ARTICLE IN PRESS
Journal of Financial Economics 82 (2006) 251–288
www.elsevier.com/locate/jfec
Market transparency, liquidity externalities, and
institutional trading costs in corporate bonds$
Hendrik Bessembindera,Ã, William Maxwellb, Kumar Venkataramanc
aDavid Eccles School of Business, University of Utah, Salt Lake City, UT, 84112, USA
bEller School of Management, University of Arizona, Tucson, AZ, 85721, USA
cCox School of Business, Southern Methodist University, Dallas, TX, 75275, USA
Received 21 April 2005; received in revised form 8 September 2005; accepted 3 October 2005
Available online 18 May 2006
Abstract
We develop a simple model of the effect of public transaction reporting on trade execution
costs and test it using a sample of institutional trades in corporate bonds, before and after
initiation of the TRACE reporting system. Trade execution costs fell approximately 50% for bonds
eligible for TRACE transaction reporting, and 20% for bonds not eligible for TRACE
reporting, suggesting the presence of a ‘‘liquidity externality.’’ The key results are robust to
changes in variables, such as interest rate volatility and trading activity that might also affect
execution costs. Market shares and the cost advantage to large dealers decreased post-TRACE.
$The authors thank Yakov Amihud, Tihomir Asparouhova, Shmuel Baruch, Ekkehart Boehmer, Cliff Ball,
Jeremy Graveline, Joel Hasbrouck, Mike Lemmon, Ananth Madhavan, Mark Shenkman, Hans Stoll, Rex
Thompson, Christopher Vincent, Arthur Warga, an anonymous referee, and seminar participants at Vanderbilt
University, the NBER Microstructure Conference, the University of Virginia, Claremont-McKenna College,
Brigham Young University, Indiana University, Southern Methodist University, the University of Houston, the
Stockholm School of Economics, the Northern Finance Association Meetings, and the University of Utah for
valuable comments. Thanks are also due to Lehman Brothers and MarketAxess for the provision of data. The
NAIC bond pricing data and Fixed Income Securities database are obtained from the Fixed Income Research
Program of the Bauer College of Business at the University of Houston. The second author acknowledges
financial support provided by Moody’s Credit Market Research Fund. Earlier versions of this paper were titled
‘‘Optimal Market Transparency: Evidence from the Initiation of Trade Reporting in Corporate Bonds’’ and
‘‘Market Transparency and Institutional Trading Costs.’’
ÃCorresponding author.
E-mail address: finhb@business.utah.edu (H. Bessembinder).
0304-405X/$ - see front matter r 2006 Elsevier B.V. All rights reserved.
doi:10.1016/j.jfineco.2005.10.002

---

ARTICLE IN PRESS
252
H. Bessembinder et al. / Journal of Financial Economics 82 (2006) 251–288
These results indicate that market design can have first-order effects, even for sophisticated
institutional customers.
r 2006 Elsevier B.V. All rights reserved.
JEL classification: D82; G14; G19
Keywords: Corporate bonds; Institutional trading costs; Market transparency
1. Introduction
Security markets vary greatly in their transparency, that is, in the amount of information
regarding market conditions made public on a timely basis. Equity markets generally
disseminate continuous pre-trade information, such as best quotations and, in some cases,
information about unexecuted limit orders, and immediately report prices and sizes of
completed trades. Most futures markets report trades, but do not disseminate pre-trade
information. Foreign exchange markets disseminate only non-binding ‘‘indicative’’
quotations to the public, and do not report transactions at all. Corporate bond markets
were traditionally similarly opaque, with quotations available only to a few market
professionals, and with no public transaction reporting.
Market transparency has been the subject of a handful of studies, but neither the
theoretical predictions nor the empirical evidence is conclusive as to whether market
quality is enhanced by increased transparency. In this paper, we develop a simple model of
how more precise estimation of values occasioned by increased transparency can affect
trade execution costs, and test its implications by estimating institutional trading costs in
corporate bonds when the National Association of Securities Dealers (NASD) began to
publicly report transactions in approximately 500 corporate bond issues through its trade
reporting and compliance engine (TRACE) on July 1, 2002. The initiation of TRACE
transaction reporting provides a potentially powerful experiment for assessing whether
transparency is important to market quality, because the corporate bond markets were
quite opaque prior to TRACE. In contrast, previous empirical work on transparency
studied relatively small changes in the already quite transparent equity markets.
Improved transparency has the potential to reduce trade execution costs for corporate
bonds, for at least two reasons. First, transparency might reduce market-maker rents.
Pagano and Roell (1996) emphasize that opaque markets tend to benefit relatively well
informed dealers in their negotiations with customers. Also, transactions reported through
the TRACE system can be monitored by self-regulatory agencies and by the United States
Securities and Exchange Commission (SEC). The potential role of increased transparency
in improving customers’ ability to control and evaluate trade execution costs has been
emphasized by Annette Nazareth, Director of the Division of Market Regulation of the
SEC:
For investors as well as regulators, the difficulty lies in establishing the prevailing
market price for a bond. This generally is the base line that is used to assess whether a
mark-up (trade execution cost) is reasonabley Improved transparency will enable
investors to better determine the fair price of a bond. This will make them better able
to protect themselves against unfair pricingy’’ (excerpted from testimony before the

