v_acharya

Insider Trading in Credit Derivatives





Viral V. Acharya and Timothy C. Johnson1





DRAFT





May, 2005









1 Both authors are at London Business School, Regent’s Park, London - NW1 4SA, United

Kingdom. Tel: +44 (0)20 7262 5050, Fax: +44 (0)20 7724 3317, e-mails: vacharya@london.edu,

tjohnson@london.edu. Viral Acharya is also a Research Affiliate of the Centre for Economic Policy

Research (CEPR). A part of this work was undertaken while Timothy Johnson was visiting MIT.

Pedro Saffi and Jason Sturgess provided research assistance. We are indebted to Sreedhar Bharath,

Anand Srinivasan, Amir Sufi, and Ilya Strebulaev for sharing data with us, and to the Grant

Approval Committee at Fondation Banque de France, seminar participants at Goldman Sachs

Asset Management (GSAM), David Goldreich, Jose Liberti, and Lucie Tepla for useful inputs. We

acknowledge financial support in the form of a grant for this project from Fondation Banque de

France. This is a preliminary draft. Comments are welcome.

Abstract



Insider trading in the credit derivatives market has become a significant concern for regula-

tors and participants. This paper attempts to quantify the problem. Using news reflected

in the stock market as a benchmark for public information, we report evidence of signifi-

cant incremental information revelation in the CDS market, consistent with the occurrence

of insider trading. We show that the degree of this activity increases with the number of

banks having lending/monitoring relations with a given firm, and is robust to controls for

non-informational trade. Furthermore, information revelation in the CDS market is asym-

metric, consisting exclusively of bad news, consistent with hedging activity by banks with

loan exposure and private information. We find no evidence, however, that the degree of

insider activity adversely affects prices or liquidity in either the equity or credit markets. If

anything, the reverse appears to be true.







Keywords: adverse selection, default, bank relationship, credit default

swaps, asset pricing.

JEL CLASSIFICATIONS: G12, G13, G14, G20, D8

‘ ‘[B]anks must not use private knowledge about corporate clients to trade instruments

such as credit default swaps (CDS), says a report drawn up by five bodies including the

International Swaps and Derivatives Association and the Loan Market Association... The

warning highlights the challenges credit derivatives pose to banks and regulators trying to

build a functioning market infrastructure... [M]any banks and institutions are trading CDS

instruments in the same companies they finance - sometimes because they want to reduce

the risks to their own balance sheets.” (Financial Times, April 25, 2005 - ‘Banks warned on

insider trading threat posed by market for credit derivatives’)







1 Introduction

Credit derivatives have been perhaps the most important and successful financial innovation

of the last decade. The use of credit derivatives has been suggested as an important reason

for the observed robustness of banks and financial institutions to the historically high levels

of corporate defaults all over the globe during the period 2000 to 2004.1 In addition, markets

for credit derivatives have helped banks create synthetic liquidity in their otherwise illiquid

loan portfolios.2 Not surprisingly, the growth in the size of this market remains unabated

as products are expanding to emerging markets, and indices such as iBoxx and iTraxx are

becoming industry benchmarks for credit conditions.

If credit derivatives are to provide seamlessly the insurance and liquidity-creation roles

for banks, then the liquidity of credit derivatives itself becomes a desirable feature of mar-

1

For instance, Alan Greenspan, Chairman of the Federal Reserve of the United States, in his speech

“Economic Flexibility” before Her Majesty (HM)’s Treasury Enterprise Conference (London, 26 January

2004), mentions with reference to credit derivatives: “The new instruments of risk dispersion have enabled

the largest and most sophisticated banks in their credit-granting role to divest themselves of much credit risk

by passing it to institutions with far less leverage. These increasingly complex financial instruments have

contributed, especially over the recent stressful period, to the development of a far more flexible, efficient,

and hence resilient financial system than existed just a quarter-century ago. One prominent example was

the response of financial markets to a burgeoning and then deflating telecommunications sector.”

2

In an important recent example, Citigroup had distributed a large portion of its exposure to Enron

through issuance of credit-linked notes at regular intervals (in the two-year period preceding default of

Enron). The final effect of Enron’s collapse on the balance sheet of Citigroup was as a result small compared

to the size of its loan exposures to Enron.







1

kets. Credit derivatives however, like all forms of insurance, are subject to moral hazard (see,

Duffee and Zhou, 2001) and asymmetric information risks. In this paper, we are concerned

with the latter of these risks. Specifically, if some creditor of Company X has private infor-

mation about the likelihood of default, or can itself influence default, then this creditor may

try to exploit its privileged information by buying credit insurance from a less well-informed

counterparty. Or if loan officers, who deal directly with a bank’s borrowers, pass on inside

information to the traders buying credit derivatives, the institution on the other side of the

trade may get a rotten deal.

Indeed, some recent episodes in the credit derivatives markets reveal that this issue may

be real with potentially important implications for the efficiency of credit-risk transfer across

institutions. In striking recent episodes, managers of Pacific Investment Management Co.

(PIMCO), the largest bond investor in the U.S., have cited several cases of insider trading

in credits such as Household International Inc., AT&T Wireless, and Sprint Corp.: “Credit

default markets are a mechanism with which friendly commercial bankers ... can profit by

betraying and destroying their clients through the use of inside information,” and “... firms

with large lending departments would always come in and buy protection at exactly the

right moment.”3 The underlying hypothesis behind these complaints is that first, the price

at which default protection is bought conveys potentially valuable inside information about

likelihood of default, and second, these prices may get affected also by temporary buying

pressures. This has raised the interesting possibility that the actual default of a corporation

and thus the prices of its securities are themselves affected by activities in credit derivatives

markets, in particular, by insider or strategic trading.

Of course, insider trading problems potentially exist in most markets. But the credit

derivatives market may be especially vulnerable since, almost by definition, most of the

3

Such events have been acknowledged often in press. See, for example, the article in The Economist,

“Pass the Parcel – Credit Derivatives,” January 18, 2003. A similar observation has also been made with

respect to potentially a predatory form of trading by hedge funds in the credit derivatives market: “It was

the hedge funds creating a credit event by forcing the bond price down and trying to get the rating agencies

to downgrade the company to benefit themselves: they were scaring everyone into selling. (Henry Snedel,

December 5, 2002, in Deals and Deal Makers, Wall Street Journal).









2

major players are insiders. A large number of banks and financial institutions act as inter-

mediaries: they quote prices for credit derivatives written on corporations to which they have

loan exposures.4 From an academic standpoint, this renders the credit derivatives market a

particularly attractive laboratory for testing of hypotheses pertaining to insider-trading risk:

first, the potentially informed players are well-identified, namely the banks; second, the na-

ture of private information is unambiguous, specifically the identification of credit risk; and,

third, unlike the cash market, daily data on prices of most widely traded credit derivatives

is available for the past four years.

Given this motivation, we study empirically the market for trading in credit default

swaps (CDS), the most active credit derivative instrument, in order to answer the following

question: Is there evidence of actual insider trading in credit derivatives markets?

We use data on the CDS levels and bid-ask spreads for a cross-section of North American

firms over the period January 2001 through October 2004, and on the banking relation-

ships of these firms. Using news reflected in the stock market as a benchmark for public

information, we report evidence of significant incremental information revelation in the CDS

market, consistent with the occurrence of insider trading. We show that the degree of this

activity increases with the number of banks having lending (monitoring) relations with a

given firm. Furthermore, this finding is robust to controls for non-informational trade. Cru-

cially, information revelation in the CDS market is asymmetric, consisting exclusively of bad

news, consistent with hedging activity by banks with loan exposure and private information.

Finally, we show that the information flow from the CDS market to the stock market is

robust when studied over sub-samples where we expect a priori that insider trading would

be a more significant issue: entities that experience credit deterioration during the sample

4

The fact that banks make money from their customers by reselling the default swaps to other customers,

including other banks, is now well-recognized. For example, Bloomberg Markets Credit Swaps, High Risks

Few Rules, reports that on June 6, 2003 J.P.Morgan was offering $209,000 to buy credit default protection

on a $10 million loan to Altria Group Inc.’s Philip Morris tobacco unit and was offering to sell that same

contract to anyone for $221,000 – a difference of 5 percent - according to Bloomberg data. On the same day,

J.P.Morgan was offering to pay $9,000 to sellers of protection on $10 million of newspaper publisher Gannett

Co.’s debt while charging $21,000 to anyone who wanted to buy the same protection, a difference of more

than 100 percent.







3

period, whose CDS levels are generally high, and whose credit ratings are low.

If insider trading does exist in these markets, then it is possible that market makers will be

less willing to make prices in these derivatives in situations where they perceive the likelihood

of private information to be high. In particular, in volatile conditions or when default risks

rise, the risk of insider trading may rise, resulting in a loss of liquidity precisely when hedging

needs are greatest. To investigate this, we also study whether liquidity providers in the CDS

markets fear insider trading, and, in turn, charge greater bid-ask spreads for credits and in

times where insider-trading risk is greater? That is, should insider-trading risk in the CDS

markets be a significant concern for liquidity and orderly functioning of these markets?

