Loan term interdependences and imformation asymmertries

Loan term interdependences and information asymmetries Phu Quoc Pham1 (The Department of Accounting and Finance, Monash University) Michael Skully (The Department of Accounting and Finance, Monash University) Shrimal Perera (The Department of Accounting and Finance, Monash University) (Preliminary draft: Please not quote, comments are welcome) Abstract This study examines whether loan terms are jointly determined and if information asymmetries’ effects on them across revolving and term loans A simultaneous equation model is applied for revolving and term loans made by US commercial banks to US corporate borrowers from 1987 to 2009 The findings suggest that loan terms are jointly determined and that information asymmetries’ effects on them differently across loan types The research contributes literature in several ways It first provides evidence the determinants of largest set of loan terms in a simultaneous context therefore may better reflect reality This is also the first examines the effect of information asymmetries on loan terms, especially on covenants and loan size, in such a context Lastly, this is the first study that distinguishes revolving and term loans to inspect these issues Key words: Loan terms, information asymmetry, revolving loans, term loans JEL classification: G2  PhD Student Email: PhuQuoc.Pham@monash.edu  1 Introduction This paper studies the interaction of loan terms and how their selection addresses information asymmetries in revolving and term loans The interdependencies between debt features are theoretically recognized in many studies Banks and borrowers are suggested to trade-off contract terms as a “package” for an optimal contract (Melnik and Plaut, 1986) This idea was first proposed by Dennis, Nandy, and Sharpe (2000) Their research argues against focusing on a single contract term if these terms are jointly designed Others subsequently also supported the idea of loan term interdependency (e.g, Brick and Palia, 2007; Chava, Livdan, and Purnanandam, 2009; Wittenberg-Moerman, 2009; Bharath, Dahiya, Saunders, and Srinivasan, 2011) These studies, however, only allow for the jointness between some loan terms and are inconsistent Their conflicting results might be explained by their exclusion of different loan terms from their models (Steijvers and Voordeckers, 2009) Moreover, no study examines whether all five key loan terms (i.e., price, collateral, maturity, covenants, and size) are simultaneously determined Hence, the interdependences between loan price, collateral and maturity themselves when interacting with covenants and loan size remain unknown Therefore, an examination of the interdependences between all five key loan terms might be necessary This paper is motivated to address this gap by examining these interdependences and also adds to the literature by investigating whether the effects of information asymmetries on loan terms differ as these terms are simultaneously designed The simultaneous equation studies on the independencies between loan terms not focus on how information asymmetries affect them whereas others that investigate the effect of information asymmetries on loan terms not consider their jointness (Dennis, et al., 2000) These issues might differ depending different types of loans Although many studies omit the differences in revolving and term loans (Strahan, 1999; Coleman, Esho, and Sharpe, 2002; Gottesman and Roberts, 2004, 2007), none considers these issues across them Using a sample of 12,207 revolving and 3,972 term loans made by US commercial banks to US borrowers (excluding banks and non-bank financial institutions) from January 1987 to 31 December 2009, this study examines these three gaps in the literature It further supports the argument that loan terms are jointly determined and provides evidence  For example, Merton (1974), Myers (1977), Smith and Warner (1979), Bester (1985), Flannery (1986), Chan  and Thakor (1987), Milde and Riley (1988), Boot, Thakor, and Udell (1991), and Diamond (1993).  on how information asymmetries affect loan terms in the context of such interdependent relationships This study contributes to bank lending literature in several vital ways This is the first studies whether there is the joint determinant of all five key loan terms which is believed to bring the findings closer to the reality The study also is the first that officially examines information asymmetries’ effects on loan covenants and size It also conducts robust tests on how information asymmetries impact loan price, collateral and maturity in the context of lager set of loan terms’ interaction This research also the first examines these above issues across both revolving and term loans The rest of this paper is structured as follows Section reviews the literature on loan term determinants and information asymmetry effects to hypothesize the research Data and empirical method are described in Section Empirical results and conclusion are presented in Section and respectively Literature review and hypotheses’ development 2.1 The jointness of loan terms Many studies acknowledge the joint determinants of some loan terms For instance, banks screen their borrower quality by designing interest rates and collateral simultaneously (Bester, 1985) Lenders may increase interest rate and decrease maturity to address risk due to adverse selection and moral hazard (Wittenberg-Moerman, 2009) The interrelationship between collateral and maturity may be also substituted (Leeth and Scott, 1989; Degryse and van Cayseele, 2000; Voordeckers and Steijvers, 2006), or complementary (Chan and Kanatas, 1985; Besanko and Thakor, 1987b, 1987a) Alternatively, maturity can be substituted by covenants as long-term loans are riskier than short-term ones then they require more covenants (S C Myers, 1977) Lenders may employ an even greater combination of loan terms to address information asymmetries For instance, non-price terms might be used for information asymmetries and use price terms for the remaining risk (Strahan, 1999) Similarly, maturity, covenants, collateral and price might be simultaneously designed to obtain an optimal solution in presence of information asymmetries and credit risk (Steijvers and Voordeckers, 2009) Where collateral reduce their risk exposure, lenders can then extend the loan’s maturity or lower the interest rate (Ortiz-Molina and Penas, 2008) These ideas are consistent with the Melnik and Plaut (1986) view that borrowers tradeoff between different loan terms offered by lenders to achieve their optional contracts Therefore, it is likely that these loan terms (i.e., price, collateral, maturity, covenants and size) are jointly determined Thus, based on the literature, the first hypothesis is as follow: H.1: The relationships between loan contract terms (i.e., price, collateral, maturity, covenants, and size) are both bidirectional and unidirectional