar month. As the distance to default in-
creases with the reference entity’s credit quality, the following
hypothesis is expected to hold.
4 7
See Chordia et al. (2002), Chordia and Subrahmanyam (2004), Coval and Stafford
(2007), Greenwood and Vayanos (2010), Bollen and Whaley (2004), Amin et al. (2004)
and Garleanu et al. (2009).
Hypothesis 1. Changes in CDS prices are negatively related to
changes in the distance to default (DTD).
We capture a firm’s credit risk also through leverage (LEV) and
the firm’s credit rating (CR). LEV is calculated as the ratio of book
value of debt over the sum of the market value of equity and book
value of debt. The credit rating variable (CR) is the average of the
ratings assigned to the same company by Moody’s, Standard &
Poor’s and Fitch. After mapping the letter ratings to a numeric scale
(with higher numbers corresponding to lower ratings as described
in Table A.1 in the Appendix), the average rating is calculated as a
simple average of the corresponding numerical ratings. Since credit
risk typically increases with poorer credit ratings and higher lever-
age, the following hypothesis is expected to hold:
Hypothesis 2. Changes in CDS prices are positively related to
changes in credit rating (CR) and leverage (LEV).
The bid-ask spread has been extensively used in the literature to
measure the liquidity of equity, debt and OTC derivative markets.
The spread captures the cost of executing small size trades (for order
book markets) or normal size trades (for quote driven markets). It is
considered a good approximation of the costs incurred by liquidity
providers such as order processing costs, adverse selection costs,
inventory costs,5 and is affected by the level of competition among
market makers. While the relative value of the bid-ask spread is com-
monly used in the equity market, it is still an open debate whether rel-
ative or absolute bid-ask spread should be used in the context of the
CDS market. In our opinion, Pires et al. (2011) convincingly show that
the bid-ask spread of CDS prices is already a proportional measure and
does not need to be divided by the CDS price. They then show that
there is a positive and significant relationship between CDS prices
and absolute bid-ask spreads.6 This result is in line with the findings
reported by Bongaerts et al. (2011). Conversely, Acharya and Johnson
(2007) and Tang and Yan (2007) use relative bid-ask spread and reach
different conclusions. Acharya and Johnson (2007) do not detect any
influence of the bid-ask spread on CDS prices, whereas Tang and Yan
(2007) find only a weak relationship.
The positive effect of the bid-ask spread on the CDS price is con-
sistent with the assumption that liquidity providers are typically
protection sellers who demand a premium when facing illiquidity.
Evidence of a liquidity premium demanded by CDS sellers is also
found in Tang and Yan (2007). Bongaerts et al. (2011) provide an
alternative explanation and argue that, when short selling is taken
into account ‘‘illiquid assets can trade at higher prices than liquid
assets’’.
To capture the relationship between illiquidity and CDS prices
we proxy illiquidity with the daily time-weighted average bid-
ask spread (TBAS). Each bid-ask spread observed within a trading
day is weighted by the number of seconds it remains available.
Next, we take simple monthly averages of the daily average
spreads.7 As mentioned above, the bid-ask spread is a broad illiquid-
ity measure as it reflects inventory costs, processing costs and asym-
metric information. Later we will introduce a firm specific trading
intensity measure (TI) to isolate the impact of asymmetric informa-
tion as a source of illiquidity. By using TBAS and TI in isolation and
jointly in our regression analysis we seek to capture the influence
on CDS prices of illiquidity stemming from asymmetric information
on one side and the remaining factors (inventory costs and process-
5
See Amihud and Mendelson (1980), Ho and Stoll (1981), Copeland and Galai
(1983), Easley and O’Hara (1987) and Glosten and Milgrom (1985).
6
Similarly, when studying liquidity effects in the corporate bond market, Chen
et al. (2007) find that there is a positive relationship between the yield spread and the
bid-ask spread.
Instead of the quoted bid-ask spread we could have computed the effective bid-
ask spread. However, this measure would have led to several missing observations
due to the often low level of trading activity in our sample.
 F. Corò et al. / Journal of Banking & Finance 37 (2013) 5511–5525
ing costs) on the other, and their relative importance. Similarly, we
further introduce proxies to detect commonality in liquidity (indus-
try average bid-ask spread, IBAS, and industry average asymmetric
information, ASY) with the aim of disentangling and comparing firm
specific and systematic illiquidity effects on CDS price changes. As
for TBAS, we test the following hypothesis:
Hypothesis 3. Changes in CDS prices are positively related to
changes in the time-weighted absolute bid-ask spread (TBAS).
Bollen and Whaley (2004) study the effect of demand pressure
on the implied volatility for index and individual stock options.
Garleanu et al. (2009) provide a theoretical model that shows
how option prices are affected by the corresponding demand level.
It is reasonable to expect that demand pressure could be a signifi-
cant factor in explaining price changes in the credit derivative mar-
ket as well. The obvious justification for this assumption is that
liquidity suppliers (protections sellers) are not willing to sell an
unlimited amount of a particular CDS contract at the current price.
Following Bollen et al.’s (2004) work, we define demand pressure
(DP) for each company and for each month as the difference be-
tween the number of buyer-initiated trades and the number of
seller-initiated trades. If a firm has no trades over a particular
month then the demand pressure is set to zero. Assuming that
liquidity suppliers demand a liquidity premium as a response to
an excessive demand of credit protection, the following hypothesis
is expected to hold:
Hypothesis 4. Changes in CDS prices are positively related to
changes in the demand pressure (DP).
A natural candidate to measure the effects of asymmetric infor-
mation on CDS prices at the firm level would be the k, that is, the
permanent trade impact discussed in Section 3.1. However, k can-
not be derived reliably over time for a number of companies in our
sample due to thin trading. So, we use k as an industry aggregate
measure of asymmetric information, as explained later in this Sec-
tion. At the firm level instead, we employ an alternative asymmet-
ric information proxy, the trading intensity (TI). For each company
and each month in the sample period, TI is calculated as the aver-
age number of daily trades in that month. Chordia et al. (2000) ar-
gue that the number of trades is a good indicator of asymmetric
information in the market. They suggest that this could be the case
because informed traders try to conceal their trading by splitting
large orders into smaller ones hence increasing the number of exe-
cutions. A theoretical justification for using a proxy of trading
intensity as a determinant of equity prices is offered by Easley
and O’Hara (1992) and corroborating empirical evidence is pro-
vided by Dufour and Engle (2000) who show that the price impact
of equity trades is larger when trading intensity is greater. Their
intuition is that uninformed traders are expected to trade indepen-
dently of the existence of information and hence should exhibit a
uniform trading presence in the market over time whereas in-
formed traders access the market only when they possess valuable
information. As a result, liquidity providers revise their quotations
by a greater magnitude when trading intensity is higher as they be-
lieve they are more likely to face informed traders.8 In the context
8
The existing literature offers a couple of additional approaches for measuring the
level of information asymmetry in the market. Easley et al. (1996) devise an
econometric model which uses the number of buy and sell trades to estimate the
probability of informed trading (PIN). Easley et al. (2002) show that PIN is particularly
important in explaining the dynamic of the equity returns and that assets with larger
returns have higher PIN. In our case, however, the implementation of this measure is
not feasible because often firms have very thin trading. An alternative approach is
proposed by Acharya and Johnson (2007) who suggest measuring information
asymmetries by calculating the number of banking relationships that each company
has. However, they find no effect of information asymmetries on CDS price levels.
5515
Table 2
Summary of hypotheses.
Explanatory variable Expected sign
DDTD 
DCR +
DLEV +
DTBAS +
DP +
DTI +
DASY +
DIBAS +
QI +
DSTOXX 
DVSTOXX +
DEU YIELD +/
The table shows a summary of the assumed signs for the relationships between CDS
price changes and credit, liquidity and macro-economic factors. DTD is the distance
to default; LEV is leverage; CR is credit rating; TBAS is the time-weighted absolute
bid-ask spread; ASY is an industry-wide asymmetric information index; TI is the
number of executed trades; DP is the demand pressure; IBAS is the cross sectional
bid-ask spread for the industry sectors considered in our sample (that is financials,
consumer goods, telecommunications and consumer services), QI is the quote
imbalance, STOXX is a benchmark stock index for the Euro area, VSTOXX is the
stock index’s implied volatility and EU YIELD is the 10-year Euro area government
bond yield.
of the CDS market, protection sellers may charge a higher liquidity
premium (higher offering price) when trading activity increases be-
cause this implies higher adverse selection costs. Given these consid-
erations, the following hypothesis is expected to hold:
Hypothesis 5. Changes in CDS prices are positively related to
changes in trading intensity (TI).
An imbalance between bid and ask quotes for a particular firm
may reveal a momentum effect whereby when a dealer posts a
more aggressive bid or ask quote other dealers compete to revise
quotes in the same direction and outdo the first revision. This phe-
nomenon is supported by conversations we held with a senior CDS
trader that employs the GFI platform. We capture momentum with
the quote imbalance (QI), which is defined as the difference be-
tween the number of bid quote updates and the number of ask
quote updates for a single name over a given month. This is similar
to the ‘‘net buying interest’’ employed by Tang and Yan (2010). A
predominance of buy quote revisions may indicate greater compe-
tition to buy CDSs which would push CSD prices up. Conversely, a
predominance of sell quote revisions would indicate decreasing
prices.
Hypothesis 6. Changes in CDS prices are positively related to the
quote imbalance (QI).
While TBAS and TI are firm specific measures of liquidity, it may
be reasonable to explore the presence of a systematic liquidity
component. In the stock market, Chordia et al. (2000) and Has-
brouck and Seppi (2001) find evidence of commonality in liquidity.
Following their lead, we investigate whether changes in firm-spe-
cific liquidity measures (e.g. TBAS) remain significant after intro-
ducing industry-wide liquidity proxies. Our monthly aggregate
liquidity measure for a given industry sector is the average of the
monthly time weighted bid-ask spread across all the firms in that
sector (IBAS). In addition, we measure the specific impact of asym-
metric information with an industry-wide (persistent) trade im-
pact indicator, ASY, based on the ks estimated with model (1).
For this purpose, we cluster observations by industry and execute
daily rolling regressions based on a time window of 30 days. ASY is
calculated as the monthly average of the daily ks estimated over
the sample period. Our testable hypothesis is as follows:
5516 F. Corò et al. / Journal of Banking & Finance 37 (2013) 5511–5525

