Bank relationships and their effects on firm performance around the Asian financial crisis: evidence from Taiwan
Table I. Summary Statistics
This table shows the summary statistics of the bank relationship
variables and related firm characteristics. N is the number of
observations. ROA is return on assets. NUM is the number of bank
relationships, i.e., the number of banks from which a firm borrows.
FNUM and DNUM are the number of relationships with foreign banks
and domestic banks, respectively. BLR is the ratio of bank loans
to total loans. MULTIPLE equals one for firms with multiple bank
relationships, and zero otherwise. FOREIGN equals one for firms
with foreign bank relationships, and zero otherwise. REP is a proxy
for bank quality. It equals 1,008 minus the weighted average of the
lending banks' ranking in The Banker magazine's top 1,000 list. A
higher value of REP indicates better quality. AGE is the number of
years since the firm was founded. ASSET is the firm's total assets
in thousands of dollars (NT dollars). DEP is the ratio of depreciation
expense to total sales, a proxy for growth opportunities. MNC equals
one for multinational corporations based on a list prepared by the
Ministry of Economic Affairs, and zero otherwise. LEVERAGE is the
debt-to-total assets ratio. BOARD equals one if lending banks are
represented on the board of directors, and zero otherwise. CEO equals
one if the chief executive officer is also the chairman of the board,
and zero otherwise. FAMILY equals one for firms controlled by a family
group, and zero otherwise. GROUP equals one if the firm belongs to an
industrial group, and zero otherwise. PUBLIC equals one for firms with
issued public debt, and zero otherwise. ICR is the interest coverage
ratio. The last column reports the t-statistics (z-statistics) of
differences between means (medians) of the variables between the
pre-crisis (1994-1996) and the crisis period (1997-1998).
1994-1996
(N = 534)
Variable Mean Median Std. Dev.
ROA (%) 5.58 5.26 5.89
NUM 9.47 8 7.37
FNUM 2.98 1 4.15
DNUM 6.49 5 4.85
BLR (%) 34.00 34.10 19.00
MULTIPLE 0.94 1 0.238
FOREIGN 0.63 1 0.483
REP 743 735 102.48
AGE 26.76 27 9.59
ASSET 12,445,931 6,047,712 19,760,752
DEP (%) 18.75 17.10 11.76
MNC 0.746 1 0.436
LVRG. (%) 41.20 40.65 13.57
BOARD 0.072 0 0.258
CEO 0.181 0 0.385
FAMILY 0.527 1 0.499
GROUP 0.718 1 0.451
PUBLIC 0.286 0 0.452
ICR 9.252 3.92 2.83
1997-1998
(N = 356)
Variable Mean Median Std. Dev.
ROA (%) 3.28 3.91 8.71
NUM 9.69 8 7.65
FNUM 2.14 1 3.48
DNUM 7.55 6 5.98
BLR (%) 31.80 30.10 20.00
MULTIPLE 0.92 1 0.265
FOREIGN 0.537 1 0.499
REP 733 739 108.11
AGE 29.26 29 9.57
ASSET 17,621,264 8,633,233 26,652,434
DEP (%) 19.66 17.50 14.11
MNC 0.746 1 0.436
LVRG. (%) 41.96 41.86 14.76
BOARD 0.051 0 0.220
CEO 0.181 0 0.385
FAMILY 0.534 1 0.450
GROUP 0.718 1 0.451
PUBLIC 0.438 1 0.497
ICR 9.06 2.75 3.49
Variable t-Statistic Z-Statistic
ROA (%) -4.347 *** -0.637 ***
NUM 0.424 0.156
FNUM -3.271 *** -2.905 ***
DNUM 2.786 *** 1.415
BLR (%) -1.861 * -1.370
MULTIPLE -0.914 ** -0.935
FOREIGN -1.163 -2.735 ***
REP -0.313 0.274
AGE 3.812 *** 3.379 ***
ASSET 3.134 *** 4.513 ***
DEP (%) 1.008 0.410
MNC -- --
LVRG. (%) 0.776 1.231
BOARD -1.280 -1.239
CEO -- --
FAMILY 1.923 0.192
GROUP -- --
PUBLIC 4.707 *** 4.639 ***
ICR -0.086 -3.178 ***
*** Significant at the 0.01 level.
** Significant at the 0.05 level.
* Significant at the 0.10 level.
Table II. Firm Performance for Firms with High Values
and Low Values of Bank Relationship Variables
This table shows the average return on assets for two groups of firms.