The signs of the various loan term coefficients might be explained by different theories The signalling theory suggests that if good borrowers will pledge collateral to signal their good credit ability and thus receive a lower price.3 In contrast, the risk observed theory claims that high-risk borrowers are more likely to pledge collateral and still pay a high price because their collateral might offset not enough While most empirical studies support signalling theory,4 some also find support for the risk observed theory (Berger and Udell, 1990; John, Lynch, and Puri, 2003) The two alternative theories on the relationship between loan price and maturity are the trade-off theory and the credit quality theory (Gottesman and Roberts, 2004) The trade-off theory presumes a positive relationship between price and maturity This comes from both the lender and borrower objectives For lenders, to avoid an agency problem with borrowers, they tend to make shorter loans For borrowers, short-term loans present liquidity problems when the loan is due so they prefer longer-term loans Lenders will issue long-term loans but only at a higher price Some empirical studies such as Coleman, Esho, and Sharpe (2002) and Bharath et al., (2011) support for this theory In contrast, the credit quality theory suggests a positive relationship between price and maturity On one hand, because the borrowers’ risk increases with the longer-term loans, lenders charge higher prices to reduce their risk exposure On the other hand, low risk borrowers may seek shorter loans to signal their high quality and so receive lower prices (e.g., Flannery (1986) and Kale and Noe (1990)) Some empirical studies also support for the latter theory (e.g., Strahan (1999), Dennis et al., (2000), and Wittenberg-Moerman ) The relationship between loan price and covenants can be explained by the signalling theory of covenants and the agency theory of covenants (Demiroglu and James, 2010) The For instance, Bester (1985), Chan and Kanatas (1985), Chan, Greenbaum, and Thakor (1986), and Besanko  and Thakor (1987b, 1987a).  For example, Berger and Udell (1995), Strahan (1999), Dennis et al., (2000), and Pozzolo (2002).  signalling theory claims that borrowers have knowledge about their future performances and will try not to break the lenders’ covenants Therefore, their acceptance of those covenants may reduce the loan price In contrast, agency theory predicts that because of agency cost problems related to the borrowers’ poor past performances, both higher prices and covenants might be used Therefore, the relationship between price and covenants is positive Demiroglu and James (2010) support the former view whereas Wahrenburg and Steffen (2008) hold for the latter The answers are also unclear on the relationship between the price and size of a loan Traditional finance theory claims that the larger the loan size then the greater the lender risk exposure Therefore, lenders may charge more for larger loans to compensate for their higher risk (Moore and Craigwell, 2003) In addition, borrowers can signal their high quality by accepting a smaller loan for a lower price (Milde and Riley, 1988).5 In contrast, the negative relationship between loan price and loan size may also result as large loans are usually granted to larger and well-organized firms with less risk and less information asymmetry Conversely, small loans are granted to small firms with more risk and high asymmetric information problems Therefore, lenders require high prices on small loans to recover their risk and managerial costs The lower economics of scale in making small loans may also cause lenders to charge higher prices Finally, as small firms have fewer sources of financing, they may accept these loan higher prices given their lack of other alternatives Most empirical studies find that loan size and price have a negative relationship.6 The relationship between collateral and maturity is ambiguous The positive relationship between collateral and maturity might have two explanations (Leeth and Scott, 1989) Firstly, short-term loans have an advantage over long-term loans as they prevent borrower asset substitution, so loans with longer terms are more likely to require collateral Secondly, the longer the loan maturity, the riskier the loan is and so long-term loans are more likely to need collateral In contrast, the collateral value may decline over the loan period and so be of less use with longer loans, therefore, short term loans are more likely to be secured (Stulz and Johnson, 1985) Empirical evidence is inconclusive in regards to this relationship:  In contrast, Cressy and Toivanen (2001) show that the better the borrowers the larger loans and the lower  rates they receive.  For example, Berger and Udell (1990), Booth (1992), Beatty, Ramesh, and Weber (2002), and Wittenberg‐ Moerman  (2009).  Some studies report a positive relationships7 and others find negative ones (e.g., Scott and Smith (1986) and Boot et al., (1991)) Collateral and covenants offer no theory on their relationship Most studies, however, find that collateral and covenants have an inverse relationship.8 The negative relationship between collateral and loan size could be explained as larger loans are often lent to large firms with lower risk exposure (Boot, et al., 1991) The signalling theory, however, argues that good borrowers signal their high quality by pledging collateral to borrow larger loans (Schwartz, 1989) Moreover, the increase in loan size may affect firm’s leverage (Leeth and Scott, 1989; Jaffee and Stiglitz, 1990) which may raise its possibility of default Hence, larger loans are more likely to require collateral Empirical studies, however, find mixed results on the relationship between collateral and loan size: positive relationship (Leeth and Scott (1989)) and negative one (Scott and Smith (1986)) As agency cost increases with debt maturity, longer term loans are more likely to have covenants (Kare, 1996) This is because (financial) covenants help reduce asymmetric information problems between lenders and borrowers and hence loans with (financial) covenants can be for longer term (Wittenberg-Moerman, 2009) Some empirical studies also find a positive relationship between maturity and covenants (S C Myers, 1977; Berger and Udell, 2003; Billett, King, and Mauer, 2007) The maturity and loan size relationship is also mixed Some argue that larger loans are safer than smaller ones and, therefore, can have longer term (Lee, 2004) This argument is consistent with Boot et al., (1991) Others claim that maturity and loan size are complimentary in solving the information asymmetry and credit risk problems so the relationship is positive (e.g., Strahan (1999), and Cressy and Toivanen (2001)) Since large loans are usually made to large and well established firms, they may be less likely to have covenants (Apitado and Millington, 1992) But this is inconsistent with Citron et al., (1997) Therefore, the relationship between covenant and loan size has not been conclusively confirmed  For  example,  Harhoff  and  Körting  (1998),  Degryse  and  van  Cayseele  (2000),  Dennis  et  al.,  (2000),  and  Voordeckers and Steijvers (2006).   