國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
67
Table 2
Descriptive Statistics for Variables Used in Regression Analyses Panel A:Descriptive statistics by IFRS adopters (treatment sample)
IFRS borrowers pre-IFRS period post-IFRS period
N Mean Std P25 Med P75 N Mean Med N Mean Med
Dependent Variable
Proportion_lead
721 16.76% 15.78% 5.41% 10.06% 21.05% 512 16.52% 10.00% 209 17.38% 10.37%Herfindahal_index
721 1,696.70 1,548.44 577.43 1,104.89 2,243.67 512 1,667.60 1,087.49 209 1,773.91 1,200.00Num_lender
721 13.93 10.53 6.00 11.00 20.00 512 14.86 12.00 209 11.46 10.00Cross_country
721 9.32 8.63 3.00 7.00 14.00 512 9.81 7.00 209 8.03 6.00Cross_region
721 3.48 4.05 1.00 2.00 5.00 512 3.61 2.00 209 3.13 2.00Financial_covenant
721 0.10 0.30 0.00 0.00 0.00 512 0.11 0.00 209 0.09 0.00Loan-specific Variable
Loan_size
721 865,440,737 1,717,976,229 69,000,000 242,849,513 818,646,544 512 916,914,439 239,178,651 209 728,867,298 249,729,480Loan_maturity
721 4.11 2.25 3.00 4.91 5.00 512 4.14 5.00 209 4.07 4.00Spread
504 0.0956 0.0797 0.0425 0.0687 0.1350 376 0.0959 0.0700 128 0.0948 0.0670Performance_pricing
721 0.21 0.41 0.00 0.00 0.00 512 0. 26 0.00 209 0.09 0.00Acc_based_pp
721 0.07 0.26 0.00 0.00 0.00 512 0.09 0.00 209 0.02 0.00Secured
168 0.48 0.50 0.00 0.00 1.00 121 0.43 0.00 47 0.61 1.00Term
721 0.34 0.47 0.00 0.00 1.00 512 0.29 0.00 209 0.46 0.00Revolver
721 0.46 0.49 0.00 0.00 1.00 512 0.49 1.00 209 0.36 0.00Borrower-specific Variable
Log(Firm_size)
721 21.46 1.89 20.12 21.43 22.81 512 21.52 21.51 209 21.30 21.30LEV
721 0.20 0.17 0.09 0.19 0.28 512 0.20 0.18 209 0.21 0.22ROA
721 0.04 0.06 0.01 0.04 0.07 512 0.03 0.03 209 0.06 0.06US_listing
721 0.21 0.40 0.00 0.00 0.00 512 0.22 0.00 209 0.18 0.00‧
Panel B: Descriptive statistics by non-IFRS adopters (benchmark sample)Non-IFRS borrowers pre-IFRS period post-IFRS period
N Mean Std P25 Med P75 N Mean Med N Mean Med
Dependent Variable
Proportion_lead
1,498 32.12% 17.68% 18.71% 30.00% 42.00% 885 34.12% 33.33% 613 29.22% 27.27%Herfindahal_index
1,498 2,429.06 1,480.79 1,331.65 2,185.75 3,253.12 885 2,599.11 2,336.00 613 2,182.99 1,837.50Num_lender
1,498 6.64 4.93 3.0 5.0 8.0 885 7.01 5.00 613 6.12 5.00Loan_size
1,498 37,301,671 221,495,017 287,907 802,056 3,986,700 885 26,576,805 719,037 613 52,820,360 1,002,836Loan_maturity
1,498 3.05 2.47 1.00 3.00 5.00 885 2.80 1.08 613 3.40 3.00Proportion_lead is defined as the fraction of the loan retained by the lead arranger. The lead lender is designated if the following descriptions receive in the DealScan database: lead bank, lead manager, lead agent, admin agent, arranger, agent, and bookrunner. Herfindahal_index measures as the sum of the squared percentage ownership of each lender in the loan. Num_lender is defined as the total number of lenders in the loan syndicate. Financial_covenant is defined as an indicator variable taking the value of one if financial covenants imposed by the debt
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
69
agreement and 0 otherwise. Cross_country is defined as the number of lenders in the syndicate whose nationality is different from that of the borrower. Cross_region is defined as the number of lenders in the syndicate whose region is different from that of the borrower. Worldwide syndicated loan market is divided into the three regions: Europe, North America, and Australasia (refer to Carey and Nini [2006]). Loan_size is the total dollar value of the loan converted to US dollar. Loan_maturity is the number of years to maturity of the loan. Spread is based on the All-in-Drawn-Spread measure reported by DealScan. This measure is equal to the amount the borrower pays in basis points over LIBOR or LIBOR equivalents for each dollar drawn down.