Our answer to these questions are not in the affirmative. We find no evidence that the

degree of insider activity adversely affects prices or liquidity in either the equity or credit

markets. If anything, the reverse appears to be true: CDS markets for corporate entities

that have a large number of banking relationships tend to have smaller bid-ask spreads on

average. Though we find this result puzzling, we provide a few candidate explanations in the

paper. These explanations are based on the potential endogeneity of insider trading to the

liquidity of the CDS market, and on the possibility that insiders play a liquidity-provision

role in relatively opaque CDS market in order to learn about the order flow (in the spirit of

Bloomfield and O’Hara, 1999, 2000).

The institutional response to complaints of insider trading in CDS markets has been to

issue voluntary guidelines on information sharing inside banks (issued by agencies such as

the Joint Market Practices Forum and the Financial Stability Forum in North America), and

on how “material non-public information should be handled by credit portfolio managers in

European Union member states” (issued by International Swap Dealers’ Association (ISDA),

the Bond Market Association, and the Loan Market Association).5

Our results suggest that while the complaints against insider trading in CDS markets may

5

As cited earlier, some of the recent debate has explicitly referred to insider-trading risk as proxied in

our paper: “Use of material non-public information by securities and credit derivative market participants

will be embraced by these new laws on insider dealing. Knowledge gleaned from bank lending relationships

is the main area of scrutiny, so far as the use of non-public information in these markets is concerned.”

(Creditflux, January 2005, www.creditflux.com)







4

be well-founded, it is unclear whether there should be any regulatory concern and action to

curb insider trading. In particular, we do not find evidence to support the hypothesis that

insider trading in CDS markets is harmful per se to the overall liquidity of these markets.

Indeed, our results suggest that regulatory restrictions on the activity of commercial banks,

the potentially informed players in these markets, may result in a reduction in the overall

liquidity of these markets: for example, insiders may withdraw their liquidity-provision role

if they do not stand to reap any benefits from their being informed.

The remainder of the paper is organized as follows. Section 2 summarizes the related

literature. Section 3 describes the data sets we pool together for our analysis. Section 4

presents our tests to detect evidence of insider trading in CDS markets and the results.

Section 5 examines the effect of insider-trading risk in the CDS market of a corporate entity

on the liquidity of CDS and equity markets for that entity. Section 6 concludes.







2 Related Literature

To date, we are not aware of a study that has examined the set of issues we study concerning

insider trading in credit markets and its effect on liquidity of these markets. Insider trading

has been the focus of a large body of literature on equity markets which has found that

insider trading lowers liquidity and increases trading costs (Easley, Kiefer, O’Hara, Paperman

(1996), Chun & Charoenwong (1998), Bettis, Coles & Lemmon (2000), Brockman & Chung

(2003), and Fishe & Robe (2004), among others), that insider trading raises cost of equity

capital (Bhattacharya & Daouk (2002)), and that insider trading increases volatility (Du &

Wei (2003)).

There has been somewhat of a surge in studying liquidity in credit markets (Fleming

& Remolana (1997, 1999), Fleming (2001), Schultz (2001), Hotchkiss & Ronen (2002), and

Chordia, Sarkar, Subrahmanyam (2003)), but this has been contained on studying liquidity

in the cash (bond) market and not the credit derivatives markets. The few existing studies

on credit derivatives (e.g., Blanco, Brennan and Marsh, 2003, and Longstaff, Mithal and





5

Neis, 2003) focus primarily on explaining the basis between the CDS and the cash (bond)

markets.

Crucially, most of the papers studying liquidity in credit markets focus on how quickly

the information in equity markets gets incorporated into credit (cash) markets, but not

the reverse channel of information flow. Examining the flow of information from the credit

(derivatives) markets to the equity markets is key to detecting insider trading in the context

of our paper, and also constitutes the most important contribution of our paper. Finally, our

study is also unique in its integration of two different strands of literature (and indirectly

in its synthesis of the respective datasets): banking – the specialness of banks in possessing

private information about their borrowers; and, market microstructure – the strategic use of

private information by insider potentially affecting prices and bid-ask spreads.







3 Data

Our primary data set for this study comes from daily closing quotes for the most widely

traded CDS names from January 1, 2001 through October 20, 2004. The bid-ask quotes are

obtained from CreditTrade, an online trading platform for credit derivatives, that attracts a

significant portion of trade and quotes in these markets and has participation from all the

top players. The CDS levels are calculated as mid prices based on the bid-ask quotes. By and

large, the quotes are for standard trades, that is, for 10 million USD notional transaction for

a CDS of five year maturity with the underlying reference credit of the firm being of Senior

Unsecured category. In a few quotes, the reference credit is of Subordinated category. The

CreditTrade data has also been employed by Blanco, Brennan and Marsh (2003) in their

study of the basis between CDS and the cash markets. We focus on a total of 79 North

American corporate credits, the so called “benchmarks” in CreditTrade terminology. The

names of these entities are listed in Appendix A. Matching information for stock market

data and balance-sheet data is obtained for these firms from CRSP and Compustat. While

limited in a time-series and a cross-sectional sense, this is the best data available on CDS





6

prices. Furthermore, the sample period is in fact one of the most interesting ones in the

history of corporate defaults since it witnessed defaults of large U.S. corporations such as

WorldCom, Enron and Sprint, and significant credit deterioration for others such as Tyco.

Our next data set is that on banking relationships for these firms. This data was obtained

using the Loan Pricing Corporation (LPC)’s DealScan database on loans made to large

publicly traded firms (as all the firms we study are). For each borrower on a given date, we

look back as far as 1996 for any outstanding syndicated loan facilities undertaken by this

borrower, as well as its affiliated and predecessor companies. Based on the lead arrangers

retained for these deals, we identify the number of unique banks with which the borrower has

an on-going relationship. Our approach to calculating the number of bank relationships is

similar in spirit to the one described in detail in Bharath, Dahiya, Saunders and Srinivasan

(2004). As acknowledged by these authors, LPC data does not give detailed bank-by-bank

break-up for a loan facility. It is thus not possible to come up with a precise loan-size-weighted

measure of the number of banking relationships that would incorporate actual exposure data

(without making some ad-hoc assumption such as assigning the entire amount of the loan

facility to each lead bank).

While calculating the number of bank relationships for a firm, we focus on the top 100

large, commercial and investment banks which together account for most of the market share

in syndicated lending in the LPC database over our sample period. These banks are listed in

Appendix B. An important reason for focusing on these largest banks is linked to the specific

question of insider trading in CDS markets that we wish to explore. The small commercial

banks are not perceived to be major players in the CDS markets. Hence, accounting for

relationships of a borrower with small banks would overstate the actual number of potential

insiders for the borrower. Finally, an additional advantage of focusing on large banks is that

most of these banks are also underwriters (who may also have access to private information

about the borrowers) whereby the need to compute separate underwriting relationships is

obviated.

The large amount of merger and acquisition activity in the banking sector over our sample





7

period, and to some extent in the corporate sector, requires us to make careful adjustments

while employing the LPC data. The data on bank merger and acquisition activity is obtained

from the data employed and generously provided to us by Bharath et al. (2004), Sufi (2004),

and from the web-site of Federal Reserve Bank of Chicago. The data on corporate merger

and acquisition activity is obtained from the SDC Platinum database. Bank relationships

are computed at the level of parent banks, and are assumed to merge whenever a merger

takes place at the bank level or at the level of the borrowing firms. Similarly, at the level of

each firm, the relationships of the parent and the subsidiary entities are also included in our

relationship measure.

Table 1 provides summary statistics for the credits in our sample. The median CDS

level is 81 basis points (bps) with a high of 2400 bps (for Enron close to its bankruptcy).

Interestingly, though the median bid-ask spread in CDS markets is 20 bps, the high is 2000

bps representing a trading cost that is comparable in magnitude to the highest observed CDS

level. The median firm in our sample has a Moody’s (S&P) credit rating of Baa1 (BBB+),

a market capitalization of 15.82 billion USD, book debt of 8.88 billion USD, and book debt

to assets ratio of 21%. The median number of firms for which quotes are observed on a day

is 46, about 60% of the total number of firms in our sample.

In terms of banking relationships, the median number of lead banking relationships is 16

with the median for all banking relationships being 29. There is substantial cross-sectional

variation: there are firms with no banking relationships (only public debt) and firms with as

many as 50 lead-bank relationships. Most relationships arise from loan facilities, the median

number of facilities being 4 and the median amount per firm being about 4 billion USD. The

average relationship length across various banking relationships has a median of 4.1 years in

the sample.