For  example,  Schwartz  (1989),  Citron  (1992),  Citron,  Robbie,  and  Wright  ((1997),  and  Niskanen  and  Niskanen (2004).  2.2 Information asymmetries’ effects on loan terms Many previous studies suppose that lenders charge higher prices and restrict non-price terms for high information asymmetries borrowers (1981; 1991; Easley, Hvidkjaer, and O'Hara, 2002; Easley and O'Hara, 2004; Ackert, Huang, and Ramírez, 2007) These studies, nevertheless, not address the jointness between different loan terms Lenders not only use a specific loan term to address information asymmetries but also use different ones to so (Steijvers and Voordeckers, 2009) This, therefore, may produce the trade- off between loan terms in dealing with information asymmetries to achieve the optimal loan contract (Melnik and Plaut, 1986) Hence, the second hypothesis is as follows: H2: There is the trade-off between loan terms in addressing borrower information asymmetries Data and empirical method 3.1 Data and sample procedure The data for this research comes from four databases The loan and lender variables are extracted from the Loan Pricing Corporation’s (LPC) DealScan Borrower variables are mainly from Compustat and some proxies for information asymmetries such as forecast dispersion, forecast error and the number of analysts that follow the borrower are from Thomson I/B/E/S Finally, the macroeconomic factors are obtained from Datastream (Thomson Reuters) The sample selection process is summarised in Appendix It is worth noting, however, an inconsistency between the year of DealScan loans and those of Compustat borrowing firms The borrower’s financial information therefore cannot be obtained by simply matching its loan year in DealScan with its previous Compustat year This is because their fiscal years ending might differ with the calendar ones Compustat assigns to the previous calendar years for any fiscal years ending between January and May and current calendar for any fiscal years ending between June and December (Standard & Poor’s, 2011, p 36) For example, Global Imaging Systems’ (GISX) fiscal years ending on 31/03/2005 and a loan on 11/03/2005 Because the date 31/03 is between January and May, Compustat assigns that fiscal year as 2004 In this case, the lender(s) of the loan on 11/03/2005 could not use the Compustat borrower financial information of the fiscal year 2004 This research avoids this problem by following Bharath, et al (2011) method Hence any DealScan facility in a calendar year t with its active date at least six months after its firm’s Compustat fiscal year ending month in calendar year t, the Compustat financial information for that fiscal year is used If a DealScan facility active date is less than six months after the fiscal year ending month, the information of the firm’s Compustat fiscal year ending in calendar year t−1 is used (Bharath, et al., 2011) 3.2 Empirical model In this research, the system of normalisation is chosen based on prior studies and the practice in the banking industry Theoretically, lenders should design all loan terms as a package (Plaut, 1985; Melnik and Plaut, 1986) In reality, lenders may negotiate n-1 loan terms to form a loan contract (Dennis, et al., 2000; Bharath, et al., 2011) Loan price is determined by non-price terms and among other factors and non-price terms (i.e., collateral, maturity, covenants, and loan price) are simultaneously determined Loan price, collateral, maturity and covenants are normally considered as endogenous variables While loan size is often considered as an exogenous variable,9 others argue that it is an endogenous variable (e.g., Ham and Melnik, 1987; Milde and Riley, 1988; Bae and Goyal, 2009; Santos, 2011) Accordingly, this research considers that all these key loan variables are endogenous Loan price is first determined by other loan terms Then each loan non-price term is determined by the other terms As previous studies suggest, these terms are also affected by information asymmetries, prior banking relationships, borrower characteristics, lender characteristics, other loan features and macroeconomics factors (see: Dennis, et al., 2000; Coleman, et al., 2002; Berger, Scott Frame, and Ioannidou, 2011; Bharath, et al., 2011) The model is stated as follows: Index Index Index Index  For example, Murphy (1984), Leeth and Scott (1989), Berger and Udell (1990), Boot et al., (1991), Harhoff  and  Körting  (1998),  Elsas  and  Krahne  (2000)  and  Gonas  et  al.,  (2004)  (Booth  and  Booth,  2006)  and  Ross  (2010).  Index Where: X : (Constant, Profitability, Leverage, Market to Book, Log(1+ Coverage), Top Ten Lenders, Syndicated, Tranche, Loan Purposes (Repayment-Recap, Acquisitions and Other Purposes), LIBOR, Interest Volatility, Term Premium, GDP) X : (Constant, Profitability, Leverage, Market to Book, Tangibility, Loan Concentration, Top Ten Lenders, Loan Purposes, GDP) X : (Constant, Profitability, Leverage, Market to Book, Log(Asset Maturity), Earning Variance, Tax Asset, Utilities, Top Ten Lenders, Loan Purposes, GDP) X : (Constant, Profitability, Leverage, Market to Book, Cash Flow Volatility, Top Ten Lenders, Loan Purposes, GDP) X : (Constant, Firm Size, Profitability, Leverage, Market to Book, Current Ratio, Top Ten Lenders, Syndicated, Loan Purposes, GDP) The model, therefore, contains five simultaneous equations of loan terms in which loan price has unidirectional relationship with non-price terms and non-price terms have bidirectional relationships each other While AISD, Maturity and Log(Loan Size) are continuous variables, Collateral is a discrete variable and Covenant Index is a count variable All equations are over-identified Loan price is represented by All-In-Spread-Drawn (AISD) which is the sum of coupon spread paid over London Inter-Bank Offered Rate (LIBOR) and annual fee for each dollar drawn For loans without LIBOR spread quoted, DealScan utilises the minimum spread and then adding or subtracting a LIBOR differential corresponding to the spread used (Loan Pricing Corporation, 1994, p.19, as cited in Daniels and Ramírez, 2008).10 The presence of collateral (Collateral) is measured by a dummy variable which is unity if the loan is secured, and zero otherwise Loan maturity (Maturity) is the facility length (years) Loan covenants is measured by a covenant index (Covenant Index) which is constructed 10 See Hubbard et al. (2002, p. 562, footnote 7 ), for the differential applied for each kind of spreads.  