---

ARTICLE IN PRESS
H. Bessembinder et al. / Journal of Financial Economics 82 (2006) 251–288
253
United States Senate Committee of Banking, Housing, and Urban Affairs, June 17,
2004).
In addition to potential reductions in market-maker rents, improved transparency could
decrease market-making costs. Naik et al. (1999) develop a model implying that improved
transparency in a dealer market can improve inventory risk sharing, thereby decreasing
inventory carrying costs.
Among the more important recent studies of bond markets, Green et al. (2004) and
Harris and Piwowar (2005) examine trades in municipal bonds. Each study reports that
small trades pay much larger percentage trading costs than large trades, and each set of
authors conjectures that this may occur because unsophisticated small investors cannot
readily evaluate the trading costs they pay in the opaque market for municipal bonds.
However, neither study provides direct evidence on the relation between market
transparency and trading costs.
This study provides direct evidence on the issue by analyzing trade execution costs for
institutional (insurance company) transactions in corporate bonds before and after the
introduction of transaction reporting for corporate bonds through TRACE. The results
indicate average reductions in one-way trading costs for corporate bonds subject to
TRACE transaction reporting of five to eight basis points. These estimated reductions in
trading costs average 40–60% of pre-TRACE trading cost estimates, and equate to
approximately $2,000 per trade in the present sample of insurance company transactions.
Extrapolating beyond the present sample, we estimate market wide trading cost reductions
of roughly $1 billion per year after the initiation of TRACE transaction reporting. The key
empirical results are robust to the inclusion of control variables such as interest rate
volatility and bond market trading activity that might alternately explain variation in trade
execution costs. We also find that trading activity is less concentrated with large dealers
and that the large dealer cost advantage previously documented by Schultz (2001) is
reduced post-TRACE. Collectively, these results indicate that the public reporting of
corporate bond trades has had first-order effects on market quality, even for the relatively
sophisticated institutional traders that are the focus of this study.
In a contemporaneous paper, Edwards et al. (2006) examine determinants of cross-
sectional variation in trade execution costs for corporate bonds using a comprehensive but
proprietary database of transactions during 2003. Among other findings, they report that
one-way transaction costs for those bonds whose trades are publicly disseminated through
TRACE are one to four basis points lower, after controlling for other relevant factors.
Similar point estimates are reported by Goldstein et al. (2005), who examine trading costs
for BBB-rated bond issues during 2003, using a proprietary database provided by NASD.
A distinction between this analysis and that provided by Edwards et al., and Goldstein
et al., is that we estimate the effect of TRACE reporting on bond market quality around
the time that public reporting of trades through TRACE was first initiated, on July 1, 2002,
while Edwards et al., and Goldstein et al., consider the impact of transparency on
execution costs during 2003, after transaction reporting had already been introduced for
about 500 bond issues. This distinction is important if transaction reporting for some bond
issues also improves market quality for other issues. The model we develop implies that
this should indeed be the case. Consistent with this reasoning, Amihud et al. (1997)
document that an improvement in the trading mechanism used for a subset of Tel Aviv
Stock Exchange securities led to enhanced liquidity not only for the affected stocks, but