Table 3
Regression results on trade impact.

Whole sample Pre-crisis Crisis

Model (a) Model (b) Model (a) Model (b) Model (a) Model (b)
Constant 0.079*** 0.086*** 0.039*** 0.043*** 0.113*** 0.119***
(18.007) (18.010) (9.344) (11.594) (16.543) (15.595)
Signed Trade 0.429*** 0.623*** 0.387*** 0.461*** 0.521*** 0.855***
(8.953) (8.000) (8.346) (6.902) (7.882) (6.545)
Signed Trade(lag1) 0.082*** 0.080*** 0.027** 0.027*** 0.097*** 0.096***
(5.182) (5.073) (2.552) (2.763) (4.001) (3.869)
Signed Trade(lag2) 0.029* 0.029* 0.003 0.004 0.034 0.032
(1.795) (1.758) (0.321) (0.353) (1.219) (1.146)
Signed BAS 0.051*** 0.030*** 0.01 0.003 0.056*** 0.034***
(6.361) (3.365) (1.342) (0.288) (7.137) (3.277)
S  Signed Trade 0.034 0.032 0.107
(0.286) (0.335) (0.510)
S  Signed BAS 0.036** 0.03 0.021
(2.460) (1.430) (1.288)
L  Signed Trade 0.125 0.241*** 0.09
(0.886) (2.865) (0.521)
L  Signed BAS 0.006 0.074 0.022
(0.209) (1.598) (0.712)
IGSigned Trade 0.167* 0.061 0.369**
(1.757) (0.802) (2.329)
IGSigned BAS 0.011 0.003 0.021
(0.805) (0.254) (1.390)
Observations 24,713 24,713 11,054 11,054 13,659 13,659
Adjusted R2 0.112 0.119 0.126 0.133 0.119 0.125

This table shows estimates of trade impact based on microstructure model (1) for the whole sample, the pre-crisis period and the crisis period. The sample period is January 1,
2006 to July 31, 2009.The pre-crisis period goes from January 2006 to March 2007. The crisis periods is from April 2007 to July 2009. The dependent variable is the difference,
for reference entity i, between the mid-quote 60 s after the trade that takes place at time t and the mid-quote at time t. The explanatory variables are Qt (Signed Trade) is the
direction of the trade at t and takes the value +1 if the trade is buyer-initiated, 1 if it is seller-initiated and 0 if it is not classifiable. The model is extended to include the
interaction term Qi,tSt1 (Signed BAS) which is the product between the direction of trade and the bid ask spread before time t. The above variables are also interacted with the
following variables: (a) dummy variables L and S which take value 1 when the firms fall in the large and small capitalisation sub-samples respectively. Small, and large
capitalisation sub-samples correspond to the first and third terciles of the sample distribution stratified by the firm’s market capitalisation; and (b) a dummy variable IG
which takes the value 1 when the firm has an investment grade rating. The panel estimation is carried out by using firm-level fixed-effects and panel-robust standard errors
(clustering at the firm level) to control for autocorrelation and heteroskedasticity. t-Statistics are given in parenthesis.
*
Significance at 10% level.
**
Significance at 5% level.
***
Significance at 1% level.

Hypothesis 7. Changes in CDS prices are positively related to
changes in industry wide information asymmetries (ASY) and
industry wide bid-ask spread (IBAS).
Finally, we hypothesise that CDS prices may be affected by busi-
ness climate and macro-economic factors which we proxy with the
return of the European stock market (STOXX), the market’s implied
volatility (VSTOXX) and the level of 10-year interest rates in the
Euro area (EU YIELD). Specifically, as a benchmark for the European
stock market we employ the Euro Stoxx 50 index which includes
blue chip companies that are sector leaders in the Eurozone. Im-
plied volatility is measured with the VSTOXX index obtained from
options written on the Euro Stoxx 50.9 The interest rate variable is
sourced from the ECB and is based on 10-year government bonds of
Euro-area sovereigns whose credit rating is triple A.
The above variables cannot be unequivocally categorised as
credit or liquidity factors. Rather, as indicators of the macro-eco-
nomic environment, they may influence both default risk and asset
liquidity. On one hand, stock index return and stock index volatil-
ity can be related to credit risk as suggested by Collin-Dufresne
et al. (2001) (a similar interpretation is found in Acharya and John-
son, 2007). Stock index return can be used as a measure of changes
in business climate. Even if the behaviour of the stock market may
not impact directly on the default probability of a firm, it may still
influence credit losses by its effect on recovery rates. Positive stock
9
STOXX and VSTOXX, together with about 300 other indices, are supplied by
STOXX Ltd. which is jointly owned by Deutsche Bourse and SIX Group.
returns may indicate healthy business conditions and higher
recoveries in case of bankruptcy. Moreover, the Merton (1974)
model of a firm’s debt establishes a link between stock volatility
and credit risk, whereby higher volatility increases the probability
of default.
The negative relationship between stock return and credit risk
and the positive relationship between stock volatility and credit
risk illustrated above can also be established, with the same sign,
with respect to liquidity risk, and ultimately CDS prices. Bru-
nnermeier and Pedersen (2009) find that stock return affect the
availability of capital that traders can use to fund their positions.
Capital is more easily available when market return is positive.
This favours asset liquidity as traders face fewer constraints in
their transactions. Hence, higher stock return is expected to be in-
versely related to asset illiquidity. On the contrary, asset illiquidity
increases with stock volatility because higher uncertainty in the
market pushes margin requirements upward together with trad-
ers’ need for capital.
The level of interest rates can similarly be interpreted as a credit
or liquidity factor. It is a credit factor, as higher interest rates in-
crease the drift of the firm value process that determines the firm’s
default probability. An increase in the drift causes the default prob-
ability (and hence credit risk) to fall (see Longstaff and Schwartz,
1995; Collin-Dufresne et al., 2001). In contrast, from a macro-eco-
nomic perspective, interest rates do not have a pre-determined ef-
fect on credit and liquidity due to their ambivalent role on growth.
On one hand, higher interest rates can stifle growth by increasing
borrowing costs. In this sense, they would be positively related
 F. Corò et al. / Journal of Banking & Finance 37 (2013) 5511–5525 5517

Table 4
Credit and liquidity determinants of CDS price changes.