Performance_pricing is defined as an indicator variable taking the value of one if the loan contract incorporates a performance pricing provision and 0 otherwise. Acc_based_pp is defined as an indicator variable taking the value of one if the performance pricing provision based on an accounting ratio and 0 otherwise. Secured is defined as an indicator variable taking the value of one if the loan is secured with collateral and 0 otherwise. Term is defined as an indicator variable taking the value of one if the loan’s type belongs to term loan and 0 otherwise. Revolver is defined as an indicator variable taking the value of one if the loan’s type belongs to lines of credit and 0 otherwise. Log(Firm_size) is defined as the natural log of the borrower’s book value of the total assets. LEV is defined as the ratio of the borrower’s long-term debt to total assets. ROA is defined as the borrower’s earnings before interests and taxes (EBIT) scaled by average total assets at the time of the deal. US_listing is defined as an indicator variable taking the value of one if the borrower issue ADR and 0 otherwise.
‧
Effect of mandatory IFRS reporting on the loan syndicate structure (H1)
Model Model 1 Model 2 Model 3
Dependent Variable
Proportion_lead Herfindahal_index Num_lender
Independent Variable Coef.
t-value
Coef.t-value
Coef.t-value
1. This table examines the effect of mandatory IFRS reporting on the loan syndicate structure using the difference-in-differences approach. In model 1, Proportion_lead is regressed on the interested interaction term, Mandatory_Adopter*Post_mandatory, and control variables. In model 2, Herfindahal_index is regressed on the interested interaction term, Mandatory_Adopter*Post_mandatory, and control variables. In model 3, Num_lender is regressed on the interested interaction term, Mandatory_Adopter*Post_mandatory, and control variables. Mandatory_Adopter is defined as an indicator variable taking the value of one if a firm adopts IFRS in countries requiring mandatory IFRS reporting and 0 otherwise. Post_mandatory is defined as an indicator variable takes the value of one if a firm-year observation falls into 2005 or later and 0 otherwise. Most countries, except for Singapore, in my sample do not mandatorily adopt IFRS until 2005, when it becomes mandatory to do so. All variables are defined in appendix A.
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
71 2. Regressions include year fixed effects.
3. ***,**, and * denote significance at the 1%, 5%, and 10% levels, respectively (two-tailed test).
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
72
Table 4
Effect of mandatory IFRS reporting on the presence of foreign lenders in the loan syndicate structure (H2)
Model Model 1 Model 2
Dependent Variable
Cross_country Cross_region
Independent Variable Coef.