---

ARTICLE IN PRESS
254
H. Bessembinder et al. / Journal of Financial Economics 82 (2006) 251–288
also for correlated stocks with no change in trading mechanism. They coined the phrase
liquidity externality, which they attributed to ‘‘the fact that improved value discovery for
one security facilitates value discovery for the other (correlated) security.’’
If, as our model implies, a similar liquidity externality exists for corporate bonds, then
an analysis of trade execution costs after TRACE initiation will understate the importance
of trade transparency in reducing transaction costs. A liquidity externality seems
particularly plausible for corporate bonds, since market practitioners often estimate the
value of non-traded bonds based on ‘‘matrix’’ pricing that incorporates bond
characteristics and observed prices for bonds that do trade.1 Improved information about
market transactions in some bonds should allow more accurate valuation and better
monitoring of trade execution costs for non-reported bonds as well.
Consistent with the existence of liquidity externalities, we document that one-way
trading costs for non-TRACE-eligible bonds decreased by about three and a half basis
points on average after transaction reporting through TRACE was initiated in July 2002.
For non-TRACE-eligible bonds issued by firms in the same industry as a firm with at least
one bond issue eligible for TRACE reporting, the reductions in one-way trading costs are
larger, averaging about five basis points. More to the point, the estimated five to eight basis
point reduction in trading costs for TRACE-eligible bonds reported here is substantially
larger than the 2.1 basis point cross-sectional estimate for large trades reported by
Edwards et al.
The difference in the point estimates between this study and Edwards et al., points out a
broader issue in the literature, namely the use of matched firm or control samples in
general. While a matched approach is usually the preferred methodology, in applications
where the effects of a treatment spill over to the general population it will not be possible
to measure the effect of the treatment by comparing treated and non-treated observations.
Finding larger trading cost reductions in the present sample is all the more striking since
we measure trading costs for institutional transactions. If opaqueness is primarily a
problem for naı¨ve individual investors then we should observe little or no effect of TRACE
reporting. In contrast, the substantial effects documented here support the conclusion that
transparency is important to institutional customers as well.
This paper is organized as follows. Section 2 reviews recent papers on bond markets and
on market transparency. In Section 3 we present a theoretical model that examines the
potential role of transaction reporting on bond valuation and trading costs. Section 4
discusses methods for obtaining trading cost estimates for corporate bonds, while Section 5
describes the available data and some implementation issues that arise. Section 6 presents
the key empirical results and tests of robustness, while Section 7 concludes.
2. The recent literature on bond markets and transparency
2.1. Recent studies of the bond markets
The increasing availability of data has spurred a substantial volume of recent research
focused on bond markets. Hotchkiss and Ronen (2002) study a sample of 55 high-yield
1Both buy and sell side bond traders typically have matrix pricing information available on their computer
displays. Sell-side traders can also view an automatic matrix valuation for any bond based on observed trades in
bonds that are similar in terms of credit rating and maturity.

---

ARTICLE IN PRESS
H. Bessembinder et al. / Journal of Financial Economics 82 (2006) 251–288
255
bonds. The source of their data was the fixed income pricing system (FIPS), a predecessor
to TRACE, which disseminated hourly summary reports on the pricing of a select set of
high-yield bonds. They focus on the relative information efficiency of stock and bond
markets, reporting that stock price changes do not systematically lead bond price changes,
and that market quality as measured by pricing errors in the Hasbrouck (1993) framework
is similar across stocks and bonds.
Schultz (2001) provides estimates of trading costs for a large sample of corporate bonds.
He obtains a dataset of insurance company trades in corporate bonds from Capital Access
International (CAI). For the period January 1995 to April 1997 he estimates average
round-trip trade execution costs of about 27 basis points. Schultz also reports that active
institutions pay less than inactive institutions, which suggests that trading costs in the
relatively opaque pre-TRACE bond markets depended in part on the customer’s degree of
sophistication and familiarity with the bond markets.
Chakravarty and Sarkar (2003) also study the CAI database of insurance company bond
transactions and report average trading costs for municipal bonds of 23 basis points,
compared to 21 basis points for corporate bonds and 8 basis points for Treasury bonds.
They also report that in the cross-section, spreads rise with bond maturity and credit risk,
and fall with trading volume. Hong and Warga (2000) compare insurance company trades
in the CAI database to trades on the NYSE Automated Bond System, reporting similar
bid-ask spreads on each. Chen et al. (2003) adapt the methodology of Lesmond et al.
(1999) to examine the liquidity of corporate bonds, using proprietary data from
Bloomberg, and document that their liquidity measure can explain 16% of the variation
across bonds in yield spreads. This finding is potentially important because it implies that
market liquidity not only determines transactions costs, but may also affect the valuation
of the bonds themselves.
In a related study, Edwards et al. (2006) examine trading costs for corporate bonds.
Their non-public sample includes all transactions reported to the NASD during calendar
year 2003, including those made public through TRACE and those not made public.
Edwards et al., document several important empirical regularities regarding corporate
bond trading costs. First, they report that corporate bond trading costs also decrease with
trade size, a result that the authors attribute to small traders’ lack of sophistication in
combination with limited transparency. Second, they provide evidence as to cross-sectional
variation in corporate bond trading costs, documenting that costs increase with time from
issue, and decrease with better credit rating, issue size, bond complexity, the presence of a
floating interest rate and a previous private equity issue. Third, they address the role of
transparency, reporting that in the cross-section trade execution costs are lower for bonds
whose trades are publicly disseminated through TRACE, after controlling for variation in
other characteristics. In particular, their Table 5 reports point estimates indicating that
one-way trade execution costs are reduced by 0.9 basis points for $10,000 trades, 2.9 basis
points for $20,000 trades, 3.8 basis points for $100,000 trades, and 2.1 basis points for $1
million trades. Edwards et al., also report broadly similar point estimates from a time-
series experiment, with execution costs for a set of bonds phased into the public
dissemination of trade reports during 2003 declining by about three to four basis points on
average. Given the result obtained here that liquidity externalities exist for corporate
bonds, the point estimates provided by Edwards et al., may be viewed as quantifying the
effect of public dissemination of trade information through TRACE conditional on
transaction prices for other bonds already being available through TRACE, and therefore