Since most firms in our sample are reasonably large, trading in their equity market is

by and large liquid: the median volume is 2.65 million shares per day and the median

turnover is 5.6% of the outstanding shares per day. The median of average (annualized)

stock return volatility based on daily stock returns is 39% with a high of 83% and a low of





8

24%. Interestingly, while there are firms with no banking relationships, there are also firms

with no public bond issues outstanding, that is, our sample includes firms that are entirely

bank-financed as far as debt financing is concerned. The median number of public bond

issues for firms in our sample is 9 with a high of 71.6







4 Is there evidence of actual insider trading?

In this section, we discuss our primary tests aimed at understanding the nature of insider-

trading risk in the credit default swap market.





4.1 Preliminary evidence



In order to address the question of whether the claims of insider trading in CDS markets

are actually valid, we conduct some preliminary tests. Specifically, we examine the cross-

correlation structure of information revelation in CDS and equity markets. Our maintained

assumption is that stock market reaction is coincident with public release of information

about the firm. Hence, if there is in fact insider trading in CDS markets, then we should

at least sometimes observe a flow of information from CDS markets to equity markets.

Examining this possibility also offers a nice counterpart to the much-studied response of

bond prices to stock moves. The literature here has generally found a slow adjustment of

bond prices to stock moves in that bond returns have a negative cross-correlation with up

to several days’ lag of stock returns.7

We provide a graphical representation of cross-correlation structure between daily stock

returns and CDS changes in Figure 1. Specifically, the figure shows the cross-correlation

between percent changes in CDS prices at time t and stock returns at time t + k as a

6

Our data on corporate bonds comes from the data employed and generously provided to us by Schaefer

and Strebulaev (2003). The data includes corporate bonds that are included either in the Merrill Lynch

Corporate Master index or the Merrill Lynch Corporate High Yield index. These indices include most rated

publicly issued U.S. corporate bonds.

7

This finding of prior literature is however most likely due to poor quality of bond data. A recent study

by Hotchkiss and Ronen (2002) which employs daily data on corporate bond prices from NASDAQ finds the

bond market to be informationally as efficient as the stock market.





9

Figure 1: Cross Correlation of Stock Returns and CDS Changes.

Whole Sample

0.05



0



−0.05



−0.1



−0.15



−0.2

−5 −4 −3 −2 −1 0 1 2 3 4 5



WorldCom, Enron, Sprint, Tyco

0.1



0



−0.1



−0.2



−0.3



−0.4

−5 −4 −3 −2 −1 0 1 2 3 4 5

lead/lag (days)

The figure shows the cross-correlation between percent changes in CDS prices at time t and stock returns

at time t + k as a function of k. In each panel the cross-correlations for individual firms are averaged across

firms.





function of k = −5, −4, . . . , +5. In the first panel, results are shown for the whole sample,

whereas in the second panel, results are shown for the four large cases of corporate distress

in our sample: WorldCom, Enron, Sprint, and Tyco. In each panel, the cross-correlations

for individual firms are averaged across firms.

In each of the two panels, there is a negative cross-correlation between CDS changes and

lagged equity returns, reflecting a flow of information from equity market to CDS market:

when stock returns are positive, there is a decrease in contemporaneous and subsequent price

of default protection. In the panel where the entire sample is employed, the cross-correlation

between CDS changes and future equity returns is essentially zero. In striking contrast, in

the panel with four firms of our sample that experienced significant deterioration in credit

quality, this cross-correlation structure is different: there is on average a negative cross-





10

correlation between CDS changes and future equity returns, the correlation being highest in

magnitude for future date t + 1. In other words, for firms for which there was adverse credit

information during our sample, we do find evidence of an information flow from today’s CDS

price changed to future stock returns. Indeed, for these firms the flow of information is both

ways between these markets.

This preliminary evidence is consistent with the hypothesis that if insiders are active in

our sample, then firms that experienced severe credit deterioration should exhibit significant

cross-correlation of CDS changes with future stock returns. In other words, insiders appear

to be exploiting information in the CDS market only when there is significant negative

information.

Since studying the “pure” effect of a CDS market innovation at time t on the stock-

market return at time t + k should also control for CDS innovations between t and t + k, we

now conduct more rigorous econometric tests of the hypotheses concerning insider trading.





4.2 Econometric analysis



We seek to establish rigorously whether or to what extent credit markets actually acquire

information prior to its revelation to the market as a whole. As already discussed, our

identifying assumption is that “the market as a whole” means the stock market. Since our

sample consists entirely of liquid, large, and well-followed companies, there is every reason

to suspect that public information is rapidly incorporated in their equity prices. Hence any

predictable flow of information from credit markets to stocks is highly likely to stem from

the availability of non-public information to participants in the credit markets.

The section proceeds as follows. First we describe our methodology for isolating inno-

vations in the CDS prices. Then we establish the basic findings that, while there is little

unconditional spillover of these innovations to the stock market, there are strong condi-

tional effects linked to certain firms at certain times. We examine alternative explanations,

particularly those based on the extent of uninformed trade. We then employ a variety of

econometric techniques to test the robustness of these conditional effects and further refine





11

our understanding of where they are coming from.





4.2.1 Constructing innovations in CDS market



Since credit and equity markets are highly dependent, the first step in our analysis is to

regress changes in our CDS prices on contemporaneous stock returns in order to extract the

residual component. We do this by means of separate time-series regressions for each firm,

also including five lags of both the CDS changes and the stock returns to absorb any lagged

information transmission within the credit market.

Two points about this step are worth clarifying. First, the CDS changes are defined as log

differences in credit spread, or, implicitly, percentages of percentages. Second, we anticipate

that the relationship between these changes and stock returns should be inherently nonlinear.

This can be seen in the context of any structural model of credit spreads. To illustrate, Figure

2 plots the elasticity of credit spreads with respect to stock prices under the Merton (1973)

model of risky debt. This elasticity represents the theoretical relationship that should, under

that model, relate percentage changes in stock prices to percentage changes in credit spreads.

The figure shows that there is a roughly linear relationship between this elasticity and the

inverse level of the credit spread itself.

Guided by this, our specification of expected CDS returns includes interactions of the

stock returns (both contemporaneous and lagged) with the inverse CDS level. We could

impose the functional form of the Merton (1973) or any other structural model in this stage.

But we choose to remain agnostic about the degree of the nonlinearity. As a result, we are,

in effect purging the credit market innovations of any level-dependent dynamics. This affects

our ability to separately identify such level-dependent effects in the cross-market dynamics

studied below.

To summarize, for each firm, a regression is run of CDS percentage changes on a constant,

five lags of CDS percentage changes, the contemporaneous stock return, the product of that

return and the inverse CDS level, and five lags of the latter two terms. We view the residuals

from each of these regressions as independent news arriving in the credit markets, which is





12

Figure 2: Merton (1973) model credit-spread elasticity.

0

sigma = 0.125

sigma = 0.25

sigma = 0.375

CDS return elasticity w.r.t. Stock return









−0.5









−1









−1.5

10 15 20 25 30 35 40 45 50

1 / CDS Level

The figure plots the elasticity of the credit spread s with respect to equity value E under the Merton (1973)

model. Specifically, zero-coupon debt with face value of 100 maturing in 5 years is considered, with risk-free

interest rate set to 4%. Three values for firm-value volatility are considered: 12.5%, 25% and 37.5%. The

firm value V is varied in order to generate different values of the credit spread s and equity value E. The

ds dE

credit-spread elasticity is calculated as ( dV /s) divided by ( dV /E). This elasticity is plotted against the

reciprocal of the credit spread s for different values of V .





either not relevant or simply not appreciated by the stock markets at the time.

It is worth noting that these credit innovations are not small. Although the debt and eq-

uity returns are highly correlated, and although our specification errs on the side of imposing

too few restrictions, the R2 from our regressions are mostly in the single digits.8





4.2.2 Information flows from credit markets to stock markets



We now present tests of our main hypothesis: that the degree of insider trading in the CDS

market is a function of the number of informed participants in that market. To do so, we

8

This, in itself, is a somewhat surprising finding which perhaps points to the inherent limitations on the

explanatory power of structural models.





13

run regressions of the form:



stock returnt = a0 + [b0 + b1 (number of insiders)](CDS innovation)t−1 + t







where the CDS innovation is constructed as described above.

As outlined in the data section, our primary candidate measure for the number of in-

formed players is the number of banks having outstanding syndicated loan commitments to

each firm.9 This number varies substantially across firms and over time. To exploit this, we

start by estimating the model in a panel data setting.