We classify subgroups depending on the mean or median of each bank
relationship variable. A firm belongs to the high (low) group if its
bank relationship variable is higher than (lower than or equal to)
the sample mean or median. NUM is the number of bank relationships,
i.e., the number of banks from which a firm borrows. FNUM and DNUM
are the number of relationships with foreign and domestic banks,
respectively. BLR is ratio of bank loans to total loans. REP is a
proxy for bank quality. It equals 1,008 minus the weighted average
of the lending banks' ranking in The Banker magazine's top 1,000
list. The values are the average return on assets for the groups.
The t-statistics are for the mean difference in return on assets
between groups. Numbers in parentheses are numbers of observations.
Pre-Crisis Period (1994-1996)
Low High t-statistic
NUM
By mean 5.64% (321) 5.49% (213) 0.29
By median 5.68% (304) 5.45% (230) 0.43
FNUM
By mean 4.95% (336) 6.66% (198) -3.27 *
By median 4.77% (280) 6.47% (254) -3.38 *
DNUM
By mean 6.32% (319) 4.48% (215) 3.57 *
By median 6.61% (273) 4.50% (261) 4.16 *
BLR
By mean 6.08% (269) 5.08% (265) 1.98 ***
By median 6.08% (268) 5.07% (266) 1.96 ***
REP
By mean 4.83% (270) 6.34% (264) -2.97 *
By median 4.81% (268) 6.35% (266) -3.04 *
Crisis Period (1997-1998)
Low High t-statistic
NUM
By mean 4.11% (204) 2.15% (152) 1.99 **
By median 4.23% (191) 2.17% (163) 2.17 **
FNUM
By mean 2.97% (251) 4.04% (103) -0.88
By median 2.69% (219) 4.23% (135) -1.51
DNUM
By mean 4.51% (215) 1.38% (139) 3.07 ***
By median 4.56% (190) 1.80% (164) 2.93 ***
BLR
By mean 4.28% (183) 2.21% (171) 2.26 **
By median 4.99% (176) 1.60% (178) 3.74 ***
REP
By mean 1.43% (169) 4.97% (185) -3.83 ***
By median 1.47% (179) 5.30% (175) -3.83 ***
*** Significant at the 0.01 level.
** Significant at the 0.05 level.
* Significant at the 0.10 level.
Table III. Simultaneous Equation Estimates of the Relation Between
Choice of Single- or Multiple-Bank Relationships and Firm Performance
This table shows two-stage estimates of the relation between firm
performance and the choice of single- or multiple-bank relationships. N
is the number of observations. In the first stage, an OLS regression
relates the return on assets to a set of exogenous variables. In
addition, we use a probit regression to regress the bank relationship
dummy (MULTIPLE = 1 for multiple-bank relationships, and zero
otherwise) on the same set of exogenous variables using a probit
regression. We then use the fitted values obtained from the first
stages' probit regression (Fitted MULTIPLE) and OLS (Fitted ROA) as the
explanatory variables in the second-stage estimations. AGE is the
natural log of the number of years since the firm was founded. SIZE is
the natural log of the firm's total assets. DEP is the ratio of
depreciation expense to total sales, a proxy for growth opportunities.
MNC equals one for multinational corporations, and zero otherwise.
LEVERAGE is the debt-to-total-assets ratio. BOARD equals one if
lending banks are represented on the board of directors, and zero
otherwise. CEO equals one if the chief executive officer is also the
chairman of the board, and zero otherwise. FAMILY equals one for firms
controlled by a family group, and zero otherwise. GROUP equals one if
the firm belongs to an industrial group, and zero otherwise. PUBLIC
equals one for firms with issued public debt, and zero otherwise.
ICR is the interest coverage ratio. LBLLP is the loan loss provision
ratio of the largest lending bank. LBLEV and LBROE are the debt-equity
ratio and return on equity of the largest lending bank,
respectively. t-statistics are in parentheses.