---

ARTICLE IN PRESS
256
H. Bessembinder et al. / Journal of Financial Economics 82 (2006) 251–288
providing a lower bound on the overall effect of TRACE reporting on corporate bond
execution costs.
2.2. Studies of market transparency
Security market transparency refers to the amount of information regarding market
conditions and transactions made public on a timely basis. Transparency is often
categorized as pre-trade transparency, which concerns the dissemination of quotations or
other indications of trading interest (such as unexecuted orders in the limit order book),
and post-trade transparency, which concerns the dissemination of data about completed
trades. Markets that disseminate little or no price data are referred to as being opaque, or
non-transparent.
Madhavan (1995) models a competitive dealer market in which all dealers are required to
publicly disclose trades, and compares the resulting equilibrium to that obtained when trade
disclosure is not mandatory for some or all dealers. His analysis confirms the existence of an
equilibrium where dealers choose to not disclose trades, because they profit from the
associated reduction in price competition. These dealers will also post narrower spreads at the
initiation of trading in order to attract informative order flow. Large informed and liquidity
traders who spread their orders across multiple periods benefit from the lack of trade
disclosure in this model, since they can transact with dealers who are unaware of their previous
trades. However, since the model assumes a competitive market, dealers break even on
average, and these gains come at the expense of small noise traders who arrive later.
Pagano and Roell (1996) focus on the effect of trade disclosure on the strategies of price
setters (market makers or limit order traders). Their model implies that liquidity traders
will pay lower average trading costs in the more transparent market, though not
necessarily for all trade sizes. Naik et al. (1999) develop a model implying that improved
transparency in a dealer market can improve inventory risk sharing, thereby decreasing
inventory carrying costs. Consistent with these predictions, Flood et al. (1999) provide
experimental evidence that pre-trade transparency reduces bid-ask spreads.
However, some theoretical analyses predict that less transparent markets might improve
liquidity. In particular, Bloomfield and O’Hara (1999) argue that an opaque market might
give market makers incentives to quote narrow bid-ask spreads, because the order flow
attracted by narrow spreads contains valuable information about market fundamentals,
and Bloomfield and O’Hara (2000) provide experimental evidence generally consistent
with this reasoning. The empirical evidence from actual asset markets is inconclusive, in
part because structural changes in the transparency of actual markets are rare. Gemmill
(1996) examines the London Stock Exchange after two changes in required post-trade
transparency for large block trades, and does not detect any change in liquidity.
Madhavan et al. (2005) examine the liquidity of the Toronto Stock Exchange when during
1990 it began to publicly disseminate its limit order book, and document increased
execution costs and greater price volatility after the increase in pre-trade transparency.
Boehmer et al. (2005) examine the effect when the New York Stock Exchange began to
disseminate limit order book information in January 2002. They document that limit order
traders are able to use the information to refine their strategies and, in contrast to the
findings of Madhavan et al., report improved liquidity as measured by transaction costs
and the informational efficiency of prices. The inconclusive results obtained from these
empirical studies could reflect that they studied changes in transparency in equity markets