(a) (b) (c) (d) (e) (f) (g) (h) (i)

*** *** *** *** *** *** * *
Constant 1.93 1.31 2.06 1.94 1.23 2.12 0.56 0.67 0.5
(8.7) (7.42) (8.89) (8.66) (2.71) (10.02) (1.83) (1.93) (1.31)
DDTD 4.21*** 2.34** 4.21*** 4.21*** 3.96*** 3.95*** 1.44 0.57 0.52
(3.19) (2.60) (3.21) (3.20) (3.09) (3.08) (1.61) (0.86) (0.77)
DCR 6.24 2.65 6.17 6.25 6.01 6.02 7.47 1.17 1.5
(0.76) (0.58) (0.75) (0.76) (0.74) (0.76) (0.88) (0.26) (0.33)
DLEV 2.26*** 1.88*** 2.25*** 2.24*** 2.22*** 2.27*** 1.16*** 1.25*** 1.32***
(3.37) (4.03) (3.36) (3.34) (3.35) (3.42) (2.75) (3.68) (3.64)
DTBAS 3.69*** 3.24*** 3.21***
(10.15) (10.22) (10.26)
DP 0.18 0.07 0.05
(1.61) (0.91) (0.58)
DTI 0.05* 0.03 0.03
(1.98) (1.48) (1.43)
QI 0.30*** 0.14*** 0.14***
(7.24) (4.34) (4.31)
DASY 8.34*** 7.54*** 7.12***
(6.7) (6.49) (6.19)
DIBAS 5.04*** 3.01*** 2.68***
(8.54) (8.16) (5.87)
DEU YIELD 16.61***
(7.17)
DVSTOXX 0.03
(0.16)
Observations 2908 2908 2908 2908 2908 2908 2908 2908 2908
Adjusted R2 0.045 0.386 0.045 0.045 0.06 0.072 0.182 0.458 0.463

The dependent variable of the panel regressions shown in this table is the change in the mid-quote price. The explanatory variables are the change in the distance to default
(DDTD), the change in leverage (DLEV), the change in credit rating (DCR), the change in the time-weighted absolute bid-ask spread (DTBAS), the change in the industrial
information asymmetries index (DASY), the change in the number of executed trades (DTI), the buying pressure (DP), the change in the industry-wide bid-ask spread (DIBAS),
the quote imbalance (QI), the change in implied volatility of a benchmark stock index for the Euro area (DVSTOXX) and the change 10-year Euro area government bond yield
(DEU YIELD). Dependent variable and explanatory variables are sampled on a monthly basis. The sample period is January 1, 2006 to July 31, 2009. The panel estimation is
carried out by using firm-level fixed-effects and panel-robust standard errors (clustering at the firm level) to control for autocorrelation and heteroskedasticity. t-Statistics are
given in parenthesis.
*
Significance at 10% level.
**
Significance at 5% level.
***
Significance at 1% level.

to default risk and, through higher funding costs for traders, they
may also be linked to higher illiquidity.10 On the other hand, higher
(nominal) interest rates, if driven by higher real interest rates, can be
an indication of positive growth expectations on the part of bond
traders (Alexopoulou et al., 2009). In this case, higher interest rates
would be associated with lower default risk and lower illiquidity.
As a result, in the hypothesis below, we leave the relationship be-
tween EU YIELD and CDS prices undetermined. Through the influ-
ence of macro-economic variables on credit risk and liquidity risk,
the following relationship with CDS price changes is expected to
hold:
Hypothesis 8. Changes in CDS prices are negatively related to
changes in stock index return (STOXX) and positively related to
changes in implied volatility (VSTOXX). The relationship between
CDS prices and 10-year interest rates (EU YIELD) can be either
positive or negative.
A summary of all our assumptions is reported in Table 2. To test
the above hypotheses we run a panel regression of changes in CDS
prices on changes of the credit and liquidity proxies we have intro-
duced. The dependent variable of the regression is the change in
the monthly average of daily CDS mid-quotes. The daily mid-quote
of any given CDS contract is the weighted average of all the intra-
10
The positive relationship between government bond yields and the risks of
default and illiquidity would be stronger in a scenario of heightened sovereign risk as
it has been experienced in the Euro-zone since late 2009. However, as our sample
ends in August 2009, we do not expect sovereign risk to have a noticeable impact on
our results, especially given that the Euro government bond yield we employ in our
analysis is built by considering only triple-A rated sovereigns.
day mid-quotes where the weights are the number of seconds
each quotation remains outstanding divided by the total length
of the trading day in seconds. Then, we estimate the following
model,
DCDSit ¼ ai þ b1 DDTDit þ b2 DCRit þ b3 DLEVit þ b4 DTBASit
þ b5 DPit þ b6 DTIit þ b7 QIit þ b8 DASYst þ b9 DIBASst
þ b10 DVSTOXXt þ b11 DEUYIELDt þ uit ð2Þ
where i = 1, ... , N indexes the firms, t = 1, . . . , T denotes time in
months and s a specific industry sector. It should be noted that
the variable STOXX has not been included in the regression to avoid
multicollinearity problems in that we find it to be highly correlated
with VSTOXX, as we shall show later. Regression (2) is estimated
with a standard ‘‘within transformation’’, like regression (1), where
all the terms in the panel are expressed as differences from the
respective firm level means. As the demeaned model will have a
zero intercept, we recover the regression constant by adding the
respective panel means to all the terms in (2). Due to the presence
of heteroskedasticity and autocorrelation, we adjust regression
coefficients’ t-statistics by calculating panel-robust standard errors.
Summary statistics and pair-wise correlations for the depen-
dent variable and covariates in regression (2) are given in
Table A.2 in the Appendix A. All variables are calculated on a
monthly basis. In panel A, we show summary statistics of the vari-
ables in levels. We can see that although there seems to be a bal-
ance between buyer initiated trades and seller initiated trades
(the median value of the demand pressure, DP, is 0), quote revi-
sions are more frequent on the bid side (quote imbalance, QI, has
a median of 8). This may imply that dealers need to compete more
to buy credit protection, which may be the result of hedging
5518 F. Corò et al. / Journal of Banking & Finance 37 (2013) 5511–5525

Table 5
Credit and liquidity determinants of credit spreads before and during the 2007–2009 financial crisis.

(a) (b) (c) (d) (e) (f) (g) (h) (i)

Panel A: Pre-crisis period
Constant 0.96*** 0.77*** 1.13*** 1.00*** 2.93*** 0.96*** 0.76*** 2.64*** 2.65***
(10.78) (8.82) (5.69) (10.56) (7.94) (10.46) (6.43) (6.79) (8.20)
DDTD 0.29 0.15 0.34 0.31 0.33 0.29 0.38 0.25 0.24
(0.53) (0.28) (0.64) (0.57) (0.65) (0.53) (0.71) (0.51) (0.48)
DCR 3.88 3.76 3.71 3.85 3.78 3.88 3.76 3.43 3.43
(0.87) (0.88) (0.83) (0.87) (0.95) (0.88) (0.86) (0.90) (0.90)
DLEV 0.33 0.22 0.37 0.33 0.23 0.33 0.22 0.19 0.19
(1.26) (0.86) (1.53) (1.28) (0.81) (1.26) (0.88) (0.79) (0.82)
DTBAS 4.64*** 4.52*** 4.52***
(5.15) (5.05) (5.05)
DP 0.15 0.15 0.15
(1.12) (1.47) (1.43)
DTI 0.04* 0.04* 0.04*
(1.73) (1.83) (1.79)
QI 0.17*** 0.14*** 0.14***
(5.66) (5.69) (5.42)
DASY 0.28 2.67 2.67
(0.15) (1.53) (1.48)
DIBAS 3.36*** 0.17 0.05
(3.83) (0.24) (0.04)
DEU YIELD 0.66
(0.23)
DVSTOXX 0.02
(0.11)
Observations 781 781 781 781 781 781 781 781 781
Adjusted R2 0.009 0.17 0.017 0.017 0.078 0.007 0.034 0.239 0.238

Panel B: Crisis period

Constant 2.20*** 1.58*** 2.44*** 2.27*** 1.26** 2.45*** 0.68 0.38 0.22
(6.02) (5.79) (6.32) (5.89) (2.11) (6.9) (1.45) (0.82) (0.43)
DDTD 7.91*** 4.62*** 7.93*** 7.92*** 7.49*** 7.53*** 3.13** 1.65* 1.33
(4.92) (4.23) (5.02) (4.93) (4.72) (4.78) (2.62) (1.90) (1.49)
DCR 12.43 1.19 12.41 12.46 12.53 12.04 13.27 0.56 0.22
(1.12) (0.19) (1.12) (1.12) (1.13) (1.13) (1.14) (0.09) (0.04)
DLEV 2.30*** 1.95*** 2.30*** 2.26*** 2.27*** 2.32*** 1.21** 1.32*** 1.40***
(3.07) (3.7) (3.07) (3.02) (3.07) (3.12) (2.58) (3.43) (3.37)
DTBAS 3.64*** 3.22*** 3.19***
(10.07) (10.16) (10.2)
DP 0.44** 0.23* 0.16
(2.22) (1.71) (1.17)
DTI 0.08* 0.07** 0.07**
(1.66) (2.14) (2.19)
QI 0.35*** 0.16*** 0.16***
(7.25) (3.96) (4.06)
DASY 8.29*** 7.60*** 7.11***
(6.67) (6.53) (6.16)
DIBAS 4.89*** 2.89*** 2.42***
(8.42) (7.91) (5.25)
DEU YIELD 20.18***
(7.61)
DVSTOXX 0.02
(0.1)
Observations 2127 2127 2127 2127 2127 2127 2127 2127 2127
Adjusted R2 0.058 0.392 0.058 0.057 0.073 0.085 0.185 0.461 0.467