following Bradley and Roberts (2004), Vu (2008), and Demiroglu and James (2010).11 Loan size is represented by Log(Loan Size) which is the natural logarithm of the facility size Other loan characteristics are measured by the variables such as Syndicated, Tranche, loan purposes (Repayment-Recap, Acquisitions and Other Purposes), and Numbers of Lenders Syndicated is assigned one if loan distribution method is reported “Syndicated” and zero otherwise Tranche is coded one, if there is more than one facility in this deal but otherwise zero If loan specific purpose is reported “DebtRepayment” or “Recapitalization then” then Repayment-Recap is one, and zero otherwise If loan specific purpose is reported “LBO_MBO” or “AcquisLine” then Acquisitions is one; otherwise zero If loan specific purpose is not DebtRepayment or Recapitalization or LBO_MBO or CorpPurposes or “WorkingCapital” then Others are one; otherwise zero (Dennis, et al., 2000) Number of Lenders is the total lenders involve in the loan The lender characteristic (Top Ten Lenders) is constructed from DealScan by yearly ranking market shares This ranking on a yearly basis avoids the name change problems in mergers or acquisitions Banking relationship (BRDummy) is assigned one if the current lead lender(s) also lead any previous loans within last five years This research regards lead arrangers as lead lenders For any loans without information on lead arrangers, then the lead role lenders are treated as lead lenders (Sufi, 2007) Information asymmetries (IA ) measured by five proxies namely R&D to Assets, Log((Firm Age)), Forecast Dispersion, Forecast Error, and number of analyst following the borrower (Analysts) R&D to Assets is research and development expenses scaled by sales Log(Firm Age) is natural logarithm of the duration in years from the first year the company has the name on the Compustat database to the fiscal year (Sufi, 2009) Forecast Dispersion is the ratio of EPS forecast standard deviation to its absolute mean forecast (Atiase and Bamber, 1994) Forecast Error is calculated following Drobetz, Grüninger, and Hirschvogl (2010) which is the natural logarithm of one plus absolute value of (actual and forecasted EPS scaled by it median) To aggregate the full effects of all proxies an information asymmetry index is constructed following (Lin, Pantzalis, and Park, 2009; Drobetz, et al., 11  See Appendix 2.  10 Firm Size Market to Book Profitability Leverage Log(1+ Coverage) Tangibility Loan Concentration Log(Asset Maturity) Cash Flow Volatility Earning Variance Tax_Asset Current Ratio Number of Lenders LIBOR Interest Volatility Term Premium GDP Growth [14] [15] [16] [17] [18] [19] [20] [14] ‐0.06 0.22 0.17 ‐0.09 0.17 ‐0.58 [15] [16] [17] [18] [19] [20] 0.13 ‐0.26 0.48 ‐0.14 0.16 0.18 0.13 0.41 0.02 ‐0.61 0.26 ‐0.33 ‐0.14 0.33 ‐0.08 [21] [22] [23] [24] [25] [26] [27] [28] [29] [30] 0.32 ‐0.41 ‐0.40 ‐0.01 ‐0.37 0.55 ‐0.13 0.03 0.03 ‐0.09 ‐0.13 0.21 0.14 0.50 0.13 ‐0.02 0.06 ‐0.07 ‐0.11 0.09 0.39 ‐0.11 ‐0.08 0.20 ‐0.16 0.14 0.02 ‐0.03 ‐0.05 0.02 0.32 ‐0.16 ‐0.15 ‐0.29 ‐0.23 0.12 ‐0.03 ‐0.01 0.02 0.01 ‐0.21 0.10 0.06 0.55 0.28 ‐0.06 ‐0.01 ‐0.05 ‐0.08 0.00 0.83 ‐0.08 ‐0.03 ‐0.11 ‐0.37 0.07 0.00 0.01 0.02 0.03 ‐0.18 0.24 0.23 0.18 0.37 ‐0.14 0.08 ‐0.11 ‐0.11 0.10 [29] [30] [27] [28] [29] [30] [27] ‐0.25 ‐0.72 0.19 [28] LIBOR Interest Volatility Term Premium GDP Growth 0.61 ‐0.56 ‐0.39 26 [21] [22] [23] [24] ‐0.16 ‐0.16 ‐0.13 ‐0.30 0.15 ‐0.02 0.00 0.01 0.02 0.79 ‐0.01 0.16 ‐0.21 0.03 0.03 0.03 0.00 ‐0.06 0.16 ‐0.21 0.02 0.04 0.04 0.01 0.12 0.02 0.09 ‐0.05 ‐0.09 0.04 (Table 2 continued) [25] [26] ‐0.23 0.04 ‐0.03 ‐0.03 0.02 ‐0.11 0.01 0.04 ‐0.04 Table 3: Correlation matrix for term loans Bold texts indicate statistically significant at 1% level or better [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13] AISD [1] Collateral [2] 0.26 Maturity [3] ‐0.07 0.16 Log(Loan Size) [4] ‐0.20 ‐0.08 0.25 Covenant Index [5] 0.06 0.43 0.20 0.29 R&D to Assets [6] 0.07 0.08 ‐0.12 ‐0.28 ‐0.13 Log(Firm Age) [7] ‐0.08 ‐0.09 ‐0.04 0.22 0.03 ‐0.12 Forecast Dispersion [8] 0.13 0.03 0.03 0.03 0.00 ‐0.02 0.00 Forecast Error [9] 0.22 0.05 ‐0.05 ‐0.06 ‐0.03 0.01 ‐0.05 0.56 Analysts [10] ‐0.22 ‐0.16 0.04 0.46 0.00 ‐0.01 0.24 ‐0.02 ‐0.12 IA Index [11] 0.21 0.11 0.02 ‐0.24 ‐0.09 0.35 ‐0.46 0.35 0.54 ‐0.19 IA Index_1 [12] 0.25 0.15 ‐0.01 ‐0.38 ‐0.07 0.31 ‐0.49 0.30 0.50 ‐0.50 0.93 BR Dummy [13] ‐0.13 ‐0.09 ‐0.01 0.19 0.03 ‐0.05 0.09 0.05 0.00 0.12 ‐0.04 ‐0.08 Firm Size [14] ‐0.18 ‐0.19 0.06 0.78 0.10 ‐0.28 0.35 0.04 ‐0.03 0.55 ‐0.25 ‐0.42 0.21 Market to Book [15] ‐0.13 0.00 0.04 ‐0.01 0.02 0.24 ‐0.15 ‐0.07 ‐0.10 0.19 0.02 ‐0.04 ‐0.04 Profitability [16] ‐0.18 ‐0.08 0.14 0.27 0.07 ‐0.38 0.06 ‐0.02 ‐0.08 0.21 ‐0.16 ‐0.21 0.06 Leverage [17] 0.00 0.01 0.11 0.31 0.08 ‐0.27 ‐0.08 0.10 0.07 ‐0.04 0.04 0.05 0.11 Log(1+ Coverage) [18] ‐0.26 ‐0.08 0.02 ‐0.08 ‐0.02 0.21 0.02 ‐0.18 ‐0.24 0.09 ‐0.18 ‐0.17 ‐0.06 Tangibility [19] 0.02 ‐0.04 ‐0.01 0.11 ‐0.05 ‐0.20 0.05 0.09 0.08 0.14 0.00 ‐0.05 0.02 1 Loan Concentration [20] ‐0.07 0.09 0.24 0.25 0.20 0.04 ‐0.16 ‐0.04 ‐0.07 ‐0.07 0.03 0.07 ‐0.11 Log(Asset Maturity) [21] ‐0.01 ‐0.06 0.01 0.22 ‐0.04 ‐0.15 0.05 0.08 0.06 0.18 ‐0.01 ‐0.07 0.04 Cash Flow Volatility [22] 0.16 0.13 ‐0.11 ‐0.32 ‐0.07 0.38 ‐0.14 0.00 0.07 ‐0.12 0.25 0.26 ‐0.09 Earning Variance [23] 0.16 0.11 ‐0.13 ‐0.36 ‐0.11 0.38 ‐0.16 0.03 0.11 ‐0.16 0.27 0.29 ‐0.06 Tax_Asset [24] ‐0.17 ‐0.04 0.09 0.03 0.03 ‐0.01 ‐0.03 ‐0.17 ‐0.25 0.10 ‐0.27 ‐0.26 ‐0.03 Current Ratio [25] ‐0.05 0.03 0.04 ‐0.18 ‐0.03 0.18 ‐0.11 ‐0.05 ‐0.05 ‐0.10 0.12 0.15 ‐0.10 Number of Lenders [26] ‐0.17 0.02 0.14 0.47 0.28 ‐0.13 0.10 ‐0.01 ‐0.07 0.24 ‐0.14 ‐0.21 0.17 LIBOR [27] ‐0.13 0.00 0.07 ‐0.15 ‐0.16 0.04 ‐0.14 ‐0.05 0.01 ‐0.04 0.11 0.11 ‐0.02 Interest Volatility [28] 0.20 ‐0.06 ‐0.16 ‐0.07 ‐0.01 0.03 0.05 0.01 0.03 ‐0.03 ‐0.02 0.00 ‐0.05 Term Premium [29] 0.21 ‐0.04 ‐0.15 ‐0.07 ‐0.01 0.00 0.06 0.03 0.02 ‐0.04 ‐0.03 0.00 ‐0.02 GDP Growth [30] ‐0.19 0.08 0.14 ‐0.09 0.07 0.01 ‐0.16 ‐0.02 ‐0.02 ‐0.06 0.02 0.03 ‐0.04 27 (Table 3 continued) [14] [15] [16] [17] [18] [19] [20] [21] [22] Firm Size [14] Market to Book [15] ‐0.15 Profitability [16] 0.24 0.05 Leverage [17] 0.28 ‐0.11 0.25 Log(1+ Coverage) [18] ‐0.20 0.38 0.16 ‐0.53 Tangibility [19] 0.17 ‐0.13 0.17 0.17 ‐0.11 Loan Concentration Log(Asset Maturity) [20] ‐0.28 0.27 0.05 ‐0.25 0.42 ‐0.13 [21] 0.31 ‐0.12 0.12 0.20 ‐0.19 0.78 ‐0.11 Cash Flow Volatility [22] ‐0.37 0.26 ‐0.27 ‐0.19 0.15 ‐0.10 0.11 ‐0.12 [23] [24] [25] [26] Earning Variance [23] ‐0.40 0.18 ‐0.23 ‐0.21 0.14 ‐0.08 0.08 ‐0.16 0.75 Tax_Asset [24] ‐0.11 0.39 0.19 ‐0.19 0.54 ‐0.11 0.27 ‐0.16 0.00 ‐0.02 Current Ratio [25] ‐0.27 0.13 ‐0.16 ‐0.23 0.33 ‐0.25 0.34 ‐0.10 0.12 0.13 0.11 Number of Lenders [26] 0.42 ‐0.02 0.12 0.15 ‐0.06 0.02 0.02 0.11 ‐0.16 ‐0.20 0.02 ‐0.10 LIBOR [27] ‐0.20 0.08 ‐0.01 ‐0.07 0.06 0.01 0.08 ‐0.04 0.06 0.07 0.10 0.07 ‐0.06 Interest Volatility [28] 0.01 ‐0.07 ‐0.05 ‐0.04 ‐0.04 0.03 ‐0.11 0.01 0.03 0.03 ‐0.03 ‐0.04 ‐0.04 Term Premium [29] 0.02 ‐0.14 ‐0.04 0.00 ‐0.11 0.04 ‐0.15 0.03 0.01 0.02 ‐0.09 ‐0.04 ‐0.01 GDP Growth [30] ‐0.16 0.05 ‐0.03 0.02 0.02 ‐0.01 0.09 0.00 0.05 0.06 0.04 0.08 0.06 [27] [28] [29] LIBOR [27] Interest