---

ARTICLE IN PRESS
H. Bessembinder et al. / Journal of Financial Economics 82 (2006) 251–288
257
that were already quite transparent. The present analysis, in contrast, focuses on a market
that was almost entirely opaque prior to the initiation of transaction reporting.
2.3. The transparency of the bond markets
The corporate bond market traditionally reported trades only to the parties involved, so
investors could not compare their own execution price to other transactions. Even
institutional investors had to invest significant time and effort to obtain market
information, and were limited in their ability to compare their transaction prices to those
of other investors. Limited information regarding current prices, in the form of
‘‘indicative’’ quotes, was available to institutional investors through a messaging system
provided by Bloomberg. Investors could use this system to indicate interest in buying or
selling a particular issue in an effort to solicit bids or offers, or could telephone dealers for
quotes. The situation was even more difficult for individual investors, who were precluded
from accessing virtually all real-time market information.
In an effort to bring greater transparency to the bond markets and provide additional
regulatory oversight, the United States Securities and Exchange Commission (SEC) on
January 31, 2001 approved rules requiring the NASD to report all over-the-counter
secondary market transactions in a specified set of corporate bonds. The requirement
initially applied to a set of 498 bonds with issuance size of $1 billion or greater, and was
implemented July 1, 2002.
During the 2002 sample period NASD members were required to report all corporate
bond transactions to the TRACE system within 1 h and 15 min. For each trade the member
is required to report bond identification (CUSIP or NASD symbol), the date and time of
execution, trade size, trade price, yield, and a buy or sell indication. However, not all of the
reported information is disseminated to the public: investors receive bond identification,
the date and time of execution, and the price and yield for bonds specified as TRACE-
eligible. Trade size is provided for investment-grade bonds if the par value transacted was
$5 million or less, otherwise an indicator variable denotes a trade of more than $5 million.
Investors can access the trade information on the NASD website without charge, but with
a four-hour delay. The information is also retransmitted without delay via third-party
vendors to subscribing investors. Institutional investors typically rely on a third-party
vendor to disseminate the pricing information in an easily accessible and useable format,
with MarketAxess apparently being the most widely used.
3. A simple model of market transparency and price efficiency
3.1. Estimating bond values
In this section, we present a model that examines how improved precision in valuation
affects transaction costs in corporate bonds. The model considers the case of two bonds,
but the intuition is readily generalized to the case of multiple bonds. Assume that the time t
transaction price for bond j is
Pjt ¼ EtðV jÞ þ aSjQjt,
(1)
where Qjt is an indicator variable that equals 1 for buyer-initiated trades and À1 for seller-
initiated trades, Et(Vj) is the estimated value of bond j at time t, conditional on knowledge