t-value
Coef.t-value
Intercept
-20.48 -13.56*** -9.73 -12.65***Mandatory_Adopter
2.63 6.22*** 0.10 0.47Post_mandatory
-3.08 -3.54*** -0.88 -1.99**Mandatory_Adopter*Post_mandatory
-1.47 -2.91*** -0.36 -1.43Log(Loan_size)
2.46 16.62*** 1.13 15.00***Loan_maturity
-0.14 -2.93*** -0.08 -3.34***Term
0.88 2.64*** 0.42 2.46**Revolver
0.10 0.33 -0.04 -0.29Prior_lead
0.19 0.84 0.07 0.67Syndicate_relationship
-3.84 -2.40** -1.96 -2.40**Log(Firm_size)
0.46 5.81*** 0.22 5.56***LEV
-3.37 -4.52*** -1.75 -4.61***ROA
0.09 0.05 0.80 0.86US_listing
2.50 6.53*** 1.61 8.26***Year fixed effects Yes Yes
N 2,219 2,219
Adj. R2 53.69% 42.98%
1. This table examines the effect of mandatory IFRS reporting on the presence of foreign lenders in the loan syndicate structure using the difference-in-differences approach.
In model 1, Cross_country is regressed on the interested interaction term, Mandatory_Adopter*Post_mandatory, and control variables. In model 2, Cross_region is regressed on the interested interaction term, Mandatory_Adopter*Post_mandatory, and control variables. Mandatory_Adopter is defined as an indicator variable taking the value of one if a firm adopts IFRS in countries requiring mandatory IFRS reporting and 0 otherwise. Post_mandatory is defined as an indicator variable takes the value of one if a firm-year observation falls into 2005 or later and 0 otherwise. Most countries, except for Singapore, in my sample do not mandatorily adopt IFRS until 2005, when it becomes mandatory to do so. All variables are defined in appendix A.
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
73 2. Regressions include year fixed effects.
3. ***,**, and * denote significance at the 1%, 5%, and 10% levels, respectively (two-tailed test).
‧
Effect of mandatory IFRS reporting on the adoption of financial covenants (H3)
Model Model 1 Model 2 Model 3 Model 4
Dependent Variable
Financial_covenant BS_covenant IS_covenant Relative_fin_covenant
Independent Variable Coef.
Wald Test
Coef.Wald Test
Coef.Wald Test
Coef.t-value
Intercept
3.01 5.56** 3.88 7.26*** 3.83 7.29*** 0.40 5.17***Mandatory_Adopter
0.12 0.10 0.58 1.55 0.30 0.48 -0.04 -2.19**Post_mandatory
2.64 16.93*** 3.56 25.90*** 1.98 5.41** 0.15 3.34***Mandatory_Adopter*Post_mandatory
-1.81 16.86*** -2.21 17.39*** -2.07 17.51*** -0.08 -3.06***Log(Loan_size)
0.49 17.40*** 0.38 9.05*** 0.51 16.21*** 0.03 4.42*** regression estimation of the probability that financial covenants are included in a debt contract. The dependent variable is regressed on the interested interaction term, Mandatory_Adopter*Post_mandatory, and control variables. Mandatory_Adopter is defined as an indicator variable taking the value of one if a firm adopts IFRS in countries requiring mandatory IFRS reporting and 0 otherwise. Post_mandatory is defined as an indicator variable takes the value of one if a firm-year observation falls into 2005 or later and 0 otherwise.Most countries, except for Singapore, in my sample do not mandatorily adopt IFRS until 2005, when it becomes mandatory to do so. All variables are defined in appendix A.
2. Regressions include year fixed effects.
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
75
3. ***,**, and * denote significance at the 1%, 5%, and 10% levels, respectively (two-tailed test).