The table exhibits the estimation results of regression (2) before and after the crisis. The dependent variable is the change in the mid-quote price. The explanatory variables
are the change in the distance to default (DDTD), the change in leverage (DLEV), the change in credit rating (DCR), the change in the time-weighted absolute bid-ask spread
(DTBAS), the change in the number of executed trades (DTI), the demand pressure (DP), the change in the industrial information asymmetries index (DASY), the change in the
industry-wide bid-ask spread (DIBAS), the quote imbalance (QI), the change in implied volatility of a benchmark stock index for the Euro area (DVSTOXX) and the change 10-
year Euro area government bond yield (DEU YIELD). Dependent variable and explanatory variables are sampled on a monthly basis. The pre-crisis period goes from January
2006 to March 2007. The crisis periods is from April 2007 to July 2009. The panel estimation is carried out by using firm-level fixed-effects and panel-robust standard errors
(clustering at the firm level) to control for autocorrelation and heteroskedasticity. t-Statistics are given in parenthesis.
*
Significance at 10% level.
**
Significance at 5% level.
***
Significance at 1% level.

pressure to cover for the positions they have taken when providing
liquidity as protection sellers. Surprisingly, we observe that some-
times the distance to default is negative as evidenced by the
minimum value of the DTD variable. We have explored the issue
and found that this happens exclusively for the financial institu-
tions in our sample. The distance to default is defined as the differ-
ence between the expected log value of a reference entity’s assets
and the default trigger, represented by reference entity’s log liabil-
ities.11 Financial institutions were amongst the worst hit by the
11
Following Vassalou and Xing (2004) which is in turn based on the KMV default
model (see, for example, Crosbie, 2003), we define the default trigger as the log value
of short term liabilities plus 50% of long term liabilities.
 F. Corò et al. / Journal of Banking & Finance 37 (2013) 5511–5525 5519

financial crisis, so much so that their expected (log) asset values, in Table 6
some cases, fell below the default trigger. Credit and liquidity determinants of credit spreads with crisis interaction effects.

The correlations between changes in the CDS price and the Coefficient t-Value
credit, liquidity and macro-variables over the whole sample, and Constant 2.07 ***
(4.24)
the pre-crisis and crisis sub-samples, are reported in the second DDTD 0.07 (0.15)
column of panel B. These correlations broadly confirm all the DCR 3.76 (0.84)
hypotheses introduced in the previous section. The correlation be- DLEV 0.20 (0.79)
DTBAS 4.83*** (4.50)
tween DCDS and changes in credit ratings, DCR, in the pre-crisis DP 0.15 (1.50)
period is not in line with our assumptions, though it is in the crisis DTI 0.04** (2.05)
period. Upon closer inspection of the data, we found that this is due DASY 2.21 (1.22)
to two large CDS changes for a single firm. If these outliers were DIBAS 0.02 (0.02)
QI 0.10*** (3.49)
eliminated the correlation would take the expected sign. Indeed,
DEU YIELD 1.36 (0.48)
our regression results and robustness tests show that DCR is never DVSTOXX 0.03 (0.16)
statistically significant in the pre-crisis period. It is noteworthy Crisis 1.81** (2.28)
that VSTOXX and STOXX have a correlation of 90% and 71% in Crisis  DDTD 1.22 (1.24)
the pre-crisis and crisis periods, respectively. As mentioned earlier, Crisis  DCR 3.60 (0.47)
Crisis  DLEV 1.23*** (2.80)
in our regression analysis we retain only VSTOXX. However, con-
Crisis  DTBAS 1.63 (1.52)
sistent results (unreported) are obtained when using STOXX. Crisis  DP 0.32* (1.94)
As previously reported, most variables in regression (2) are in Crisis  DTI 0.03 (0.85)
first differences as we find that CDS prices and our covariates, with Crisis  DASY 9.35*** (4.61)
Crisis  DIBAS 2.42** (2.06)
the exception of the demand pressure, quote imbalance and indus-
Crisis  QI 0.06 (1.54)
try-wide asymmetric information proxy, are non-stationary. The Crisis  DEU YIELD 18.60*** (4.64)
stationarity of the variables is assessed through a Fisher-type test Crisis  DVSTOXX 0.01 (0.05)
(Choi, 2001).12 Choi (2001) shows that, when the number of firms Observations 2908
is finite, alternative panel unit root test statistics can be based on Adjusted R2 0.465
the inverse of a normal distribution or the inverse of a logistic distri-
The dependent variable of the panel regressions shown in this table is the change in
bution. We use both approaches and test the null hypothesis that all the mid-quote price. The explanatory variables are the change in the distance to
firms contain a unit root versus the alternative hypothesis that at default (DDTD), the change in leverage (DLEV), the change in credit rating (DCR),
least some firms are stationary. Results are shown in Table A.3 in the change in the time-weighted absolute bid-ask spread (DTBAS), the change in
the Appendix A. It can be seen that when the variables are consid- the industrial information asymmetries index (DASY), the change in the number of
executed trades (DTI), the buying pressure (DP), the change in the industry-wide
ered in levels the null hypothesis is not rejected which means that
bid-ask spread (DIBAS), the quote imbalance (QI), the change in implied volatility of
firms exhibit non-stationarity. On the other hand, when taking the a benchmark stock index for the Euro area (DVSTOXX), the change 10-year Euro
first difference of the variables (with the exclusion of DP and QI area government bond yield (DEU YIELD) and a crisis dummy (Crisis). Dependent
which are used in levels in our regression) the null hypothesis of unit variable and explanatory variables are sampled on a monthly basis. The sample
roots is always rejected. period is January 2006 to July 2009. The crisis dummy identifies the sub-sample
from April 2007. The panel estimation is carried out by using firm-level fixed-effects
and panel-robust standard errors (clustering at the firm level) to control for auto-
4. Results correlation and heteroskedasticity. t-Statistics are given in parenthesis.
*
Significance at 10% level.
**
Significance at 5% level.
We first report the results of the trade impact of order flow in ***
Significance at 1% level.
the CDS market, estimated with regression (1). The findings are
shown in Table 3 for the entire sample and two sub-periods: the
pre-crisis and the crisis periods. We identify the beginning of the regardless of market conditions. An F-test on k reveals that it is sta-
financial crisis with the bankruptcy of New Century Financial (the tistically significant at the 0.1% level in all cases. The value of k
second largest US subprime lender) on April 22, 2007 which is indicates by how much, on average, the CDS midpoint adjusts up-
one of the first precursors of the subprime debacle. In the Table, ward (downward) after a purchase (sale), while taking into account
model a refers to regression (1), while model b is an expanded ver- delayed effects of past trades. As one would expect, the k during
sion that includes interactions of the explanatory variables with the crisis (0.727 for model b) is markedly higher than the k in
dummies. The dummies capture firm-specific characteristics that the pre-crisis period (0.430). This shows that price corrections fol-
are normally related to the quality of the information available lowing a trade are stronger in crisis periods as the extent of the
on a given firm, namely, its credit rating and market capitalisation. information asymmetry between liquidity providers and informed
The credit quality dummy, IG, takes the value 1 when the firm has traders widens.
an investment grade rating. For market capitalisation we employ Higher uncertainty about the price of the CDS in periods of mar-
two dummies, L and S, which take value 1 when the firms fall in ket turmoil is also reflected in larger bid-ask spreads. We recognise
the large firm and small firm sub-samples respectively. These that a larger spread will necessarily imply a greater trade impact.
sub-samples correspond to the first and third terciles of the sample Indeed, the interaction between the signed trade variable Qt and
distribution stratified by the firms’ market capitalisation. the bid ask spread is positive and statistically significant in the cri-
As a measure of persistent trading effects on the CDS price we sis period. From a study of the equity market, Hasbrouck (1991)
use k, that is, the sum of the coefficients of the variable Signed Trade finds that companies with larger market capitalisation and higher
and its lags. We find k to be positive for model a and model b, over credit rating tend to have a lower persistent price impact of order
the whole sample as well as the pre-crisis and crisis sub-samples. flow. Hasbrouck concludes that blue chip stocks have lower
This is a strong indication of the presence of informed trading information asymmetry due to the larger quantity of information
available in the market for those firms. In agreement with these
12
This is a modified version of the traditional augmented Dickey–Fuller (ADF) test findings, we observe that CDS trades on large companies tend to
which is suitable for unbalanced panels. Before running the test, we subtract the have a lower price impact. But this correction is statistically signif-
cross-sectional averages from the series, as Levin et al. (2002) suggest that this icant only in the pre-crisis period. This suggests that the greater
procedure mitigates the impact of cross-sectional dependence.
5520 F. Corò et al. / Journal of Banking & Finance 37 (2013) 5511–5525

Table 7
Robustness tests.

