; Institutional Investment in Syndicated Loans
Documents
Resources
Learning Center
Upload
Plans & pricing Sign in
Sign Out
Your Federal Quarterly Tax Payments are due April 15th Get Help Now >>

Institutional Investment in Syndicated Loans

VIEWS: 12 PAGES: 63

  • pg 1
									             Institutional Investment in Syndicated Loans




                                         Abstract

         A recent innovation in the syndicated loan market has been the arrival of
institutional investors, including hedge funds, hybrid funds and collateralized loan/debt
obligation (CLD/CLO) managers. This paper focuses on the institutional loan market, an
area of little prior empirical research, and finds that institutional loans have higher mean
primary yield spreads (about 50bps) than bank loans, ceteris paribus. Moreover,
institutional loans behave differently than bank loans on the secondary market, as
evidenced by higher first trading day return, greater price volatility, higher liquidity, and
a shorter holding period by their original lenders. Following information-based theories
in the IPO literature (Rock (1986), and Chemmanur (1993) among others), we conjecture
that the higher yield on institutional loans may serve as compensation for attracting
sufficient demand and also ensure the continuous participation of the institutional
investors, who are the less informed investors in the syndicated loan market.
1 Introduction

     The past decade has seen significant changes in the makeup of the syndicated loan

market as growing numbers of institutional investors, including prime funds, hedge

funds, hybrid funds, collateralized loan/debt obligations (CLOs/CDOs) managers and
                                                                1
insurance companies have entered the market.                        According to the Loan Pricing

Corporation, the volume of institutional issuance has shifted from a being a footnote in

the early 1990s to being a considerable portion of the current syndicated leveraged loan

market (as shown in Figure 1).2 Clearly, the entrance of institutional investors has been

instrumental to the increased liquidity in the secondary loan market .This has been

attributed to institutional investors having higher liquidity requirements for their assets,

based on their investment objectives and constraints.3

     FIGURE 1 GOES HERE

     A classic leveraged loan deal has a multi-tranche structure, consisting of a credit

revolver and several classes of term loans. Loans are structured to accommodate two

primary types of syndicated lenders: banks and institutional investors.4 Typically, the

credit revolver and term loan A (TLA), which are called “pro rata debt”, are designed to

be sold to retail commercial banks while term loans B, C, and D (TLB, TLC and TLD),

1
  A hedge fund is an aggressively managed fund that takes positions usually in speculative investments and
focuses on secondary opportunities and hybrids in the loan market. Hybrid refers to loans that, due to their
position in the capital structure or due to the borrower’s credit risk, are outside the normal risk-return
spectrum of senior, secured loans. Examples include second liens, first liens, second priority, loans to high-
risk companies. A prime fund is a mutual fund that invests in leveraged loans. A CLO (Collateralized loan
obligation) is a structured vehicle that invests in loan assets with the goal of arbitraging the spreads
between the asset and liabilities tranches.
2
  As defined by LPC, the syndicated loan market comprises of the investment grade loan market, the
leveraged loan market, and the other market. Nearly all institutional loans fall in the leveraged loan market
segment.
3
  For example, one of the investment constraints of a prime fund is that its asset holding must be marked-
to-market.
4
  For a detailed description of the process and structure of this market, please refer to Miller (2004).
which are called “institutional debt”, are structured to be sold to institutional investors.

The overwhelming majority of institutional loans are leveraged, non-investment grade

loans (with credit rating below BBB-) with longer maturity (over 5 years, on average).5

     Despite the increasing importance of institutional lending in the corporate loan

market, to the best of my knowledge, no academic research paper has addressed this

sector. Drawing on the primary loan market database—Dealscan and a unique secondary

loan market database—Secondary Market Pricing Service (hereafter SMPS), this paper

presents an original analysis of institutional loans in terms of their primary pricing and

secondary market performance. Institutional investors participate in the loan market both

through the primary syndicated loan distribution process and secondary market trading.

As the most active loan traders, they promote mutual interaction between the primary and

secondary loan markets and improve loan market efficiency (Taylor and Yang (2004),

Yago and McCarthy (2004), Miller (2004)). However, the impact of institutional

investors on the loan market and on loan borrowers remains an open empirical question.

     Does the participation of institutional investors narrow or widen the premia paid on

loans? Do institutional tranches behave differently than bank tranches on the secondary

loan trading market? Our empirical tests indicate that institutional loan tranches are

initially, in almost all cases, priced at a higher yield than bank loan tranches, ceteris

paribus. Moreover, we find that the first-day trading prices on institutional loans resold

on the secondary market carry lower average discounts than comparable bank loans. It

appears that institutional lenders are systematically granted higher rates during the

primary loan distribution process and this premium is not eliminated during secondary

5
  The other three key market segments include the leveraged market, investment grade loan market and
secondary loan market. LPC classifies institutional loans as one of the four key loan market segments and
provides periodic analytical statistics for this segment.
loan trading. This is a striking finding that this paper first documents and attempts to

explain. For ease of exposition, we will henceforth refer to this as the ‘high spread

puzzle’.

    We conjecture that the lead agents of syndicated loans need to compensate

institutional investors by offering a higher spread to attract sufficient participation to

clear the market. A price sweetener would also motivate and invite costly information

production by less informed investors and ensure better secondary market performance

for the loan if it were resold by the original lenders. Further, a wide margin offered to

institutional investors can effectively prevent syndication failure and serves as

compensation for revealing their aggregate demand. We term this conjecture the “spread

compensation hypothesis”: institutional loan tranches are associated with higher interest

rates than bank loan tranches. The extra spread margin offered to original institutional

investors is also reflected by a relatively higher (less discounted) initial trading price on

the secondary loan resale market.

       There are obvious parallels between the loan syndication process and the

distribution process in Initial Public Offerings (IPOs) of equity. Both involve lead

underwriters (in IPOs) /lead arrangers (in loan syndications) as delegated agents

marketing and allocating the securities (equity/debt) issues. Like lead underwriters of

IPOs, lead arrangers of syndicated loans also need to gather information from the market

and guarantee to provide full funding under a firm commitment. Potential information

asymmetry problems exist in both financing practices. In loan syndications, lead banks,

as major agents hired by borrowers, originate and underwrite the loan while funding a
portion of the loan. They are perceived to be informed investors. Other participants in the

loan syndications are perceived to be relatively less informed.

       The information asymmetry between lead lenders and loan participants has been

widely discussed in the syndicated loan literature. Simons (1993), Dennis and Mullineaux

(2000), Panyagometh and Roberts (2005) and Sufi (2006) among others, study how the

principal amount of syndicated loans is split between lead lenders and other loan

participants to mitigate the costs of information asymmetry. The consensus is that lead

banks tend to retain a larger portion of lower quality loans while syndicating out a greater

fraction of better quality loans. However, the information asymmetry problem may also

be addressed in the loan tranche design. There is no theoretical model in the area of loan

syndications to reconcile the price gap between institutional loans and bank loans.

The enormous evidence in the IPO literature, however, has shown that new equity issues

are generally underpriced (Reilly and Hatfield (1969), Logue (1973) and Ibbotson (1975),

Ritter (1984), Hanley (1993) and Loughran and Ritter (2002) among others. Founded on

an assumption of information asymmetry between informed and uninformed investors,

Rock’s (1986) adverse selection model predicts that IPO underpricing serves as a way to

attract sufficient demand and provides compensation to uninformed investors for their

continuous participation in the IPO market. Chemmanur (1993) suggests that quality

firms tend to employ underpricing to invite information production and enhance

aftermarket equity prices. Welch’s (1992) informational cascade model shows that a

successful IPO relies on a successful initial sales effort. Since the full-subscription price

is never known ex ante, underpricing serves as a defense against the possibility of
issuance failure.6 Moreover, the dynamic information acquisition model proposed by

Benveniste and Spinte (1989) reveals that IPO underpricing compensates investors for

their truthfully revealing the aggregate demand to underwriters during the book-building

phase. In addition, Loughran and Ritter (2002) put forth a prospect theory and argue that

IPO underpricing helps underwriters with discretionary power to earn side-payments

produced from investors’ rent-seeking behaviors.

         IPO underpricing is quite similar to our findings of an association between

institutional loans and higher initial loan yield spreads as well as higher first-day loan

resale prices. Therefore, an explanation borrowed from the IPO literature might help to

understand this phenomenon.

         Another potential explanation draws on the corporate governance literature which

suggests that institutional shareholders are active and efficient firm monitors. A great

deal of evidence has shown the efficacy of institutional shareholders in the corporate

governance context (Demsetz (1983), Demsetz and Lehn (1985), Shleifer and Vishny

(1986), Allen, Bernardo and Welch (2000), Gorton and Schmidt (2000), Morck et al.

(2000), Hartzell and Starks (2003) and Roberts and Yuan (2006) among others).

Institutional investors are often referred to as “smart money” and their participation

delivers a positive message about the firm’s financial health

     Private debtholders also perform monitoring roles in mitigating agency and

information costs. Not only do agency problems exist between management and

shareholders, they also occur between management and debtholders (Jensen and

Meckling (1976)). It is well accepted that lenders have both the incentive and the

6
 Rock (1986) also provides a detailed discussion of the full-subscription price. Full-subscription price is
defined as the price at which the issuer can rely on selling all the shares in the bad state, as well as in the
good state.
capability to act as effective firm monitors. Therefore, agency risk could also be reduced

by debtholders’ monitoring efforts, leading to an increase in firm value. Assuming

institutions also possess superior monitoring and asset selection skills, one may provide

an alternative explanation for the ‘high spread puzzle’. That is, firms would like to offer a

lower price (higher yield) to attract more institutional investors to certify the quality of

their loans. However, our test results, coupled with evidence from the secondary market

lead to a tentative conclusion that inviting ‘smart money’ and monitoring services from

institutional investors are not the primary reasons for the observed differential pricing.

    The rest of the paperis organized as follows. Section 2 positions this paper in the

literature through reviewing other related work and then develops testable hypotheses.

Section 3 explains the methodology while Section 4 describes the sample selection

process and presents the main results. Section 5 offers robustness checks. Section 6

discusses the competing governance argument for the ‘high spread puzzle’. Section 7

concludes.




2 Literature Review and Testing Hypothesis Development

2.1 Why Are Institutional Loan Tranches Priced at a Higher Spread than Bank

Tranches?


    The issue of asymmetric information between lead lenders and participants in a loan

syndicate has been among the most hotly debated topics in the syndicated loan literature.

Simons (1993), Dennis and Mullineaux (2000), Panyagometh and Roberts (2005) and

Sufi (2006) among others, highlight the problem of asymmetric information in the
primary loan distribution process and discuss how syndicated loans are structured to

mitigate the problem. In a loan syndication, one or two lead banks, hired by a firm as

major agents (known as arrangers), originate, underwrite, administer and provide

monitoring services while funding the loan (themselves) during the life of the contract (as

long as they do not sell out their holdings). They are perceived as being informed

investors, possessing private information about the borrowing firm while other

participants in the loan syndicate are assumed to be relatively less informed investors.

    During the primary loan distribution process, the arrangers need to gather

information from the market and then price a deal at a spread and fee that clears the

market. As discussed above, this process is, to some extent, similar to the book building

process undertaken by equity underwriters in the IPO market, although loan distribution

is a confidential and private offering made only to qualified commercial banks and other

accredited investors.

    The evidence we will present in the following sections indicates that institutional

loan tranches are, in almost all cases, priced at a higher yield than bank loan tranches

ceteris paribus. This striking, “high spread puzzle” is the focus of this paper. The

institutional loan market has not been subjected to extensive investigation in the

academic literature and therefore there exists no directly applicable theoretical model to

explain the observed phenomenon. However, an intuitive and novel explanation

borrowed from the IPO literature provides a way for us to understand the puzzle.

    A number of studies document that new equity issues are generally underpriced

(Reilly and Hatfield (1969), Logue (1973) and Ibbotson (1975), Ritter (1984), Hanley

(1993) and Loughran and Ritter (2002), among others). Rock’s (1986) adverse selection
model is a well-accepted theoretical model for the IPO underpricing problem. This model

relies on the existence of information asymmetry between informed investors and

uninformed investors in the equity market. According to Rock (1986), informed

investors, taking advantage of superior information, crowd out uninformed investors in

good issues and leave uninformed investors to purchase bad issues. When pro rata

rationing is adopted in the primary stock distribution process, the probability of receiving

bad issues surpasses the probability of receiving good issues for the uninformed investor.

Realizing this ‘winner’s curse’, uninformed investors revise their valuations of new

issues downward, resulting in higher required returns on all IPO investments. As a result,

firms are forced to underprice their shares to prevent the uninformed investors from

withdrawing from the IPO market.

     Following this line of studies, Chemmanur (1993) puts forth another information-

based model to explain IPO underpricing. He suggests that quality firms will underprice

to maximize outsider information production and achieve a higher aftermarket price with

a view to the future proceeds from seasoned equity offerings. Another relevant model

proposed by Welch (1992) is called the “informational cascade model”, in which he

shows that a successful IPO relies on the initial sales effort. Given that the full-

subscription price is never known ex ante, underpricing serves as a defense against the

possibility of issuance failure.7 Moreover, the dynamic information acquisition model

proposed by Benveniste and Spinte (1998) reveals that IPO underpricing compensates

investors for truthfully revealing their aggregate demand to underwriters during the book-

building phase. They argue that issuers do not mind “leaving the money on the table” in


7
  Rock (1986) also provides a detailed discussion of the full-subscription price defined as the price at which
the issuer can rely on selling all the shares in the bad state, as well as in the good state.
view of the huge net wealth gain available to the senior executives who hold original

shares. In particular, they point out that underpricing also assists underwriters who

receive side-payments produced from investors’ rent-seeking behaviors.

    If we assume that the lead agents in a syndicated loan have private information about

the quality of the borrower and that institutional investors are relatively less informed and

cannot estimate the probability of default or value a loan as accurately as lead banks do

(in most of the cases, lead banks are commercial banks), then the theories of IPO

underpricing offer a way to explain why the participation of institutional investors widens

the spread on a loan holding all else constant.

    We conjecture that lead banks have to compensate institutional investors by offering

a higher spread to attract enough participation to clear the market (Rock (1986)). A price

sweetener would also motivate and invite costly information production by less informed

investors and ensure better secondary market performance for the loan if it is resold by

original lenders (Chemmanur (1993)). Furthermore, a wide-margin offered to

institutional investors can effectively prevent syndication failure (Welch (1992)) and

serves as compensation to institutional investors for revealing their aggregate demand

(Benveniste and Spinte (1998)).

    Finally, Miller (2004) points out that as commercial banks move toward a portfolio

management approach, they recognize that loans are not the most attractive investments

in terms of risk and return on a stand-alone basis. Instead of focusing solely on spread

income, commercial banks seek other sources of revenue from relationship banking: for

example other relationship-generated profits from bonds, equities and M&A consulting

and money-management services. In contrast to commercial banks, institutional investors
focus largely on spread income and price loans based on the risk and return principle. In

this regard, Louhgran and Ritter’s (2002) explanation for IPO underpricing is applicable.

They argue that issuers do not mind “leaving the money on the table” in view of the huge

net wealth gain available to the senior executives who hold original shares. Of relevance

to loan pricing, they also point out that underpricing allows underwriters to receive side-

payments. The widened spread associated with institutional loan tranches and a dual-

pricing scheme could benefit lead syndicated agent (s) by creating other fee-driven

business opportunities for lead banks, since the participation of institutional investors

allows them exposure to a wider credit spectrum.

2.2 Hypothesis Development: Spread Compensation Hypothesis

A. Impact on Primary Loan Pricing and Distribution of Credit Risk of the Loan Market

    With the aim of inviting information production and exploiting more business

opportunities, a lead bank(s) proposes to structure one or more institutional tranches with

a bonus interest rate, packaged up with other bank tranches within a loan deal. Firms with

less bargaining power (i.e., those firms which otherwise would not have been able to

secure a bank loan at all) are forced to accept the proposal and compensate the relatively

less informed investors by providing them a higher yield, thereby ensuring the receipt of

the needed funds. If this conjecture is tenable, then we would expect borrowing costs to

increase as a result of inviting institutional investor participation in a loan syndicate.

Specifically, at the facility level, institutional loan tranches on average will have wider-

margins than bank tranches ceteris paribus. At the deal level, the average loan yield

spread of a loan deal will be higher when institutional tranches are present.
Prediction 1: Primary Loan Pricing

     Our spread compensation hypothesis suggests a positive relationship between the

institutional loan indicator and initial loan yield spread, both at the facility level and at

the deal level.