Table 2 show several versions of the specification. Column (A) shows that, uncondition-

ally, there is a statistically significant spillover of yesterday’s CDS innovation to today’s stock

return. In column (B) we see that this unconditional effect is completely absorbed by the

inclusion of the number of banking relationships. For a firm with 10 relationships there is

essentially no spillover, whereas for one with 50 relationships the spillover is around 4%. The

tests in columns (C) and (D) distinguish between the effect of positive and negative CDS

innovations. In (C) there is no difference in the unconditional response: both are about the

same as that found in (A). However, the conditional responses are quite different. Only pos-

itive CDS shocks (i.e., bad news) are now found to have a significant impact on future stock

prices. The coefficient b+ is around eight times larger than b− and implies approximately

1 1



6% spillover for a firm with 50 relationships.

The last two columns augment the specification with lagged CDS innovations, so that

the response term looks like

5

ak [b0 + b1 (number of insiders)t−k ](CDS innovation)t−k .

k=1



This specification allows us to check whether the leading relationship we have identified

is merely a transient effect, possibly due to some short-term price pressure from the hedging

activity of debt-market participants in the stock market. Indeed there is some evidence that

9

The definition used in the main tests considers only lead-bank relationships and restricts attention to

the top 100 banks by market share in overall syndicated lending. Recall that the measure also includes

relationships of parent and subsidiary entities.





14

the unconditional effect is partially reversed: the coefficients a2 to a5 imply that around

a quarter of the effect dissipates within a week (although the coefficients are estimated

imprecisely). By contrast, the conditional effect appears to be robust to the inclusion of lags,

with day-two and day-three reversals being more than outweighed by additional continuation

effects from days four and five.10 Because the lag effects are poorly estimated, we will

continue to check below for the impact of weekly effects in exploring alternative hypotheses.

We conclude from the results in Table 2 that there is, indeed, prima facie evidence that

informed trading exists in the credit

derivatives markets. The fact that the information flow occurs primarily for negative

news is also consistent with the interpretation of hedging activity by asymmetrically informed

banks with positive loan exposures. We now explore alternative specifications designed to

further test and refine this interpretation.





4.3 Alternative hypotheses



We have suggested that the evidence uncovered so far is consistent with banks using their

monitoring role to uncover relevant credit information about borrowers and then engaging in

informed hedging when negative information arises. We now consider some alternative po-

tential explanations and, in the process, establish more clearly how the lead-lag relationship

operates.

Table 3 shows panel regressions, similar to those in Table 2, which model the CDS

innovation component as a function of other controls. That is,





(stock return)t = a0 + [b0 + b1 (number of banks) + b (other controls)](CDS innovation)t−1



+ t.

10

This specification is estimated by non-linear least squares and the ak terms are measured relative to a1

which is set equal to one for purposes of identification. We calculate p values for all specifications (in square

brackets) using a within-period bootstrap which accounts for any cross-sectional autocorrelation. For the

non-linear specification, the ak coefficients are not identified under the null of no effect. For these terms the

p-values measure the bootstrap probability that an unrestricted lag term exceeds the sample estimates.









15

We limit the specification here to one lag of the CDS innovation, but, as before, include five

own lags of each stock’s return.

A first natural concern is that our measure of bank relationships is simply picking up the

quantity of a firm’s borrowing. That is, debt markets may, as a whole, be better informed

if there are more participants of every kind. If this is so, the role of bank relationships may

have nothing to do with their differential ability to gather information. Specifications (A)

- (D) in Table 3 include other measures of the size of firms’ debt market. We include the

overall equity value of the firm and the book value of its debt (both in logs). Also we count

the number of public bond issues of each firm at each date, and include this count along with

an indicator variable if any of these issues is a convertible bond. We also employ the total

number of syndicated loan facilities and the face value of these. None of these variables is

significant, nor does any affect the size of the bank relationship coefficient. The coefficient

on the number of bond issues is economically large, but its sign is positive, meaning that

there is less lagged information flow for firms with more non-bank debt. We interpret this

as indicating that cross-market arbitrage is probably more active in these names, i.e., that

any informational advantage by debt markets is quickly exploited.

Specification (E) includes a refinement of our definition of bank relationships. Hitherto,

following Sufi (2004), we have viewed lead arrangers as the most likely to acquire non-

public information about borrowers. But, as stressed by Lee and Mullineaux (2004), each

participating bank has full rights and responsibilities for monitoring, and is entitled to share

any information acquired by any delegated member.11 When we add the number of non-

lead participating banks, we actually find that their influence is apparently larger than that

of leads. Both are in the same direction, and the lead effect is not much diminished by

the inclusion of non-leads. A firm with 20 lead banks and 20 participants would have an

information spillover effect of around 7%.

Another hypothesis about our results is that the lead-lag relationship is due to the relative

liquidities of debt and equity markets for each firm. Under this view, slower information

11

Further, there is a regulatory requirement that each participating bank perform independent due-

diligence for each loan.





16

transmission may be an artifact of relatively less liquid stock markets for companies which

have many banks. Specifications (F) and (G) include proxies for stock and CDS market

activity and liquidity. For the stock market we include share volume and turnover, as well

as Amihud’s (2002) measure of market impact. The only direct proxy for credit market

liquidity we have is the bid/ask spread in our CDS data, which we measure as a percentage

of the mid-market CDS level.12 The inclusion of these quantities again fails to diminish the

role of banks in explaining the lead-lag effect, and even increases its statistical significance.

Stock volume alone appears to be an additional significant conditioning variable. Curiously,

its negative sign implies that bond lags matter more for more highly traded stocks. This

may be due to unmodeled heteroscedasticity: volume is known to be positively associated

with volatility. Both are driven by the quantity of news released about a given stock. The

result here then could simply be telling us that there is more information flow when there is

more information.





4.4 Robustness



4.4.1 Sub-samples



In this section, we check the robustness of the information flow from the CDS to equity

markets to different sub-samples based on credit conditions and on bank relationships. In

addition, the specifications in this section allow the own-lag effects on stock returns to vary

across the subsamples, to ensure that any apparent CDS-lag effects are not artifacts of

unmodeled dynamics in the share price itself.

First, we allow for the unconditional information flow to vary between sub-samples formed

on the basis of whether (i) firms experienced significant credit deterioration on some day

during our sample period; (ii) firms experienced a general widening of credit spread during

our sample period; and, (iii) credit rating of firms is low.

12

The spreads are close to a linear function of CDS levels as they should be, mechanically, since a matched

purchase and sale in the CDS markets is still itself exposed to default risk.









17

Specifically, we estimate the panel specification

5

(stock return)t = a0 + [b0,k + bD · (Credit-condition Dummy)t ](CDS innovation)t−k

0,k

k=1

5

+ [c0,k + cD · (Credit-condition Dummy)t ](stock return)t−k + t .

0,k

k=1





We estimate this specification for three dummies in Table 4. In specification (A), the dummy

is one if the firm experienced a one-day decline in credit spread level exceeding 50 basis

points between time t and end of the sample period. In specification (B), the dummy is

one whenever the firm’s credit spread level remained at a level greater than 100 basis points

between time t and end of the sample period. Finally, in specification (C), the dummy is

one if the credit rating of the firm at time t was low (A3/A- or worse).

Table 4 shows that the evidence is consistent with there being greater information flow

from the CDS to equity markets for those firms which experienced, or were more likely to

experience, “credit” events in future. In each of the three specifications, there is no uncondi-

5

tional flow from CDS to equity markets ( k=1 b0,k is essentially zero). However, conditional

on the credit-condition dummy, the flow is present. For firms that actually experienced

credit deterioration (specifications A and B), the sum of the coefficients on lagged CDS in-

5 5

novations, k=1 (b0,k + bD ), is negative, and the flow measure (

0,k

D

k=1 b0,k ), is both negative

and statistically significant. For firms that are more likely to experience credit deterioration

(the low-rated firms), the effect is negative but not as statistically significant. Overall, this

reinforces the earlier finding that CDS markets reveal more information about adverse credit

developments than about improvement in credit conditions. Note that examining the sum

of the coefficients on the five lags also confirms that the information flow we have detected

is permanent.

Next, we test if the conditional effect of bank relationships identified in Tables 2 and 3

is robust across sub-samples of firms with high and low number of relationships.

5

(stock return)t = a0 + [b0,k + bD · (Relationship Dummy)t ](CDS innovation)t−k

0,k

k=1







18

5

+ [c0,k + cD · (Relationship Dummy)t ](stock return)t−k + t .

0,k

k=1





We estimate three specifications in Table 5, defining the dummy to be one when the

number of banks is above median (specification A), and when in addition the one-lagged

CDS innovation is positive (specification B) and negative (specification C). While the first

specification merely checks whether the effect of CDS innovations on stock markets is negative

only for firms with a large number of banks, the next two specifications also check whether

the effect of large number of banks is restricted to days with adverse credit news (as in Table

2).