Weekly data Monthly data

Pre-crisis Crisis Pre-crisis Crisis Pre-crisis Crisis
Constant 0.93*** 0.15 0.95*** 0.42*** 3.07*** 0.59
(9.38) (1.32) (9.50) (3.42) (8.72) (1.19)
DDTD 0.09 0.75* 0.05 0.91** 0.17 1.05
(0.29) (1.73) (0.19) (2.20) (0.36) (1.24)
DCR 1.15 13.35*** 1.1 13.88*** 3.4 0.19
(1.13) (2.94) (1.07) (3.26) (0.89) (0.03)
DLEV 0.11 1.69*** 0.17 1.44** 0.18 1.22***
(0.35) (2.82) (0.59) (2.59) (0.77) (3.37)
DTBAS 0.89*** 1.27*** 0.89*** 1.26*** 4.49*** 3.11***
(3.12) (10.79) (3.12) (10.59) (4.98) (10.01)
DP 0.14 0.16 0.14 0.17 0.14 0.19
(1.64) (1.25) (1.62) (1.33) (1.36) (1.39)
DTI 0 0.04 0 0.02 0.04* 0.05
(0.14) (0.97) (0.16) (0.6) (1.82) (1.43)
QI 0.20*** 0.22*** 0.20*** 0.21*** 0.14*** 0.17***
(7.17) (7.03) (7.28) (6.91) (5.75) (4.29)
DASY 2.9 2.58*** 2.87 2.18*** 2.46 5.81***
(1.51) (5.35) (1.47) (4.65) (1.36) (5.23)
DIBAS 0.18 1.22*** 0.2 1.09*** 2.55 0.43
(0.67) (7.97) (0.8) (7.31) (1.54) (1.13)
DEU YIELD 2.37* 38.40***
(1.79) (9.59)
DVSTOXX 0.15** 0.89***
(2.55) (7.16)
DUS YIELD 2.88** 22.79*** 5.83* 30.40***
(2.64) (8.28) (1.81) (9.92)
DVIX 0.06 1.39*** 0.47 0.91***
(1.05) (7.99) (1.09) (5.06)
Observations 2667 6088 2667 6088 781 2127
Adjusted R2 0.095 0.25 0.093 0.262 0.241 0.487

The dependent variable is the change in the mid-quote price. The explanatory variables are the change in the distance to default (DDTD), the change in leverage (DLEV), the
change in credit rating (DCR), the change in the time-weighted absolute bid-ask spread (DTBAS), the change in the number of executed trades (DTI), the demand pressure
(DP), the change in the industrial information asymmetries index (DASY), the change in the industry-wide bid-ask spread (DIBAS), the quote imbalance (QI), the change in
implied volatility of a benchmark stock index for the Euro area (DVSTOXX), the change 10-year Euro area government bond yield (DEU YIELD), the change in the VIX index
(DVIX), the change 10-year US government bond yield (DUS YIELD). Dependent variable and explanatory variables are sampled on a monthly basis. The pre-crisis period goes
from January 1, 2006 to April 21, 2007. The crisis periods is from April 22, 2007 to July 31, 2009. The panel estimation is carried out by using firm-level fixed-effects and panel-
robust standard errors (clustering at the firm level) to control for autocorrelation and heteroskedasticity. t-Statistics are given in parenthesis.
*
Significance at 10% level.
**
Significance at 5% level.
***
Significance at 1% level.

availability of information does not help in a crisis, either because
the quality of public information deteriorates or because the inside
information problem becomes more acute. Interestingly, during
the crisis it is credit quality rather than firm size that affects the
magnitude of trade impact. The negative and statistically signifi-
cant sign of the investment grade variable (interacted with the sign
trade variable) indicates that, in the crisis period, dealers revise
quotes more aggressively for sub-investment grade companies.
The finding implies that asymmetric information is more severe
in the case of low credit quality firms when the market is charac-
terised by higher uncertainty.
In Table 4 we report the estimation results for regression (2).
The full model in column i shows that the credit variable DLEV
and liquidity variables DTBAS, DASY, DIBAS and QI are statistically
significant and have the expected sign. Of the macro-variables,
DEU YIELD is strongly significant with a negative sign which is in
line with previous literature on credit spreads in the bond and
CDS markets. VSTOXX is strongly significant and with the correct
sign only when regressed without liquidity variables but becomes
insignificant when DIBAS is introduced in the regression
(unreported results). This suggests a strong relationship between
liquidity risk and macro-economic conditions. The positive coeffi-
cient of the credit variable DLEV confirms that higher default risk
is associated with an increase in CDS prices, as one would expect.
On the other hand, CDS price changes are also positively related to
illiquidity. Specifically, the positive coefficient of DTBAS (3.21) is
consistent with the findings of Pires et al. (2011).13 However, differ-
ently from Pires et al. (2011), we do not ascribe this effect to infor-
mation asymmetries alone, but to the interplay among transaction
costs, liquidity providers’ competition as well as information asym-
metries. The price impact of information asymmetries has been
widely investigated in the equity market but less so in the CDS mar-
ket. Easley et al. (2002) find that the probability of informed trading
(PIN) is a relevant factor in explaining cross-sectional equity returns.
On the other hand, Acharya and Johnson (2007) use bank relation-
ships as a proxy for informed trading and find it does not influence
CDS prices. These authors suggest employing intraday data and
microstructure models to shed light on the price impact of informed
trading in the CDS market. As indicated by the microstructure anal-
ysis conducted with regression (1) and reported in Table 3, indeed
asymmetric information measured with k, the trade impact of the or-
der flow, has a positive and statistically significant effect on CDS
prices. Unfortunately, thin trading does not allow us to estimate a
firm-specific trade impact variable for a sufficiently large number
of firms. However, we are able to build a k-based asymmetric infor-
mation proxy at the industry level. In Table 4 we show that the
industry aggregate asymmetric information variable, DASY, plays a
significant role in explaining CDS price dynamics. Importantly, the
13
Pires et al. (2011) propose a quantile regression to analyse the determinants of
the CDS spread. However, while they focus on levels, we consider first differences as
we detect non-stationarity in our variables during the crisis period.
 F. Corò et al. / Journal of Banking & Finance 37 (2013) 5511–5525 5521

Table 8
Credit and liquidity determinants of credit spreads in periods of high and low risk.

Bid-ask spread CDS Vol CDS Price

Low High Low High Low High
Constant 0.299 1.668** 1.309*** 0.886 0.915*** 1.171
(1.174) (2.293) (4.799) (1.147) (4.148) (1.633)
DDTD 0.226 2.966** 0.128 1.825 0.262 3.009**
(0.500) (2.141) (0.320) (1.419) (0.617) (2.134)
DCR 2.282 2.291 0.548 2.784 1.885 2.905
(0.732) (0.276) (0.192) (0.318) (0.606) (0.366)
DLEV 0.077 1.957*** 0.004 1.919*** 0.036 2.038***
(0.680) (3.255) (0.014) (3.293) (0.285) (3.263)
DTBAS 3.798*** 3.253*** 3.487*** 3.234*** 3.799*** 3.257***
(5.107) (9.978) (9.398) (9.321) (5.262) (9.979)
DP 0.042 0.285 0.033 0.193 0.045 0.287
(0.568) (1.082) (0.462) (0.826) (0.627) (1.040)
DTI 0.029 0.174** 0.028* 0.065 0.035* 0.126*
(1.382) (2.621) (1.823) (1.001) (1.843) (1.768)
QI 0.055*** 0.196*** 0.107*** 0.155*** 0.056*** 0.213***
(2.900) (3.420) (5.776) (2.902) (3.264) (3.752)
DASY 3.142 7.092*** 1.432 7.462*** 4.484 7.250***
(1.111) (5.990) (0.848) (5.833) (1.547) (6.123)
DIBAS 3.820*** 2.142*** 5.301*** 2.125*** 4.041*** 2.160***
(4.467) (4.438) (9.633) (4.241) (4.772) (4.358)
DEU YIELD 2.698 26.613*** 2.646* 28.173*** 3.388** 24.793***
(1.621) (6.196) (1.779) (5.587) (2.237) (5.783)
DVSTOXX 0.049 0.051 0.270* 0.039 0.070 0.066
(0.296) (0.234) (1.841) (0.180) (0.422) (0.291)
Observations 1439 1469 1439 1469 1439 1469
Adjusted R2 0.480 0.474 0.551 0.466 0.509 0.474