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
76
Table 6
Effect of mandatory IFRS reporting conditional on the legal origin (H4a)
Model Model 1 Model 2 Model 3 Model 4 Model 5 Model 6
Dependent Variable Proportion_lead Herfindahal_index Num_lender Cross_country Cross_region Financial_covenant
Independent Variable Coef. t-value Coef. t-value Coef. t-value Coef. t-value Coef. t-value Coef. Wald
Intercept 88.62 18.01*** 8108.5 18.24*** -19.58 -9.57*** -20.39 -13.51*** -9.57 -12.60*** 3.58 7.50***
Mandatory_Adopter -2.87 -2.08** -138.6 -1.11 2.92 5.11*** 2.60 6.17*** 0.05 0.24 0.05 0.02
Post_mandatory 6.85 2.35** 644.5 2.45** -4.55 -3.76*** -3.47 -3.88*** -1.64 -3.66*** 2.04 9.63***
Mandatory_Adopter*Post_mandatory 11.20 5.47*** 916.5 4.96*** -3.90 -4.59*** -2.19 -3.50*** -1.80 -5.71*** -3.23 20.17***
Mandatory_Adopter*Post_mandatory*Common_law -8.90 -3.96*** -745.3 -3.67*** 1.95 2.10** 1.33 1.94* 2.66 7.67*** 2.17 9.25***
Log(Loan_size) -6.35 -13.16*** -358.0 -8.20*** 2.02 10.10*** 2.46 16.64*** 1.13 15.25*** 0.51 18.86***
Loan_maturity -0.34 -2.12** -15.9 -1.09 -0.02 -0.43 -0.13 -2.67*** -0.06 -2.41** 0.08 5.08**
Term -2.89 -2.63*** -336.7 -3.39*** 1.55 3.39*** 0.83 2.48** 0.31 1.88* 0.15 0.28
Revolver -0.26 -0.26 -235.7 -2.55** 0.32 0.77 0.06 0.22 -0.11 -0.75 -0.62 4.74**
Performance_pricing 1.92 41.23***
Prior_lead -0.00 -0.01 -107.5 -1.61 0.31 1.03 0.20 0.92 0.11 0.97 -0.05 0.07
Syndicate_relationship -8.69 -1.66* -1347.4 -2.85*** 1.24 0.57 -3.81 -2.38* -1.90 -2.36** -2.45 1.71
Log(Firm_size) -0.84 -3.24*** -156.5 -6.67*** 0.77 7.13*** 0.45 5.73*** 0.21 5.33*** -0.48 43.79***
LEV -1.13 -0.47 156.5 0.71 -2.72 -2.69*** -3.28 -4.39*** -1.56 -4.17*** -1.85 6.76***
ROA -0.84 -0.14 493.8 0.92 -3.03 -1.33 0.05 0.03 0.72 0.79 1.54 1.17
US_listing 0.96 0.77 75.1 0.66 2.31 4.45*** 2.55 6.64*** 1.70 8.82*** 1.29 19.11***
Year fixed effects Yes Yes Yes Yes Yes Yes
N 2,219 2,219 2,219 2,219 2,219 2,219
Adj. R2 32.00% 20.03% 37.27% 53.74% 44.45% 44.36%
1. This table examines the effect of mandatory IFRS reporting conditional on the legal origin, common law or code law, using the difference-in-differences approach. In model 1, Proportion_lead is regressed on the interested interaction term, Mandatory_Adopter*Post_mandatory*Common_law, and control variables. In model 2, Herfindahal_index is regressed on the interested interaction term, Mandatory_Adopter*Post_mandatory*Common_law, and control variables. In model 3, Num_lender is regressed on the interested interaction term, Mandatory_Adopter*Post_mandatory*Common_law, and control variables. In model 4, Cross_country is regressed on the
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
77
interested interaction term, Mandatory_Adopter*Post_mandatory*Common_law, and control variables. In model 5, Cross_region is regressed on the interested interaction term, Mandatory_Adopter*Post_mandatory*Common_law, and control variables. In model 6, Financial_covenant is regressed on the interested interaction term, Mandatory_Adopter*Post_mandatory*Common_law, and control variables. Mandatory_Adopter is defined as an indicator variable taking the value of one if a firm adopts IFRS in countries requiring mandatory IFRS reporting and 0 otherwise. Post_mandatory is defined as an indicator variable takes the value of one if a firm-year observation falls into 2005 or later and 0 otherwise. Common_law is defined as an indicator variable taking the value of one if the firm is domiciled in a country with a common-law legal system and 0 otherwise. Most countries, except for Singapore, in my sample do not mandatorily adopt IFRS until 2005, when it becomes mandatory to do so. All variables are defined in appendix A.