Model 1
1994-1996 1997-1998
(N = 534) (N = 356)
ROA MULTIPLE ROA MULTIPLE
Variable (OLS) (Probit) (OLS) (Probit)
Intercept 2.89 -2.19 -0.74 -4.68 **
(1.14) (-1.50) (-0.14) (2.18)
Fitted MULTIPLE 0.07 -0.46
(0.20) (-0.63)
Fitted ROA 0.10 0.01
(0.70) (0.09)
AGE -1.71 *** 0.08 -0.60 -0.34
(-2.88) (0.23) (-0.50) -0.87
SIZE 1.90 *** 0.03 2.13 *** 0.67 *
(7.15) (0.12) (3.79) (2.58)
DEP -0.09 *** -0.11 ***
(-4.53) (-3.79)
MNC 0.16 0.03
(0.61) (0.11)
LEVERAGE -17.09 *** 4.63 ** -25.51 *** 2.44
(-8.33) (2.01) (-8.21) (1.08)
BOARD 1.56 -0.41 -1.28 0.22
(1.64) (-0.90) (-0.58) (0.32)
CEO -0.46 1.10
(-0.75) (0.99)
FAMILY 0.62 ** 0.87 ***
(2.73) (2.94)
GROUP 0.29 -0.32
(1.23) (-1.10)
PUBLIC 0.02 -0.43
(0.07) (-1.26)
ICR -0.01 -0.01
(-1.00) (-1.17)
LBLLP -0.07 0.10
(-0.40) (0.41)
LBLEV 0.03 * (0.02)
(1.87) (1.06)
LBROE 0.01 (-0.00)
(0.79) (-0.16)
Adjusted [R.sup.2] 0.24 0.24
F-statistic 24.20 16.16
Pseudo [R.sup.2] 0.74 0.69
Log-Likelihood -95.94 -74.74
*** Significant at the 0.01 level.
** Significant at the 0.05 level.
* Significant at the 0.10 level.
Table IV. Simultaneous Equation Estimates of Relation Between Firm
Performance and Five Bank Relationship Variables (1994-1996)
This table shows 2SLS estimates of the relation between firm
performance and five bank relationship variables for the period
1994-1996. N is the number of observations. ROA is return on assets.
LNUM is the natural log of the number of total bank relationships.
LFNUM and LDNUM are the natural log of the number of relationships with
foreign and domestic banks, respectively. BLR is the ratio of bank
loans to total loans. LREP is a proxy for bank quality. It equals the
natural log of 1,008 minus the weighted average of the lending banks'
ranking in The Banker magazine's top 1,000 list. A higher value of LREP
indicates better quality. AGE is the natural log of the number of years
since the firm was founded. SIZE is the natural log of the firm's total
assets. DEP is the ratio of depreciation expense to total sales, a
proxy for growth opportunities. MNC equals one for multinational
corporations, and zero otherwise. LEVERAGE is the debt-to-total-assets
ratio. BOARD equals one if lending banks are represented on the board
of directors, and zero otherwise. CEO equals one if the chief executive
officer is also chair of the board. FAMILY equals one for firms
controlled by a family group, and zero otherwise. GROUP equals one if
the firm belongs to an industrial group, and zero otherwise. PUBLIC
equals one for fines with issued public debt, and zero otherwise. ICR
is the interest coverage ratio. LBLLP is the loan loss provision ratio
of the largest lending bank. LBLEV and LBROE are the debt-equity ratio
and return on equity of the largest lending bank, respectively.
t-statistics are in parentheses.
Model 2 Model 3
(N = 534) (N = 534)
ROA LNUM ROA LFNUM
Intercept 1.77 -0.54 6.18 ** -1.41 *
(0.63) (-1.44) (1.98) (-1.72)
ROA 0.07 * -0.24 ***
(-1.95) (-2.90)
LNUM -1.67
(-0.75)
LFNUM 1.58 *
(1.77)
LDNUM
BLR
LREP
AGE -1.69 *** -0.13 -1.56 *** -0.52 **
(-2.85) (-1.29) (-2.62) (-2.36)
SIZE 2.21 *** 0.34 *** 1.33 *** 0.70 ***
(4.64) (4.55) (3.19) (4.29)
DEP -0.08 *** -0.11 ***