     Furthermore, as we have argued, this practice allows banks exposure to a wider

credit spectrum thanks to the introduction of the dual-pricing scheme and the institutional

risk-sharing mechanism during the primary loan syndication process. As a result, we

conjecture that the corporate loan market is able to accommodate more leveraged loans as

previously un-bankable firms obtain opportunities to tap banks and other institutional

fund providers for syndicated loans. Concretely, an increase in institutional lending

should be matched by an increase in the overall leveraged issuance in the loan market.

This conjecture is consistent with the statistics provided by LPC. As shown in Figure 2,

leveraged loan lending has climbed dramatically from $28.18 billion in 1993

(representing 9% of the entire syndicated lending market), to $328.58 billion in 2003

(representing 35% of the whole syndicated lending market). Ranson (2003) highlights the

issue of redistribution of credit risk in the corporate loan market and points out that the

rapid expansion of the leveraged loan market in the last decade was paralleled by a

dramatic growth in institutional loan issuance. According to Ranson (2003), the

importance of institutional investors to the leveraged-loan market even extends beyond

the institutional loan tranches8, with institutional investment accounting for 48% of the

leveraged loan market by 2001.

B. Impact on Secondary Loan Trading



8
  Institutional investors not only subscribe the institutional tranches, but also take pieces of the bank loan
tranches (often Term Loan A) (See Miller (2004)).
     Altman, Gande and Saunders (2003) compare the informational efficiency of the

secondary loan market and the secondary corporate bond market and conclude that the

secondary loan market is more efficient in terms of price reaction to earnings and default

announcements. Allen and Gottesman (2004) study the informational efficiency of equity

markets as compared to the syndicated loan markets and reveal a tendency for the two

markets to be co-integrated. If the secondary loan market is efficient and the secondary

mark-to-market price accurately reflects all loan-related risks and represents the fair

value of the loan, then our spread compensation hypothesis also implies that the average

initial loan resale return of institutional loans should be higher than that of bank tranches

since the market forces of demand and supply should eventually eliminate the pre-market

spread premium of institutional loans in the aftermarket.9                    10
                                                                                   In addition, Miller (2004)

indicates that institutional investors tend to rebalance their portfolios more actively than

commercial banks since they opportunistically participate in loans, either seeking to

maximize the arbitrage between assets and liabilities (in the case of CLOs) or seek to

maximize the total return of their investment while having to meet their clients’ cash

withdrawal demands (as is the case for different kinds of mutual funds). Taylor and Yang

(2004) point out that it is difficult to determine whether institutional investors cause the

change in liquidity, or whether improved liquidity and higher yield attract the wide-

participation of institutional investors. In 2000, the SEC requested bank loan mutual fund

managers to use mark-to-market data instead of an estimated price to value their


9
  Initial loan resale return refers to the loan resale return on the first trading day. Initial loan resale return =
The First-Day Trading Price-100. We use the mean of average bid and ask quotation of a loan indicated in
the SMPS database to proxy for the true transaction price.
10
   Since prepayment options severely limit upside potential, loans usually do not appreciate much above
par. It should not be a surprise if the average initial loan resale return is a negative number. However, we
conjecture that institutional loans should outperform the average bank loans due to the associated
premarket spread premium.
syndicated loan portfolios.11 This policy change has effectively improved the dynamic

interaction between the primary and secondary loan market.

     Given different characteristics and investment policies of institutional investors, we

conjecture that they tend to rebalance their portfolios more often and thus create greater

price volatility. Also institutional loans tend to be resold faster than bank tranches (i.e.,

the lock-up period is shorter).12 Furthermore, we also conjecture that institutional

investors produce a higher level of liquidity in terms of an average narrower bid-ask

spread and a larger number of indicated quotations. It is worth noting that institutional

loans represented a significant portion of the secondary loan market (more than 30%)

although they constitute less than 12% of the whole syndicated loan issuance in 2003

according to the LPC.

     Prediction 2: Secondary Market Performance of Institutional Loans

     Our spread compensation hypothesis also predicts that institutional loans are

associated with a higher initial loan resale return than bank loans. Moreover,

institutional loans have shorter lock-up periods, associated with higher price volatility,

and present a higher level of liquidity.



3 Methodology

3.1 Testing Prediction 1

     Both univariate and multivariate tests are used to test our spread compensation

hypothesis. To test prediction 1, we examine whether the yield spread on a loan that is


11
   Mark-to-market data refers to the indicated bid and ask quotations submitted by loan dealers (see Miller
2004).
12
   We define lock-up as the number of days between the loan origination and the day that it is resold, even
though there are no legal requirements preventing it from being resold earlier (unlike the IPO lock-up).
structured to be sold to institutional investors is higher than that on an otherwise

equivalent bank loan. The Dealscan database provides an indicator variable identifying

whether a loan is originally designed to be sold to institutional investors, and we use it to

document whether this particular type of loan tranche has distinguishing features. Our

tests are conducted at both the facility and at the deal levels. In this way, we can identify

whether individual institutional loans are systematically priced higher and whether the

average spread of a loan deal is systematically higher due to the inclusion of one or

multiple institutional tranches.

     The formal regression analysis allows us to simultaneously control for different

factors that may contaminate the impact of loan structure design (institutional loan

tranches vs. bank loan tranches) on primary loan pricing. Drawing on well-established

loan pricing models developed in the prior literature, we are able to control many such

factors. In our regression analysis we add an institutional dummy (institutional

participation dummy) at the facility level (deal level) in the loan yield pricing model to

test whether institutional loans and the participation of institutional investors are

associated with higher loan yield spreads. The regression models take the following

forms:

     (1) At the facility level:

     Initial   loan   yield   spread=F   (Institutional   Dummy,   Borrower   Characteristics,   Loan

     Characteristics);13

     (2) At the deal level:

     The average initial loan yield spread of a loan deal= F (Institutional Participation Dummy,

     Borrower Characteristics, Loan Characteristics).14

13
   Following the prior literature, we use variable ‘All in Drawn Spread’ obtained from the Dealscan
database as the regressand in this model.
     The determinants of initial loan pricing and other major contract loan terms have

been widely studied in the literature which focuses on how asymmetric information,

agency costs of debt, signaling, reputation effects, taxes, credit quality and lender-

borrower relationships affect debt contract terms (Stulz and Johnson (1985), Diamond

(1993), Peterson and Rajan (1994), Berger and Udell (1995), Dennis, Nandy and Sharpe

(2000), Gottesman and Roberts (2004), among others). A thorough review of the theories

of debt contract design can be found in Dennis, Nandy and Sharpe (2000).

     The central variables in this paperare the institutional loan dummy and institutional

participation dummy at the facility and deal levels, respectively. We expect that

institutional loan tranches are priced higher than bank loan tranches holding all else

constant. In our tests, we control for an array of determinants of initial loan pricing that

have been proposed in the prior literature. We group these determinants under three

categories with the expected sign on each variable given in parentheses:

     (1) Ex ante Borrower Characteristics (All accounting variables are obtained at the

fiscal year end prior to the year of loan origination.)

     M/B (+): Market to book ratio. A higher M/B ratio implies an elevated level of the

borrower’s asymmetric information. We expect a positive sign on M/B ratio.

     Leverage (+): Leverage ratio. Based on Merton (1974) default risk model, the

interest rate charged by lenders should be increasing in the borrower’s leverage. We

expect a positive sign on the leverage ratio.

     Credit rating (-): Altman Z score or investment grade, non-investment grade and not

rated dummy. Altman’s Z score is a composite indicator of the likelihood of distress,


14
  The average initial loan yield spread is an average of loan yield spread of a deal weighted by facility
amount.
based on various balance sheet and market-based variables. The investment grade dummy

takes the value of 1 if the S&P long-term debt rating is BBB or above and 0 otherwise.

The not-rated dummy takes the value of 1 if the S&P long-term debt credit rating is

missing and 0 otherwise. The omitted category is non-investment grade loans.

    (2) Loan Features

    Maturity (uncertain): Maturity is measured as the loan facility duration in years.

Gottesman and Roberts (2004) make a distinction between the tradeoff hypothesis and

the credit quality hypothesis. The former holds that lenders are compensated for lending

with longer maturity, implying a positive relationship between maturity and loan yield.

The latter hypothesis suggests lenders limit their risk exposure by forcing riskier

borrowers to take shorter terms while charging higher rates, resulting in a negative

relationship between loan spread and maturity.

    Secured Status (+): This variable takes the value 1 if the loan is secured and 0

otherwise.

    Loan concentration (uncertain): Loan Concentration is defined as the natural

logarithm of deal size in U.S. dollars over deal size plus total debt. This variable is a

proxy for the lender-borrower relationship. Diamond’s (1991) model implies that a

strengthened lender-borrower relationship attenuates the adverse selection problem and

eliminates the need for monitoring. Therefore, a negative relationship between this

variable and loan yield is implied. However, Rajan (1992) proposes that relationship

banking could induce an information monopoly and hold-up problem. In this case, a

positive relationship between the lender-borrower relationship and loan yield spread is

expected.
     Loan Purpose: Loan purpose is divided into five broad categories: LBO loans,

takeover loans, repayment/recapitalization loans, general corporate purpose loans and

other loans. We utilize dummy variables for the different categories of loans. The omitted

category is general corporate purpose loan.

     Loan/facility size (uncertain): Size is defined as the natural logarithm of total

deal/facility size in U.S dollars.

     Revolver: This dummy variable takes value of 1 if the loan is a credit revolver and 0

if it is a term loan.

     Financial Covenants: Industry practitioners believe that higher yields are granted to

institutional loans because they have longer maturity and loose financial covenants.15

Therefore we also control for the impact of differing financial covenants on loan pricing.

This variable reflects 15 different types of financial covenants including the maximum

levels of capital expenditure, debt to EBITDA ratio, debt to equity ratio, senior leverage

coverage ratio, senior debt to EBITDA, loan value, leverage ratio, and debt to net worth.

Likewise, controls for the minimum levels for the interest coverage ratio, quick ratio,

fixed charge coverage ratio, EBITDA, debt service coverage ratio, current ratio, cash

interest coverage ratio are included in the regressions. We normalize maximum capital

expenditure, maximum loan value and minimum EBITDA by total assets. If a loan deal

does not have certain types of financial covenants, we assign the value of 0 for the

variables corresponding to the types of financial covenants. The omitted category is

minimum quick ratio.

     (3) Other Control Variables.


15
  I thank Timothy Perry of Credit Suisse First Boston (Houston) for the helpful discussion and comments
on the relevant issues.
         Loan resale constraints: There are 3 types of constraints: (1) borrower consent, (2)

agent consent, and (3) minimum assignment size.16 Mullineax and Pyles (2004) provide a

detailed discussion on the three types of assignment constraints. They note that since

these constraints reduce the liquidity associated with the loan, loan buyers require

compensation for liquidity risk.

         LIBOR (uncertain): We use LIBOR as a proxy for the risk-free rate. Merton’s

(1974) model indicates that the risk-free rate should be positively associated with loan

yield spread. However, Angbazo, Mei and Saunders (1998) argue that since loan interest

rates are sticky and typically lag behind interest rate adjustments, a negative relationship

could be found between the two.

         Interest Volatility (+): Interest Volatility is used as a proxy for interest rate risk. It

is calculated as a 12-month moving average of the standard deviation of the monthly

yields on 10-year U.S. T-bonds at deal month end. Merton’s (1974) model implies that

this variable should have a positive sign.

         Term premium (uncertain): This variable is a proxy for the slope of the yield

curve. It is calculated as the 12-month average for the deal year of the yield differential

between 10-year and 1-year U.S. Treasury bonds.

3.2 Testing Prediction 2

     Prediction 2 is also tested using both univariate tests and multivariate regressions.

The univariate tests examine the average initial loan resale price, the average lock-up

period, the average aftermarket price volatility, and the liquidity of bank loans compared

16
  According to LPC’s definition, the agent consent constraint refers to the case in which lead banks’
agreement is required in order for an institution to trade all or portion of a loan to another entity. Borrower
consent requires borrower’s agreement in order for an institution to trade all or portion of a loan to another
entity. Minimum assignment size stipulates the minimum amount of a facility that can be traded under an
assignment.
to institutional loans. Results reveal that, on average, institutional loans are resold at a

higher price than bank loans on the first trading day and have a shorter mean lock-up

period. In addition, the aftermarket prices of institutional loans are more volatile than

those of bank loans, while the resold institutional loans appear to be more liquid than the

resold bank loans.17 In addition, we construct an initial loan resale return model and

further investigate whether the mean initial loan resale return on institutional loans is

higher than that on bank loans controlling for other potential determinants. The model is

in the following form:

     Raw (Market Adjusted) Initial Loan Resale Return =F (Institutional Dummy, Ex ante loan

     risks, Lock-up Period, Time to Maturity, Interest Rate Risks, Other Loan Features, Market

     Awareness, Firm Size)18

     (1) Institutional Dummy (+)

     The variable of interest in this regression is the institutional dummy. Under our

spread compensation hypothesis, institutional loans should be resold at a relatively higher

price given that they were granted a bonus spread in the primary loan issuance process

and the demand and supply forces will eliminate the extra spread compensation in the

aftermarket.

     (2) Ex ante loan risks (uncertain):

     We use all-in-drawn spread as a proxy for the ex ante loan risks. The determinants of

loan yield spread have been extensively studied in the prior literature (Saunders (1996),



17
   We use number of indicated quotations and bid ask spread to measure the level of liquidity of a loan
facility.
18
   We use both the raw initial loan resale return and the market index adjusted initial loan resale return as
dependent variable in this model. We thank Ruth Young of the Loan Syndication Trading Association for
providing us with the Standard & Poor’s/LSTA Leveraged Loan Index Returns information. This Index
comprises all loans that fit the inclusion criteria and that have marks from the LSTA/LPC mark-to-market
service.
Dennis et al. (2000), Coleman et al. (2004), among others). The unanimous view on the

initial loan spread is that it reflects and measures all ex ante risk factors.

     (3) Lock-up period (-):

    This variable is the time (measured in years) between the loan’s origination day and

the first loan trading day. We introduce the lock-up period variable in our initial loan

resale return model to control for the migration risk proposed by Altman (1998).

Migration risk refers to the risk of loss due to a change in an obligation’s credit rating.

Altman (1998) points out that firms’ credit ratings tend to be downgraded within three

years of new debt issuance. Accordingly, we conjecture that the longer an original lender

holds the loan, the greater the migration risk and therefore the lower the loan resale price.

    (4) Time to maturity (+):

    This variable is calculated as the time (in years) between loan maturity and the first

loan trading day.

    (5) Interest rate risks (uncertain):

    Loans are floating rate assets and subject to interest rate risk. We use three variables

to control for interest rate risk, all collected in the month during which the loan was first

resold on the secondary market. Three-month LIBOR is used as a proxy for the risk-free

interest rate. Interest volatility is calculated as a 12-month moving average of the

standard deviation of the monthly yields on 10-year U.S. T-bonds. We measure the slope

of the yield curve with the term premium, calculated as the 12-month average for the deal

year of the yield differential between 10 year and 1 year U.S. bonds.

    (5) Other loan features (uncertain):
      Other loan features include secured status, which takes the value of 1 if the loan is

secured or 0 otherwise, credit revolver dummy, number of lenders (log of the number of

lenders), facility size (log of facility amount), facility ratio (facility amount/ deal amount)

and      loan      purpose        dummies         (including        leveraged        buyouts,        takeovers,

repayment/recapitalization and other purposes; the omitted case is general corporate

purpose loan). If the loan market is efficient and the loan price has incorporated all the

information contained in these ex ante loan features, then none of these variables should

have a significant coefficient estimate.

      (6) Market awareness (uncertain):

      We use the log of number of quotations on the first trading day as an indicator of the

market awareness of the loan:the higher the number of quotations on the first trading day,

the higher the level of cognizance the loan received from the market.

(7) Firm size (uncertain):

      We use the log of sales at close to control for the size of the firm.19 Again if the loan

market is efficient, we should not expect to see a significant sign associated with firm

size.




4 Sample Selection and Test Results

4.1 Data Used in Testing Prediction 1

      Our sample of loans is obtained from the Dealscan database between 1995 and 2004.