2. Regressions include year fixed effects.
3. ***,**, and * denote significance at the 1%, 5%, and 10% levels, respectively (two-tailed test).
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
78
Table 7
Effect of mandatory IFRS reporting conditional on the creditor right (H4b)
Model Model 1 Model 2 Model 3 Model 4 Model 5 Model 6
Dependent Variable Proportion_lead Herfindahal_index Num_lender Cross_country Cross_region Financial_covenant
Independent Variable Coef. t-value Coef. t-value Coef. t-value Coef. t-value Coef. t-value Coef. Wald
Intercept 89.34 18.16*** 8171.7 18.42*** -19.57 -9.57*** -20.27 -13.49*** -9.73 -12.65*** 3.74 7.80***
Mandatory_Adopter -2.87 -2.09** -135.6 -1.09 2.90 5.07*** 2.57 6.11*** 0.08 0.40 0.00 0.00
Post_mandatory 4.46 1.57 446.8 1.74* -3.99 -3.38*** -3.07 -3.54*** -0.89 -2.01** 2.40 11.84***
Mandatory_Adopter*Post_mandatory 12.71 5.29*** 1176.9 5.43*** -3.63 -3.64*** -1.92 -2.62*** -0.81 -2.16** -5.23 21.59***
Mandatory_Adopter*Post_mandatory*Creditor_right -2.87 -3.74*** -300.5 -4.34*** 0.37 1.18 0.23 0.99 0.20 1.67* 1.15 15.09***
Log(Loan_size) -6.40 -13.25*** -363.3 -8.34*** 2.02 10.06*** 2.45 16.62*** 1.13 15.03*** 0.54 20.39***
Loan_maturity -0.26 -1.67* -9.9 -0.69 -0.04 -0.64 -0.14 -2.86*** -0.08 -3.30*** 0.09 5.35**
Term -2.98 -2.72*** -340.4 -3.43*** 1.63 3.58*** 0.92 2.76*** 0.41 2.39** 0.12 0.18
Revolver -0.27 -0.26 -233.3 -2.53** 0.37 0.89 0.12 0.40 -0.05 -0.35 -0.64 4.90**
Performance_pricing 1.95 41.75***
Prior_lead 0.13 0.18 -94.5 -1.42 0.31 1.03 0.22 1.01 0.08 0.70 -0.09 0.21
Syndicate_relationship -8.32 -1.59 -1311.6 -2.78*** 1.18 0.55 -3.85 -2.41** -1.97 -2.41** -2.43 1.70
Log(Firm_size) -0.87 -3.39*** -159.5 -6.81*** 0.77 7.19*** 0.45 5.78*** 0.22 5.54*** -0.50 43.66***
LEV -0.72 -0.30 185.6 0.85 -2.91 -2.88*** -3.46 -4.66*** -1.75 -4.60*** -1.91 7.17***
ROA -0.39 -0.07 557.0 1.03 -3.55 -1.43 -0.27 -0.15 0.80 0.86 1.16 0.63
US_listing 1.38 1.10 112.7 1.00 2.25 4.33*** 2.52 6.59*** 1.62 8.29*** 1.34 19.68***
Year fixed effects Yes Yes Yes Yes Yes Yes
N 2,215 2,215 2,215 2,215 2,215 2,215
Adj. R2 31.97% 20.25% 36.95% 53.53% 42.95% 44.31%
1. This table examines the effect of mandatory IFRS reporting conditional on the creditor right using the difference-in-differences approach. In model 1, Proportion_lead is regressed on the interested interaction term, Mandatory_Adopter*Post_mandatory*Creditor_right, and control variables. In model 2, Herfindahal_index is regressed on the interested interaction term, Mandatory_Adopter*Post_mandatory*Creditor_right, and control variables. In model 3, Num_lender is regressed on the interested interaction term, Mandatory_Adopter*Post_mandatory*Creditor_right, and control variables. In model 4, Cross_country is regressed on the interested interaction term,
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
79
Mandatory_Adopter*Post_mandatory*Creditor_right, and control variables. In model 5, Cross_region is regressed on the interested interaction term, Mandatory_Adopter*Post_mandatory*Creditor_right, and control variables. In model 6, Financial_covenant is regressed on the interested interaction term, Mandatory_Adopter*Post_mandatory*Creditor_right, and control variables. Mandatory_Adopter is defined as an indicator variable taking the value of one if a firm adopts IFRS in countries requiring mandatory IFRS reporting and 0 otherwise. Post_mandatory is defined as an indicator variable takes the value of one if a firm-year observation falls into 2005 or later and 0 otherwise. Creditor_right is the proxy for the creditor rights protected in a country, derived annually at the country level from La Porta et al.