Volatility [28] ‐0.23 Term Premium [29] ‐0.71 0.61 GDP Growth [30] 0.17 ‐0.55 ‐0.37 [30] 1 28 Table 4: The regression results for five loan term equations for revolving loans Panel A, B, C, D, and E are results for AISD, Collateral, Maturity, Covenant Index, and Log(Loan Size) equations respectively The information asymmetry proxies are R&D to Sales, Log(Firm Age), Forecast Dispersion Forecast Error, IA Index and IA Index_1in Column (2), (3), (4), (5), (6), (7) and (8)) respectively Standard errors in parentheses and ***, **, and * indicate significantly different than zero at the 1%, 5%, and 10% levels, respectively Numbers in parentheses are standard errors corrected for heteroscedasticity Panel A: Regression results for AISD equation (1) (2) (3) (4) (5) (6) (7) (8) Fitted Collateral 2.353*** 2.279*** 2.313*** 2.087*** 2.294*** 2.029*** 2.000*** (0.095) (0.100) (0.105) (0.101) (0.095) (0.109) (0.110) ‐ Fitted Maturity ‐0.200*** ‐0.193*** ‐0.199*** ‐0.176*** ‐0.216*** 0.173*** ‐0.193*** (0.030) (0.028) (0.029) (0.030) (0.031) (0.029) (0.029) Fitted Covenant Index ‐0.024 ‐0.035* ‐0.045** ‐0.016 ‐0.028 ‐0.030 ‐0.025 (0.020) (0.020) (0.021) (0.020) (0.020) (0.021) (0.021) ‐ Fitted Log(Loan Size) ‐0.050*** ‐0.056*** ‐0.040*** ‐0.066*** ‐0.043*** 0.056*** ‐0.041*** (0.014) (0.013) (0.014) (0.014) (0.016) (0.013) (0.013) R&D to Assets 1.096*** (0.298) Log(Firm Age) ‐0.024* (0.014) Forecast Dispersion 0.205*** (0.064) Forecast Error 0.222*** (0.042) Analysts ‐0.001 (0.002) IA Index 0.423*** (0.075) IA Index_1 0.536*** (0.096) BR Dummy ‐0.030 ‐0.031* ‐0.050*** ‐0.034* ‐0.039** ‐0.046** ‐0.049*** (0.018) (0.018) (0.019) (0.019) (0.019) (0.019) (0.019) ‐ Profitability ‐0.193*** ‐0.213*** ‐0.195*** ‐0.128* ‐0.145** 0.199*** ‐0.168*** (0.065) (0.065) (0.066) (0.066) (0.068) (0.065) (0.065) Leverage 0.679*** 0.629*** 0.796*** 0.722*** 0.734*** 0.819*** 0.828*** (0.084) (0.083) (0.089) (0.086) (0.088) (0.088) (0.088) ‐ Market to Book ‐0.074*** ‐0.066*** ‐0.075*** ‐0.076*** ‐0.079*** 0.075*** ‐0.071*** (0.010) (0.010) (0.010) (0.010) (0.011) (0.010) (0.010) Log(1 + Coverage) ‐0.034** ‐0.042*** 0.009 ‐0.006 ‐0.004 ‐0.005 ‐0.003 (0.016) (0.016) (0.016) (0.016) (0.016) (0.017) (0.016) Top Ten Lenders 0.086*** 0.078*** 0.062*** 0.047** 0.058*** 0.047** 0.046** (0.021) (0.020) (0.021) (0.021) (0.021) (0.021) (0.021) ‐ Syndicated ‐0.133** ‐0.132** ‐0.168** ‐0.103 ‐0.098 0.179*** ‐0.183*** (0.054) (0.054) (0.068) (0.063) (0.062) (0.069) (0.069) Tranche 0.079*** 0.091*** 0.109*** 0.096*** 0.111*** 0.107*** 0.120*** (0.021) (0.021) (0.021) (0.020) (0.023) (0.020) (0.021) Repayment ‐ Recap ‐0.059** ‐0.044 ‐0.008 ‐0.029 ‐0.032 0.008 0.011 (0.028) (0.028) (0.029) (0.028) (0.029) (0.029) (0.029) Acquisitions 0.059 0.081* 0.048 0.060 0.054 0.089* 0.088* (0.047) (0.047) (0.051) (0.050) (0.050) (0.050) (0.051) ‐ Other Purposes ‐0.247*** ‐0.234*** ‐0.261*** ‐0.241*** ‐0.283*** 0.234*** ‐0.254*** (0.038) (0.036) (0.037) (0.037) (0.039) (0.036) (0.036) 29 LIBOR Interest Volatility Term Premium GDP Growth Constant ‐ 0.030*** ‐0.030*** (0.010) (0.010) ‐0.654 ‐1.046 (1.010) (1.010) (Table 4 continued) ‐0.000 ‐0.003 (0.020) (0.020) ‐ 0.112*** ‐0.114*** (0.010) (0.010) 1.631*** 1.603*** (0.150) (0.149) ‐0.001 (0.010) ‐1.033 (0.934) ‐0.009 (0.010) ‐0.821 (0.899) ‐0.017* (0.010) ‐0.785 (0.986) ‐0.015 (0.010) ‐1.122 (0.978) ‐0.009 (0.010) ‐1.387 (1.004) 0.045** (0.019) 0.031 (0.019) 0.016 (0.020) 0.025 (0.019) 0.025 (0.019) ‐0.094*** (0.009) 1.682*** (0.134) ‐0.099*** (0.009) 1.863*** (0.162) ‐0.100*** (0.010) 1.663*** (0.150) ‐0.103*** (0.009) 1.711*** (0.141) ‐0.101*** (0.009) 1.745*** (0.146) 7,274 0.512 0.510 7,538 0.523 0.521 7,642 0.501 0.500 7,191 0.513 0.512 7,191 0.512 0.511 (4) 0.157*** (0.038) 0.349*** (0.029) ‐0.299*** (0.018) (5) 0.123*** (0.037) 0.336*** (0.029) ‐0.299*** (0.017) (6) 0.105*** (0.037) 0.343*** (0.029) ‐0.310*** (0.020) (7) 0.141*** (0.038) 0.354*** (0.029) ‐0.261*** (0.019) (8) 0.136*** (0.038) 0.348*** (0.029) ‐0.221*** (0.020) Observations 8,629 8,629 R‐squared 0.515 0.524 Adjusted R‐squared 0.514 0.523 Panel B: Regression results for Collateral equation (1) (2) (3) Fitted Maturity 0.081** 0.084** (0.036) (0.036) Fitted Covenant Index 0.346*** 0.340*** (0.026) (0.026) Fitted Log(Loan Size) ‐0.317*** ‐0.285*** (0.015) (0.016) R&D to Assets 1.094** (0.534) Log(Firm Age) ‐0.169*** (0.022) Forecast Dispersion 0.833*** (0.094) Forecast Error 0.541*** (0.057) Analysts ‐0.001 (0.003) IA Index 1.419*** (0.114) IA Index_1 BR Dummy Profitability Leverage Market to Book Tangibility Loan Concentration Top Ten Lenders ‐0.003 (0.032) ‐0.343** (0.144) 0.930*** (0.117) ‐0.080*** (0.020) ‐0.103 (0.079) 1.044*** (0.109) ‐0.155*** (0.035) ‐0.004 (0.032) ‐0.370** (0.144) 0.741*** (0.118) ‐0.088*** (0.020) ‐0.104 (0.078) 0.782*** (0.116) ‐0.155*** (0.035) 0.039 (0.036) ‐0.129 (0.161) 0.834*** (0.132) ‐0.063*** (0.023) ‐0.218** (0.089) 0.946*** (0.126) ‐0.165*** (0.039) 30 0.014 (0.035) ‐0.078 (0.155) 0.924*** (0.129) ‐0.053** (0.022) ‐0.218** (0.086) 0.996*** (0.121) ‐0.162*** (0.037) 0.011 (0.035) ‐0.255 (0.157) 1.006*** (0.129) ‐0.066*** (0.022) ‐0.139* (0.084) 1.033*** (0.125) ‐0.148*** (0.037) 0.029 (0.037) ‐0.225 (0.158) 0.828*** (0.131) ‐0.066*** (0.023) ‐0.118 (0.089) 0.821*** (0.128) ‐0.161*** (0.039) 1.698*** (0.141) 0.029 (0.037) ‐0.132 (0.159) 0.738*** (0.133) ‐0.044* (0.023) ‐0.098 (0.089) 0.671*** (0.130) ‐0.166*** (0.039) Repayment ‐ Recap Acquisitions Other Purposes GDP Growth Constant 0.070 (0.044) 0.164* (0.085) ‐0.053 (0.054) ‐0.094*** (0.011) ‐0.087 (0.082) 0.056 (0.044) 0.167* (0.086) ‐0.040 (0.055) ‐0.098*** (0.011) 0.413*** (0.099) Observations 9,390 9,390 Pseudo R‐squared 0.229 0.239 Panel C: Regression results for Maturity equation (1) (2) (3) Fitted Collateral 2.160*** 1.911*** (0.191) (0.199) Fitted Covenant Index 0.195*** 0.232*** (0.031) (0.031) Fitted Log(Loan Size) 0.388*** 0.359*** (0.020) (0.020) R&D to Assets ‐0.625 (0.576) Log(Firm Age) 0.074*** (0.027) Forecast Dispersion 0.013 (0.051) 0.137 (0.094) 0.019 (0.059) ‐0.097*** (0.012) ‐0.538*** (0.100) 0.024 (0.049) 0.106 (0.093) ‐0.008 (0.058) ‐0.091*** (0.012) ‐0.457*** (0.094) 0.040 (0.049) 0.134 (0.091) ‐0.044 (0.057) ‐0.102*** (0.012) ‐0.252*** (0.090) 7,868 0.245 8,173 0.241 8,284 0.230 (Table 4 continued) ‐0.017 ‐0.007 (0.051) (0.051) 0.103 0.115 (0.096) (0.097) 0.001 ‐0.000 (0.060) (0.060) ‐0.094*** ‐0.092*** (0.012) (0.012) ‐1.258*** ‐1.550*** (0.123) (0.140) 7,781 0.251 (4) 2.868*** (0.231) 0.117*** (0.038) 0.402*** (0.022) (5) 2.634*** (0.218) 0.155*** (0.035) 0.400*** (0.021) (6) 2.357*** (0.215) 0.169*** (0.034) 0.425*** (0.023) (7) 2.570*** (0.234) 0.165*** (0.038) 0.383*** (0.022) ‐0.371*** (0.064) Analysts ‐0.010*** (0.004) IA Index ‐0.372** (0.154) IA Index_1 Profitability Leverage Market to Book Log(Asset Maturity) Earning Variance Tax Asset Utilities (8) 2.544*** (0.244) 0.172*** (0.038) 0.378*** (0.021) ‐0.508*** (0.099) Forecast Error BR Dummy 7,781 0.251 ‐0.088** (0.036) 