Our choice of 1995 follows from the recognition that institutional loans came into broad


19
   The sales at close refer to borrower’s sales reported in its financial statements at the time of the closure of
a loan deal.
usage only from the mid-1990s. There are initially 58,821 U.S. loan facilities in the data

set. We further screen the sample by the following criteria: (1) We only keep those loan

facilities with LIBOR as the base rate (16,172 facilities are removed); (2) we delete those

facilities identified as banker acceptances, bridge loans, leases, FRN, SLC, CD, Bonds,

Notes, Guidance lines, traded letters of credit, multi-option facilities, other loans or

undisclosed (2,147 facilities are removed). The screening process leaves us with 40,502

observations. Next we manually match our sample with the Compustat database by

company name, ticker and the deal active date for each loan facility. Financial ratios are

retrieved on the last fiscal year end prior to the year of deal origination. This means we

only retain the loan facilities of public companies. Our final sample includes 10,471 loan

deals involving 2,932 firms. There is institutional participation in 915 (8.74%) of these

deals, representing 24.6% of the whole Dealscan sample during the period. At the facility

level, there are 14,448 loan facilities, of which 945 tranches (6.54%) are institutional.

4.2 Data Used in Testing Prediction 2

    To test Prediction 2, we combine Dealscan with the SMPS database to obtain first-

day loan trading information and retrieve the primary loan issuance information for each

loan facility. The SMPS database consists of daily loan bid-ask quotations and the

number of quotes for each loan facility traded on the secondary market. Several papers

have used this database to study the efficiency of the loan resale market relative to other

capital markets (Altman, Gande and Saunders (2004), Allen and Gottesman (2005)).

Following convention, we use the average of the mean bid and mean ask quotes as a
proxy for the real loan transaction price.20 SMPS contains the daily quote information of

5,101 loan facilities from 1998 to 2004.

     We extract the trading information on the first trading day for these 5,101 loan

facilities and match with the 40,502 Dealscan samples used for testing Prediction 1.

Finally, we match this sample with the loan-market-index data by loan origination day

and first loan trading day respectively. This matching process leaves us 3,170 loan

facilities based on which we set up and test our initial loan resale return model.

4.3 Test Results

A. Summary Statistics and Univariate Tests for Prediction 1

        The sample statistics of all the dependent and independent variables in the loan

pricing model at facility level are reported in Table 1.

TABLE 1 GOES HERE

     Our sample statistics are comparable with those reported in prior studies, with the

all-in-spread drawn over LIBOR averaging 1.843 percent and the average maturity being

47 months. In comparison, Sufi (2006), using Dealscan, reports a mean all-in-spread

drawn of 1.59%, with the mean loan maturity being 37 months. The mean M/B ratio is

1.772, with a mean loan concentration ratio of 0.411 and a mean facility ratio of 0.663.

Using the SDC syndicated loan database, Coleman, Esho and Sharpe (2004) report an

average maturity of 49 months, an average loan yield spread over LIBOR of 1.268%, an

average M/B ratio of 1.793, an average loan concentration ratio of 0.415 and mean

facility ratio of 0.685.



20
  Transaction prices are not available in the syndicated bank loan market. However, LPC collects the bid
and ask prices from major loan dealers on a daily basis. Internal LPC studies suggest that transactions
prices for par loans (priced above 90) are close to the mean of the mean bid and mean ask quotes.
    We divide our sample into 2 groups: institutional loan tranches vs. bank loan

tranches at facility level and deals that involve institutional participation vs. those that do

not. The sample statistics are reported for 2 distinct groups at facility (deal) level in Table

2 (Table 3). The mean difference t-test statistics for the two groups are shown in the last

column.

TABLE 2 AND TABLE 3 GO HERE

    As shown in Tables 2 and 3, institutional loans are, on average, associated with

higher yield, longer maturity and are extended to smaller firms with lower credit ratings

and lower Altman Z scores. They fund more leveraged buy-outs and takeovers than do

bank loans. These results are confirmed at both facility and deal levels. At the facility

level, the mean spread charged to bank tranches is 1.72 % over LIBOR while the mean

spread charged to institutional tranches is 3.07%. At the deal level, the weighted average

spread of loan deals that only contain bank tranches is 1.50% over LIBOR, but 2.89% for

deals involving institutional tranches is 2.89%.

    In addition, the mean difference test results shown in the last column of Table 2 and

Table 3 indicate that in most cases the mean of institutional loan tranches is statistically

different from that of bank loan tranches. Loan deals that contain tranches designed for

institutional investors are also distinct from deals that do not contain such tranches.

Although the analysis here is unconditional, there is strong evidence, confirmed with

conditional analysis below, that institutional loans and loan deals that contain institutional

tranches are charged higher interest rates ceteris paribus.

B. Multivariate Tests for Prediction 1

TABLE 4 GOES HERE
    The facility level loan pricing regression results based on the entire sample from

1995-2004 are reported in Table 4 and the institutional loan dummy is positive and

significant across several alternative loan pricing models. The magnitude of the estimated

coefficients on the institutional loan dummy implies that on average institutional loans

are compensated with an additional 46 basis points (in Model 5) spread over otherwise

equivalent bank loans. This result directly supports the spread compensation hypothesis.

    To confirm the effect of the institutional loan dummy, we exclude this variable in

Model 1 and reintroduce it in Model 2. Compared to Model 1, the adjusted R square of

Model 2 improves by 0.93%. A comparison of Model 1 and Model 2 reveals that adding

the institutional loan dummy does not significantly impact the magnitude and

significance of the other coefficient estimates on borrower’s ex ante risk characteristics

and loan features, suggesting that the institutional loan variable is a complement, not a

substitute, for the other explanatory variables that have been identified in the prior

literature.

        The significance, magnitudes and signs of coefficient estimates are generally

consistent across different models and the adjusted R square of around 58% is

comparable with other empirical studies (e.g. Coleman et al. (2004) and Hao and Roberts

(2004)).

TABLE 5 GOES HERE

    The deal-level loan pricing regression results based on the entire sample from 1995-

2004 are reported in Table 5. In these models, we control for 15 types of financial

covenants and investigate whether the impact of the institutional participation dummy on

the loan pricing fades away. As shown in Table 5, the average cost of borrowing for deals
that contain institutional tranches is significantly higher than that for deals that do not

contain such tranches holding ex ante firm risks, loan risks and tightness of financial

covenants constant. This result holds across different models. Our tests demonstrate that

firms whose loans incorporate institutional tranches have to pay an additional 55 basis

points to complete a deal holding all else constant.

    In brief, both univariate and multivariate tests show that the institutional loan dummy

(institutional participation dummy) is positively and significantly related to loan yield

spread, consistent with the spread compensation hypothesis. The results lead us to

conclude that firms have to pay a higher interest rate when inviting institutional lenders to

participate in a loan syndicate.

C. Summary Statistics and Univariate Tests for Prediction 2

    In this section, we report additional evidence from the secondary loan resale market

that supports Prediction 2 under the spread compensation hypothesis. We show how

institutional loans are distinguishable in terms of secondary loan resale return, liquidity

and price volatility. Our univariate test results demonstrate that institutional loans are

resold at higher prices, are more liquid, have shorter lock-up periods and have higher

price volatility.

TABLE 6 GOES HERE

    Table 6 reports the standard deviation of loan prices and average daily quotes of all

loan facilities that are traded on the secondary market. The statistics are shown for

institutional tranches and bank tranches respectively. There are 4,186 loan facilities out of

which 2,858 are bank loan tranches and 1,328 are institutional tranches. As shown in the

Table 6, the prices of institutional loans appear to be more volatile than those of bank

loans. The mean standard deviation of prices is 4.321 for institutional loans and 2.578 for
bank loans. In addition, the mean daily number of quotations of institutional loans

(2.868) is also higher than bank loan tranches (1.851). The mean difference tests shown

in the last column confirm that the means are significantly different between institutional

loans and bank loans. Our evidence indicates that institutional loans are more liquid and

volatile on the secondary loan trading market.

    Turning to the loan resale return model, Table 7 reports the summary statistics for the

dependent and independent variables.

TABLE 7 GOES HERE

    The dependent variable is market-adjusted initial loan resale return. The mean of this

variable is -2.12% for bank loan tranches, but the corresponding number for institutional

loan tranches is 0.16%. The average raw initial loan resale return corresponding to bank

loan tranches is -3.96%, whereas the corresponding number for institutional loans is -

0.70%.

    The unconditional test results point to institutional loans outperforming both bank

loans and the broader loan market index.         This result is further confirmed in our

conditional multivariate tests. Additionally, the average number of quotations received on

the first trading day is 1.37 for institutional loans and 1.23 for bank loans and the mean

lock-up period for bank loans is 314.18 days but only 133.55 days for institutional loans.

Clearly, the original lenders resell institutional loan tranches much faster. Moreover,

compared to bank loans traded on the secondary market, institutional loans are associated

with higher yields, longer loan lives and are extended to relatively smaller firms. Also, a

higher proportion of institutional loans is used to fund LBO endeavors. Finally, all the

above results have correspondingly significant t-statistics in the mean difference tests.
    In a nutshell, the univariate tests results are consistent with Prediction 2 of the spread

compensation hypothesis. It seems that the additional spread awarded to institutional

investors is diminished by the demand and supply forces through secondary trading. We

verify this result using multivariate regressions in the next section.

D. Multivariate Tests for Prediction 2

TABLE 8 GOES HERE

    In this section, we set up an initial loan resale return model and show how

institutional loans perform on the first day of trading in the secondary market. Table 8

shows both the raw initial loan resale regression results and the market-adjusted initial

loan resale regression results based on our 1998-2004 sample period. In general, the

institutional loan dummy variable is significant at 5% across different models with a

coefficient estimate of 0.51-0.66. This indicates that the mean initial loan resale return of

institutional loans surpasses that of bank loans by 51 to 66 basis points, confirming the

univariate test results and providing support for Prediction 2 of our spread compensation

hypothesis.

    Consistent with our hypothesis, the lock-up period appears to be a highly negatively

significant explanatory variable directly supporting Altman’s migration risk theory. The

sooner a loan is resold the lower the migration risk associated with it. However, the ex

ante risks (proxied by all-in-drawn spread) are statistically significant across all the

models. If the loan price incorporated all the information, then we would not have seen a

significant relation between ex ante loan risks and loan resale return. Our finding thus

leads to a tentative inference that the loan market is basically a lemons market where

assets are subject to a lemons discount and it could be the case that riskier assets are

subject to a greater lemons discount.
      Finally, the time to maturity of a loan at the point of the loan resale is a significantly

positive explanatory variable. This is not surprising as a longer-lived loan can be resold at

a higher price. Moreover, the regression results indicate that credit revolvers are resold at

lower prices on the first trading day. Concerns about the possible endogeneity of the

lock-up period and other possible explanatory variables are addressed in the robustness

tests in the next section.

      In brief, the multivariate exercises provide direct support for Prediction 2 of the

spread compensation hypothesis.




5 Robustness Tests

5.1 Re-estimating the Loan Pricing Model in a Simultaneous Framework

      Dennis, Nandy and Sharpe (2000) model loan pricing and other loan contract terms

within a simultaneous decision framework and point out that ignoring the simultaneity

problems could produce biased and inconsistent estimates. To address the simultaneity

issue, we re-estimate our loan pricing model allowing maturity and loan pricing to be

simultaneously determined.21 Following the approach in our OLS tests, we construct the

simultaneous equations at both facility and deal levels. Table 3.9 reports the estimation

results.

TABLE 3.9 GOES HERE

     Compared with the loan pricing model estimated with OLS regressions at the facility

level (Table 4, Model 3), the coefficient estimate on maturity changes from -0.07 to -0.16


21
  The simultaneous model we use here is similar to Dennis, Nandy and Sharpe (2000) but addresses only
the simultaneity of loan pricing and maturity.
and significance is also enhanced under the simultaneous framework. The positive and

significant association between the institutional loan dummy and loan yield spread is

robust under the simultaneous framework. At the deal level, compared with OLS

estimators (Table 5, Model 3), the coefficient estimate on maturity changes from -0.13 to

0.17 and significance is enhanced under the simultaneous framework. Although the

coefficient estimate on loan participation falls from 0.56 to 0.35, the sign and significance

of the loan participation dummy does not change in the simultaneous equations.

Moreover, the coefficient estimates on the other firm characteristics and loan features in

the loan pricing model remain unchanged. We conclude that the positive relationship

between loan yield spread and institutional loan (institutional participation) is robust

when allowing loan pricing and loan maturity to be simultaneously determined and our

spread compensation hypothesis receives support under the simultaneous regression

framework.

5.2 Controlling the Endogeneity of Lock-up Period in the Loan Resale Model

    Institutional investors focus largely on loan-specific return and pay much less

attention to relationship-generated revenue than do commercial banks. We hypothesize

that institutional investors tend to rebalance their loan portfolios more often and make

loan resale decisions more quickly than banks. The lock-up period of a loan may be

directly associated with the loan type, but it might also be predetermined by ex ante loan

risks and some market factors such as interest-rate risk and the term premium. Simply

treating the lock-up period as an exogenous variable might induce a potential bias. This

raises a question of whether the excess loan resale return associated with institutional
loans continues to hold after we control for endogeneity, or if the excess loan resale

return is simply a result of the shorter lock up period and lower migration risk.

    To address these concerns, we adopt a 2-stage least squares framework in which the

lock-up period is endogenously determined. In the first step, we estimate a lock-up period

model using a set of explanatory variables including the institutional loan dummy, ex

ante loan risks and some market risk factors; in the second stage, we re-estimate the

initial loan resale model using the 2-stage-least-squares estimation technique and allow

the lock-up period and (market adjusted) initial loan resale return to be endogenously

determined.

    The lock-up period model is in the following form:

    Lock-up period= F (institutional dummy, ex ante loan risks, interest rate risk during the period).

TABLE 10 GOES HERE

    Consistent with the univariate test results (Table 7), the coefficient estimate on the

institutional loan dummy is highly significant and negative, implying that institutional

loans tend to be resold faster by original syndicated lenders than bank loans holding the

ex ante loan risks, maturity and other market conditions constant. The negative

coefficient estimate (-0.19) on all-in-drawn spread indicates that the higher the loan risks

the more quickly the loan is resold on the secondary market. On average, if the all-in-

drawn spread increases by 100 basis points, the lock-up period of the loan will be 0.19

years shorter. This result provides indirect support for the lower quality loan disposal

hypothesis proposed in Kamstra, Roberts and Shao (2006). The lower the quality of the

loan, the higher the probability of its being resold on the secondary market. In addition,

the higher the level of interest rates at loan origination, the sooner the loan is resold. A
positive yield curve predicts a faster loan resale while higher interest rate risk predicts the

opposite.

    Using a 2-stage-least-squares estimation technique, we re-estimate the (market-

adjusted) loan resale return models and lock-up period model within a simultaneous

framework. The estimation results are shown in Table 11.

TABLE 11 GOES HERE

    The first two columns report the regression results of a simultaneous system of

regressions including equations for the initial loan resale return and the lock-up period.

The last two columns report the market-adjusted initial loan resale return equation and

the lock-up period equation. Compared to the results of the OLS regression (Model 1

from Table 8), the simultaneous model yields similar coefficient estimates on the

institutional loan dummy, but the significance falls to the 10% level. The coefficient

estimates on lock-up period remain almost the same and the significance level is

enhanced under the simultaneous regression framework. All the other coefficient

estimates generally remain the same in terms of magnitude, significance and sign. We

conclude that the higher (market-adjusted) initial loan resale return associated with

institutional loans is at least partly due to the shorter lock-up period and lower migration

risks associated with them when they first come to the market. However, institutional

loans are still resold at a higher (or less discounted) price relative to bank loans holding

ex ante loan risks, time to maturity, controlling for market conditions at time of loan

trading.
6 A Competing Explanation for the Institutional Loan Spread Premium

Puzzle---Inviting ‘Smart Money’ and Efficient Monitors

    The arguments in the preceding sections rest on the assumption that institutional

investors are at an informational disadvantage relative to banks. This allows us to borrow

from the IPO literature to explain the observed difference in pricing between bank and

institutional loan tranches. However, the corporate governance literature points to a

potential alternative explanation. In the context of corporate governance, institutional

shareholders are viewed as playing a role as firm monitors. Institutional shareholders are

considered to be active and efficient firm monitors and their participation in loan deals

may be correlated with higher firm value.