[1998]. Most countries, except for Singapore, in my sample do not mandatorily adopt IFRS until 2005, when it becomes mandatory to do so. All variables are defined in appendix A.
2. Regressions include year fixed effects.
3. ***,**, and * denote significance at the 1%, 5%, and 10% levels, respectively (two-tailed test).
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
80
Table 8
Effect of mandatory IFRS reporting on other attributes of a debt contract
Model Model 1 Model 2 Model 3
Dependent Variable
Log(Loan_size) Loan_maturity Spread
Independent Variable Coef.
t-value
Coef.t-value
Coef.t-value
Intercept
-0.39 -1.84* 1.13 1.74* 0.314 8.79***Mandatory_Adopter
2.04 48.08*** 0.92 5.10*** -0.016 -1.67*Post_mandatory
0.28 2.23** -0.61 -1.64 0.012 0.62Mandatory_Adopter*Post_mandatory
-0.40 -5.58*** -1.22 -5.62*** 0.012 1.13Log(Loan_size)
0.23 3.67*** 0.013 3.54***Loan_maturity
0.02 3.67*** 0.004 3.95***Term
0.04 0.96 2.43 17.89*** -0.008 -1.06Revolver
0.07 1.58 0.28 2.09** -0.028 -3.90***Prior_lead
-0.06 -1.88* -0.48 -4.99*** -0.012 -2.24**Syndicate_relationship
-0.50 -2.21** -0.65 -0.94 -0.058 -1.15Log(Firm_size)
0.29 31.21*** -0.07 -2.05** -0.015 -8.02***LEV
0.38 3.55*** 2.15 6.71*** 0.042 2.46**ROA
1.38 5.28*** 2.88 3.65*** -0.088 -2.28**US_listing
0.23 4.23*** 0.10 0.64 0.005 0.71Year fixed effects Yes Yes Yes
N 2,219 2,219 1,366
Adj. R2 78.41% 31.33% 22.93%
1. This table examines the effect of mandatory IFRS reporting on other attributes of a debt contract using the difference-in-differences approach. In model 1, Log(Loan_size) is regressed on the interested interaction term, Mandatory_Adopter*Post_mandatory, and control variables. In model 2, Loan_maturity is regressed on the interested interaction term, Mandatory_Adopter*Post_mandatory, and control variables. In model 3, Spread is regressed on the interested interaction term, Mandatory_Adopter*Post_mandatory, and control variables. Mandatory_Adopter is defined as an indicator variable taking the value of one if a firm adopts IFRS in countries requiring mandatory IFRS reporting and 0 otherwise. Post_mandatory is defined as an indicator variable takes the value of one if a firm-year observation falls into 2005 or later and 0 otherwise. Most countries, except for Singapore, in my sample do not mandatorily adopt IFRS until 2005, when it becomes mandatory to do so. All variables are
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
81 defined in appendix A.
2. Regressions include year fixed effects.
3. ***,**, and * denote significance at the 1%, 5%, and 10% levels, respectively (two-tailed test).