These hypotheses are supported in this non-linear conditional tests as well. Estimates in

specification (A) clarify that the negative flow from CDS innovations to stock markets arises

for firms with number of bank relationships that are above the sample median, but not for

5 5

the remaining firms. In particular, k=1 (b0,k + bD ) is negative, and the flow (

0,k

D

k=1 b0,k ) is

negative and statistically significant. This is consistent with the panel estimates of Table 2

where the information flow was found to be close to zero for firms with around 10 banking

5

relationships. Somewhat surprisingly, the flow ( k=1 b0,k ) is positive (albeit small) and sta-

tistically significant for these remaining firms. In specification (B), we see that the negative

flow for firms with above-median relationships is in fact twice as large on days with positive

changes in CDS level. In contrast, specification (C) reveals that there is no significant flow

for these firms on days with negative changes in CDS level.





4.4.2 Cross-sectional analysis



A feature of the panel estimates described so far is that they force all firms to have the

same dynamic properties, except in so far as captured by the conditioning introduced in

the lagged-response terms. In this section, we estimate separate dynamics for each firm and

then study the cross-firm variation in response to credit market information. This analysis

addresses the possibility that our previous finding of a significant conditional effect from

the CDS innovations was actually driven by uncaptured variation in the other terms (the





19

intercept and stock lag coefficients).

Here we follow the methodology of Hou and Moskowitz (2005) who study cross-firm

variation in lagged response to market news. Specifically, for each firm i we run the time-

series regression

5

(stock return)i,t = af

i + bf (CDS innovation)i,t−k + t ,

i,k

k=1



continuing to include five lags of stock return on the right hand side. We then define a

measure of the information flow from CDS market to the stock market for firm i as

5

θi = bf .

i,k

k=1



For firms for which the information flow is large and permanent, θ should be large and

negative; if the information flow is partially reversed within five days, then θ should be less

negative; and, θ should be close to zero for firms for which there is not much information

flow in the first place. Panel A of Table 6 shows the summary statistics for the estimated

θi ’s. The mean is 0.0043 and statistically insignificant. That is, there is not much of an

unconditional effect once the full dynamics are allowed to be firm specific.

Next we sort our firms into quintiles based on their lagged response and examine the

average firm characteristics of each. Panel B of Table 6 reveals that the main evidence of

insider activity is confined to the lowest θ quintile. Firms in this set are on average larger,

more actively traded, and somewhat more volatile than the sample as a whole. (Note that

the table reports medians, and is thus not sensitive to individual outliers.) Neither credit

rating nor leverage varies much across quintiles. However credit risk, as measured by CDS

level, does rise monotonically as θ falls, echoing the observation above that acquisition of

non-public information may respond to the incentive represented by increased risk to bank

portfolios. Finally, in line with our primary findings, number of bank relationships also varies

monotonically with θ.

To check that the cross-firm dynamic characteristics do indeed preserve our previous

panel results, we run a single cross-sectional regressions of θi ’s on time-series averages of





20

firm-specific characteristics, most notably bank relationships. Specification (A) in Panel

C confirms that banks with more lead-bank relationships have a more negative θi . The

estimated coefficient on lead banks implies that a firm with 50 lead-bank relationships has

13% information spillover from CDS innovations to stock returns. Specifications (B) through

(E) show that the effects of CDS level and firm rating are insignificant, irrespective of whether

lead banks are included in the specification or not.

The failure of CDS level in explaining the cross-sectional variation of θi ’s is at odds with

the median summaries in Panel B. However, this can be rationalized once the within-quintile

variation in CDS levels is examined. For the lowest quintile firms, the standard deviation of

mean CDS levels (across firms) is 123 bps, and for the highest quintile, this figure is 169.35.

These variability measures are 75% and 250% of the median CDS levels for these quintiles,

respectively. In contrast, the within-quintile variation in the number of bank relationships

is much smaller. For quintile 1, the variability in number of banks is 10.4 (40% of median

banks), and for quintile 5, the corresponding variability is 11.7 (95% of median banks). In

particular, the lowest quintile firms, for which thetai ’s are the most negative, are primarily

firms with large number of bank relationships.

This confirms that the number of insiders, proxied by the number of lead bank relation-

ships, is the singular critical determinant of the cross-sectional variation in the permanent

information flow from CDS markets to equity markets, as measured by θ.



Put together, these robustness checks provide convincing evidence of information flow

from CDS markets to stock markets, in times when the potential of inside information with

relationship banks is likely to be high, and for firms where the number of such relationships

is large.







5 The impact of information asymmetry

Does insider trading in the credit derivatives markets matter? As discussed in the introduc-

tion, there is abundant evidence that participants, regulators, and industry bodies all regard



21

it as a serious threat to the integrity of the market. Having isolated some strong predictors

of relative insider activity, we are now in a position to offer evidence on this topic.

To do so, we need to first consider what is actually perceived as being under threat.

Our interpretation of the industry view is that the situation is analogous to the classic moral

hazard problem in any other insurance market. The threat of informed purchase of insurance

leads to a lemon’s problem in which insurance premia are set too high and the quantity of

insurance written in equilibrium is too low.

Since we have no information on the amount of credit risk insured for any of our names

and, anyway, cannot hope to gauge the amount of such transfer that is efficient, our approach

is to try to detect the effect of information asymmetry on bid and ask prices in the CDS

market. If the threat of informed trading drives a wedge between the reservation prices of

buyers and sellers, then the effect is the same as in classic microstructure models in finance

(where uninformed market makers face potentially informed buyers and sellers). Unlike most

microstructure settings, the one-sided-ness of the threat in the CDS market, further implies

that we should potentially see an effect in the levels of prices, i.e., insurance may be too

expensive.

It is worth noting, at this point, that the evidence of informed trading uncovered above is

not necessarily evidence of asymmetric information within the credit markets. An alternative

interpretation is that the CDS market as a whole is better informed, in certain circumstances,

than other investors. Under this view, the threat of asymmetric information, if it exists,

might be to the liquidity of the stock market. Accordingly, we test for such effects as well.

Table 7 shows regressions of stock and CDS liquidity measures on some standard control

variables as well as on bank relationships, our measure of the prevalence of non-public infor-

mation in the credit market. Results are shown for both panel regressions with time fixed

effects, and for Fama-MacBeth (1973) regressions. In the latter case, standard errors are

corrected for autocorrelation of up to six months; in the former case, the reported t statistics

are clustered at the firm level.

Both methodologies lead to the same conclusions. In the first two columns, stock liquidity





22

is shown to be essentially insensitive to the bank relationship variable. More interestingly,

the second two columns indicate that these relationships do affect credit market illiquidity

– but with a negative sign. Hence, to the extent that more banks implies more informed

players, the evidence suggests that this leads to, or is associated with, narrower spreads and

greater liquidity provision. In unreported tests, we find this result to be robust for inclusion

of the proxies employed in Table 3 for overall activity in a firm’s debt, suggesting that there is

a possibly endogenous connection between being informed and choosing to provide liquidity,

or between prevalence of insider-trading activity with availability of liquidity.

Table 8 investigates whether the risk of informed trading shows up in the cost of credit

insurance, i.e. the CDS levels themselves. In principle, the moral hazard effect could raise

these prices even if there is no direct effect on liquidity. However the first two columns

establish that, controlling for known determinants of credit spreads, bank relationships play

no additional role. Finally we check whether our one direct measure of liquidity, the bid/ask

spreads of the default swaps, influence prices directly. This test allows for the possibility that,

in using bank relationships, we have simply failed to isolate a valid proxy for asymmetric

information. However, the regressions in the rightmost columns show no influence of spreads

on levels.13

To summarize, returning to the policy question at issue, we find no evidence that the

presence of informed insiders adversely affects liquidity provision or raises the price of credit

insurance.





5.1 Possible explanations



If insider-trading risk does not have an effect on liquidity in a market, then it is not surprising

that it does not affect the level of prices in that market either. Hence, it suffices in our context

to try and understand why insider-trading risk in CDS markets, proxied by the number of

bank relationships, does not increase illiquidity (as measured by bid-ask spreads) and perhaps

13

We have replicated these results using numerous additional controls, including non-linear terms, struc-

tural estimates of credit risk, accounting predictors of default, and firm transparency ratings.







23

is even associated with greater liquidity. We provide a few candidate explanations that are

consistent with our findings and discuss their relative merits.

The first candidate explanation is based on the observation that even informed players in

markets need to know the nature of the order flow (the pattern of uninformed trades) in order

to strategically time their information release and reap rewards thereof. CDS markets are

relatively opaque. In particular, quotes posted at CreditTrade, the on-line brokerage service

we have obtained our data from, are anonymous: until a trade hits a quote, the counterparty

information is not revealed to the two involved parties. Bloomfield and O’Hara (1999,

2000) have shown through trading experiments that in opaque markets, the informed players

emerge as liquidity-providers and post narrower bid-ask spreads (compared to other agents

and more transparent markets). This is consistent with the informed players learning about

liquidity for a strategic reason and in the process providing liquidity to the market.14 Banks

having relationships with the corporate entities we have examined are also the intermediaries

of the CDS markets. The proposed explanation can thus rationalize simultaneously the

greater information release for CDS names that have more relationship banks, and their

enjoying greater liquidity compared to names with fewer relationship banks.