    The impact of institutional shareholdings on firm value and the efficacy of

shareholder activism as a corporate governance mechanism have been widely discussed

in the corporate governance literature. Institutional investors are often referred to as

“smart money”. Their involvement in an equity investment delivers a message about the

company’s health and financial future. Allen, Bernardo and Welch (2000) argue that

institutional investors have advantages in detecting firm quality.

    Demsetz (1983), and Demsetz and Lehn (1985) argue that ownership structure is an

endogenous outcome of firms’ value maximization process and high ownership

concentration motivates the efficient monitoring of management and alleviates the

shirking problem. Shleifer and Vishny (1986), Hill and Snell (1988), Agrawal and

Mandelker (1990), Gorton and Schmidt (2000), Morck et al. (2000), and Gedajlovic and

Shapiro (2002) are among those who document a positive relationship between

ownership concentration and firm performance across different countries. Smith (1996)
analyzes whether institutional shareholder activism is effective as a source of monitoring

using a sample of firms targeted by CalPERS. He documents an improvement in firms’

operating performance and shareholder wealth following the adoption of CalPERs’

proposals. Hartzell and Starks (2003) argue that institutional shareholders play a

monitoring role in alleviating agency problems, demonstrating the positive impact of

institutional shareholdings on the pay-for-performance sensitivity of executive

compensation.

     Despite the extensive discussion of the efficacy of institutional equity ownership as a

corporate governance mechanism, very few papers have studied the role of institutional

debt holdings in restraining the agency costs arising from the conflict of interest between

shareholders and debtholders.22 It is well known that loans are special in that lenders

(especially banks) are quasi-insiders of the firm, and are able to provide monitoring at

lower costs. As a result, firm value is improved due to debtholders’ monitoring services

and mitigated agency costs. If we assume that institutions share a willingness to monitor,

and also possess some skill in identifying quality investments, then firms would have

incentive to invite such institutional debtholders. Accordingly, the higher yield offered to

institutional loan investors could be explained as an incentive to attract institutional debt

holding.

     If we are willing to further assume that most institutional shareholders are also the

firm’s institutional loan holders, then institutional participation undoubtedly helps to

mitigate the conflict of interests between shareholders and debtholders and reduces the


22
   Roberts and Yuan (2006) study the impact of institutional equity ownership on firms’ loan borrowing
cost and point out that if loan lenders charge higher interest rates on debt issued by firms with concentrated
institutional ownership to cover agency costs between shareholders and debtholders, the benefit of such
ownership could be offset.
agency cost of debt. Roberts and (2006) provide evidence that institutional shareholdings

are negatively correlated with loan yield spreads and they argue that the lower cost of

loans associated with higher ownership concentration is due to the fact that institutional

investors play an active role in reducing the level of firm risk.

    The difficulty with the foregoing lies in the evidence from the aftermarket. If

institutional debtholders are indeed institutional shareholders, or for some other reasons

they have the skills and the willingness to effectively intervene in the corporate

governance process, then we should not have observed institutional loans having higher

price volatility or shorter hold-up periods before the first resale activity. In the context of

corporate governance, institutional loan investors are expected to be faithful long-term

debtholders.

    To test whether inviting ‘smart money’ constitutes a primary incentive for firms to

offer a higher yield to institutional loans, we conduct a simplified empirical test by

introducing institutional shareholdings and an interaction variable linking institutional

shareholdings and the institutional loan dummy to the loan pricing model.

TABLE 12 GOES HERE

    From Table 12, we find that institutional ownership is a negative and significant

determinant of loan pricing. This result is consistent with the findings in Roberts and

Yuan (2006). After controlling for institutional shareholdings, the institutional loan

dummy continues to be positive, significant and of similar magnitude. . However, the

term reflecting the interaction between the institutional loan dummy and institutional

shareholdings is insignificant, indicating that the introduction of institutional loan holders

neither enhances nor weaken the impact of institutional ownership concentration on loan
pricing. Although we do not have direct evidence showing that institutional loan holders

are independent of the firm’s institutional shareholders, our test results suggest that the

institutional loan holders do not play a role in the corporate governance process. Our

tentative conclusion is that inviting ‘smart money’ and obtaining monitoring services

from institutional investors are not the primary incentives for firms to pay out higher

interest rates on institutional tranches.




7 Conclusion

        Despite the rapid development of institutional loan issuance during the last decade

and the exceptional performance of institutional loans on the secondary loan market, this

important and distinctive type of loan has not received enough attention from academic

researchers. This paper takes a first step toward understanding the institutional loan

market. Our empirical tests reveal that institutional loans are priced differently from bank

loans and they are always granted a higher initial yield. On average, institutional loans

enjoy a positive spread differential of 50 basis points compared to bank loans. We call

this the “high spread puzzle”. Furthermore, institutional loan facilities outperform bank

loan facilities on the first trading day. The initial loan resale return averages -3.96% for

bank loans and only -0.70% for institutional loans. After adjusting for the loan market

index return, the initial loan resale returns for bank loans averages -2.12% while that for

institutional loans is positive (0.16%). Additionally, we find that prices for institutional

loan facilities are more volatile and the average number of quotations for institutional

loans is higher than that for bank loans., Original lenders tend to resell the institutional

tranches sooner than bank tranches. Although institutional issuance only takes 10% of the
whole syndicated loan issuance, institutional loan tranches comprise 1/3 of the secondary

loan resale market. The empirical results suggest that institutional investors help improve

loan market liquidity.

       Assuming that institutional investors are relatively less informed relative to banks,

several information-based theories explaining the IPO underpricing problem offer a way

to understand the widened spreads granted to institutional loan investors. We note the

similarity between the loan syndication process and new equity distribution process and

briefly reviewed several major theories underlying the IPO underpricing problem

including Rock’s (1986) winner’s curse model, Welch’s (1992) information cascade

model, Chemmanur’s (1993) information production model, Benveniste and Spinte’s

(1998) dynamic information acquisition models and Loughran and Ritter’s (2002)

prospect theory. Applying the intuition underlying these theories to the syndicated loan

market, we address the ‘high spread puzzle’ hypothesizing that lead banks must

compensate institutional investors by offering a higher spread to attract sufficient

participation and clear the market. A price sweetener would also motivate costly

information production by less informed investors and ensure better secondary market

performance of a loan if it is to be resold by the original lenders. Furthermore, the wide

margin offered to institutional investors can effectively prevent syndication failure and

serves as compensation to institutional investors for revealing their aggregate demand. A

natural consequence of this “spread compensation hypothesis” is that that institutional

loan tranches are priced at higher spreads (lower price) than bank tranches, holding all

else constant. Further, the extra spread margin offered to institutional loans raises the

market’s cognizance and results in better secondary market performance.
       Finally, we discuss a competing explanation for the “high spread’ puzzle that

suggests that borrowers pay higher yields to invite institutional loan investors to certify

the quality of the loan, since institutional investors are considered effective monitors and

sources of ‘smart money’. If a firm’s institutional shareholders are also its debtholders,

then institutional loan holdings may help to mitigate the agency costs arising from

shareholders and debtholders’ conflicting interests. However, our simplified empirical

test results lead us to tentatively deny this type of conjecture.

       Overall, our evidence implies that the entry of institutional investors into the

corporate loan market has helped improve the transparency and liquidity of the secondary

loan market, at the expense of increased overall loan borrowing costs. As the corporate

loan market is gradually evolving from a relationship-driven, private credit market with

only informed lenders to a price-driven, hybrid capital market involving both informed

and less informed lenders, corporate borrowers now have to pay a higher interest rate to

raise their required funds. While this may be seen as a downside to the structural changes

occurring in the loan market, such changes make the loan market more capable of

accommodating relatively riskier borrowers (highly leveraged loans) while exposing

commercial banks to a wider business range.
                                      References

Agrawal, A., G.N., Mandelker, 1990. Large shareholders and the monitoring of
managers: The case of anti-takeover amendments. Journal of Financial and Quantitative
Analysis 25,143-161.

Allen, Franklin, Antonio E. Bernardo, and Ivo Welch, 2000, The Journal of Finance 55,
6, 2499-2536.

Allen, L., and A. Gottesman, 2005, The informational efficiency of the equity market as
compared to the syndicated bank loan market, Working Paper.

Altman, E.I., A. Gande and A. Saunders, 2003, Informational efficiency of loans versus
bonds: evidence from secondary market prices, NYU Working Paper.

Angbazo, L., J. Mei, and A. Saunders, 1998, Credit spreads in the market for highly
leveraged transaction loans, Journal of Banking and Finance 22, 1249-1282.

Benveniste, L. and Spindt, P., 1989, How Investment Bankers Determine the Offer Price
and Allocation of New Issues, Journal of Financial Economics, 42, 223-255.

Berger, Allen N., and Gregory F. Udell, 1995, Relationship lending and lines of credit in
small firm finance, Journal of Business 68, 351-381.

Chemmanur, Thomas J., 1993, The pricing of initial public offerings: A dynamic model
with information production, Journal of Finance XLVIII, 1, 285-304.

Coleman, A.D., F. N. Esho, and I.G. Sharpe, 2004, Do bank characteristics influence loan
contract terms?, Working paper, School of Banking and Finance, University of New
South Wales, Australia.

Demsetz, Harold, 1983. The structure of ownership and the theory of the firm. Journal of
Law and Economics 26, 375-390.

Demsetz, H., and K., Lehn, 1985. The structure of corporate ownership: Causes and
consequences. Journal of Political Economy 93, 1155-1177.

Dennis, Steven A., and Donald J. Mullineaux, 2000, Syndicated loans, Journal of
Financial Intermediation 9 (2000), pp.404-426.

Dennis, Steven A., D. Nandy, and I. G. Sharpe, 2000, The determinants of contract terms
in bank revolving credit agreements, Journal of financial and Quantitative Analysis 35, 1,
P87-110.

Diamond, D., 1993, Seniority and the maturity structure and liquidity risk, Quarterly
Journal of Economics, 53, 341-368.
Gedajlovic E., and Shapiro D., 2002, Ownership structure and firm profitability in Japan.
Academy of Management Journal 45: 565–575.

Gottesman, A., and Gordon S. Roberts, 2004, Maturity and corporate loan pricing,
Financial Review 39 (February), 55-77.

Gorton G, Schmidt A. 2000. Universal banking and the performance of German firms.
Journal of Financial Economics 58, 29–80.

Hanley, K. W., 1993, The underpricing of Initial Public Offerings and the Partial
Adjustment Phenomenon, Journal of Financial Economics, 34, 231-250.

Hartzell, Jay C., and Laura T. Starks, 2003, Institutional investors and executive
compensation, THE Journal of Finance LVIII, 6, 2351-2374.

Hill C., and Snell S.,1988, External control, corporate strategy, and firm performance in
research-intensive industries,Strategic Management Journal 9(6): 577–590.

Ibbotson, R. G., 1975, Price Performance of Common Stock New Issues, Journal of
Financial Economics, 2, 235-272.

Logue, D., 1973, On the Pricing of unseasoned equity issues, 1965-69, Journal of
Financial and Quantitative Analysis, 8, 91-103.

Loughran, T. and Ritter, J. R., 2002, Why Don’t Issuer Get Upset About Leaving Money
on the Table in IPOs?, The Review of Financial Studies, 15, No. 2, 413-443.

Merton, R.C., 1974, On the pricing of corporate debt: The risk structure of interest rates,
Journal of Finance 29, 449-470.

Miller, Steven C., 2004, A syndicated loan primer, Standard & Poor’s: A Guide to the
Loan Market.

Morck R, Nakamura M, Shivdasani A. 2000. Banks, ownership structure, and firm value
in Japan, Journal of Business 73: 539–567.

Mullineaux. D., and M. Pyles, 2004, Constraints on loan resales, Working Paper,
University of Kentucky.

Ranson, Brian J., 2003, The growing importance of leveraged loans, The Journal of
Lending & Credit Risk Management 2003 May.

Panyagometh, Kamphol, and Gordon S. Roberts, 2005, Private information, agency
problems and determinants of loan syndications: Evidence from 1987-1999, Working
Paper, Schulich School of Business, York University.
Peterson, M., and R. Rajan, 1994, The benefits of lending relationships: Evidence from
small business data, Journal of Finance 49, 3-37.

Reily, F. and Hatfield, K, 1969, Investor Experience with New Stock Issues, Financial
Analysts Journal, 73-80.

Ritter, J. R., 1984, The ‘Hot Issue’ Market of 1980, Journal of Business, 57, 215-240.

Roberts, G. S. and Yuan, L. Z., 2006, The role of institutional ownership in corporate
governance: A new perspective, Working Paper, Schulich School of Business, York
University.

Rock, K., 1986, Why New Issues Are Underpriced, Journal of Financial Economics, 15,
187-212.

Saunder, A., 1996, Credit spread in the market for highly leveraged transactions loans,
Working Paper, New York University.

Shleifer,A. and R. Vishny, 1986. Large shareholders and corporate control. Journal of
Political Economy 94, 461-488.

Smith, M. P., 1996. Shareholder activism by institutional investors: Evidence from
CalPERS, Journal of Finance 51, 227-252.

Simons, K., 1993, Why do banks syndicate loans? New England Economics Review
Federal Reserve Bank Boston, 45-52.

Stulz, R., and H. Johnson., 1985, An analysis of secured debt, Journal of Financial and
Quantitative Analysis, P501-521.

Sufi A, 2005, Agency and renegotiation on corporate finance: Evidence form syndicated
loans, Working Paper.

Taylor, A. A., and R. Yang, 2004, The evolution of the loan market, Loan Syndication
Trading Association, Document.

Welch, I., 1992, Sequential Sales, Learning, and Cascades, The Journal of Finance, 47,
No. 2, 695-732.

Yago, G., and D. McCarthy, 2004, The U.S. leveraged loan market: A primer, Milken
Institution Research Report.
 Figure 1: Issuance of Leveraged Loan vs. Institutional Loan (in $Bil).


350                                                                                                                                  328.6 324.3
                                                                                         320.2
                                        Leveraged                                                   310.0
300
                                                                            273.0
                                        Institutional                                                                 264.5
  Loan Issuance ($Bils.)




250
                                                                                                             218.0
                                                                   194.0
200


150                                                       134.8                                                                                  137.4
                                                                                                                                         118.4
                                               101.3
100                               81.1
                                                                                                                            63.8
                                                                                             55.3     50.0
                                                                                 44.9
50 28.2                                                                                                        31.9
                                                                     25.1
                                                   7.6      13.5
                           2.5         5.6
 0




                                                                                                                                                  LTM1Q04
                           1993


                                       1994


                                                   1995


                                                            1996


                                                                     1997


                                                                                 1998


                                                                                             1999


                                                                                                      2000


                                                                                                               2001


                                                                                                                            2002

 1200                                                                                                                                    2003
                                              O ther
 1000                                         Leveraged
                                              I-G rade
  800

  600

  400

  200

                           0
                                                                                                                                                  T 3 1 4
                                   1

                                               2

                                                     3

                                                             4

                                                                    5

                                                                             6

                                                                                         7

                                                                                               8

                                                                                                       9

                                                                                                               0

                                                                                                                        1

                                                                                                                                     2

                                                                                                                                            3

                                                                                                                                                 L M -3 -0
                                  9

                                              9

                                                    9

                                                            9

                                                                   9

                                                                            9

                                                                                        9

                                                                                              9

                                                                                                      9

                                                                                                             '0

                                                                                                                      '0

                                                                                                                                   '0

                                                                                                                                          '0




 Source: LPC Webinars syndicated loan market research presentation by Meredith Coffey (2004).
Table1 1: Sample statistics for the whole sample used in the facility level loan pricing regression (1995-2004). This table presents
summary statistics for the dependent variable all-in-drawn spread (measured in percentage) and for the independent variables in the facility
level loan pricing regression models. The loan agreements were originated during the period January 1995 - December 2004. The dependent
variable is the initial all-in-spread drawn which is defined as the percentage coupon spread over LIBOR plus the annual fee and the upfront fee
paid by borrowers for each dollar drawn down. We include M/B (Market/Book), Leverage (Total Debt/Assets), Altman Z-score
(3.3*EBIT/SALES+SALES/TA+1.4*RE/TA+1.2*WC/TA, where EBIT is the earning before interests and taxes, SALES is the total sales of
the year, TA is the total assets, WC is the working capital and RE is the retained earning), or alternatively Investment Grade (taking the value
of 1 if the S&P long term debt rating is BBB or above and 0 otherwise) and Not Rated dummy (taking the value of 1 if the S&P long term debt
credit rating is missing and 0 otherwise), which serve as proxies for two groups of borrower characteristics: borrower risk and information
costs. Firm characteristics are acquired at the fiscal year end prior to the deal year. We also include other loan contract terms such as loan
facility duration in years (maturity), and secured status taking the value of 1 for secured loan or 0 otherwise. Loan Concentration is defined as
ln(deal/(deal+TD)) where deal is the deal amount and TD is the total debt including long term debt and current liabilities. Revolver is the
dummy variable taking the value of 1 for credit revolver and 0 for term loan. We control for facility size (log of facility amount), facility ratio
(facility amount/ deal amount), Firm Size (log of firm’s total assets), Risk free interest rate (LIBOR), Interest rate risk (Interest volatility), term
premium, year dummy and industry dummy. LIBOR is the deal month end 3-month LIBOR rate. Interest Volatility is a 12-month moving
average of the standard deviation of the monthly yields on 10-year U.S. T-bonds at deal month end and term premium is the 12-month average
for the deal year of the yield differential between 10 years and 1 year U.S. bonds. Facility Ratio is defined as log of facility amount over the
deal amount. Assignment Size specifies the minimum assignment size for loan resale in a dollar amount stipulated in the loan contract. Agent
consent takes the value of 1 if the consent from lead banks is required for selling the loan or 0 otherwise while the borrower consent takes the
value of 1 if the consent from borrower is required for selling the loan or 0 otherwise. We identify 10 categories of industries.