---

ARTICLE IN PRESS
258
H. Bessembinder et al. / Journal of Financial Economics 82 (2006) 251–288
of Qjt, and aSj is the one-way transaction cost paid by customers to transact in bond j.
More precisely and as discussed further in Section 4, aSj is the non-information portion of
the bid-ask spread, attributable to inventory costs, order processing costs and, possibly,
market-maker rents. Bond value is estimated imperfectly, with valuation error defined as
jt  V j À EtðV jÞ, where jt$Nð0; s2 Þ. We posit that transaction costs increase with the
j
variance of the valuation errors, for two complementary reasons. First, greater valuation
errors increase the inventory-related risks of market making, for which risk-averse market
makers must be compensated in equilibrium. Second, greater valuation errors can increase
the market power of dealers, increasing the likelihood that economic rents can be extracted
from less informed customers. The market maker in bond j observes noisy signals of value
for bond j and also for bond k. Suppressing time subscripts, the signals are
Y j  V j þ Uj,
(2a)
Y k  V k þ Uk;
(2b)
where U j$Nð0; s2 Þ and U
Þ. We posit that improved market transparency due
Uj
k$N ð0; s2
Uk
to TRACE reporting will improve the precision of these signals by reducing the variance of
the noise components Uj and Uk. In particular, observing past transactions helps the
dealer, as well as customers, make more accurate inferences about bond value. Let V
denote the unconditional mean valuation for all bonds, and define individual bond
valuation factors as
I j  V j À V ,
(3a)
I k  V k À V ;
(3b)
where I j$Nð0; s2 Þ and I
Þ. We assume that all random variables except I
Ij
k $N ð0; s2
Ik
j and Ik
are independent of each other. The covariance between Ij and Ik, denoted sIj,Ik, allows for
the possibility that common factors affect the values of both bonds j and k.
Given joint normality, the projection theorem implies that a market maker’s optimal
estimate of value is obtained as a linear projection of value on observed signals. Letting bjj
and bjk denote the weights that are placed on the bond j and bond k signals, respectively,
the linear estimate of bond j value is
EðV jÞ ¼ bjjY j þ bjkY k;
(4)
implying that the valuation error can be expressed as
j  V j À EðV jÞ ¼ Vð1 À bjj À bjkÞ þ Ijð1 À bjjÞ À bjjUj À bjkðIk þ UkÞ;
(5)
with variance:
s2 ¼ s2 ð1 À b
b2 þ ðs2 þ s2 Þb2 À 2ð1 À b
j
Ij
jj Þ2 þ s2
Uj jj
Ik
Uk
jk
jj Þðbjk Þ sIj;Ik .
(6)
Notably, the larger the covariance between bond values, the lower is the variance of the
bond k price error, implying lower spreads when bond values have a greater common
component, ceteris paribus. The projection theorem implies that the linear coefficients are
given as
bjj ¼ A=B
(7a)

---

ARTICLE IN PRESS
H. Bessembinder et al. / Journal of Financial Economics 82 (2006) 251–288
259
and
bjk ¼ 1 À A=B;
(7b)
where A  ðs2 þ s2 þ s2 À s2
Þ and B  A þ s2 . Noting that B exceeds A by the noise
Ij
Ik
Uk
Ij;Ik
Uj
in the bond j signal, s2 , we observe that the bond j market maker optimally places
Uj
increasing weight on the bond k value signal as the bond j signal becomes less precise.
Further, an increase in the covariance of bond j and k values, sIj,Ik, reduces the optimal
weight in the own bond signal (bjj) and increase weight in the related bond signal (bjk).
Substituting (7a) and (7b) into (6), the variance of the valuation error can be expressed
as
s2 ¼ Að1 À A=BÞ
(8)
j
3.2. The effect of noise reductions due to TRACE
As noted above, we posit that bond transaction reporting can affect bond j trading costs
by improving the precision with which bond values are estimated, i.e., by reducing noise in
the value signal for bond j, s2 , and in the related bond k, s2 . If transaction reporting
Uj
Uk
through TRACE were to eliminate the noise in the bond j signal ðs2 ¼ 0Þ, then A ¼ B and
Uj
s2 ¼ 0. In contrast, if TRACE eliminated the noise in the related bond k signal ðs2 ¼ 0Þ,
j
Uk
then the terms A and B are each reduced, and s2 is reduced but remains positive. Knowing
j
the exact value of the related bond k does not give the exact value of bond j, but it does
reduce the variance of valuation errors.
However, we do not anticipate that TRACE reporting will entirely eliminate noise in
valuation. Transaction prices are still made public with a delay, and in any case differ from
underlying values due to bid-ask bounce. (The public TRACE data does not include a buy-
sell indicator, so it is not possible to know with certainty whether a particular transaction
price lies above or below bond value.) To examine the effect of a TRACE-induced
reduction in noise, differentiate (8) with respect to s2 and s2 , respectively, to get
Uj
Uk
qs2 =qs2 ¼ ðA=BÞ240,
(9a)
j
Uj
qs2 =qs2 ¼ ½1 À ðA=BÞŠ240.
(9b)
j
Uj
The right sides of both (9a) and (9b) are positive, implying that trade reporting will
reduce valuation errors and spreads for bond j, if either own bond j prices are reported
(9a), or if related bond k prices are reported (9b). The latter is a formalization of the
liquidity spillover argument.
Interestingly, the effect of bond j transaction reporting on bond j spreads is not
necessarily stronger than the externality effect of bond k transaction reporting on bond j
spreads. The latter effect can be greater if the ratio of A to B is sufficiently small. This ratio
declines with the noise in the bond j signal, s2 . In the case of a bond whose own value is
Uj
estimated with a great deal of noise, it could actually be more beneficial to reduce noise in
the signal on a related bond that is already valued more precisely than to reduce the noise
in the own bond value signal.
Finally, the magnitude of the liquidity externality is greater if bonds are more similar in
the sense that sIj,Ik is greater. To demonstrate this, differentiate the right side of (9b) with