The table exhibits the estimation results of regression (2) on low/high risk sub-samples created with three market-wide risk indicators (average bid-ask spread, average CDS
price and average CDS price volatility). The high/low risk sub-samples are built by grouping firm-level observations according to whether they are above or below the median
value of their market-wide averages. The dependent variable is the change in the mid-quote price. The explanatory variables are the change in the distance to default (DDTD),
the change in leverage (DLEV), the change in credit rating (DCR), the change in the time-weighted absolute bid-ask spread (DTBAS), the change in the number of executed
trades (DTI), the demand pressure (DP), the change in the industrial information asymmetries index (DASY), the change in the industry-wide bid-ask spread (DIBAS), the
quote imbalance (QI), the change in implied volatility of a benchmark stock index for the Euro area (DVSTOXX) and the change 10-year Euro area government bond yield
(DEU YIELD). Dependent variable and explanatory variables are sampled on a monthly basis. The sample period is from January 1, 2006 to July 31, 2009. The panel estimation
is carried out by using firm-level fixed-effects and panel-robust standard errors (clustering at the firm level) to control for autocorrelation and heteroskedasticity. t-Statistics
are given in parenthesis.
*
Significance at 10% level.
**
Significance at 5% level.
***
Significance at 1%, level.

positive and significant coefficients of DIBAS as well as DASY show
the strong linkages between aggregate market liquidity and CDS
prices. This confirms the findings of Tang and Yan (2007) who also
detect commonality in liquidity in the CDS market. The fact that
the demand pressure is not statistically significant is puzzling. The
failure to detect demand pressure effects on CDS prices could depend
on the fact that DP is approximated using the difference between the
number of buy and the number of sell trades and not actual order
flows. This would be a noisy estimator of demand pressure if there
is a great deal of variation in the size of buy and sell trades.
An analysis of the goodness of fit of regression (2) when one
liquidity variable is added in turn to the credit variables (models
from b to g in Table 4) suggests that the firm-specific and indus-
try-wide bid-ask spreads have the greatest explanatory power
(+34.1% and +13.7% adjusted R-squared respectively), followed by
the asymmetric information variable DASY (+2.7%). This confirms
the prominence of the aggregate liquidity measures and the impor-
tance of taking their effect into account when pricing CDS con-
tracts. On the other hand, the combined explanatory power of
the credit variables, DDTD, DCR and DLEV, appears to be low in
our sample as their adjusted R-squared is only 4.5%. This suggests
that the common practice of interpreting changes in CDS prices as
mainly driven by changes in default risk may be misleading and
could generate substantial errors in the assessment of the credit
risk of individual assets especially in periods of high market
uncertainty. Indeed, when we repeat the analysis on the pre-crisis
and crisis sub-samples (reported in Table 5) our conclusions are
broadly confirmed. The combined explanatory power of the credit
variables is only marginally higher during the crisis, and still sub-
stantially lower than that of the liquidity proxies based on firm-
specific and industry-wide bid-ask spreads.
The results for the pre-crisis and crisis analysis reported in Ta-
ble 5 are striking. Before the crisis (Panel A, model i), none of the
credit variables as well as industry-wide liquidity variables, DASY
and DIBAS, are statistically significant. During the crisis (Panel B,
model i), leverage, DLEV, the industry bid-ask spread, DIBAS, and
the industry asymmetric information measure, DASY, and the level
of interest rate, DEU YIELD, become strongly significant. In partic-
ular, the increased significance of industry-wide variables and the
macro-variable highlights the presence of common factors in CDS
spreads during the crisis. This is also confirmed when we compute
the variance explained by the first principal component of CDS
price changes, which doubles from the pre-crisis period (29.4%)
to the crisis period (60.7%). Overall, the results for the whole sam-
ple appear to be mainly driven by the crisis. In addition, in the cri-
sis period the firm specific asymmetric information indicator DTI is
also strongly significant (while it is only mildly significant before
the crisis). These results clearly suggest that informed trading be-
comes particularly pervasive during the crisis, both at the firm le-
vel and at the industry level. Indeed, adverse selection costs due to
asymmetric information become more of a concern for dealers dur-
ing the crisis when default losses are more likely to materialise.
The finding is consistent with the well-known adverse selection
problem in insurance markets. Acharya and Johnson (2007) aptly
state that ‘‘[t]he threat of informed purchase of [default] insurance
leads to a lemon’s problem in which insurance premia are set too
5522 F. Corò et al. / Journal of Banking & Finance 37 (2013) 5511–5525
high and the quantity of insurance written in equilibrium is too
low’’. It is not surprising that the credit variables are not significant
before the crisis as CDS spreads and credit spreads in the bond
market were ‘‘artificially’’ low in that period.14 Changes in credit
spreads seamless sensitive to their fundamental drivers when mar-
ket conditions are benign. Interestingly, the amplified sensitivity of
CDS price changes to risk factors in crisis periods has also been de-
tected in the sovereign CDS market, as shown by Fender et al (2012).
5. Robustness tests
We run a series of tests to check the robustness of our findings.
In Table 6 we report a panel regression on the whole sample as in
Table 4 but with the addition of interaction effects for all explana-
tory variable with a crisis dummy. This enables us to check
whether there is a statistically significant change in the value of
the regression coefficients during the crisis period. Again we con-
firm that the influence of leverage and the industry-wide liquidity
variables on CDS prices changes increases markedly as markets be-
come more volatile. In addition the negative relationship between
CDS price change and the level of interest rates in the Euro area
also becomes stronger. Furthermore, as pointed out by Fender
et al. (2012) lower frequency data tend to exhibit higher correla-
tions than higher frequency data. This may lead one to give empha-
sis to relationships that may not be as strong when looking at
variables defined over shorter horizons. Therefore, we re-compute
our main regressions by using weekly data. Results are reported in
Table 7. Indeed the explanatory power of our covariates falls but all
main relationships and their significance remain broadly unaltered
before and during the crisis.15 In addition, following Fender et al.
(2012) we consider the impact of global macro-economic factors
on CDS price changes. To capture these effects we use the stock vol-
atility in the US market, proxied by the VIX index, and the level of US
interest rates, with both weekly and monthly data. Results are also
reported in Table 7. Again the significance of the credit and liquidity
variables before and during the crisis is confirmed and the explana-
tory power of the regressions remains broadly similar.
The crisis was characterised by periods of acute market stress as
well as phases of relative calmness where CDS prices and bid-ask
spreads fell considerably from their peaks. Additionally, there is
no general consensus on the starting date of the crisis. To assess
the robustness of our findings with respect to our definition of
the crisis period we consider alternative approaches for separating
phases of turmoil and relative calm. In particular, we split the sam-
ple in periods with high and low market uncertainty, where uncer-
tainty is measured using three market-wide variables: the average
CDS price level, average CDS price volatility and average CDS bid-
ask spread. As in Beber et al. (2009), high/low uncertainty sub-
samples are built by grouping firm-level observations according
to whether they are above or below the median value of their mar-
ket-wide averages. The results are shown in Table 8. Our previous
conclusions are confirmed. As during the crisis, leverage and asym-
14
Acharya et al. (2009), pp. 12–13, report that ‘‘There is almost universal agreement
that the fundamental cause of the crisis was the combination of a credit boom and a
housing bubble . . . [T]here was just a fundamental mispricing in capital markets – risk
premiums were too low and long term volatility reflected a false belief that future
short term volatility would stay at its current low levels. This mispricing necessarily
implied low credit spreads and inflated prices of risky assets.’’
15
We do not perform our analysis on daily data as in Fender et al. (2012) because of
the low frequency of our trades for several of the names in our sample. Selecting only
those firms for which we have trades on a daily basis consistently over the whole
sample would drastically reduce the number of observations and impair the
meaningfulness of our analysis. The alternative of retaining all firms regardless of
trading activity will lead to a highly unbalanced panel where relationships are driven
at any point in time by a constantly changing pool of firms. This again would not
allow us to capture the fundamental relationships among covariates and CDS price
changes due to the high level of noise.
metric information effects are always stronger and highly signifi-
cant in the high uncertainty period. The firm specific bid-ask
spread variable (DTBAS) remains a critical determinants of CDS
price changes, as it is significant in all sub-samples, alongside the
industry-wide bid-ask spread (DIBAS).
6. Conclusion
In this paper, we explore the credit and liquidity determinants of
CDS price movements and investigate how their role changed as a
result of the 2007–2009 ‘‘Great Recession,’’ by many indicators
the worst period of market turmoil since the Great Depression. In
general, we observe that liquidity effects dominate CDS price vari-
ations. Although firm-specific credit risk effects are very significant
during the crisis, they have a lower explanatory power than liquid-
ity effects, a confirmation of the commonly held view that the
severity of the crisis was primarily due to liquidity factors. This sug-
gests that CDS price changes may not be accurate indicators of
changes in default risk, even in periods of high uncertainty, contrary
to accepted wisdom in the industry. Interestingly, liquidity effects
are the only ones that preserve a statistically significant explana-
tory power in the pre-crisis period. Macro-economic effects also
play a role although a minor one. Banks, insurance companies, reg-
ulators and rating agencies closely monitor CDS price changes to as-
sess firm specific and market-wide variations in default risk. In light
of our findings, these industry players may need to rethink how
they interpret the signals they receive from the CDS market.
In addition, we find that informed trading has a more promi-
nent impact on CDS prices during the crisis, which indirectly sup-
ports concerns that insider information may influence CDS price
formation (Acharya and Johnson, 2007). In this context, our paper
is one of the first to use intraday data to measure the level of asym-
metric information in the CDS market. We observe that the order
flow has a significant impact on CDS prices and its extent varies
depending on the sample period, firm characteristics and the size
of the bid-ask spread. In particular, the trade impact increases after
the onset of the financial crisis for CDS contracts with higher bid-
ask spread and it decreases for high credit quality firms. Before
the crisis, on the other hand, the size of the firm appears to affect
the price impact, with larger firms attracting a lower impact. This
is in line with the view that public information is more readily
available for larger firms. However, during the crisis this informa-
tional advantage seems to vanish. We conjecture that the result
may be due to the lower quality of public information during a cri-
sis or the greater severity of the asymmetric information problem.
Our results highlight the importance of systematic liquidity, espe-
cially in the Great Recession period. This finding corroborates sim-
ilar evidence provided for the stock market (Hasbrouck and Seppi,
2001). This lends support to the great emphasis placed by regula-
tors and governments on tackling market illiquidity, for example,
by increasing transparency in the CDS market, via central clearing
arrangements, and through the standardisation of CDS contracts.
An interesting avenue of future research would be an assessment
of the impact of these new policy measures and of their ability to
ensure the operation of financial markets in periods of sustained stress.
Acknowledgements
We thank Nadia Benbouzid, Jason Wei, Paola Zerilli, and the
referee for their suggestions. We are grateful for the comments
received by participants at the 2012 International Conference of
the Financial Engineering and Banking Society, London; the
2012 International Risk Management Conference, Rome; the
2012 International Finance and Banking Society (IFABS) Confer-
ence, Valencia; and the 2012 Northern Finance Association meet-
ing, Niagara Falls.
 F. Corò et al. / Journal of Banking & Finance 37 (2013) 5511–5525 5523