  Variable*                              Mean                   Median                Standard              Minimum               Maximum
                                                                                      Deviation
 All-in-Drawn Spread                      1.843                 1.750                 1.168                 0.150                 10.500

 MB                                      1.772                  1.444                 1.177                 0.510                 22.161

 Altman Z score                           1.812                 1.823                 1.427                 -29.690               8.484

 Leverage                                0.369                  0.352                 0.199                 0.004                 0.964

 Investment Grade                        0.188                  0.000                 0.391                 0.000                 1.000

 Not Rated                               0.498                  0.000                 0.500                 0.000                 1.000

 Maturity (in years)                      47.081                48.000                22.927                1.000                 163.000

 Secured Status                          0.664                  1.000                 0.472                 0.000                 1.000

 Loan Concentration                       0.411                 0.400                 0.213                 0.001                 0.996

 Revolver                                0.729                  1.000                 0.444                 0.000                 1.000

 Facility Size ($M)                      18.715                 18.826                1.267                 12.583                23.942

 Facility Ratio                          0.663                  0.692                 0.328                 0.011                 1.000

 LIBOR                                   3.977                  5.065                 2.147                 0.116                 6.838

 Term Premium                            1.287                  0.858                 1.022                 -0.095                2.904

 Interest Volatility                     0.382                  0.365                 0.081                 0.254                 0.625

 Total Asset                              3135.830              726.174               8849.160              1.648                 242223.000

 Company Consent                         0.942                  1.000                 0.233                 0.000                 1.000

 Agent Consent                            0.967                 1.000                 0.180                 0.000                 1.000

 Assignment Size ($)                      5687895.740           5000000.000           3627169.110           0.000                 50000000.000

 LBO                                     0.015                  0.000                 0.122                 0.000                 1.000

 Takeover                                0.224                  0.000                 0.417                 0.000                 1.000

 Repay/Recapitalization                   0.274                 0.000                 0.446                 0.000                 1.000

 Other purpose                           0.127                  0.000                 0.333                 0.000                 1.000

*There are total 5064 non-missing observations used in the facility level loan pricing regressions.
Table 2 Sample statistics for the dependent and independent variables used in the loan pricing model (at facility level), partitioned by
institutional tranches and bank tranches. This table presents summary statistics for the dependent variable all-in-drawn spread (measured in
percentage) and for the independent variables in the facility level loan pricing regression models. The loan agreements were originated during
the period January 1995 - December 2004. We classify the sample by institutional loan dummy. Institutional loan dummy takes a value of 1 for
institutional loans and 0 for bank loans. The dependent variable is the initial all-in-spread drawn which is defined as the percentage coupon
spread over LIBOR plus the annual fee and the upfront fee paid by borrowers for each dollar drawn down. We include M/B (Market/Book),
Leverage (Total Debt/Assets), Altman Z-score (3.3*EBIT/SALES+SALES/TA+1.4*RE/TA+1.2*WC/TA, where EBIT is the earning before
interests and taxes, SALES is the total sales of the year, TA is the total assets, WC is the working capital and RE is the retained earning), or
alternatively Investment Grade (taking the value of 1 if the S&P long term debt rating is BBB or above and 0 otherwise) and Not Rated dummy
(taking the value of 1 if the S&P long term debt credit rating is missing and 0 otherwise), which serve as proxies for two groups of borrower
characteristics: borrower risk and information costs. Firm characteristics are acquired at the fiscal year end prior to the deal year. We also
include other loan contract terms such as loan facility duration in years (maturity), and secured status taking the value of 1 for secured loan or 0
otherwise. Loan Concentration is defined as ln(deal/(deal+TD)) where deal is the deal amount and TD is the total debt including long term debt
and current liabilities. Revolver is the dummy variable taking the value of 1 for credit revolver and 0 for term loan. We control for facility size
(log of facility amount), facility ratio (facility amount/ deal amount), Firm Size (log of firm’s total assets), Risk free interest rate (LIBOR),
Interest rate risk (Interest volatility), term premium, year dummy and industry dummy. LIBOR is the deal month end 3-month LIBOR rate.
Interest Volatility is a 12-month moving average of the standard deviation of the monthly yields on 10-year U.S. T-bonds at deal month end
and term premium is the 12-month average for the deal year of the yield differential between 10 years and 1 year U.S. bonds. Facility Ratio is
defined as log of facility amount over the deal amount. Assignment Size specifies the minimum assignment size for loan resale in a dollar
amount stipulated in the loan contract. Agent consent takes the value of 1 if the consent from lead banks is required for selling the loan or 0
otherwise while the borrower consent takes the value of 1 if the consent from borrower is required for selling the loan or 0 otherwise. We
identify 10 categories of industries. The t-statistics presented in the last column are the difference in mean between the two samples, bank loan
sample minus the institutional loan sample. ***, **, * indicate significance at 1%, 5%, and 10% levels, respectively.

                                                                           Standard
        Variables             *        Mean             Median                               Minimum          Maximum                 T-test
                                                                           Deviation
 All-in-Drawn Spread          0         1.72              1.50                1.10              0.15             10.00              -24.83***
                              1         3.07              3.00                1.11              1.00             10.50
           MB                 0         1.78              1.45                1.20              0.51             22.16               2.67***
                              1         1.65              1.39                0.97              0.55              7.16
     Altman Z score           0         1.85              1.85                1.44             -29.69             8.48               6.79***

                              1         1.44              1.45                1.20              -6.07             5.39                    -7.5***
        Leverage              0         0.36              0.34                0.20              0.00              0.96
                              1         0.44              0.42                0.20              0.02              0.88
   Investment Grade           0         0.20              0.00                0.40              0.00              1.00              14.88***
                              1         0.04              0.00                0.20              0.00              1.00
       Not Rated              0         0.51              1.00                0.50              0.00              1.00               5.64***
                              1         0.38              0.00                0.48              0.00              1.00
   Maturity (in years)        0         1.15              1.39                0.66              -2.48             2.61              -25.6***
                              1         1.70              1.79                0.41              -0.88             2.30
     Secured Status           0         0.63              1.00                0.48              0.00              1.00              -38.36***
                              1         0.98              1.00                0.12              0.00              1.00
  Loan Concentration          0         0.41              0.40                0.21              0.00              1.00               -3.8***
                              1         0.45              0.43                0.21              0.02              0.99
        Revolver              0         0.80              1.00                0.40              0.00              1.00              136.27***
                              1         0.00              0.00                0.00              0.00              0.00
   Facility Size ($M)         0        18.71             18.76                1.28              12.58            23.94                -0.67
                              1        18.75             18.83                1.18              13.82            21.31
      Facility Ratio          0         0.68              0.75                0.33              0.01              1.00              17.02***
                              1         0.46              0.42                0.26              0.03              1.00
         LIBOR                0         4.04              5.17                2.14              0.12              6.84               6.27***
                              1         3.37              2.40                2.17              0.12              6.84
     Term Premium             0         1.26              0.83                1.01              -0.10             2.90              -5.15***
                              1         1.54              1.54                1.09              -0.09             2.90
   Interest Volatility        0         0.38              0.36                0.08              0.25              0.63               4.64***
                              1         0.37              0.36                0.07              0.25              0.60
      Total Assets            0       3220.04            712.93             9124.21             1.65           242223.00             3.37***
                              1       2278.69            838.81             5220.31             7.96            48792.00
   Company Consent            0         0.94              1.00                0.24              0.00              1.00                 -2**
                           1        0.96            1.00             0.20             0.00            1.00
    Agent Consent          0        0.97            1.00             0.18             0.00            1.00             -2.09**
                           1        0.98            1.00             0.14             0.00            1.00
  Assignment Size ($)      0    5888342.88       5000000.00       3680008.70          0.00        50000000.00         19.48***
                           1    3647582.78       5000000.00       2159984.47          0.00        10000000.00
         LBO               0        0.01            0.00             0.11             0.00            1.00            -3.41***
                           1        0.05            0.00             0.21             0.00            1.00
       Takeover            0        0.21            0.00             0.41             0.00            1.00            -4.56***
                           1        0.32            0.00             0.47             0.00            1.00
Repay/Recapitalization     0        0.28            0.00             0.45             0.00            1.00               1.4
                           1        0.25            0.00             0.43             0.00            1.00
     Other purpose         0        0.13            0.00             0.34             0.00            1.00             8.7***
                           1        0.04            0.00             0.20             0.00            1.00

* This column shows the indicator 0 for the bank loan tranches and 1 for the institutional loan. We have 4611 non-missing bank loan
observations and 453 non-missing institutional loan observations used in our facility level loan pricing model.
Table 3 Sample statistics for the dependent and independent variables used in the loan pricing model (at deal level), partitioned by
institutional participation deal and non-institutional participation deal. This table presents summary statistics for the dependent variable
all-in-drawn spread (measured in percentage) and for the independent variables in the deal level loan pricing regression models. The loan
agreements were originated during the period January 1995 - December 2004. We classify the sample by institutional participation dummy.
Institutional participation dummy takes a value of 1 for deals that involve institutional investors and 0 otherwise. The dependent variable is the
facility amount weighted average initial all-in-spread drawn of the deal which is defined as the percentage coupon spread over LIBOR plus the
annual fee and the upfront fee paid by borrowers for each dollar drawn down. We include M/B (Market/Book), Leverage (Total Debt/Assets),
Altman Z-score (3.3*EBIT/SALES+SALES/TA+1.4*RE/TA+1.2*WC/TA, where EBIT is the earning before interests and taxes, SALES is the
total sales of the year, TA is the total assets, WC is the working capital and RE is the retained earning), or alternatively Investment Grade
(taking the value of 1 if the S&P long term debt rating is BBB or above and 0 otherwise) and Not Rated dummy (taking the value of 1 if the
S&P long term debt credit rating is missing and 0 otherwise), which serve as proxies for two groups of borrower characteristics: borrower risk
and information costs. Firm characteristics are acquired at the fiscal year end prior to the deal year. We also include other loan contract terms
such as loan deal average duration in years (facility amount weighted average maturity), and secured status taking the value of 1 for secured
loan or 0 otherwise. Loan Concentration is defined as ln(deal/(deal+TD)) where deal is the deal amount and TD is the total debt including long
term debt and current liabilities. Revolver is the dummy variable taking the value of 1 for credit revolver and 0 for term loan. We control for
facility size (log of facility amount), facility ratio (facility amount/ deal amount), Firm Size (log of firm’s total assets), Risk free interest rate
(LIBOR), Interest rate risk (Interest volatility), term premium, year dummy and industry dummy. LIBOR is the deal month end 3-month
LIBOR rate. Interest Volatility is a 12-month moving average of the standard deviation of the monthly yields on 10-year U.S. T-bonds at deal
month end and term premium is the 12-month average for the deal year of the yield differential between 10 years and 1 year U.S. bonds.
Facility Ratio is defined as log of facility amount over the deal amount. Assignment Size specifies the minimum assignment size for loan resale
in a dollar amount stipulated in the loan contract. Agent consent takes the value of 1 if the consent from lead banks is required for selling the
loan or 0 otherwise while the borrower consent takes the value of 1 if the consent from borrower is required for selling the loan or 0 otherwise.
We identify 10 categories of industries. We also control for the impact of different financial covenants on the loan pricing. This variable
reflects 15 different types of financial covenant including maximum capital expenditure, maximum debt to EBITDA ratio, maximum debt to
equity ratio, maximum senior leverage coverage ratio, maximum senior debt to EBITDA, maximum loan value, maximum leverage ratio,
maximum debt to net worth, minimum interest coverage ratio, minimum quick ratio, minimum fixed charge coverage ratio, minimum
EBITDA, minimum debt service coverage ratio, minimum current ratio, minimum cash interest coverage ratio. We normalize maximum capital
expenditure, maximum loan value and minimum EBITDA by firm’s total asset. If a loan deal does not have certain types of financial
covenants, we assign the value of 0 for the variables corresponding to the types of financial covenants. The omitted case is minimum quick
ratio. The t-statistics presented in the last column are the difference in mean between the two samples, non-insitutional participation sample
minus the institutional participation sample. ***, **, * indicate significance at 1%, 5%, and 10% levels, respectively.