The second candidate explanation relies on the possibility that the informed are perhaps

aware of the nature of uninformed trading, but merely time their trades or choose the

location of their trades so as to minimize their price-impact and trading costs. If having

more relationship banks also induces greater uninformed trading on average (e.g., due to

portfolio-rebalancing or regulatory-arbitrage reasons), then there would be greater liquidity

for CDS names with more bank relationships, and simultaneously a greater information

release. While this simple endogeneity argument is appealing at first blush, it is at odds

with our panel results in Table 3 that the bid-ask spread in the CDS market has no effect on

the extent of information flow from the CDS to the stock markets. Nevertheless, a complete

test would recognize the endogeneity of the bid-ask spread and instrument for it suitably.

The final explanation we propose is based on the idea of competition amongst informed

14

See also Hong and Rady (2002) for strategic timing of trades by the informed when they are uncertain

about the exact nature of uninformed trading.





24

agents. Holden and Subrahmanyam (1992) show that if a large number of informed agents

receive the same piece of information about an asset’s value, then they trade aggressively,

revealing instantaneously most of this information into prices. The depth of the market in

the limiting case becomes infinite as the informed compete and erode each other’s profits. It

is possible that all relationship banks receive the same quantum of credit information about

the underlying CDS name, and in an attempt to capitalize on this information, reveal all of

this information into the market. This would also generate greater information release and

greater liquidity being associated with CDS names that have more bank relationships. Note,

however, that in our specific setting, we would expect the competing informed banks to also

avail of trading in the stock markets, an outcome of which would be a coincident release

of information in the CDS and stock markets. Since our CDS innovations are orthogonal

to contemporaneous stock-market innovations, the information flow from CDS innovations

to future stock-market changes is necessarily due to information revealed only in the CDS

markets.15

The data and tests we have employed so far cannot shed conclusive light on the relative

merit of these candidate explanations. The general finding of liquidity rising with information

asymmetry runs counter to much of accepted wisdom in market microstructure. So further

investigation of this topic is certainly warranted.







6 Conclusion

In this paper, we provided empirical evidence that there is information flow from the credit

default swaps markets to equity markets and this flow is permanent and more significant

for entities that have a greater number of bank relationships. This information flow is

concentrated on days with negative credit news, and for entities that experience or are more

likely to experience adverse credit events. These findings are consistent with the existence

of insider trading in these markets. However, we do not find evidence that this form of

15

Examining this last explanation makes it clear that the first two explanations implicitly rely on imperfect

competition between informed players in the CDS markets.





25

insider-trading risk affects adversely the liquidity provision in the credit derivatives or the

equity markets of these entities.

Our study is the first in the literature to examine insider trading in credit markets and

its effect on liquidity of these markets. Our findings also constitute a first step towards

understanding whether there is a case for the current regulatory response to complaints of

insider trading in these markets: the response has been to limit the use of material non-

public information gleaned from bank lending relationships. Our view is that such limits

on trading by banks need to be cautiously re-examined since such limits may potentially

endanger the liquidity-provision role that banks seem to play in the CDS markets.

Future work employing better proxies of insider-trading risk, perhaps using intra-day

data on actual transactions in the CDS market, and enlarging the sample to include non-

US and sovereign credits would be valuable in shedding further light on these issues. We

are currently in the process of acquiring intra-day data on actual CDS transactions with

information on whether the involved counterparties have banking relationships with the

underlying corporate entity. This data should help us enrich our analysis by providing more

direct tests of the effect of actual insider trades on market prices and liquidity.









26

References

Bettis, Coles and Lemmon, Corporate policies restricting trading by insiders, 2000 Journal

of Financial Economics, 57, 2, 191-220.



Bhattacharya and Daouk, The World Price of Insider Trading, 2002 Journal of Finance,

57, 1, 75 - 108



Bharath, Sreedhar, Sandeep Dahiya, Anthony Saunders, and Anand Srinivasan, So What

Do I Get? The Banks View of Lending Relationships, Working Paper, University of

Michigan, 2004.



Blanco, Robert, Simon Brennan and Ian W. Marsh, An Empirical Analysis of the Dy-

namic Relationship between Investment Grade Bonds and Credit Default Swaps, 2003,

forthcoming, Journal of Finance.



Bloomberg Markets, Credit Swaps High Risks, Few Rules, August 2003.



Bloomfield, Robert and Maureen O’Hara, “Market Transparency: Who Wins and Who

Loses?,” Review of Financial Studies, 1999, 12(1), pp. 5–35.



Bloomfield, Robert and Maureen O’Hara, “Can Transparent Markets Survive?,” Journal

of Financial Economics, 2000, 55, pp. 425–59.



Brockman and Chung, Investor Protection and Firm Liquidity, 2005, forthcoming, Journal

of Finance



Chordia, Sarkar, Subrahmanyam, An Empirical Analysis of Stock and Bond Market Liq-

uidity, 2003, Review of Financial Studies, 18, 85 - 129.



Chun and Charoenwong, Insider Trading and the Bid-Ask Spread, 1998, Financial Review



Du and Wei, Does Insider Trading Raise Market Volatility?, 2004 Economic Journal, 114,

498, 916-942.





27

Duffee, G. and C. Zhou, Credit Derivatives in Banking: Useful Tools for Managing Risk?,

2001 Journal of Monetary Economics, 48, 25–54.



Easley, Kiefer, and O’Hara, One Day in the Life of a Very Common Stock, 1997, Review of

Financial Studies, 10, 3, 805-835.



Easley, Kiefer, O’Hara, and Paperman (1996); Liquidity, Information, and Infrequently

Traded Stocks, 1996, Journal of Finance, 51, 4, 1405–1436.



Fama, Eugene, and James D. MacBeth (1973), Risk, Return and Equilibrium: Empirical

Tests, Journal of Political Economy, 81, 607–636.



Fishe and Robe, The impact of illegal insider trading in dealer and specialist markets:

evidence from a natural experiment, 2004, Journal of Financial Economics, 71, 3,

461-488



Fleming, Michael J., and Eli M. Remolona, Price Formation and Liquidity in the U.S.

Treasury Market: The Response to Public Information, 1999 Journal of Finance, 54,

5, 1901 - 1915.



Fleming, Michael J., Measuring Treasury Market Liquidity, 2001, Federal Reserve Bank of

New York Staff Paper 133.



Holden, Craig and Avanidhar Subrahmanyam, Long-Lived Private Information and Imper-

fect Competition, Journal of Finance, March 1992.



Hong, Harrison and Sven Rady, Strategic Trading and Learning About Liquidity, Journal

of Financial Markets, 2002.



Hotchkiss and Ronen, The Informational Efficiency of the Corporate Bond Market: An

Intraday Analysis, 2002, Review of Financial Studies, 15, 1325 - 1354.



Hou, K. and T. Moskowitz, Market Frictions, Price Delay, and the Cross-Section of Ex-

pected Returns, 2005, forthcoming, Review of Financial Studies.



28

Lee, Sang Whi, and Donald J. Mullineaux, Monitoring, Financial Distress, and the Struc-

ture of Commercial Lending Syndicates, Financial Management, 2004, 33(3).



Longstaff, Francis A., Sanjay Mithal and Eric Neis, Corporate Yield Spreads: Default Risk

or Liquidity?, 2003, forthcoming, Journal of Finance.



Merton, Robert C., 1974, ”On the Pricing of Corporate Debt: The Risk Structure of Interest

Rates”, Journal of Finance, Vol. 29, pp. 449-470.



Sneder, Henny, A Market Backfires, and Investors Pay Though Greenspan praised idea,

Credit-Derivative Trading Mutates; Instead of Stability Instability, December 5, 2002,

in Deals and Deal Makers, Wall Street Journal.



Schaefer, Stephen and Ilya Strebulaev, Structural Models of Credit Risk are Useful: Evi-

dence from Hedge Ratios on Corporate Bonds, 2003 Working Paper, London Business

School.



Schultz, Paul, Corporate Bond Trading Costs: A Peek Behind the Curtain, 2001 Journal

of Finance, 56, 2, 677 - 698.



Sufi, Amir, Information Asymmetry and Financing Arrangements: Evidence from Syndi-

cated Loans, 2004, Working Paper, MIT Department of Economics.



The Economist, Whos Carrying the Can? August 14, 2003.