0.497*** (0.156) 0.053 (0.126) ‐0.000 (0.024) 0.059** (0.024) ‐4.428*** (0.490) 6.168*** (0.773) ‐0.224*** (0.058) ‐0.091** (0.037) 0.531*** (0.156) 0.141 (0.126) ‐0.002 (0.024) 0.051** (0.024) ‐4.009*** (0.481) 6.049*** (0.774) ‐0.238*** (0.057) ‐0.136*** (0.040) 0.421** (0.167) ‐0.300** (0.140) ‐0.020 (0.026) 0.090*** (0.026) ‐5.035*** (0.572) 5.894*** (0.847) ‐0.202*** (0.062) 31 ‐0.120*** (0.039) 0.399** (0.166) ‐0.184 (0.137) ‐0.026 (0.025) 0.077*** (0.026) ‐4.902*** (0.549) 6.148*** (0.824) ‐0.199*** (0.061) ‐0.117*** (0.039) 0.588*** (0.166) ‐0.204 (0.136) ‐0.005 (0.025) 0.068*** (0.026) ‐4.874*** (0.547) 6.332*** (0.830) ‐0.204*** (0.060) ‐0.132*** (0.040) 0.462*** (0.167) ‐0.297** (0.142) ‐0.021 (0.026) 0.075*** (0.026) ‐4.817*** (0.587) 5.864*** (0.854) ‐0.227*** (0.062) ‐0.293 (0.193) ‐0.132*** (0.040) 0.449*** (0.168) ‐0.293** (0.142) ‐0.025 (0.026) 0.075*** (0.026) ‐4.918*** (0.587) 6.053*** (0.851) ‐0.220*** (0.061) Top Ten Lenders Repayment ‐ Recap Acquisitions Other Purposes GDP Growth Constant Observations R‐squared (Table 4 continued) 0.312*** 0.312*** (0.044) (0.044) ‐0.027 ‐0.031 (0.057) (0.057) ‐0.243** ‐0.246** (0.118) (0.118) ‐1.052*** ‐1.053*** (0.050) (0.050) 0.189*** 0.188*** (0.013) (0.013) 0.803*** 0.791*** (0.140) (0.154) 0.328*** (0.040) ‐0.050 (0.051) ‐0.096 (0.103) ‐1.024*** (0.047) 0.169*** (0.012) 0.489*** (0.129) 0.307*** (0.040) ‐0.028 (0.051) ‐0.061 (0.103) ‐1.044*** (0.047) 0.165*** (0.012) 0.406** (0.169) 0.329*** (0.043) ‐0.041 (0.057) ‐0.255** (0.117) ‐1.038*** (0.050) 0.193*** (0.013) 0.543*** (0.134) 0.323*** (0.043) ‐0.069 (0.055) ‐0.245** (0.114) ‐1.044*** (0.049) 0.181*** (0.013) 0.564*** (0.130) 0.311*** (0.042) ‐0.074 (0.055) ‐0.206* (0.114) ‐1.028*** (0.049) 0.185*** (0.013) 0.509*** (0.131) 8,884 0.201 8,884 0.198 7,476 0.215 7,756 0.210 7,862 0.210 7,393 0.215 7,393 0.215 0.208 0.208 0.213 0.213 Adjusted R‐squared 0.200 0.196 0.213 Panel D: Regression results for Covenant Index equation (1) (2) (3) (4) Fitted Collateral 1.786*** 1.749*** 1.823*** (0.092) (0.093) (0.100) Fitted Maturity 0.260*** 0.275*** 0.222*** (0.023) (0.022) (0.024) Fitted Log(Loan Size) ‐0.004 ‐0.010 ‐0.051*** (0.016) (0.016) (0.017) R&D to Assets ‐1.076** (0.443) Log(Firm Age) 0.013 (0.016) Forecast Dispersion ‐0.375*** (0.059) Forecast Error (5) 1.732*** (0.096) 0.254*** (0.023) ‐0.048*** (0.017) (6) 1.601*** (0.097) 0.241*** (0.023) ‐0.020 (0.018) (7) 1.781*** (0.100) 0.246*** (0.023) ‐0.077*** (0.017) ‐0.218*** (0.040) Analysts ‐0.013*** (0.002) IA Index ‐0.700*** (0.092) IA Index_1 BR Dummy Profitability Leverage Market to Book Cash Flow Volatility Top Ten Lenders Number of Lenders (8) 1.785*** (0.104) 0.257*** (0.023) ‐0.089*** (0.017) ‐0.001 (0.022) ‐0.110 (0.078) ‐0.163** (0.066) 0.021 (0.014) ‐2.178*** (0.385) ‐0.021 (0.025) 0.017*** (0.002) 0.000 (0.022) ‐0.102 (0.079) ‐0.142** (0.065) 0.014 (0.014) ‐2.091*** (0.374) ‐0.031 (0.025) 0.017*** (0.002) ‐0.010 (0.024) ‐0.158* (0.082) ‐0.066 (0.074) 0.007 (0.014) ‐2.333*** (0.419) ‐0.025 (0.026) 0.020*** (0.002) 32 ‐0.006 (0.023) ‐0.197** (0.081) ‐0.100 (0.071) ‐0.004 (0.014) ‐2.120*** (0.402) ‐0.024 (0.026) 0.019*** (0.002) ‐0.003 (0.023) ‐0.068 (0.084) ‐0.127* (0.072) 0.025* (0.014) ‐2.295*** (0.415) ‐0.035 (0.026) 0.021*** (0.002) ‐0.012 (0.024) ‐0.163** (0.083) ‐0.089 (0.075) 0.003 (0.014) ‐1.769*** (0.406) ‐0.028 (0.026) 0.021*** (0.002) ‐0.723*** (0.112) ‐0.011 (0.024) ‐0.196** (0.082) ‐0.088 (0.075) ‐0.003 (0.014) ‐1.900*** (0.410) ‐0.030 (0.026) 0.020*** (0.002) Repayment ‐ Recap Acquisitions Other Purposes GDP Growth Constant Observations Pseudo R‐squared 0.028 (0.029) ‐0.029 (0.053) 0.336*** (0.036) ‐0.006 (0.008) ‐1.227*** (0.066) 0.031 (0.029) ‐0.028 (0.053) 0.344*** (0.036) ‐0.008 (0.008) ‐1.282*** (0.091) 0.063** (0.032) ‐0.003 (0.060) 0.267*** (0.039) ‐0.004 (0.009) ‐0.853*** (0.073) 0.073** (0.031) 0.045 (0.058) 0.308*** (0.038) ‐0.012 (0.008) ‐0.911*** (0.070) 0.072** (0.031) 0.038 (0.058) 0.306*** (0.038) ‐0.010 (0.008) ‐0.922*** (0.070) 9,390 0.130 9,390 0.132 7,868 0.140 8,173 0.137 8,284 0.138 (4) ‐0.908*** (0.098) 0.757*** (0.013) ‐0.086*** (0.018) (5) ‐0.694*** (0.099) 0.740*** (0.013) ‐0.098*** (0.018) (Table 4 continued) 0.079** 0.075** (0.032) (0.032) 0.033 0.031 (0.060) (0.060) 0.288*** 0.299*** (0.039) (0.039) ‐0.007 ‐0.008 (0.009) (0.009) ‐0.509*** ‐0.451*** (0.082) (0.090) 7,781 0.141 7,781 0.140 Panel E: Regression results for Log(Loan Size) equation (1) Fitted Collateral Fitted Maturity Fitted Covenant Index R&D to Assets (2) ‐0.609*** (0.080) 0.827*** (0.013) ‐0.148*** (0.016) 0.549** (0.222) Log(Firm Age) (3) ‐0.708*** (0.080) 0.823*** (0.013) ‐0.148*** (0.016) (6) ‐0.648*** (0.088) 0.749*** (0.013) ‐0.103*** (0.017) (7) ‐0.912*** (0.099) 0.739*** (0.013) ‐0.077*** (0.018) ‐0.132*** (0.011) Forecast Dispersion 0.186*** (0.044) Forecast Error 0.117*** (0.032) Analysts 0.015*** (0.002) IA Index 0.298*** (0.067) IA Index_1 BR Dummy Firm Size Profitability Leverage Market to Book Current Ratio Top Ten Lenders Syndicated (8) ‐0.766*** (0.100) 0.737*** (0.013) ‐0.093*** (0.018) 0.149*** (0.015) 0.505*** (0.008) ‐0.258*** (0.066) ‐0.585*** (0.053) 0.039*** (0.008) ‐0.092*** (0.008) ‐0.110*** (0.018) 0.557*** (0.039) 0.146*** (0.015) 0.527*** (0.008) ‐0.333*** (0.066) ‐0.632*** (0.051) 0.030*** (0.008) ‐0.092*** (0.008) ‐0.113*** (0.018) 0.566*** (0.039) 0.153*** (0.016) 0.498*** (0.009) ‐0.159** (0.068) ‐0.419*** (0.060) 0.048*** (0.009) ‐0.092*** (0.009) ‐0.100*** (0.020) 0.748*** (0.050) 33 0.148*** (0.016) 0.514*** (0.009) ‐0.139** (0.068) ‐0.503*** (0.059) 0.046*** (0.009) ‐0.091*** (0.009) ‐0.071*** (0.019) 0.539*** (0.047) 0.153*** (0.016) 0.461*** (0.010) ‐0.301*** (0.069) ‐0.406*** (0.058) 0.016* (0.009) ‐0.096*** (0.009) ‐0.078*** (0.019) 0.634*** (0.046) 0.143*** (0.016) 0.509*** (0.008) ‐0.154** (0.069) ‐0.405*** (0.061) 0.045*** (0.009) ‐0.094*** (0.009) ‐0.087*** (0.020) 0.641*** (0.049) 0.097 (0.084) 0.143*** (0.016) 0.514*** (0.008) ‐0.141** (0.068) ‐0.423*** (0.060) 0.046*** (0.009) ‐0.091*** (0.009) ‐0.082*** (0.020) 0.652*** (0.049) Repayment ‐ Recap Acquisitions Other Purposes GDP Growth Constant Observations R‐squared Adjusted R‐squared 0.126*** (0.021) 0.012 (0.043) 0.875*** (0.023) ‐0.081*** (0.005) ‐1.634*** (0.078) 0.118*** (0.021) 0.020 (0.043) 0.879*** (0.023) ‐0.086*** (0.005) ‐1.293*** (0.085) 0.131*** (0.023) 0.111** (0.051) 0.823*** (0.024) ‐0.085*** (0.006) ‐1.626*** (0.086) 0.150*** (0.023) 0.105** (0.050) 0.804*** (0.024) ‐0.075*** (0.006) ‐1.521*** (0.083) 0.134*** (0.022) 0.065 (0.050) 0.800*** (0.024) ‐0.083*** (0.006) ‐1.369*** (0.088) 9,390 0.799 0.799 9,390 0.802 0.801 7,868 0.758 0.757 8,173 0.772 0.771 8,284 0.776 0.775 34 (Table 4 continued) 0.124*** 0.129*** (0.023) (0.023) 0.127** 0.122** (0.052) (0.052) 0.808*** 0.806*** (0.025) (0.025) ‐0.083*** ‐0.080*** (0.006) (0.006) ‐1.667*** ‐1.635*** (0.082) (0.089) 7,781 0.756 0.756 7,781 0.756 0.755 Table 5: : The regression results for five loan term equations