  Variable                           *        Mean             Median           Standard          Minimum           Maximum              T-Test
                                                                                Deviation
 All-in-Drawn Spread                 0        1.498              1.250             1.016             0.150             10.000          -27.38***
                                     1        2.892              2.763             0.976             0.813             8.784
 MB                                  0        1.792              1.445             1.226             0.510             22.161           2.99***
                                     1        1.638              1.385             0.953             0.549             7.165
 Altman Z score                      0        1.938              1.930             1.428            -29.690            8.484            7.57***
                                     1        1.458              1.466             1.186             -6.068            5.388
 Leverage                            0        0.348              0.325             0.195             0.004             0.964            -8.33***
                                     1        0.435              0.425             0.202             0.018             0.882
 Investment Grade                    0        0.234              0.000             0.423             0.000             1.000            16.14***
                                     1        0.038              0.000             0.191             0.000             1.000
 Not Rated                           0        0.522              1.000             0.500             0.000             1.000            5.49***
                                     1        0.383              0.000             0.487             0.000             1.000
 Maturity (in years)                 0        1.102              1.126             0.646             -2.485            2.609           -22.77***
                                     1        1.667              1.792             0.448             -0.875            2.303
 Secured Status                      0        0.550              1.000             0.498             0.000             1.000           -38.09***
                                     1        0.981              1.000             0.136             0.000             1.000
 Loan Concentration                  0        0.373              0.355             0.210             0.001             0.996            -6.82***
                                     1        0.448              0.434             0.213             0.019             0.987
 Revolver                            0        0.823              1.000             0.382             0.000             1.000            74.08***
                                     1        0.028              0.000             0.166             0.000             1.000
 Log Deal Size                       0        19.088            19.114             1.161             15.607            23.942           -9.96***
                                     1        19.645            19.599             1.063             16.118            22.669
 LIBOR                               0        3.945              5.053             2.170             0.116             6.838            5.26***
                                     1        3.352              2.403             2.162             0.116             6.838
 Term Premium                        0        1.323              0.890             1.019             -0.095            2.904            -3.94***
                                     1        1.544              1.748             1.089             -0.085            2.904
 Interest Volatility                 0        0.385              0.365             0.083             0.254             0.625            4.76***
                                     1        0.367              0.362             0.069             0.254             0.597
Total Asset                0   3468.250   716.200    10060      1.648     242223    4.09***
                           1   2196.920   811.586     5118      7.958     48792
Company Consent            0    0.938      1.000      0.242     0.000     1.000      -2.1**
                           1    0.960      1.000      0.197     0.000     1.000
Agent Consent              0    0.962      1.000      0.191     0.000     1.000     -2.52***
                           1    0.981      1.000      0.136     0.000     1.000
Assignment Size ($)        0   6256460    5000000   3856799     0.000    50000000   20.58***
                           1   3641146    5000000   2164438     0.000    10000000
LBO                        0    0.002      0.000      0.049     0.000     1.000      -4.2***
                           1    0.045      0.000      0.207     0.000     1.000
Takeover                   0    0.156      0.000      0.363     0.000     1.000     -5.94***
                           1    0.293      0.000      0.456     0.000     1.000
Repay/Recapitalization     0    0.301      0.000      0.459     0.000     1.000      2.42**
                           1    0.246      0.000      0.431     0.000     1.000
Other Purposes             0    0.140      0.000      0.347     0.000     1.000     7.63***
                           1    0.047      0.000      0.212     0.000     1.000
Max. Capital Expenditure   0    0.015      0.000      0.048     0.000     0.750      -6***
                           1    0.034      0.006      0.065     0.000     0.479
Max Debt to EBIT           0    1.917      2.000      2.046     0.000     18.500    -16.38***
                           1    4.119      4.250      2.652     0.000     25.000
Max. Debt to Equity        0    0.005      0.000      0.103     0.000     3.750       -0.68
                           1    0.013      0.000      0.219     0.000     4.400
Max. Debt to Net Worth     0    0.094      0.000      0.472     0.000     6.200       1.74
                           1    0.056      0.000      0.409     0.000     4.000
Max. Leverage Ratio        0    0.149      0.000      0.300     0.000     5.250     6.38***
                           1    0.053      0.000      0.289     0.000     4.950
Max Loan to Value          0    0.000      0.000      0.022     0.000     1.200       -0.92
                           1    0.005      0.000      0.109     0.000     2.250
Max Senior Debt to EBIT    0    0.243      0.000      0.897     0.000     12.000    -8.97***
                           1    0.989      0.000      1.676     0.000     8.100
Max. Senior Leverage       0    0.001      0.000      0.018     0.000     0.550       -0.84
R ti                       1    0.004      0.000      0.085     0.000     1.750
Min. Cash Interest         0   5110.76     0.000    276877.21   0.000    15000000      1
C                          1    0.034      0.000      0.283     0.000     3.500
Min. Current Ratio         0    0.106      0.000      0.373     0.000     7.000      1.95**
                           1    0.073      0.000      0.314     0.000     2.250
Min. Debt Service          0    0.081      0.000      0.381     0.000     4.000       -0.74
C                          1    0.095      0.000      0.373     0.000     3.500
Min. EBITDA                0    0.006      0.000      0.034     -0.126    0.608     -3.86***
                           1    0.016      0.000      0.054     -0.031    0.367
Min. Fix Charge Coverage   0    0.611      0.000      0.849     0.000     4.000     -4.28***
                           1    0.766      1.000      0.672     0.000     4.000
Min. Interest Coverage     0    1.024      0.000      1.459     0.000     7.500     -6.76***
                           1    1.494      1.700      1.316     0.000     5.750
Min. Quick Ratio           0    0.024      0.000      0.192     0.000     6.000     6.89***
                           1    0.000      0.000      0.000     0.000     0.000
Table 4. Facility level loan pricing model (1995-2004). This table presents facility level loan pricing model results. The loan agreements were
originated during the period January 1995 - December 2004. The dependent variable is the initial all-in-spread drawn which is defined as the
percentage coupon spread over LIBOR plus the annual fee and the upfront fee paid by borrowers for each dollar drawn down. The Institutional
loan dummy takes a value of 1 for institutional loans and 0 for bank loans. We include M/B (Market/Book), Leverage (Total Debt/Assets),
Altman Z-score (3.3*EBIT/SALES+SALES/TA+1.4*RE/TA+1.2*WC/TA, where EBIT is the earning before interests and taxes, SALES is the
total sales of the year, TA is the total assets, WC is the working capital and RE is the retained earning), or alternatively Investment Grade
(taking the value of 1 if the S&P long term debt rating is BBB or above and 0 otherwise) and Not Rated dummy (taking the value of 1 if the
S&P long term debt credit rating is missing and 0 otherwise), which serve as proxies for two groups of borrower characteristics: borrower risk
and information costs. Firm characteristics are acquired at the fiscal year end prior to the deal year. We also include other loan contract terms
such as loan facility duration in years (maturity), and secured status taking the value of 1 for secured loan or 0 otherwise. Loan Concentration is
defined as ln(deal/(deal+TD)) where deal is the deal amount and TD is the total debt including long term debt and current liabilities. Revolver
is the dummy variable taking the value of 1 for credit revolver and 0 for term loan. We control for facility size (log of facility amount), facility
ratio (facility amount/ deal amount), Firm Size (log of firm’s total assets), Risk free interest rate (LIBOR), Interest rate risk (Interest volatility),
term premium, year dummy and industry dummy. LIBOR is the deal month end 3-month LIBOR rate. Interest Volatility is a 12-month moving
average of the standard deviation of the monthly yields on 10-year U.S. T-bonds at deal month end and term premium is the 12-month average
for the deal year of the yield differential between 10 years and 1 year U.S. bonds. Facility Ratio is defined as log of facility amount over the
deal amount. Assignment Size specifies the minimum assignment size for loan resale in a dollar amount stipulated in the loan contract. Agent
consent takes the value of 1 if the consent from lead banks is required for selling the loan or 0 otherwise while the borrower consent takes the
value of 1 if the consent from borrower is required for selling the loan or 0 otherwise. We identify 10 categories of industries. The
heteroskedastic corrected t-statistics are presented in parentheses. ***, **, * indicate significance at 1%, 5%, and 10% levels, respectively.

 Variables                                               Model 1              Model 2              Model 3             Model 4              Model 5
 Intercept                                             2.83                 2.97                 3.75                 4.02                 3.38
                                                      (7.43)               (8.02)              (10.02)               (8.96)               (7.15)
 Institutional Loan Dummy                                                   0.48                 0.48                 0.48                 0.46
                                                                         (7.89)***            (8.06)***            (8.06)***            (7.82)***
 M/B                                                   0.01                 0.02                 0.03                 0.02                 0.01
                                                      (1.57)               (1.83)               (2.38)               (2.11)               (1.27)
 Leverage                                              1.35                 1.34                 1.48                 1.5                  1.42
                                                      (14.21)              (14.34)             (15.17)              (15.07)              (14.30)
 Altman Z Score                                                                                  -0.1                 -0.1                 -0.1
                                                                                                (-5.96)             (-5.29)              (-5.39)
 Investment Grade                                      -0.37                -0.36
                                                     (-10.32)             (-10.61)
 Not Rated                                             -0.09                -0.09
                                                      (-2.93)              (-2.8)
 Maturity                                              -0.07                -0.09               -0.07                -0.07                -0.03
                                                      (-3.33)              (-4.28)              (-3.28)             (-3.29)              (-1.58)
 Secured Status                                        0.83                 0.82                 0.85                 0.84                 0.81
                                                      (29.92)              (30.15)             (30.42)              (30.26)              (29.78)
 Loan Concentration                                    -0.13                -0.12                -0.02                0.01                -0.03
                                                      (-2.97)              (-2.74)              (-0.35)              (0.28)              (-0.61)
 Revolver                                              -0.46                -0.32                -0.31               -0.31                -0.29
                                                     (-12.82)              (-8.08)              (-7.85)             (-7.85)              (-7.69)
 Facility Size                                         0.002                -0.02               -0.07                -0.08                -0.05
                                                      (0.08)               (-0.56)              (-2.27)             (-2.67)               (-1.8)
 Facility Ratio                                        -0.35                -0.34               -0.22                 -0.2                -0.25
                                                      (-5.45)              (-5.46)              (-3.39)             (-3.05)               (-3.9)
 LIBOR                                                 -0.14                -0.13               -0.14                -0.14                -0.04
                                                     (-12.53)             (-11.98)             (-12.63)             (-12.8)              (-1.78)
 Term Premium                                          -0.14                -0.13                -0.15               -0.15                -0.13
                                                      (-6.03)              (-5.81)              (-6.36)             (-6.48)              (-2.23)
 Interest Volatility                                   0.29                 0.29                 0.31                 0.32                 1.49
                                                      (2.11)               (2.18)               (2.32)               (2.44)               (4.73)
 Firm Size                                             -0.1                 -0.1                -0.06                -0.06                 -0.1
                                                      (-3.59)              (-3.64)              (-2.1)              (-1.79)              (-3.14)
 Agent Consent                                          0.2                  0.2                 0.19                 0.19                 0.04
                                                      (3.78)               (3.92)               (3.50)               (3.58)               (0.84)
Company Consent               -0.12      -0.12      -0.11      -0.11     -0.16
                             (-2.14)    (-2.17)      (-2)     (-2.01)    (-2.99)
Min. Assignment Size          -0.04      -0.03      -0.04      -0.03     -0.03
                             (-11.27)   (-10.69)   (-11.16)   (-10.65)   (-8.55)
LBO                           0.53       0.49       0.49       0.47       0.47
                              (7.16)     (6.62)     (6.62)    (6.37)     (6.64)
Takeover                      0.05       0.05       0.04       0.05       0.08
                              (1.41)     (1.41)     (1.30)    (1.35)     (2.24)
Repayment/Recapitalization    -0.07      -0.07      -0.07      -0.07     -0.03
                             (-1.98)    (-2.23)    (-2.15)    (-2.07)    (-0.91)
Other Purposes                -0.17      -0.16      -0.25      -0.25     -0.28
                             (-4.63)    (-4.33)    (-6.99)    (-7.07)    (-7.94)
Industry Dummy                 no         no         no         Yes       Yes
Year Dummy                     no         no         no         No        Yes
Number of Observations        5064       5064       5064       5064      5064
Adjusted R square            0.5753     0.5846     0.5922     0.5945     0.6118
Table 5 The deal level loan pricing model (1995-2004). This table presents deal level loan pricing regression results. The loan agreements
were originated during the period January 1995 - December 2004. The dependent variable is the facility amount weighted average initial all-in-
spread drawn of the deal which is defined as the percentage coupon spread over LIBOR plus the annual fee and the upfront fee paid by
borrowers for each dollar drawn down. Institutional participation dummy takes a value of 1 for deals that involve institutional investors and 0
otherwise.         We        include         M/B       (Market/Book),         Leverage       (Total       Debt/Assets),         Altman       Z-score
(3.3*EBIT/SALES+SALES/TA+1.4*RE/TA+1.2*WC/TA, where EBIT is the earning before interests and taxes, SALES is the total sales of
the year, TA is the total assets, WC is the working capital and RE is the retained earning), or alternatively Investment Grade (taking the value
of 1 if the S&P long term debt rating is BBB or above and 0 otherwise) and Not Rated dummy (taking the value of 1 if the S&P long term debt
credit rating is missing and 0 otherwise), which serve as proxies for two groups of borrower characteristics: borrower risk and information
costs. Firm characteristics are acquired at the fiscal year end prior to the deal year. We also include other loan contract terms such as loan deal
average duration in years (facility amount weighted average maturity), and secured status taking the value of 1 for secured loan or 0 otherwise.
Loan Concentration is defined as ln(deal/(deal+TD)) where deal is the deal amount and TD is the total debt including long term debt and
current liabilities. Revolver is the dummy variable taking the value of 1 for credit revolver and 0 for term loan. We control for facility size (log
of facility amount), facility ratio (facility amount/ deal amount), Firm Size (log of firm’s total assets), Risk free interest rate (LIBOR), Interest
rate risk (Interest volatility), term premium, year dummy and industry dummy. LIBOR is the deal month end 3-month LIBOR rate. Interest
Volatility is a 12-month moving average of the standard deviation of the monthly yields on 10-year U.S. T-bonds at deal month end and term
premium is the 12-month average for the deal year of the yield differential between 10 years and 1 year U.S. bonds. Facility Ratio is defined as
log of facility amount over the deal amount. Assignment Size specifies the minimum assignment size for loan resale in a dollar amount
stipulated in the loan contract. Agent consent takes the value of 1 if the consent from lead banks is required for selling the loan or 0 otherwise
while the borrower consent takes the value of 1 if the consent from borrower is required for selling the loan or 0 otherwise. We identify 10
categories of industries. We also control for the impact of different financial covenants on the loan pricing. This variable reflects 15 different
types of financial covenant including maximum capital expenditure, maximum debt to EBITDA ratio, maximum debt to equity ratio, maximum
senior leverage coverage ratio, maximum senior debt to EBITDA, maximum loan value, maximum leverage ratio, maximum debt to net worth,
minimum interest coverage ratio, minimum quick ratio, minimum fixed charge coverage ratio, minimum EBITDA, minimum debt service
coverage ratio, minimum current ratio, minimum cash interest coverage ratio. We normalize maximum capital expenditure, maximum loan
value and minimum EBITDA by firm’s total asset. If a loan deal does not have certain types of financial covenants, we assign the value of 0 for
the variables corresponding to the types of financial covenants. The omitted case is minimum quick ratio. The heterskadastic corrected t-
statistics are presented in parentheses. ***, **, * indicate significance at 1%, 5%, and 10% levels, respectively.

                  Variables                         Model 1             Model 2              Model 3             Model 4              Model 5
                  Intercept                           3.87                 4.38                5.42                 5.98                4.27
                                                   (6.63)***              (7.82)              (9.50)               (9.28)              (6.50)
    Institutional Participation Dummy                                      0.55                0.55                 0.56                0.54
                                                                        (8.84)***           (9.13)***            (9.25)***           (8.83)***
                     M/B                              0.01                 0.01                0.02                 0.02                0.01
                                                     (0.77)               (0.67)              (2.03)               (1.88)              (0.47)
                  Leverage                            1.44                 1.43                1.56                 1.59                1.41
                                                     (12.36)             (12.44)              (13.57)             (13.50)             (12.19)
               Altman Z Score                                                                  -0.09                                    -0.3
                                                                                              (-4.8)               (-4.3)              (-7.44)
              Investment Grade                        -0.31               -0.29
                                                     (-7.42)             (-7.12)
                  Not Rated                           -0.07               -0.04                                    -0.08
                                                      (-1.8)             (-1.09)
              Average Maturity                        -0.14               -0.15                -0.13               -0.13                -0.04
                                                     (-5.32)               (-6)               (-5.36)             (-5.22)              (-1.2)
               Secured Status                         0.69                 0.65                0.67                 0.67                -0.1
                                                     (22.11)             (20.97)              (20.85)             (20.66)              (-4.03)
             Loan Concentration                       -0.03               -0.02                0.08                 0.12                0.62
                                                     (-0.55)             (-0.32)              (1.46)               (2.00)             (20.19)
                  Revolver                            -0.54               -0.36                -0.34               -0.34                -0.03
                                                     (13.56)             (-8.26)              (-7.96)             (-8.04)              (-0.5)
                  Deal Size                           -0.09               -0.13                -0.19               -0.21                -0.35
                                                     (-2.22)             (-3.36)              (-4.72)             (-5.11)              (-8.27)
                   LIBOR                              -0.11                -0.1                -0.1                -0.11                -0.11
                                                     (-8.12)             (-7.93)              (-7.95)             (-8.12)              (-3.01)
               Term Premium                           -0.11                -0.1                -0.1                 -0.1                -0.03
                                                      (-3.8)             (-3.65)              (-3.71)             (-3.81)              (-0.94)
              Interest Volatility                      0.1                 0.11                 0.1                 0.11                -0.11
                                                     (0.64)               (0.72)              (0.65)               (0.76)              (-1.81)
                  Firm Size                           -0.03               -0.02                0.02                 0.04                1.52
                             (-0.77)   (-0.46)   (0.58)    (0.87)    (4.39)
      Agent Consent           0.16      0.17      0.16      0.16     -0.05
                             (2.63)    (2.89)    (2.74)    (2.78)    (-1.42)
    Company Consent           -0.1     -0.12     -0.12     -0.12      0.09
                             (-1.58)   (-1.91)   (-1.97)   (-1.92)   (1.50)
  Min. Assignment Size       -0.02     -0.02     -0.02     -0.02     -0.14
                             (-7.37)   (-6.52)   (-6.47)   (-6.2)    (-2.41)
          LBO                 0.47      0.31      0.32      0.31     -0.01
                             (3.49)    (2.89)    (2.24)    (2.19)    (-4.81)
        Takeover              0.03      0.05      0.05      0.05      0.38
                             (0.82)    (-6.52)   (1.18)    (1.28)    (2.95)
Repayment/Recapitalization   -0.05     -0.04     -0.04     -0.04      0.09
                             (-1.29)   (1.34)    (1.18)    (1.28)    (2.26)
     Other Purposes          -0.16     -0.16     -0.22     -0.21       0
                             (-3.91)   (-4.07)   (-5.53)   (-5.48)   (-0.02)
   Financial Covenants         Yes       Yes       Yes       Yes       Yes
     Industry Dummy           No        No        No        Yes       Yes
      Year Dummy              No        No        No        No        Yes
 Number of Observations      3366      3366      3366      3366      3366
    Adjusted R square        0.5976    0.6186    0.6247    0.6267    0.6342
Table 6. Standard deviation of prices and number of average daily quotes for institutional and bank loans. This table reports the
standard deviation of loan prices of number of average daily quotations for all of the secondary market samples. There are 4186 observations.
We classify the sample by institutional loan dummy. Institutional loan dummy takes a value of 1 for institutional loans and 0 for bank loans.
The t-statistics presented in the last column are the difference in mean between the two samples, bank loan sample minus the institutional loan
sample. ***, **, * indicate significance at 1%, 5%, and 10% levels, respectively.