29

Table 1: Summary Statistics



Low Median High



CDS level (mid price, BP) 13 81 2400

CDS bid-ask spread (BP) 1 20 2000

Credit rating Ba3/BB- Baa1/BBB+ AAA/Aaa2



Firm size (equity mkt val, $mm) 720 15820 412900

Firm debt (book val, $mm) 9 8874 312684

Firm leverage (debt at book val) 0.00 0.21 0.65



Average stock volume (mm shrs/day) 0.14 2.65 76.1

Average stock turnover (pct/day) 2.0 5.6 26.8

Average stock volatility (ann std dev) 0.24 0.39 0.83



Number of Bond Issues 0 9 71



Number of Bank Relationships (leads) 0 16 50

Number of Bank Relationships (all) 0 29 69

Average Relationship Length (yrs.) 0.3 4.1 7.2

Number of Active Facilities 0 4 36

Amount of Active Facilities ($mm) 0 4019 66099



Observations/day 9 46 62



The table describes the firm characteristics for our sample of credit derivatives. Sample statistics are com-

puted across all observations, except average stock trading statistics which are computed across firms.









30

Table 2:

Regressions of stock returns on CDS innovations



(A) (B) (C) (D) (E) (F)

b0 -0.0080 0.0094 -0.0056 0.0055

(2.75) (1.64) [.018] [.080]

[.003] [.206]

b1 -0.00093 -.00033

(3.55) [.028]

[.001]



b+

0 -0.0088 0.0216

(1.98) (2.63)

[.138] [.037]

b+

1 -0.0016

(4.38)

[.000]

b−

0 -0.0073 -0.0033

(1.61) (0.39)

[.107] [.650]

b−

1 -0.0002

(0.58)

[.600]

a1 1.0 1.0

[.018] [.024]

a2 -0.076 -0.470

[.718] [.150]

a3 -0.116 -0.076

[.884] [.794]

a4 -0.137 1.108

[.754] [.146]

a5 0.067 0.194

[.744] [.398]



The table shows the results of regressing daily stock returns on lagged values of CDS innovations. The lag

coefficient is modelled as b0 + b1 · (number of bank relationships) in the first two columns. The second two

allow the coefficients to differ depending on the sign of the lagged CDS innovation. The final two columns

include five lags, with each constrained to have the same values of b0 and b1 . All regressions also include an

intercept and five lags of the dependent variable. OLS t-statistics appear in parentheses. Bootstrap p-values

using within-period randomization are shown in square brackets.

31

Table 3: Controls for Uninformed Trade



(A) (B) (C) (D) (E) (F) (G)

b0 -0.0022 -0.0049 -0.0022 -0.0062 -0.0224 0.0448 0.0126

(0.05) (0.11) (0.05) (0.14) (0.51) (0.92) (0.27)

banks -0.0010 -0.0011 -0.0011 -0.0008 -0.0009 -.0009 -.0009

(2.93) (3.05) (2.11) (1.73) (2.55) (3.26) (2.39)



size 0.0003 -0.0004 -0.0003 0.0006 0.0013 0.0063 0.0057

(0.08) (0.11) (0.08) (0.18) (0.37) (1.38) (1.48)

debt 0.0009 0.0031 0.0009 0.0014 0.0038 0.0021

(0.24) (0.75) (0.23) (0.32) (0.92) (0.56)

bonds 0.0128 0.0145 0.0128 0.0134 0.0161 0.0132

(1.43) (1.60) (1.42) (1.49) (1.76) (1.46)

CB ind -0.0005

(1.39)

loans 0.0001

(0.08)

loan amt -0.0030

(0.58)

nonleads -0.0018 -0.0017

(3.55) (3.39)

volume -0.0098 -0.0066

(2.15) (2.16)

turnover 0.0024

(0.40)

ILLIQ -0.0063

(0.03)

CDS b/a 0.0071

(0.35)



The table shows the results of regressing daily stock returns on one lag of CDS innovations. The lag

coefficient is modelled as b0 + b1 · (number of bank relationships) + b (other controls). The controls are:

size (log of market capitalization); debt (log of compustat book value of debt); bonds (number of public

dollar denominated bonds of parent company); CB inf (indicator = 1 if any of the bonds counted in the

previous variable is convertible and = 0 otherwise); loans (number of active syndicated loan facilities); loan

amt (notional value of active syndicated loan facilities); nonleads (number of paericipant banks in active

syndicated loans); volume (daily stock market volume in millions); volume (daily stock market turnover);

ILLIQ (average absolute value of stock returns divided by average volume ); CDS b/a (percentage bid/ask

spread of credit default swap). All regressions also include an intercept and five lags of the dependent

variable. OLS t-statistics appear in parentheses. 32

Table 4: Robustness to credit-condition sub-samples.



(A) (B) (C)



a0 0.0003 0.0003 0.0003

(2.19) (2.21) (2.04)



5

k=1 b0,k 0.0033 0.0050 0.0111

(0.44) (0.66) (0.91)



5 D

k=1 b0,k −0.0492 −0.0465 −0.0224

(2.36) (2.52) (1.51)



5

k=1 c0,k −0.0413 −0.0399 0.0183

(3.22) (3.00) (0.97)



5 D

k=1 c0,k 0.1917 0.1338 −0.0488

(5.90) (4.68) (2.02)



The table shows OLS estimates and t-statistics for the coefficients of a regression of daily stock returns on

a constant, lagged CDS innovations, and lagged stock returns, as follows:

5

(stock return)t = a0 + [b0,k + bD · (Credit-condition Dummy)t ](CDS innovation)t−k

0,k

k=1

5

+ [c0,k + cD · (Credit-condition Dummy)t ](stock return)t−k + t .

0,k

k=1



That is, the regression also includes interaction terms of the lagged CDS innovations and stock returns with

an indicator equal to one for firm i at date t if (i) the firm experiences a credit deterioration of more than

50 basis points between date t and the end of the sample (specification A); (ii) the firm’s credit spread level

remained at a level greater than 100 basis points between time t and end of the sample period (specification

B); and, (iii) the credit rating of the firm at time t was low, that is, A3/A- or worse (specification C).









33

Table 5: Robustness to relationship sub-samples.



(A) (B) (C)



a0 0.0003 0.0005 0.0003

(2.10) (3.35) (1.81)



5

k=1 b0,k 0.0192 0.0146 −0.0035

(2.01) (1.79) (0.43)



5 D

k=1 b0,k −0.0475 −0.0714 −0.0006

(3.41) (4.17) (0.03)



5

k=1 c0,k −0.0583 −0.0211 −0.0008

(3.46) (1.56) (0.06)



5 D

k=1 c0,k 0.0909 0.0343 −0.0584

(3.86) (1.23) (1.91)



The table shows OLS estimates and t-statistics for the coefficients of a regression of daily stock returns on

a constant, lagged CDS innovations, and lagged stock returns, as follows:

5

(stock return)t = a0 + [b0,k + bD · (Relationship Dummy)t ](CDS innovation)t−k

0,k

k=1

5

+ [c0,k + cD · (Relationship Dummy)t ](stock return)t−k + t .

0,k

k=1



That is, the regression also includes interaction terms of the lagged CDS innovations and stock returns with

an indicator equal to one for firm i at date t if (i) the firm has above-median number of bank relationships

at time t (specification A); (ii) in addition to (i), the CDS innovation at date t − 1 is positive (specification

B); and, (iii) in addition to (i), the CDS innovation at date t − 1 is negative (specification C).









34

Table 6



Panel A: Properties of θ



Mean = 0.0043

t-stat = 0.4600

Min = −0.1961

Max = 0.3262





Panel A shows univariate properties of θ, the firm-specific measure of permanent information flow from CDS

innovations to stock markets. In the first stage, we run for each firm i the time-series regression



5

(stock return)i,t = af +

i bf (CDS innovation)i,t−k + t ,

i,k

k=1





continuing to include five lags of stock return on the right hand side. Then, θi is the measure of permanent

5

information flow from CDS market to the stock market for firm i, defined as θi = k=1 bf . i,k







Panel B: Properties (medians) of firms in different θ-quintiles



Q1 Q2 Q3 Q4 Q5



Average θ −11% −2% 1% 4% 8%

CDS level (mid price, BP) 185 108 101 79 68

CDS bid-ask spread (BP) 26 21 20 18 17

Credit rating 27 27 26 27 26

Firm size (equity mkt val, $mm) 28021 12477 9663 12677 13862

Firm debt (book val, $mm) 12785 7178 4136 6864 6380

Firm leverage (debt at book val) 0.28 0.32 0.33 0.33 0.27

Average stock volume (mm shrs/day) 8.07 2.10 1.37 1.71 2.49

Average stock turnover (pct/day) 5.4 6.4 4.8 5.0 5.6

Average stock volatility (ann std dev) 0.40 0.37 0.33 0.33 0.33

Number of Bond Issues 12.6 7.1 5.2 12.0 5.8

Number of Bank Relationships (leads) 26.1 18.1 14.5 10.6 12.1

Amount of Active Facilities ($mm) 3085 1429 978 1358 1050



For Panel B, firms are ranked into quintiles based on the first-stage estimates of θ, Q1 being the quintile

with the smallest (most negative) estimates, and Q5 being the quintile with the largest estimates. The

summary statistics reported for each quintile are the medians (across firms) of the time-series means of the

characteristics for each firm.