Appendix A

See Tables A.1–A.3.

Table A.1
Credit rating comparative table.

Score Fitch S&P Moody’s

1 AAA AAA AAA
2 AA+ AA+ AA1
3 AA AA AA2
4 AA AA AA3
5 A+ A+ A1
6 A A A2
7 A A A3
8 BBB+ BBB+ Baa1
9 BBB BBB Baa2
10 BBB BBB Baa3
11 BB+ BB+ Ba1
12 BB BB Ba2
13 BB BB Ba3
14 B+ B+ B1
15 B B B2
16 B B B3
17 CCC+ CCC+ Caa1
18 CCC CCC Caa2
19 CCC CCC Caa3
20 CC CC Ca
21 C C
22 D D C
5524 F. Corò et al. / Journal of Banking & Finance 37 (2013) 5511–5525

Table A.2
Summary statistics and pair-wise correlations of credit and liquidity variables.

Mean Median Std. dev. Min Max Obs.

Panel A: Summary statistics
CDS (bp) 100.6 64.0 114.1 2.8 1000.0 2908
DTD 6.3 5.6 4.9 1.8 26.7 2908
LEV 0.6 0.6 0.2 0.1 1.0 2908
CR 7.1 8.0 2.5 1.0 13.0 2908
TBAS (bp) 7.5 4.0 13.7 0.4 250.0 2908
TI 9.2 4.0 15.4 0.0 260.0 2908
DP 0.7 0.0 4.0 50.0 32.0 2908
ASY 0.4 0.3 0.7 2.4 2.9 2908
IBAS (bp) 11.9 7.2 9.6 2.4 35.9 2908
QI 10.6 8.0 15.9 123.0 133.0 2908
STOXX (%) 0.9 0.5 5.5 18.1 13.2 41
VSTOXX 26.3 22.7 12.5 14.2 63.3 41
EU YIELD (%) 4.1 4.1 0.3 3.7 4.7 41

D CDS DDTD DLEV DCR DTBAS DTI DP DASY DIBAS QI DVSTO STOXX
Panel B: Cross-sectional time series pair-wise correlations
Whole sample
DDTD 0.15*
DLEV 0.20* 0.37*
DCR 0.05 0.04 0.05
DTBAS 0.62* 0.09* 0.10* 0.08*
DTI 0.03 0.02 0.06* 0.00 0.02
DP 0.03 0.00 0.02 0.01 0.02 0.02
DASY 0.17* 0.03 0.01 0.01 0.05 0.01 0.00
DIBAS 0.42* 0.23* 0.25* 0.00 0.31* 0.07* 0.02 0.01
QI 0.12* 0.05* 0.04 0.01 0.05* 0.15* 0.01 0.02 0.15*
* * *
DVSTOXX 0.27 0.23 0.29 0.01 0.18* 0.11* 0.01 0.06* 0.72* 0.18*
* * *
STOXX 0.41 0.30 0.35 0.01 0.19* 0.08* 0.05 0.03 0.60* 0.14* 0.71*
DEU YIELD 0.27* 0.10* 0.05* 0.02 0.17* 0.02 0.04 0.10* 0.38* 0.06* 0.09* 0.20*
Pre-crisis
DDTD 0.02
DLEV 0.03 0.56*
DCR 0.10* 0.11* 0.11*
DTBAS 0.41* 0.01 0.04 0
DTI 0.01* 0.01 0 0 0.11*
DP 0.08 0 0.06 0.02 0.02 0.07
DASY 0.01 0.06 0.08 0.03 0.04 0.08 0.02
DIBAS 0.16* 0.14* 0.17* 0 0.21* 0.09 0.05 0.13*
QI 0.18* 0.03 0.06 0.07 0.03 0.24* 0.06 0.12* 0.24*
DVSTOXX 0.12* 0.26* 0.29* 0.05 0.13* 0.10* 0.04 0.21* 0.68* 0.24*
STOXX 0.16* 0.26* 0.29* 0.05 0.16* 0.03 0 0.16* 0.79* 0.27* 0.90*
DEU YIELD 0.10* 0.10* 0.05 0.04 0.13* 0.04 0.04 0 0.54* 0.09 0.05 0.06
Crisis
DDTD 0.19*
DLEV 0.21* 0.33*
DCR 0.06* 0.01 0.03
DTBAS 0.62* 0.12* 0.10* 0.09*
DTI 0.04 0.03 0.11* 0 0.03
DP 0.04 0.01 0 0.01 0.02 0.02
DASY 0.17* 0.04 0.01 0.01 0.05 0.02 0
DIBAS 0.42* 0.28* 0.26* 0 0.30* 0.11* 0.02 0.01
QI 0.13* 0.07* 0.05 0.01 0.06* 0.12* 0.05 0.02 0.17*
DVSTOXX 0.28* 0.26* 0.29* 0.01 0.18* 0.16* 0.01 0.07* 0.73* 0.18*
STOXX 0.42* 0.33* 0.35* 0.01 0.19* 0.14* 0.05 0.03 0.61* 0.13* 0.71*
DEU YIELD 0.28* 0.16* 0.06* 0.04 0.18* 0.02 0.07* 0.11* 0.40* -0.05 0.11* 0.21*

The table shows summary statistics for CDS prices and credit and liquidity determinants. All variables are computed on a monthly basis. CDS is the mid-quote CDS price; DTD
is the distance to default; LEV is leverage; CR is credit rating; TBAS is the time-weighted absolute bid-ask spread; ASY is an industry-wide asymmetric information index; TI is
the number of executed trades; DP is the demand pressure; IBAS is the cross sectional bid-ask spread for the industry sectors considered in our sample (that is financials,
consumer goods, telecommunications and consumer services), QI is the quote imbalance, STOXX is a benchmark stock index for the Euro area, VSTOXX is the stock index’s
implied volatility and EU YIELD is the 10-year Euro area government bond yield. The sample period is January 1, 2006 to July 31, 2009.The pre-crisis period goes from January
1, 2006 to April 21, 2007. The crisis period is from April 22, 2007 to July 31, 2009.
*
Significance at the 1% level.
 F. Corò et al. / Journal of Banking & Finance 37 (2013) 5511–5525 5525