  Variables      *        No. of           Mean             Median          Standard            Min.              Max.             T-test
                          Obs.                                              Deviation


  Standard       0         2858             2.578            0.388             5.802            0.000            43.094          -6.57***
  Deviation
  Of Prices
                 1         1328             4.321            0.688             8.820            0.000           144.577



 Number of       0         2858             1.851            1.233             1.411            1.000            12.607          -15.43***
  Average
   Daily
  Quotes
                 1         1328             2.868            2.091             2.199            1.000            14.550



* Indicator 0 represents institutional loan tranches while indicator 1 represents bank loan tranches.
Table 7. Summary statistics of dependent and independent variables used in initial loan resale model (1998-2004). This table presents
summary statistics for the dependent variable the raw initial loan resale return or market adjusted initial loan resale return and for all the
dependent variables used in the initial loan resale return models. The loans were traded during the period June 1998 - December 2004. The
initial loan resale return is calculated as the first day trading price-100. The market adjusted initial loan resale return is calculated as the initial
loan resale return minus the S&P/LSTA leveraged loan market index return during the same period. We classify the sample by institutional
loan dummy. Institutional loan dummy takes a value of 1 for institutional loans and 0 for bank loans. We control for initial all-in-drawn spread
as a proxy for the ex ante loan risks. The initial all-in-drawn spread is defined as the percentage coupon spread over LIBOR plus the annual fee
and the upfront fee paid by borrowers for each dollar drawn down. We include lock-up period which is defined as the time (measure in days)
between the first loan trading day and the loan’s origination day, time to maturity which is calculated as the time (in days) between loan
maturity day and the first loan trading day, risk free interest rate (LIBOR), Interest rate risk (Interest volatility), term premium. LIBOR is the
deal month end 3-month LIBOR rate. Interest Volatility is a 12-month moving average of the standard deviation of the monthly yields on 10-
year U.S. T-bonds at deal month end and term premium is the 12-month average for the deal year of the yield differential between 10 years and
1 year U.S. bonds. In addition, we also incorporate some other loan features including secured status takes the value of 1 if the loan is secured
or 0 otherwise, credit revolver dummy, number of lenders (log of number of lenders), facility size (log of facility amount), facility ratio (facility
amount/ deal amount) and loan purpose dummies (including leverage buyout loan, takeover loan, repayment/recapitalization loan and other
purposes, the omitted case is general corporate purpose loan). We use log of number of quotations on the first trading day to proxy for the
market awareness of the loan and control for the firm size (log of sales at close ) as well. The t-statistics presented in the last column are the
difference in mean between the two samples, bank loan sample minus the institutional loan sample. ***, **, * indicate significance at 1%, 5%,
and 10% levels, respectively.
            Variable                   Number          Mean                        Standard      Minimum         Maximum                 Mean
                                  *       of                           Median     Deviation                                           Difference
                                         Obs.                                                                                         t Statistics
     Initial Resale Return        0      2084          -3.96           -0.75        10.35          -99.50           20.00            -11.42***
                               1       1093          -0.70           0.125            5.75          -90.00          13.00
 Market Adjusted Return         0      2084          -2.12           -0.125           9.91          -97.25          19.68             -8.30***
                                1      1093           0.16          0.5625            5.57          -85.77          12.50
           Quotes               0      2084           1.23             1              0.71           1.00            8.00             -3.92***
                                1      1093           1.37             1              1.03           1.00           16.00
   Lock-up Period (days)        0      2084         314.18           163.5          389.59           0.00          2073.00            16.09***
                                1      1093         133.55             22           241.14           0.00          1861.00
       Secured Status           0      1616           0.92             1              0.28           0.00            1.00             -7.65***
                                1       882           0.98             1              0.14           0.00            1.00
   All-in-Drawn Spread          0      2082         275.92            275           128.36           15.00         1150.00            -12.18***
                                1      1092         327.56            325           104.79          100.00         1300.00
 Time to Maturity (Days)        0      2082         1623.49         1730.00        734.417          20.000         3567.00            -23.36***
                                1      1092         2230.77         2328.50        537.305          87.000         3407.00
     Maturity (Months)          0      2024          62.95             60            21.57           3.00           120.00            -19.25***
                                1      1040          76.89             83            17.49           6.00           119.00
          Revolver              0      2084           0.47             0              0.50           0.00            1.00             43.29***
                                1      1093           0.00             0              0.00           0.00            0.00
            LBO                 0      2084           0.16             0              0.37           0.00            1.00             -5.66***
                                1      1093           0.25             0              0.43           0.00            1.00
         Takeover               0      2084           0.27             0              0.45           0.00            1.00                0.30
                                1      1093           0.27             0              0.44           0.00            1.00
           Repay                0      2084           0.23             0              0.42           0.00            1.00                0.54
                                1      1093           0.22             0              0.41           0.00            1.00
       Other purpose            0      2084           0.11             0              0.32           0.00            1.00              5.37***
                                1      1093           0.06             0              0.23           0.00            1.00
       Sales at Close           0      1788         2789.13         714.115        8730.81           0.96         191329.00            6.88***
                                1       902         1201.45           400          3088.00           0.64         46226.00

* Indicator 0 represents institutional loan tranches while indicator 1 represents bank loan tranches.
Table 8 Initial loan resale return model (1998-2004). This table reports initial loan resale return regression model results. The loans were
traded during the period June 1998 - December 2004. The dependent variable is either the raw initial loan resale return (reported in the first
four columns) or market adjusted initial loan resale return (reported in the last four columns). The raw initial loan resale return is calculated as
the first day trading price-100. The market adjusted initial loan resale return is calculated as the initial loan resale return minus the S&P/LSTA
leveraged loan market index return during the same period. Institutional loan dummy takes a value of 1 for institutional loans and 0 for bank
loans. We control for initial all-in-drawn spread as a proxy for the ex ante loan risks. The initial all-in-drawn spread is defined as the percentage
coupon spread over LIBOR plus the annual fee and the upfront fee paid by borrowers for each dollar drawn down. We include lock-up period
which is defined as the time (measure in days) between the first loan trading day and the loan’s origination day, time to maturity which is
calculated as the time (in days) between loan maturity day and the first loan trading day, risk free interest rate (LIBOR), Interest rate risk
(Interest volatility), term premium. LIBOR is the deal month end 3-month LIBOR rate. Interest Volatility is a 12-month moving average of the
standard deviation of the monthly yields on 10-year U.S. T-bonds at deal month end and term premium is the 12-month average for the deal
year of the yield differential between 10 years and 1 year U.S. bonds. In addition, we also incorporate some other loan features including
secured status takes the value of 1 if the loan is secured or 0 otherwise, credit revolver dummy, number of lenders (log of number of lenders),
facility size (log of facility amount), facility ratio (facility amount/ deal amount) and loan purpose dummies (including leverage buyout loan,
takeover loan, repayment/recapitalization loan and other purposes, the omitted case is general corporate purpose loan). We use log of number
of quotations on the first trading day to proxy for the market awareness of the loan and control for the firm size (log of sales at close) as well.
The heterskadastic corrected t-statistics are presented in parentheses. ***, **, * indicate significance at 1%, 5%, and 10% levels, respectively.


                                        Initial Loan Resale Return                           Market Adjusted Initial Loan Resale Return
       Variable
                           Model 1        Model 2       Model 3           Model 4         Model 1        Model 2         Model 3        Model 4
       Intercept             -6.95          -4.26           -3.66           -3.94          -6.77           -3.83          -2.26           -2.41
                           (-2.71)**        (-1.11)         (-1.8)         (-2.03)       (-2.48)**        (-0.95)         (-1.1)         (-1.23)
 Institutional Loan          0.66            0.62            0.6            0.58            0.60           0.56            0.53           0.51
                           (2.55)**        (2.11)**       (2.30)**        (2.22)**        (2.35)**        (1.89)*        (2.03)**        (1.96)*
    Lock-up Period           -1.94            -2            -2.01            -2            -0.69           -0.72          -0.73           -0.71
                          (-7.61)***      (-7.7)***       (-7.7)***      (-7.56)***      (-2.74)***     (-2.81)***      (-2.83)***     (-2.69)***
 All-in-Drawn Spread         -0.4           -0.48           -0.41           -0.4           -0.57           -0.64          -0.54           -0.54
                          (-2.78)***      (-3.21)***     (-2.71)***      (-2.66)***      (-4.07)***     (-4.45)***      (-3.72)***     (-3.67)***
   Time to Maturity          0.51            0.56           0.58            0.59            0.56           0.58            0.62           0.63
                          (4.24)***       (4.49)***       (4.87)***      (4.86)***       (4.36)***      (4.44)***       (4.94)***       (4.92)***
    Secured Status           1.08            1.18                                           1.63           1.64
                             (1.3)          (1.38)                                        (1.73)*         (1.71)*
       Firm Size             0.33            0.34           0.31            0.29            0.29           0.32            0.29           0.27
                          (3.00)***        (2.46)**       (2.66)***       (2.53)**        (2.52)**       (2.25)**        (2.39)**       (2.23)**
   Number of Quotes          -1.63          -1.53           -1.49           -1.44          -1.63           -1.58          -1.51           -1.47
                          (-3.28)***      (-2.97)***       (-3)***        (-2.9)***      (-3.34)***     (-3.13)***      (-3.11)***     (-3.02)***
       Revolver              -2.02          -1.96           -1.97           -2.00          -2.06            -2            -2.01           -2.04
                          (-5.12)***      (-4.99)***       (-5)***       (-5.06)***       (-5)***       (-4.89)***      (-4.9)***      (-4.96)***
  Number of Lenders                         -0.36           -0.37           -0.37                          -0.4           -0.41           -0.41
                                           (-1.77)*       (-2.02)**       (-1.98)**                      (-1.94)*       (-2.16)**       (-2.12)**
      Facility Size                         -0.02                                                          0.02
                                            (-0.1)                                                        (0.08)
     Facility Ratio                         -0.05                                                          -0.14
                                            (-0.06)                                                       (-0.19)
        LIBOR                0.75            0.44           0.43            0.41            0.65           0.36            0.34           0.33
                           (2.49)**        (2.28)**       (2.24)**        (2.18)**         (2.07)         (1.83)*        (1.77)*         (1.73)*
    Term Premium             0.82            1.16           1.12            1.13            0.48            0.6            0.55           0.55
                             (1.1)        (3.14)***       (3.12)***      (3.18)***         (0.63)         (1.58)          (1.50)          (1.51)
  Interest Volatility        -3.75          -4.44           -4.29           -4.19          -1.97           -4.25          -4.05             -4
                            (-0.71)         (-1.56)        (-1.49)         (-1.46)        (-0.36)         (-1.47)        (-1.38)         (-1.37)
         LBO                 -0.16          -0.23           -0.22                          -0.01           -0.08          -0.08
                            (-0.36)         (-0.55)        (-0.53)                        (-0.03)         (-0.18)        (-0.17)
       Takeover              -0.11          -0.15           -0.15                          -0.08           -0.04          -0.04
                            (-0.25)         (-0.34)        (-0.34)                        (-0.18)          (-0.1)        (-0.08)
      Repayment              -0.41          -0.45           -0.44                          -0.25           -0.25          -0.24
                            (-0.94)         (-1.05)        (-1.02)                        (-0.57)         (-0.57)        (-0.53)
Other Purposes   -0.76     -0.88     -0.98              -0.74     -0.81     -0.94
                 (-1.12)   (-1.28)   (-1.41)            (-1.06)   (-1.15)   (-1.34)
Year Dummies      Yes       No         No       No        Yes       No        No       No
  No.of Obs.     2145      2145      2145      2145     2145      2145      2145      2145
Adj. R Square    0.2030    0.1943    0.1940    0.1941   0.1095    0.1086    0.1070    0.1074
Table 9 Re-estimating the second-stage loan pricing model under a simultaneous system. This table reports the coefficient estimates of the
simultaneous regression models allowing the maturity and loan pricing to be simultaneously determined. The results are reported for the facility
level regression in the first two columns and for the deal level regression in the last two columns respectively. The loan agreements were
originated during the period January 1995 - December 2004. The dependent variable for the facility level model is the spread (the initial all-in-
spread drawn) which is defined as the percentage coupon spread over LIBOR plus the annual fee and the upfront fee paid by borrowers for
each dollar drawn down. The dependent variable for the deal level regression is facility amount weighted average initial all-in-spread drawn. In
maturity regression, at facility level dependent variable is the loan facility duration (in years). At deal level, it is the facility amount weighted
average maturity of a loan deal. The Institutional loan (participation) dummy takes a value of 1 for institutional loans (participation) and 0
otherwise at facility (deal) level. We include M/B (Market/Book), Leverage (Total Debt/Assets), Altman Z-score
(3.3*EBIT/SALES+SALES/TA+1.4*RE/TA+1.2*WC/TA, where EBIT is the earning before interests and taxes, SALES is the total sales of
the year, TA is the total assets, WC is the working capital and RE is the retained earning), or alternatively Investment Grade (taking the value
of 1 if the S&P long term debt rating is BBB or above and 0 otherwise) and Not Rated dummy (taking the value of 1 if the S&P long term debt
credit rating is missing and 0 otherwise), which serve as proxies for two groups of borrower characteristics: borrower risk and information
costs. Firm characteristics are acquired at the fiscal year end prior to the deal year. We also include other loan contract terms such as loan
facility duration in years (maturity), and secured status taking the value of 1 for secured loan or 0 otherwise. Loan Concentration is defined as
ln(deal/(deal+TD)) where deal is the deal amount and TD is the total debt including long term debt and current liabilities. Revolver is the
dummy variable taking the value of 1 for credit revolver and 0 for term loan. We control for facility size (log of facility amount), facility ratio
(facility amount/ deal amount), Firm Size (log of firm’s total assets), Risk free interest rate (LIBOR), Interest rate risk (Interest volatility), term
premium, year dummy and industry dummy. LIBOR is the deal month end 3-month LIBOR rate. Interest Volatility is a 12-month moving
average of the standard deviation of the monthly yields on 10-year U.S. T-bonds at deal month end and term premium is the 12-month average
for the deal year of the yield differential between 10 years and 1 year U.S. bonds. Facility Ratio is defined as log of facility amount over the
deal amount. Assignment Size specifies the minimum assignment size for loan resale in a dollar amount stipulated in the loan contract. Agent
consent takes the value of 1 if the consent from lead banks is required for selling the loan or 0 otherwise while the borrower consent takes the
value of 1 if the consent from borrower is required for selling the loan or 0 otherwise. We identify 10 categories of industries. Tax is defined as
taxes paid scaled by TA in the year prior to the deal year. The heteroskedasticity corrected t statistics are reported in parentheses under the
coefficient estimates. ***, **, * indicate significance at 1%, 5%, and 10% levels, respectively.