35

Panel C: Second-stage cross-sectional determinants of θ



(A) (B) (C) (D) (E)



constant 0.0511 0.0110 0.0608 0.0513 -0.0017

(2.95) (0.72) (0.49) (2.65) (0.01)



banks −0.0026 −0.0026 −0.0027

(3.17) (3.12) (3.17)



CDS −4.8555 −0.1008

(∗10−5 ) (0.56) (0.01)



rating −0.0021 0.0020

(0.46) (0.456)



Panel C shows the OLS estimates and t statistics from second-stage regressions in which the first-stage

estimates of θ for different firms are regressed on firm-specific characteristics. The characteristics employed

for a given firm are the time-series averages for that firm.









36

Table 7: Illiquidity Regressions



Stock Illiq CDS % b/a



FM Panel FM Panel



size -0.0016 -0.0021 0.0480 0.0499

(3.38) (2.93) (8.59) (3.55)

volume -0.0086 -0.0083 -0.0414 -0.0429

(5.88) (2.88) (10.59) (6.02)

r1mo 0.0028 0.0002 -0.0144 -0.0364

(1.95) (0.17) (0.48) (1.42)

σ1mo 0.0130 0.0071 0.0561 0.0780

(5.01) (2.28) (1.71) (2.42)

banks -0.00004 -0.00002 -0.0029 -0.0026

(1.50) (0.37) (8.18) (2.75)

obs 667 39109 947 44932

R2 0.5266 0.3822 0.2646 0.1365



Stock and credit market illiquidity measures are the dependent variables in daily regressions using both

Fama-MacBeth (1973) regressions and panels. Stock Illiq is the lagged monthly average of absolute

returns divided by volume (c.f. Amihud (2002)). CDS % b/a is the bid-ask spread as a percentage of

the midmarket quote for our sample of credit default swaps. The controls are log market capitalization,

stock volume, one month stock return, and one month stock standard deviation. Bank relationships are

as described in the text. For the Fama-MacBeth regressions, obs is the number of cross-sections, R2 is the

arithmetic average of the R2 s from the individual regressions, and the t statistics have been corrected for

six months autocorrelation. For the panels, the specification includes time fixed-effects, and the reported t

statistics are adjusted for clustering at the firm level.









37

Table 8: Credit Spread Regressions



(A) (B)



FM Panel FM Panel



r6mo -88.76 -101.4 -84.75 -102.8

(3.60) (5.57) (3.49) (5.50)

σ6mo 390.7 542.2 370.6 524.4

(9.32) (7.15) (7.70) (6.98)

debt 16.75 11.31 14.13 14.04

(2.50) (1.07) (1.64) (1.73)

leverage 93.78 138.4 103.3 132.4

(4.77) (2.18) (3.88) (2.21)

tangible 64.89 47.29 68.63 45.04

(2.08) (1.24) (2.41) (1.27)

rating -21.42 -17.14 -20.65 -17.75

(6.17) (3.72) (5.38) (4.20)

banks -0.040 0.407

(0.11) (0.40)

bid/ask -27.50 2.67

(0.71) (0.05)

obs 891 39988 891 39988

R2 0.5867 0.5167 0.6079 0.5161



The table shows regressions of credit default swap levels on proxies for asymmetric information. The controls

are lagged 6-month equity return and standard deviation, log book value of debt, leverage using market value

of equity, tangible asset ratio, and credit rating. Bank relationships are as described in the text. The CDS

bid/ask spread is expressed as a percentage of the CDS level. For the Fama-MacBeth regressions, obs is the

number of cross-sections, R2 is the arithmetic average of the R2 s from the individual regressions, and the t

statistics have been corrected for six months autocorrelation. For the panels, the specification includes time

fixed-effects, and the reported t statistics are adjusted for clustering at the firm level.









38

Appendix A: Corporate entities with CDS and Stock market data

in our sample from Jan 2001 till Oct 2004





ALBERTSONS INC HILTON HOTELS CORP

AMR CORP INTERNATIONAL BUSINESS MACHINES CORP

AMERICAN INTERNATIONAL GROUP INTERNATIONAL PAPER CO

AOL TIME WARNER INC INTERPUBLIC GROUP COS. INC

AT&T CORP LIBERTY MEDIA CORP

AT&T WIRELESS SERVICES INC LOCKHEED MARTIN CORP

BELLSOUTH CORPORATION LUCENT TECHNOLOGIES INC

BOEING CO MARRIOTT INTERNATIONAL INC

BURLINGTON NORTHERN SANTA FE CORP MAY DEPARTMENT STORES CO

CAMPBELL SOUP CO MAYTAG CORP

CARNIVAL CORP MGM MIRAGE INC

CATERPILLAR INC MOTOROLA INC

CENDANT CORP NEIMAN MARCUS GROUP INC

CENTEX CORP NEWS AMERICA INC

CITIZENS COMMUNICATIONS CO. NORDSTROM INC

COCA-COLA ENTERPRISES INC NORFOLK SOUTHERN CORP

COMCAST CABLE COMMUNICATIONS INC NORTHROP GRUMMAN CORP

COMPAQ COMPUTER CORP OMNICOM GROUP

COOPER TIRE & RUBBER PARK PLACE ENTERTAINMENT CORP

COX COMMUNICATIONS INC PHILIP MORRIS COS INC

CSX CORP QWEST CAPITAL FUNDING INC

CVS CORP RAYTHEON CO

DANA CORP RJ REYNOLDS TOBACCO HOLDINGS INC

DEERE AND CO SAFEWAY INC

DELL INC SBC COMMUNICATIONS INC

DELPHI CORP SEARS ROEBUCK ACCEPTANCE

DELTA AIRLINES INC SOUTHWEST AIRLINES CO

DOW CHEMICAL CO SPRINT CORP

EASTMAN KODAK CO SUN MICROSYSTEMS INC

ELECTRONIC DATA SYSTEMS CORP TARGET CORP

ENRON CORP TOYS R US INC

FEDERATED DEPARTMENT STORES INC TRW INC

FEDERAL EXPRESS CORP TYCO INTERNATIONAL LTD

FORD MOTOR CREDIT CO VERIZON GLOBAL FUNDING CORP

GENERAL ELECTRIC CAPITAL CORP VIACOM INC

GENERAL MOTORS ACCEPTANCE CORP VISTEON CORP

GEORGIA-PACIFIC CORP WAL-MART STORES INC

GOODYEAR TIRE AND RUBBER CO WALT DISNEY CO

HARRAHS OPERATING CO INC WORLDCOM INC

HEWLETT-PACKARD CO







39

Appendix B: Syndicated Loan Originating/Participating Banks





ABN AMRO Bank Deutsche Bank National City Corp.

Allfirst Bank DG Bank NationsBank

ANZ Banking Group Dresdner Bank NatWest Bank

Banca Commerciale Italiana Fifth Third Bank Norddeutsche Landesbank

Banca di Roma First Chicago Corp. Northern Trust Corp.

Banca Nazionale del Lavoro First Hawaiian Bank PNC Bank

Banca Popolare di Milano First Tennessee Bank Rabobank

Banco Bilbao Vizcaya Argentaria First Union Corp. Regions Bank

Bank Brussels Lambert Firstar Bank Royal Bank of Canada

Bank of America Fleet Bank Royal Bank of Scotland

Bank of Boston Fortis Bank Sakura Bank

Bank of Hawaii Fuji Bank Salomon Smith Barney

Bank of Montreal Goldman Sachs & Co San Paolo IMI

Bank of New York Hibernia National Bank Santander Central Hispano

Bank of Nova Scotia HSBC Sanwa Bank

Bank of Tokyo-Mitsubishi HypoVereinsbank Societe Generale

BANK ONE Corp. Industrial Bank of Japan Standard Chartered Bank

Bankers Trust Co. ING Bank State Street Bank & Trust

Barclays Bank IntesaBci Sumitomo Bank

Bayerische Hypo-und Vereinsbank J.P. Morgan Suntrust

Bayerische Landesbank JP Morgan-Chase Swiss Bank Corp.

BNP Paribas KBC Bank Tokai Bank

CIBC KeyCorp Toronto Dominion Bank

CIC Banques Kredietbank International U.S. Bancorp

Citicorp Lehman Brothers UFJ Bank

Comerica Bank Lloyds Bank Union Bancorp

Commerzbank Long Term Credit Bank Union Bank of Switzerland

CoreStates Bank Mellon Bank Wachovia Bank

Credit Agricole Merrill Lynch & Co Wells Fargo Bank

Credit Lyonnais Mitsubishi Trust & Banking Westdeutsche Landesbank

Credit Suisse Mizuho Bank WestLB

Crestar Bank Morgan Stanley Westpac Banking Corp.

Dai-Ichi Kangyo Bank National Australia Bank William Street

Danske Bank









40