for term loans Panel A, B, C, D, and E are results for AISD, Collateral, Maturity, Covenant Index, and Log(Loan Size) equations respectively The information asymmetry proxies are R&D to Sales, Log(Firm Age), Forecast Dispersion Forecast Error, IA Index and IA Index_1in Column (2), (3), (4), (5), (6), (7) and (8)) respectively Standard errors in parentheses and ***, **, and * indicate significantly different than zero at the 1%, 5%, and 10% levels, respectively Numbers in parentheses are standard errors corrected for heteroscedasticity Panel A: Regression results for AISD equation (1) (2) (3) (4) (5) (6) (7) (8) Fitted Collateral 2.538*** 2.516*** 2.082*** 1.794*** 1.773*** 1.724*** 1.479*** (0.431) (0.432) (0.423) (0.398) (0.396) (0.419) (0.432) Fitted Maturity ‐0.045 ‐0.059 0.016 0.029 ‐0.065 ‐0.050 ‐0.084 (0.066) (0.067) (0.072) (0.075) (0.076) (0.075) (0.076) Fitted Covenant Index ‐0.159*** ‐0.175*** ‐0.196*** ‐0.134*** ‐0.181*** ‐0.169*** ‐0.169*** (0.041) (0.041) (0.047) (0.048) (0.046) (0.048) (0.048) Fitted Log(Loan Size) 0.005 0.020 ‐0.006 ‐0.033 0.064* 0.017 0.062* (0.026) (0.027) (0.032) (0.028) (0.034) (0.032) (0.033) R&D to Assets ‐0.276 (0.901) Log(Firm Age) ‐0.076** (0.033) Forecast Dispersion 0.168*** (0.060) Forecast Error 0.409*** (0.066) Analysts ‐0.031*** (0.007) IA Index 1.236*** (0.222) IA Index_1 1.763*** (0.285) BR Dummy ‐0.195*** ‐0.193*** ‐0.207*** ‐0.236*** ‐0.243*** ‐0.207*** ‐0.211*** (0.051) (0.051) (0.058) (0.056) (0.056) (0.058) (0.058) Profitability ‐0.866*** ‐0.917*** ‐0.562** ‐0.579** ‐0.384 ‐0.697*** ‐0.638** (0.247) (0.249) (0.263) (0.266) (0.261) (0.270) (0.268) Leverage ‐0.404*** ‐0.415*** ‐0.201 ‐0.178 ‐0.261 ‐0.017 ‐0.028 (0.155) (0.152) (0.176) (0.173) (0.165) (0.181) (0.177) Market to Book ‐0.019 ‐0.031 ‐0.059 ‐0.092** ‐0.044 ‐0.072* ‐0.051 (0.036) (0.036) (0.038) (0.037) (0.037) (0.038) (0.037) Log(1 + Coverage) ‐0.272*** ‐0.272*** ‐0.200*** ‐0.199*** ‐0.241*** ‐0.175*** ‐0.182*** (0.040) (0.040) (0.040) (0.039) (0.038) (0.040) (0.040) Top Ten Lenders ‐0.071 ‐0.081 ‐0.131** ‐0.135** ‐0.123** ‐0.128** ‐0.128** (0.062) (0.063) (0.064) (0.061) (0.060) (0.063) (0.063) Syndicated 0.069 0.089 0.290* ‐0.003 ‐0.063 0.257 0.216 (0.136) (0.136) (0.174) (0.181) (0.176) (0.173) (0.175) Tranche ‐0.318** ‐0.291** ‐0.142 ‐0.219* ‐0.094 ‐0.039 0.033 (0.126) (0.126) (0.125) (0.125) (0.130) (0.126) (0.127) Repayment ‐ Recap ‐0.013 ‐0.000 0.126 0.015 0.014 0.148 0.150 (0.082) (0.081) (0.096) (0.095) (0.095) (0.094) (0.093) Acquisitions 0.217*** 0.225*** 0.342*** 0.351*** 0.357*** 0.334*** 0.337*** (0.082) (0.081) (0.093) (0.093) (0.094) (0.093) (0.093) Other Purposes 0.015 0.039 0.135 0.079 0.106 0.168** 0.190** (0.079) (0.079) (0.083) (0.084) (0.084) (0.081) (0.081) LIBOR ‐0.095*** ‐0.106*** ‐0.103*** ‐0.088*** ‐0.090*** ‐0.125*** ‐0.132*** (0.030) (0.030) (0.034) (0.034) (0.034) (0.034) (0.034) Interest Volatility 7.658*** 7.946*** 7.645*** 6.377*** 6.720*** 7.715*** 7.710*** (2.009) (2.010) (2.191) (2.247) (2.200) (2.214) (2.208) 35 Term Premium GDP Growth Constant Observations R‐squared Adjusted R‐squared (Table 5 continued) 0.000 ‐0.012 (0.058) (0.057) ‐0.122*** ‐0.115*** (0.024) (0.024) 1.505*** 1.405*** (0.367) (0.358) ‐0.005 (0.051) ‐0.116*** (0.022) 2.224*** (0.375) ‐0.021 (0.051) ‐0.115*** (0.022) 2.515*** (0.413) 0.040 (0.057) ‐0.135*** (0.024) 1.690*** (0.383) 0.030 (0.057) ‐0.130*** (0.024) 2.197*** (0.401) 0.029 (0.056) ‐0.111*** (0.023) 2.727*** (0.443) 2,565 0.235 0.229 2,565 0.238 0.232 1,879 0.230 0.221 2,017 0.244 0.236 2,060 0.240 0.232 1,851 0.244 0.235 1,851 0.253 0.245 0.514*** (0.082) 0.253*** (0.028) ‐0.200*** (0.027) 0.447*** (0.086) 0.237*** (0.028) ‐0.211*** (0.025) 0.443*** (0.085) 0.237*** (0.028) ‐0.187*** (0.033) 0.476*** (0.087) 0.253*** (0.028) ‐0.198*** (0.029) 0.473*** (0.087) 0.252*** (0.028) ‐0.188*** (0.032) Panel B: Regression results for Collateral equation Fitted Maturity 0.344*** 0.331*** (0.080) (0.081) Fitted Covenant Index 0.224*** 0.221*** (0.026) (0.026) Fitted Log(Loan Size) ‐0.209*** ‐0.204*** (0.022) (0.023) R&D to Assets 2.052** (0.893) Log(Firm Age) ‐0.029 (0.037) Forecast Dispersion 0.013 (0.062) Forecast Error 0.128* (0.069) Analysts ‐0.010 (0.007) IA Index 0.075 (0.237) IA Index_1 BR Dummy Profitability Leverage Market to Book Tangibility Loan Concentration Top Ten Lenders Repayment ‐ Recap Acquisitions Other Purposes ‐0.054 (0.054) ‐0.579** (0.268) 0.448** (0.174) ‐0.029 (0.037) 0.338** (0.140) 0.031 (0.242) ‐0.173*** (0.066) 0.063 (0.076) ‐0.069 (0.091) 0.018 (0.074) ‐0.054 (0.054) ‐0.584** (0.272) 0.389** (0.175) ‐0.020 (0.037) 0.281** (0.138) 0.027 (0.246) ‐0.179*** (0.066) 0.062 (0.076) ‐0.074 (0.091) 0.025 (0.074) ‐0.027 (0.065) ‐0.576* (0.324) 0.238 (0.203) ‐0.007 (0.042) 0.493*** (0.166) ‐0.151 (0.267) ‐0.161** (0.073) 0.031 (0.091) ‐0.147 (0.106) ‐0.060 (0.083) 36 ‐0.024 (0.062) ‐0.592* (0.327) 0.332 (0.203) 0.015 (0.041) 0.367** (0.157) ‐0.025 (0.263) ‐0.141** (0.071) 0.078 (0.086) ‐0.140 (0.103) ‐0.010 (0.082) ‐0.025 (0.062) ‐0.598* (0.319) 0.262 (0.199) 0.037 (0.042) 0.404*** (0.153) ‐0.194 (0.263) ‐0.146** (0.070) 0.085 (0.086) ‐0.163 (0.102) ‐0.021 (0.082) ‐0.035 (0.065) ‐0.587* (0.334) 0.308 (0.205) ‐0.008 (0.043) 0.483*** (0.171) ‐0.007 (0.267) ‐0.134* (0.074) 0.038 (0.091) ‐0.130 (0.107) ‐0.053 (0.084) 0.241 (0.285) ‐0.036 (0.065) ‐0.572* (0.334) 0.277 (0.204) ‐0.003 (0.043) 0.476*** (0.171) ‐0.040 (0.263) ‐0.134* (0.074) 0.037 (0.091) ‐0.127 (0.108) ‐0.052 (0.084) GDP Growth Constant Observations Pseudo R‐squared (Table 5 continued) ‐0.065*** ‐0.064** (0.025) (0.025) ‐1.992*** ‐2.085*** (0.324) (0.324) ‐0.052** (0.021) ‐1.129*** (0.295) ‐0.053** (0.021) ‐0.949*** (0.322) ‐0.069*** (0.024) ‐2.081*** (0.334) ‐0.055** (0.023) ‐1.767*** (0.334) ‐0.050** (0.023) ‐1.705*** (0.328) 2,781 0.0825 2,781 0.0824 2,007 0.108 2,171 0.103 2,214 0.103 1,979 0.110 1,979 0.111 3.349*** (0.434) ‐0.196*** (0.053) 0.270*** (0.043) 3.058*** (0.432) ‐0.141*** (0.053) 0.254*** (0.042) 2.724*** (0.457) ‐0.134** (0.053) 0.288*** (0.045) 3.058*** (0.448) ‐0.162*** (0.055) 0.270*** (0.044) 2.932*** (0.462) ‐0.158*** (0.054) 0.286*** (0.044) Panel C: Regression results for Maturity equation Fitted Collateral 2.315*** 2.241*** (0.389) (0.395) Fitted Covenant Index ‐0.045 ‐0.049 (0.045) (0.045) Fitted Log(Loan Size) 0.229*** 0.237*** (0.037) (0.037) R&D to Assets 1.629 (1.418) Log(Firm Age) ‐0.064 (0.051) Forecast Dispersion 0.015 (0.062) Forecast Error ‐0.074 (0.088) Analysts ‐0.015 (0.010) IA Index 0.892*** (0.316) IA Index_1 BR Dummy Profitability Leverage Market to Book Log(Asset Maturity) Earning Variance Tax Asset Utilities Top Ten Lenders Repayment ‐ Recap Acquisitions Other Purposes ‐0.023 (0.079) 0.747** (0.369) 0.035 (0.193) ‐0.026 (0.055) ‐0.130** (0.051) ‐2.767*** (0.910) 4.430*** (1.514) 0.561*** (0.117) 0.368*** (0.084) 0.127 (0.106) 0.323*** (0.122) 0.213* (0.110) ‐0.025 (0.079) 0.708* (0.369) ‐0.017 (0.190) ‐0.022 (0.054) ‐0.137*** (0.050) ‐2.578*** (0.892) 