              Variable                   All in Spread Drawn            Maturity Model            All in Spread Drawn            Maturity Model
                                                 Model                                                    Model
                                         Model (Facility Level)         (Facility Level)           Model (Deal Level)             (Deal Level)

              Intercept                          3.88                         1.67                         4.8                        -3.18
                                              (11.21)***                   (17.26)***                   (8.82)***                  (-14.48)***

  (Average)All-in-Drawn Spread                                                -0.06                                                   -0.03
                                                                           (-2.36)**                                                 (-1.19)
  Institutional Loan/Institutional               0.48                                                     0.35
            Participation
                                              (10.87)***                                                (7.15)***

                M/B                              0.02                         -0.02                       0.02                        -0.04
                                                (2.00)**                   (-2.49)**                     (1.54)                     (-3.9)***

              Leverage                           1.45                         -0.19                       1.55                        -0.24
                                              (17.70)***                   (-2.75)***                  (15.62)***                  (-3.16)***

          Altman Z Score                          -0.1                        -0.03                       -0.09                       -0.02
                                              (-13.46)***                  (-2.8)***                   (-10.01)***                   (-1.64)

        (Average) Maturity                       -0.16                                                    -0.17
                                               (-8.34)***                                              (-7.12)***

           Secured Status                        0.88                         0.39                         0.7                        0.21
                                              (30.92)***                   (11.02)***                  (22.17)***                   (6.12)***

        Loan Concentration                       -0.02                                                    0.07
                                                (-0.38)                                                  (1.30)

              Revolver                           -0.31                                                    -0.42
                                               (-9.69)***                                              (-12.12)***

     Facility Size (Loan Size)                   -0.07                                                    -0.15                       0.31
                                               (-2.46)**                                               (-3.81)***                  (24.28)***

           Facility Ratio                        -0.25                        -0.34
                                               (-3.9)***                  (-10.84)***

              LIBOR                              -0.14                                                    -0.11
                                              (-12.04)***                                              (-7.97)***

          Term Premium                           -0.14                        0.03                        -0.11                       0.07
                                               (-6.25)***                  (2.80)***                   (-4.01)***                   (6.39)***

         Interest Volatility                     0.34                         0.34                        0.11                        0.35
                                                (2.43)**                   (3.06)***                     (0.68)                     (2.94)***
          Firm Size              -0.07                 -0.06        0.00                 -0.25
                               (-2.48)**             (-8.13)***     (0.00)              (-23)***

        Agent Consent            0.19                               0.16
                              (3.01)***                           (2.30)**

      Borrower Consent           -0.11                              -0.11
                               (-2.33)**                          (-2.07)**

       Assignment Size           -0.04                              -0.02
                              (-10.84)***                         (-6.53)***

            LBO                  0.48                               0.41
                              (5.38)***                           (2.88)***

          Takeover               0.04                               0.03
                                (1.29)                              (0.74)

 Repayment/Recapitalization      -0.07                              -0.05
                               (-2.32)**                           (-1.45)

       Other Purposes            -0.25                              -0.22
                              (-6.51)***                          (-5.2)***

             Tax                                       0.21                              -0.64
                                                       (0.71)                           (-1.94)*
Financial Covenants               No                    No           Yes                  No

Number of Observations                      5064                                3366
System Weighted R Square                    0.4413                             0.4864
Table 10 Lock–up period model regression results (1998-2004). This table reports the lock up period regression model results. The loans
were traded during June 1998-December 2004. The dependent variable IS lock-up period which is defined as the time (measure in years)
between the first loan trading day and the loan’s origination day, Institutional loan dummy takes a value of 1 for institutional loans and 0 for
bank loans. We control for initial all-in-drawn spread as a proxy for the ex ante loan risks. The initial all-in-drawn spread is defined as the
percentage coupon spread over LIBOR plus the annual fee and the upfront fee paid by borrowers for each dollar drawn down. We include
maturity which is facility duration (in years), risk free interest rate (LIBOR), Interest rate risk (Interest volatility), term premium. LIBOR is the
deal month end 3-month LIBOR rate. Interest Volatility is a 12-month moving average of the standard deviation of the monthly yields on 10-
year U.S. T-bonds at deal month end and term premium is the 12-month average for the deal year of the yield differential between 10 years and
1 year U.S. bonds. The heterskedasticity corrected t-statistics are presented in parentheses. ***, **, * indicate significance at 1%, 5%, and 10%
levels, respectively.


 Variable                                                  Model 1                                                 Model 2

 Intercept                                                   1.68                                                    1.17

                                                          (9.00)***                                               (4.38)***

 Institutional Loan                                         -0.44                                                   -0.44

                                                         (-12.61)***                                             (-12.61)***

 All-in-Drawn Spread                                        -0.19                                                   -0.18

                                                         (-11.41)***                                             (-10.8)***

 Maturity                                                    0.21                                                    0.25

                                                          (4.81)***                                               (5.10)***

 LIBOR                                                      -0.18                                                   -0.13

                                                          (-7.24)***                                             (-3.11)***

 Term Premium                                               -0.26                                                   -0.42

                                                          (-6.19)***                                             (-4.38)***

 Interest Volatility                                         0.85                                                    1.39

                                                           (1.93)*                                                 (2.13)**

 Year Dummies                                                 No                                                     Yes

 Number of Obs.                                              2145                                                   2145

 Adj. R Square                                              0.1509                                                 0.1685
Table 11. Re-estimating initial loan resale return model under a simultaneous equation framework. This table reports the coefficient
estimates of the simultaneous regression models allowing the initial loan resale return and lock-up period to be mutually determined. The
results are reported for the raw initial loan resale regression and the market adjusted initial loan resale return model respectively. The dependent
variable is either the raw initial loan resale return (reported in the first column) or market adjusted initial loan resale return (reported in the third
column). The raw initial loan resale return is calculated as the first day trading price-100. The market adjusted initial loan resale return is
calculated as the initial loan resale return minus the S&P/LSTA leveraged loan market index return during the same period. Institutional loan
dummy takes a value of 1 for institutional loans and 0 for bank loans. We control for initial all-in-drawn spread as a proxy for the ex ante loan
risks. The initial all-in-drawn spread is defined as the percentage coupon spread over LIBOR plus the annual fee and the upfront fee paid by
borrowers for each dollar drawn down. We include lock-up period which is defined as the time (measure in days) between the first loan trading
day and the loan’s origination day, time to maturity which is calculated as the time (in days) between loan maturity day and the first loan
trading day, risk free interest rate (LIBOR), Interest rate risk (Interest volatility), term premium. LIBOR is the deal month end 3-month LIBOR
rate. Interest Volatility is a 12-month moving average of the standard deviation of the monthly yields on 10-year U.S. T-bonds at deal month
end and term premium is the 12-month average for the deal year of the yield differential between 10 years and 1 year U.S. bonds. In addition,
we also incorporate some other loan features including secured status takes the value of 1 if the loan is secured or 0 otherwise, credit revolver
dummy, number of lenders (log of number of lenders), facility size (log of facility amount), facility ratio (facility amount/ deal amount) and
loan purpose dummies (including leverage buyout loan, takeover loan, repayment/recapitalization loan and other purposes, the omitted case is
general corporate purpose loan). We use log of number of quotations on the first trading day to proxy for the market awareness of the loan and
control for the firm size (log of sales at close) as well. The dependent variable in lock-up period model is lock-up period which is defined as the
time (measure in years) between the first loan trading day and the loan’s origination day, Institutional loan dummy takes a value of 1 for
institutional loans and 0 for bank loans. We control for initial all-in-drawn spread as a proxy for the ex ante loan risks. The initial all-in-drawn
spread is defined as the percentage coupon spread over LIBOR plus the annual fee and the upfront fee paid by borrowers for each dollar drawn
down. We include maturity which is facility duration (in years), risk free interest rate (LIBOR), Interest rate risk (Interest volatility), term
premium. LIBOR is the deal month end 3-month LIBOR rate. Interest Volatility is a 12-month moving average of the standard deviation of the
monthly yields on 10-year U.S. T-bonds at deal month end and term premium is the 12-month average for the deal year of the yield differential
between 10 years and 1 year U.S. bonds. The hetroskedastic corrected t statistics are reported in parentheses under the coefficient estimates.
***, **, * indicate significance at 1%, 5%, and 10% levels, respectively.

               Variable                     Initial Loan Resale           Lock-up Period            Market Adj. Initial           Lock-up Period
                                              Return Model                   Model                 Resale Return Model               Model


               Intercept                            -5.03                       1.68                        -4.05                       1.68
                                                 (-2.73)***                   (8.57)***                   (-2.12)**                  (8.57)***
          Institutional Loan                         0.67                       -0.44                       0.61                        -0.44
                                                   (1.93)*                   (-10.9)***                    (1.69)*                   (-10.9)***
            Lock-up Period                            -2                                                    -0.71
                                                 (-10.75)***                                             (-3.69)***
                                                    -0.41                       -0.19                       -0.57                       -0.19
        All-in-Drawn Spread
                                                 (-3.03)***                 (-11.64)***                    (-4)***                  (-11.64)***
               Maturity                                                         0.22                                                    0.22
                                                                              (4.61)***                                              (4.61)***
           Time to Maturity                          0.55                                                   0.57
                                                  (5.68)***                                              (5.75)***
            Secured Status                           1.24                                                   1.69
                                                  (2.05)**                                               (2.70)***
               Firm Size                             0.28                                                   0.27
                                                  (2.61)***                                              (2.42)***
          Number of Quotes                           -1.6                                                   -1.65
                                                 (-4.47)***                                              (-4.42)***
               Revolver                             -1.99                                                   -2.03
                                                  (-5.5)***                                               (-5.4)***
                LIBOR                                0.43                       -0.18                       0.36                        -0.18
                                                  (2.60)***                  (-8.35)***                   (2.06)**                   (-8.35)***
           Term Premium                              1.16                       -0.26                        0.6                        -0.26
                                                  (3.75)***                  (-6.48)***                    (1.88)*                   (-6.48)***
          Interest Volatility                       -4.87                       0.85                        -4.72                       0.85
                                                   (-1.74)*                   (2.31)**                     (-1.63)                    (2.31)**
                  LBO                               -0.17                                                   -0.02
                                                   (-0.35)                                                 (-0.04)
               Takeover                             -0.19                                                   -0.09
                                                   (-0.43)                                                 (-0.18)
    Repayment/Recapitalization                      -0.48                                                   -0.28
                           (-1.07)                 (-0.6)
                           -0.87                   -0.79
     Other Purposes
                           (-1.5)                  (-1.32)
    Year Dummies            Yes               No    Yes               No
 Number of Observations              2145                    2145
System Weighted R Square             0.1773                  0.1343
Table 12 Loan pricing model incorporating institutional shareholding and interaction term of institutional shareholding and
institutional loan dummy (1995-2004). This table presents facility level loan pricing model results. The loan agreements were originated
during the period January 1995 - December 2004. The dependent variable is the initial all-in-spread drawn which is defined as the percentage
coupon spread over LIBOR plus the annual fee and the upfront fee paid by borrowers for each dollar drawn down. The Institutional loan
dummy takes a value of 1 for institutional loans and 0 for bank loans. We add the institutional shareholding which is defined as the percentage
of firm’s equity shares held by institutional owners in June of the year. Interaction term is calculated as institutional shareholdings times
institutional    loan     dummy.       We      include     M/B     (Market/Book),       Leverage      (Total    Debt/Assets),     Altman       Z-score
(3.3*EBIT/SALES+SALES/TA+1.4*RE/TA+1.2*WC/TA, where EBIT is the earning before interests and taxes, SALES is the total sales of
the year, TA is the total assets, WC is the working capital and RE is the retained earning), or alternatively Investment Grade (taking the value
of 1 if the S&P long term debt rating is BBB or above and 0 otherwise) and Not Rated dummy (taking the value of 1 if the S&P long term debt
credit rating is missing and 0 otherwise), which serve as proxies for two groups of borrower characteristics: borrower risk and information
costs. Firm characteristics are acquired at the fiscal year end prior to the deal year. We also include other loan contract terms such as loan
facility duration in years (maturity), and secured status taking the value of 1 for secured loan or 0 otherwise. Loan Concentration is defined as
ln(deal/(deal+TD)) where deal is the deal amount and TD is the total debt including long term debt and current liabilities. Revolver is the
dummy variable taking the value of 1 for credit revolver and 0 for term loan. We control for facility size (log of facility amount), facility ratio
(facility amount/ deal amount), Firm Size (log of firm’s total assets), Risk free interest rate (LIBOR), Interest rate risk (Interest volatility), term
premium, year dummy and industry dummy. LIBOR is the deal month end 3-month LIBOR rate. Interest Volatility is a 12-month moving
average of the standard deviation of the monthly yields on 10-year U.S. T-bonds at deal month end and term premium is the 12-month average
for the deal year of the yield differential between 10 years and 1 year U.S. bonds. Facility Ratio is defined as log of facility amount over the
deal amount. Assignment Size specifies the minimum assignment size for loan resale in a dollar amount stipulated in the loan contract. Agent
consent takes the value of 1 if the consent from lead banks is required for selling the loan or 0 otherwise while the borrower consent takes the
value of 1 if the consent from borrower is required for selling the loan or 0 otherwise. We identify 10 categories of industries. The
heterskedasticity corrected t-statistics are presented in parentheses. ***, **, * indicate significance at 1%, 5%, and 10% levels, respectively.

             Variable                               Model 1                              Model 2                               Model 3
             Intercept                                2.76                                 3.67                                  2.76
                                                   (6.86)***                             (9.37)***                            (4.63)***
        Institutional Loan                            0.54                                  0.53                                 0.51
                                                   (4.09)***                             (4.14)***                            (3.93)***
   Institutional Shareholdings                       -0.26                                 -0.16                                 -0.24
                                                   (-4.84)***                           (-3.07)***                            (-4.45)***
         Interaction Term                            -0.13                                 -0.11                                 -0.1
                                                    (-0.62)                               (-0.54)                               (-0.51)
                M/B                                   0.02                                  0.04                                 0.01
                                                     (1.36)                              (3.75)***                              (1.13)
             Leverage                                 1.28                                  1.41                                  1.3
                                                  (12.43)***                            (14.06)***                            (12.66)***
         Altman Z Score                                                                    -0.16
                                                                                        (-13.08)***
        Investment Grade                             -0.39                                                                       -0.38
                                                  (-10.53)***                                                                 (10.49)***
             Not Rated                               -0.07                                                                       -0.05
                                                   (-2.15)**                                                                   (-1.67)*
             Maturity                                -0.08                                 -0.08                                 -0.05
                                                   (-3.8)***                            (-3.55)***                             (-2.29)**
          Secured Status                              0.82                                  0.84                                 0.77
                                                  (28.50)***                            (29.82)***                            (27.33)***
       Loan Concentration                            -0.15                                 -0.04                                 -0.14
                                                   (-3.17)***                             (-0.97)                             (-2.91)***
             Revolver                                -0.31                                 -0.31                                 -0.29
                                                   (-7.35)***                           (-7.45)***                            (-7.11)***
            Facility Size                            -0.002                                -0.04                                 -0.01
                                                    (-0.08)                               (-1.46)                               (-0.23)
           Facility Ratio                            -0.37                                 -0.25                                 -0.37
                                                   (-5.47)***                           (-3.76)***                             (-5.4)***
              LIBOR                                  -0.13                                 -0.14                                 -0.06
                                                  (-11.58)***                           (-12.08)***                            (-2.1)**
          Term Premium                               -0.14                                 -0.15                                 -0.17
                                                   (-5.61)***                           (-6.26)***                            (-2.81)***
        Interest Volatility                           0.28                                  0.3                                  1.78
                                                    (1.91)*                              (2.12)**                             (5.33)***
             Firm Size                               -0.09                                 -0.09                                 -0.11
                                                   (-2.85)***                           (-2.91)***                            (-3.47)***
      Agent Consent             0.24         0.23         0.11
                             (4.16)***    (3.96)***     (2.00)**
    Borrower Consent           -0.14         -0.14        -0.19
                             (-2.25)**     (-2.41)**    (-3.25)***
     Assignment Size           -0.03         -0.03        -0.02
                             (-9.9)***    (-10.15)***   (-6.98)***
          LBO                  0.44          0.41         0.46
                             (5.59)***    (5.11)***     (6.13)***
        Takeover               0.06          0.03         0.09
                              (1.78)*       (0.92)      (2.53)**
Repayment/Recapitalization     -0.08         -0.09        -0.04
                             (-2.33)**    (-2.55)***     (-1.16)
     Other Purposes             -0.2         -0.29        -0.23
                             (-5.41)***   (-8.04)***    (-6.43)***
    Industry Dummies            No            No           Yes
      Year Dummies              No            No           Yes
   No. of Observations         4558          4558         4558
    Adjusted R Square         0.5911        0.6014       0.6115

								
To top
;