---

ARTICLE IN PRESS
260
H. Bessembinder et al. / Journal of Financial Economics 82 (2006) 251–288
respect to sIj,Ik to obtain
@2s2 =@s2 @s
j
Uk
Ij;Ik ¼ 4½1 À ðA=BÞŠ2=B40,
(10)
which implies that the reduction in bond j valuation errors and spreads that come from a
reduction in the noise with which bond k values are estimated is greater if the bond values
are more closely related.
We summarize the implications of this model as follows. If (i) transaction costs in any
individual bond are increasing in the variance of that bond’s pricing errors, and (ii)
TRACE reporting improves the precision of value estimates, then we have the following:
H1: Direct effect. TRACE transaction reporting will reduce transaction costs for bonds
whose trades are disseminated through TRACE.
H2: Liquidity externality effect. TRACE transaction reporting for some bonds will
reduce transactions costs for other bonds, and ceteris paribus the effect will be stronger
for bonds with lower tracking error, i.e., higher covariance with TRACE-eligible bonds.
4. Measuring trading costs in bond markets
Most studies of trade execution costs have focused on equity markets, and are able to
exploit the existence of reliable quotation databases to construct measures of quoted (ask
price less bid price) and effective (trade price relative to quotation midpoint) spreads. In
contrast, data on bid and ask quotations are not broadly available for bond markets.
However, some bond transaction databases do indicate whether a dealer participated as a
buyer or a seller. As a consequence, several studies, including this one, adopt variations of
indicator variable regressions to estimate trade execution costs for bonds.
The indicator variable model used here is related to those suggested by Huang and Stoll
(1997), Madhavan et al. (1997), and Schultz (2001). Let S denote the effective one-way
spread, i.e., half the difference between the price at which dealers will sell a bond and the
price at which they will purchase the bond, initially assumed to be constant. Consistent
with prior theory, we assume that the spread contains an informational component, gS,
and a non-informational component aS that can reflect inventory costs, order processing
costs, and possible economic rents, where a(1–g). As in Section 3, Qt denotes an indicator
variable that equals 1 if the time t trade is a customer buy and À1 if it is a customer sell,
and Pt denotes a transaction price at time t, which equals the ask quote when Qt ¼ 1 and
the bid quote when Qt ¼ À1. V is the unobservable true value of the bond, and Et(V) is the
market maker’s estimate of bond value, conditional on knowledge of whether the time t
trade is a customer buy or sell. Transaction prices are given by Eq. (1), reproduced here
with the subscript identifying bond j suppressed:
Pt ¼ EtðV Þ þ aSQt.
(1a)
The market maker’s estimate of bond value is updated due to surprises in order flow,
Qt*Qt–EtÀ1(Qt), and due to public information revealed since the prior period, denoted
Zt:
EtðV Þ ¼ EtÀ1ðV Þ þ gSQÃ þ Z
t
t.
(11)
Madhavan et al. (1997), Huang and Stoll (1997), and others have documented that order
flow in equity markets is positively autocorrelated. To allow for this possibility we follow

---

Document Outline

• Market transparency, liquidity externalities, and institutional trading costs in corporate bonds
  • Introduction
  • The recent literature on bond markets and transparency
    • Recent studies of the bond markets
    • Studies of market transparency
    • The transparency of the bond markets
  • A simple model of market transparency and price efficiency
    • Estimating bond values
    • The effect of noise reductions due to TRACE
  • Measuring trading costs in bond markets
    • Discussion
  • Data description and implementation issues
    • Data sources and description
    • Estimating trade execution costs without transaction times
  • Empirical results
    • Descriptive data
    • The effect of TRACE reporting on TRACE-eligible bonds
    • The effect of TRACE reporting on non-TRACE-eligible bonds
    • Trade size analysis
    • Robustness tests
      • Alternative cost measures
      • Controlling for variation in the economic environment
    • Dealer market shares
    • Time trend
    • Expansion of TRACE-eligible bonds in 2003
  • Conclusion
  • Simulation evidence on potential bias in coefficients estimated without time stamps
  • References

---