4.138*** (1.490) 0.542*** (0.117) 0.357*** (0.084) 0.125 (0.106) 0.318*** (0.122) 0.221** (0.110) ‐0.026 (0.091) 0.762* (0.422) ‐0.123 (0.223) 0.023 (0.062) ‐0.259*** (0.055) ‐2.951** (1.215) 1.250 (1.655) 0.396*** (0.138) 0.285*** (0.093) 0.063 (0.123) 0.252* (0.151) 0.181 (0.117) 37 ‐0.047 (0.088) 0.683* (0.411) ‐0.001 (0.217) ‐0.037 (0.061) ‐0.191*** (0.055) ‐3.386*** (1.097) 3.723** (1.641) 0.435*** (0.134) 0.276*** (0.090) ‐0.003 (0.119) 0.273* (0.143) 0.183 (0.118) ‐0.046 (0.087) 0.905** (0.418) ‐0.066 (0.212) ‐0.013 (0.062) ‐0.197*** (0.055) ‐3.466*** (1.077) 3.026* (1.613) 0.431*** (0.132) 0.282*** (0.089) 0.028 (0.118) 0.335** (0.139) 0.228* (0.118) ‐0.027 (0.091) 0.652 (0.421) ‐0.054 (0.227) ‐0.000 (0.062) ‐0.250*** (0.055) ‐3.507*** (1.219) 2.641 (1.767) 0.363*** (0.140) 0.265*** (0.093) 0.036 (0.123) 0.198 (0.153) 0.169 (0.117) 1.022*** (0.382) ‐0.024 (0.091) 0.686 (0.422) ‐0.064 (0.226) 0.013 (0.062) ‐0.249*** (0.055) ‐3.378*** (1.211) 2.409 (1.753) 0.377*** (0.140) 0.266*** (0.093) 0.033 (0.123) 0.196 (0.153) 0.170 (0.117) GDP Growth Constant Observations R‐squared Adjusted R‐squared (Table 5 continued) 0.167*** 0.169*** (0.028) (0.028) 2.013*** 1.927*** (0.308) (0.319) 0.163*** (0.025) 2.317*** (0.270) 0.160*** (0.025) 2.570*** (0.319) 0.158*** (0.028) 2.304*** (0.295) 0.152*** (0.028) 2.379*** (0.279) 0.154*** (0.027) 2.495*** (0.287) 2,654 0.130 0.125 2,654 0.130 0.124 1,912 0.143 0.135 2,066 0.132 0.125 2,109 0.137 0.130 1,884 0.146 0.138 2.936*** (0.189) ‐0.279*** (0.048) 0.195*** (0.018) 2.790*** (0.194) ‐0.249*** (0.048) 0.182*** (0.020) 3.104*** (0.185) ‐0.333*** (0.048) 0.183*** (0.019) Panel D: Regression results for Covenant Index equation Fitted Collateral 2.316*** 2.324*** 3.172*** (0.160) (0.161) (0.186) Fitted Maturity ‐0.118*** ‐0.111*** ‐0.367*** (0.039) (0.039) (0.046) Fitted Log(Loan Size) 0.168*** 0.167*** 0.195*** (0.015) (0.016) (0.018) R&D to Assets ‐2.477*** (0.646) Log(Firm Age) 0.017 (0.021) Forecast Dispersion ‐0.037 (0.041) Forecast Error 0.002 (0.004) IA Index ‐0.260** (0.128) IA Index_1 Profitability Leverage Market to Book Cash Flow Volatility Top Ten Lenders Number of Lenders Repayment ‐ Recap Acquisitions Other Purposes GDP Growth 3.162*** (0.186) ‐0.336*** (0.048) 0.176*** (0.020) ‐0.120*** (0.044) Analysts BR Dummy 1,884 0.146 0.139 0.047 (0.031) 0.026 (0.122) ‐0.278*** (0.080) 0.051*** (0.019) ‐1.450*** (0.374) 0.030 (0.039) 0.002 (0.002) 0.014 (0.045) 0.079 (0.056) 0.105** (0.041) 0.022* (0.013) 0.046 (0.031) 0.062 (0.122) ‐0.219*** (0.079) 0.037* (0.019) ‐1.587*** (0.385) 0.032 (0.039) 0.002 (0.002) 0.016 (0.045) 0.083 (0.057) 0.101** (0.041) 0.024* (0.012) 0.040 (0.035) 0.292** (0.134) ‐0.354*** (0.093) 0.050** (0.022) ‐2.657*** (0.487) 0.123*** (0.042) ‐0.003* (0.002) ‐0.001 (0.050) 0.164** (0.065) 0.101** (0.045) 0.046*** (0.014) 38 0.029 (0.034) 0.198 (0.140) ‐0.327*** (0.090) 0.024 (0.021) ‐2.661*** (0.483) 0.084** (0.041) ‐0.002 (0.002) ‐0.022 (0.049) 0.146** (0.064) 0.070 (0.045) 0.033** (0.014) 0.033 (0.034) 0.234* (0.138) ‐0.329*** (0.089) 0.023 (0.021) ‐2.757*** (0.487) 0.068* (0.041) ‐0.002 (0.002) ‐0.023 (0.048) 0.159** (0.064) 0.095** (0.045) 0.027* (0.014) 0.044 (0.035) 0.255* (0.136) ‐0.346*** (0.093) 0.047** (0.021) ‐2.512*** (0.493) 0.103** (0.042) ‐0.002 (0.002) ‐0.002 (0.051) 0.154** (0.064) 0.090** (0.045) 0.040*** (0.014) ‐0.390** (0.154) 0.046 (0.035) 0.255* (0.136) ‐0.337*** (0.093) 0.043** (0.021) ‐2.546*** (0.493) 0.105** (0.042) ‐0.002 (0.002) ‐0.001 (0.051) 0.155** (0.064) 0.090** (0.045) 0.041*** (0.014) Constant Observations Pseudo R‐squared ‐0.578*** (0.127) ‐0.687*** (0.157) 0.012 (0.138) ‐0.156 (0.145) ‐0.208 (0.151) 2,781 0.0923 2,781 0.0905 2,007 0.102 2,171 0.0978 2,214 0.0949 Panel E: Regression results for Log(Loan Size) equation Fitted Collateral ‐1.502*** ‐1.502*** (0.170) (0.171) Fitted Maturity 1.187*** 1.192*** (0.030) (0.030) Fitted Covenant Index 0.205*** 0.205*** (0.018) (0.018) R&D to Assets ‐0.102 (0.526) Log(Firm Age) 0.019 (0.021) Forecast Dispersion ‐2.663*** (0.215) 1.157*** (0.034) 0.307*** (0.022) ‐2.453*** (0.208) 1.175*** (0.035) 0.308*** (0.022) ‐2.249*** (0.196) 1.182*** (0.033) 0.296*** (0.021) (Table 5 continued) 0.061 0.135 (0.139) (0.142) 1,979 0.103 1,979 0.103 ‐2.217*** (0.219) 1.126*** (0.035) 0.265*** (0.023) ‐2.133*** (0.218) 1.127*** (0.034) 0.257*** (0.022) ‐0.040 (0.038) Forecast Error 0.048 (0.037) Analysts 0.019*** (0.005) IA Index ‐0.959*** (0.121) IA Index_1 BR Dummy Firm Size Profitability Leverage Market to Book Current Ratio Top Ten Lenders Syndicated Repayment ‐ Recap Acquisitions Other Purposes GDP Growth 0.066** (0.030) 0.630*** (0.014) ‐0.653*** (0.153) ‐0.145* (0.077) 0.159*** (0.025) ‐0.077*** (0.012) ‐0.256*** (0.033) 0.051 (0.067) ‐0.154*** (0.039) ‐0.177*** (0.049) ‐0.112*** (0.043) ‐0.183*** (0.011) 0.067** (0.030) 0.627*** (0.015) ‐0.644*** (0.152) ‐0.137* (0.076) 0.161*** (0.025) ‐0.077*** (0.012) ‐0.257*** (0.033) 0.048 (0.067) ‐0.154*** (0.039) ‐0.178*** (0.049) ‐0.116*** (0.043) ‐0.183*** (0.011) 0.112*** (0.035) 0.544*** (0.019) ‐0.048 (0.162) ‐0.087 (0.096) 0.065** (0.029) ‐0.089*** (0.016) ‐0.235*** (0.037) 0.456*** (0.099) ‐0.052 (0.048) ‐0.075 (0.058) ‐0.048 (0.049) ‐0.166*** (0.012) 39 0.143*** (0.034) 0.580*** (0.017) ‐0.115 (0.158) ‐0.160* (0.092) 0.105*** (0.026) ‐0.077*** (0.015) ‐0.204*** (0.036) ‐0.013 (0.091) ‐0.043 (0.045) ‐0.178*** (0.057) ‐0.102** (0.048) ‐0.168*** (0.012) 0.137*** (0.034) 0.534*** (0.019) ‐0.400** (0.159) ‐0.084 (0.089) 0.092*** (0.027) ‐0.091*** (0.015) ‐0.213*** (0.035) ‐0.030 (0.088) ‐0.074* (0.044) ‐0.226*** (0.057) ‐0.142*** (0.047) ‐0.170*** (0.012) 0.113*** (0.035) 0.538*** (0.019) 0.077 (0.163) ‐0.090 (0.096) 0.056* (0.029) ‐0.083*** (0.016) ‐0.210*** (0.038) 0.499*** (0.100) ‐0.031 (0.048) ‐0.050 (0.059) ‐0.041 (0.050) ‐0.170*** (0.012) ‐1.270*** (0.145) 0.109*** (0.035) 0.511*** (0.019) 0.012 (0.163) ‐0.065 (0.095) 0.038 (0.029) ‐0.085*** (0.016) ‐0.213*** (0.038) 0.516*** (0.099) ‐0.028 (0.047) ‐0.057 (0.059) ‐0.047 (0.050) ‐0.172*** (0.012) Constant Observations R‐squared Adjusted R‐squared ‐5.253*** (0.154) ‐5.315*** (0.170) ‐4.539*** (0.182) ‐4.662*** (0.175) ‐4.531*** (0.167) 2,781 0.841 0.840 2,781 0.842 0.841 2,007 0.780 0.778 2,171 0.811 0.809 2,214 0.816 0.815 40 (Table 5 continued) ‐4.139*** ‐3.837*** (0.184) (0.193) 1,979 0.776 0.774 1,979 0.778 0.776 ... asymmetries’ effect on loan terms and other loan term effects respectively 4.1 Loan term interdependences The empirical results show that while the revolving and term loans’ non-price terms (i.e., collateral,... revolving and term loans We find evidence that all loan terms are determined jointly The trade-off between them differs among loan types While the revolving and term loans’ non-price terms (i.e.,... such as Leverage and Log(1+ Coverage); LIBOR and Term Premium; Log(Firm Age) and Log (Loan Size); Interest Volatility and Term Premium; and Term Premium and GDP Growth, the model may suffer from