Table A.3 Bongaerts, D., de Jong, F., Driessen, J., 2011. Derivative pricing with liquidity risk:
Panel unit root test. theory and evidence from credti default swap market. The Journal of Finance,
203–240.
Inverse normal Inverse logit Brennan, M.J., Subrahmanyam, A., 1996. Market microstructure and asset pricing:
Statistic p-value Statistic p-value on the compensation for illiquidity in stock returns. Journal of Financial
Economics 41, 441–464.
CDS 0.80 0.79 0.98 0.84
Brunnermeier, M.K., Pedersen, L.H., 2009. Market liquidity and funding liquidity.
DCDS -6.51 0.00 -6.36 0.00 Review of Financial Studies 22 (6), 2201–2238.
Buhler, W., Trapp, M., 2008. Time-Varying Credit Risk and Liquidity Premia in Bond
DTD 0.53 0.70 0.37 0.65 and CDS Markets. Working Paper.
DDTD -7.54 0.00 -7.53 0.00 Cameron, A.C., Trivedi, P.K., 2005. Microeconometrics Methods and Applications.
Cambridge University Press.
BAS 4.99 1.00 4.73 1.00 Chen, R.R., Cheng, X., Wu, L., 2005. Dynamic Interactions Between Interest Rate,
DBAS -7.93 0.00 -8.55 0.00 Credit, and Liquidity Risks: Theory and Evidence from the TermStructure of
Credit Default Swap Spreads. Working Paper.
LEV 2.43 0.99 2.30 0.99 Chen, L., Lesmond, D.A., Wei, J., 2007. Corporate yield spreads and bond liquidity.
DLEV -6.41 0.00 -6.63 0.00 The Journal of Finance 62 (1), 119–149.
Choi, I., 2001. Unit root tests for panel data. Journal of International Money and
Finance 20 (2), 249–272.
CR 5.14 1.00 5.50 1.00
Chordia, T., Subrahmanyam, A., 2004. Order imbalance and individual stock returns:
DCR -3.42 0.00 -4.55 0.00
Theory and evidence. Journal of Financial Economics 72 (3), 485–518.
Chordia, T., Roll, R., Subrahmanyam, A., 2000. Commonality in liquidity. Journal of
TI -0.66 0.25 -0.69 0.24 Financial Economics 56 (1), 3–28.
DTI -28.18 0.00 -38.65 0.00 Chordia, T., Roll, R., Subrahmanyam, A., 2002. Order imbalance, liquidity, and
market returns. Journal of Financial Economics 65 (1), 111–130.
ASY -13.97 0.00 -16.61 0.00 Collin-Dufresne, P., Goldstein, R.S., Martin, J.S., 2001. The determinants of credit
DASY -28.64 0.00 -50.05 0.00 spread changes. The Journal of Finance 56 (6), 2177–2207.
Copeland, T.E., Galai, D., 1983. Information effects on the bid-ask spread. The
DP -14.31 0.00 -15.77 0.00 Journal of Finance 38 (5), 1457–1469.
Coval, J., Stafford, E., 2007. Asset fire sales (and purchases) in equity markets.
IBAS 4.24 1.00 3.48 1.00 Journal of Financial Economics 86 (2), 479–512.
DIBAS -18.16 0.00 -20.08 0.00 Crosbie, P., 2003. Modeling Default Risk. Moody’s KMV.
Dufour, A., Engle, R.F., 2000. Time and the price impact of a trade. The Journal of
Finance 55 (6), 2467–2498.
QI -11.81 0.00 -12.78 0.00
Easley, D., O’Hara, M., 1987. Price, trade size, and information in securities markets.
Journal of Financial Economics 19 (1), 69–90.
VSTOXX 1.48 0.93 1.46 0.93 Easley, D., O’Hara, M., 1992. Time and the process of security price adjustment. The
DVSTOXX -14.12 0.00 -14.58 0.00 Journal of Finance 47 (2), 577–605.
Easley, D., Kiefer, N., O’Hara, M., Paperman, J.B., 1996. Liquidity, information, and
EU YIELD -3.15 0.00 -2.82 0.00 infrequently traded stocks. The Journal of Finance 51, 1405–1436.
DEU YIELD -12.59 0.00 -12.48 0.00 Easley, D., Hvidkjaer, S., O’Hara, M., 2002. Is information risk a determinant of asset
returns? The Journal of Finance 57, 2185–2221.
The table shows the Fisher-ADF unit root test on the dependent variable and Ericsson, J., Jacobs, K., Oviedo, R., 2009. The determinants of credit default swap
covariates of regression (2) when the variables are considered in levels and first Premia. Journal of Financial and Quantitative Analysis 44 (1), 109–132.
differences. As in Choi (2001) we consider two test statistics, the invers normal and Fender, I., Hayo, B., Neuenkirch, M., 2012. Daily pricing of emerging market
the inverse logit. CDS is the mid-quote CDS price; DTD is the distance to default; sovereign CDS before and during the global financial crisis. Journal of Banking
LEV is leverage; CR is credit rating; TBAS is the time-weighted absolute bid-ask and Finance 36 (10), 2786–2794.
spread; ASY is an industry-wide asymmetric information index; TI is the number of Garleanu, N., Pedersen, L.H., Poteshman, A., 2009. Demand-based option pricing.
executed trades; DP is the demand pressure; IBAS is the cross sectional bid-ask Review of Financial Studies 22 (10), 4259–4299.
spread for the industry sectors considered in our sample (that is financials, con- Glosten, L.R., Harris, L.E., 1988. Estimating the components of the bid/ask spread.
sumer goods, telecommunications and consumer services), QI is the quote imbal- Journal of Financial Economics 21 (1), 123–142.
Glosten, M., Milgrom, P., 1985. Bid, ask and transactions prices in a specialist market
ance, VSTOXX is the stock index’s implied volatility and EU YIELD is the 10-year
with heterogeneously informed traders. Journal of Financial Economics, 71–
Euro area government bond yield. For DP and QI first differences are not considered
100.
as the variables are employed in levels in subsequent regression analyses.
Greenwood, R., Vayanos, D., 2010. Price pressure in the government bond market.
American Economic Review 100 (2), 585–590.
References Hasbrouck, J., 1991. Measuring the information content of stock trades. The Journal
of Finance 46 (1), 179–207.
Hasbrouck, J., Seppi, D., 2001. Common factors in prices, order flows, and liquidity.
Acharya, V., Johnson, T., 2007. Insider trading in credit derivatives. Journal of Journal of Financial Economics 59, 383–411.
Financial Economics 84, 110–141. Ho, T., Stoll, H.R., 1981. Optimal dealer pricing under transactions and return
Acharya, V., Pedersen, L.H., 2005. Asset pricing with liquidity risk. Journal of uncertainty. Journal of Financial Economics 9 (1), 47–73.
Financial Economics 77, 375–410. ISDA., 2003. ISDA 2003 Credit Derivatives Definitions. Available at: http://
Acharya, V.V., Philippon, T., Richardson, M., Roubini, N., 2009. The financial crisis of www.isda.org/publications/isdacredit-deri-def-sup-comm.aspx.
2007–2009: causes and remedies. In: Acharya, V.V., Richardson, M. (Eds.), Kyle, A., 1985. Continuous auctions and insider trading. Econometrica 53, 1315–
Restoring Financial Stability – How to Repair a Failed System. Wiley Finance, 1335.
Hoboken, New Jersey. Levin, A., Lin, C.F., Chu, S.J., 2002. Unit root tests in panel data: asymptotic and
Alexopoulou, I., Andersson, M., Georgescu, O.M., 2009. An Empirical Study on the finite-sample properties. Journal of Econometrics 108 (1), 1–24.
Decoupling Movements between Corporate Bond and CDS Spreads. ECB Longstaff, F.A., Schwartz, E.S., 1995. A simple approach to valuing risky fixed and
Working Paper Series No. 1085. floating rate debt. The Journal of Finance 50 (3), 789–819.
Amihud, Y., Mendelson, H., 1980. Dealership market: Market-Making with Longstaff, F.A., Mithal, S., Neis, E., 2005. Default risk or liquidity? New evidence from
inventory. Journal of Financial Economics 8 (1), 31–53. the credit default swap market. The Journal of Finance 60 (5), 2212–2253.
Amin, K., Coval, J.D., Seyun, H.N., 2004. Index option prices and stock market Merton, R.C., 1974. On the pricing of corporate debt: The Risk structure of interest
momentum. Journal of Business 77 (4), 835–873. rates. Journal of Finance 29 (2), 449–470.
Beber, A., Brandt, M.W., Kavajecz, K., 2009. Flight-to-quality or flight-to-liquidity? Pires, P., Pereira, J.P., Martins, L.F., 2011. The Complete Picture of Credit Default
Evidence from the euro-area bond market. Review of Financial Studies 22, 925– Swap Spreads – A Quantile Regression Approach. Working Paper.
957. Tang, D.Y., Yan, H., 2007. Liquidity and Credit Default Swap Spread. Working Paper.
Blanco, R., Brennan, S., Marsh, I.W., 2005. An empirical analysis of the dynamic Tang, D.Y., Yan, H., 2010. Does the Tail Wag the Dog? The Price Impact of CDS
relation between investment-grade bonds and credit default swaps. The Journal Trading. Working Paper.
of Finance 60 (5), 2255–2281. Vassalou, M., Xing, Y., 2004. Default risk in equity returns. The Journal of Finance 59
Bollen, N.P.B., Whaley, R.E., 2004. Does net buying pressure affect the shape of (2), 831–868.
implied volatility function? The Journal of Finance 59 (2), 711–753.