(-3.92) (-4.71)
MNC 0.17 ** 0.55 **
(1.97) (2.90)
LEVERAGE -15.36 *** 0.11 -14.22 *** -4.17 ***
(-5.75) (0.17) (-9.13) (-3.08)
BOARD 1.70 * 0.20 1.34 0.42
(1.75) (1.37) (1.40) (1.31)
CEO -0.50 -0.22
(-0.84) (-0.36)
FAMILY 0.13 ** 0.09
(2.06) (0.68)
GROUP 0.01 -0.22
(0.15) (-1.33)
PUBLIC 0.15 ** 0.26
(1.92) (1.49)
ICR 0.00 0.01 **
(1.28) (2.53)
LBLLP 0.00 -0.04
(0.04) (-0.38)
LBLEV 0.00 0.01 ***
(0.47) (2.74)
LBROE 0.01 *** 0.02 **
(3.18) (2.54)
Adjusted [R.sup.2] 0.24 0.22 0.24 0.07
F-statistic 24.30*** 12.44 *** 24.57 *** 4.11 ***
Hausman -0.4 -5.16 *** 0.97 -1.08
t-statistic
Model 4 Model 5
(N = 534) (N = 534)
ROA LDNUM ROA BLR
Intercept 0.96 -0.42 20.2 *** 0.71 ***
(0.36) (-1.32) (3.06) (5.30)
ROA -0.01 -0.04 ***
(-0.25) (2.66)
LNUM
LFNUM
LDNUM -3.69 **
(-2.44)
BLR -21.87 ***
(-2.94)
LREP
AGE -1.58 *** 0.03 -2.75 *** -0.11 ***
(-2.59) (0.34) (-3.36) (-3.13)
SIZE 2.50 *** 0.16 ** 0.71 0.03
(7.05) (2.53) (1.40) (1.33)
DEP -0.09 *** -0.02
(-4.25) (-0.59)
MNC -0.07 0.02
(-0.95) (0.75)
LEVERAGE -10.25 *** 1.67 *** -10.79 *** -0.25
(-3.16) (3.20) (-3.66) (-1.21)
BOARD 2.12 ** 0.18 2.13 * 0.01 *
(2.11) (1.44) (1.79) (1.96)
CEO -0.12 -0.98
(-0.19) (-1.27)
FAMILY 0.14 *** 0.02
(2.60) -1.03
GROUP 0.04 -0.07 **
(0.64) (-2.42)
PUBLIC 0.12 * -0.05 ***
(1.81) (-1.87)
ICR -0.00 0.00
(-0.25) (1.07)
LBLLP 0.03
(0.66)
LBLEV -0.01 ***
(-4.15)
LBROE 0.01 *
(1.69)
Adjusted [R.sup.2] 0.23 0.27 0.17 0.08
F-statistic 23.76 *** 15.87 *** 16.85 *** 5.83 ***
Hausman -1.73 * -6.52 *** -3.81 *** -5.51 ***
t-statistic
Model 6
(N = 534)
ROA LREP
Intercept -18.22 ** 6.57 ***
(-2.02) (5.87)
ROA -0.03 ***
(-2.75)
LNUM
LFNUM
LDNUM
BLR
LREP 2.86 **
(2.05)
AGE 1.26 * -0.07 **
(-1.67) (-2.20)
SIZE 1.10 ** 0.07 ***
(2.26) (3.35)
DEP -0.15 ***
(-4.01)
MNC 0.09 ***
(3.18)
LEVERAGE -8.98 ** -0.73 ***
(-2.07) (-4.18)
BOARD 1.29 0.03
(1.13) (0.82)
CEO 0.02
(0.03)
FAMILY -0.01
(-0.30)
GROUP -0.01
(-0.57)
PUBLIC -0.00
(-0.06)
ICR 0.01 *
(2.23)
LBLLP
LBLEV
LBROE
Adjusted [R.sup.2] 0.17 0.04
F-statistic 16.79 *** 3.21***
Hausman 2.33 ** 2.99***
t-statistic
*** Significant at the O.Ot level
** Significant at the 0.05 level.
* Significant at the 0.10 level.
Table V. Simultaneous Equation Estimates of Relation Between Firm
Performance and Five Bank Relationship Variables (1997-1998)
This table shows 2SLS estimates of the relation between firm
performance and five bank relationship variables for the period
1997-1998. N is the number of observations. ROA is return on assets.
LNUM is the natural log of the number of total bank relationships.
LFNUM and LDNUM are the natural log of the number of relationships
with foreign and domestic banks, respectively. BLR is the ratio of
batik loans to total loans. LREP is a proxy for bank quality. It equals
the natural log of 1,008 minus the weighted average of the lending
banks' ranking in The Banker magazine's top 1,000 list. A higher value
of LREP indicates better quality. AGE is the natural log of the number
of years since the firm was founded. SIZE is the natural log of the
firm's total assets. DEP is the ratio of depreciation expense to total
sales, a proxy for growth opportunities. MNC equals one for
multinational corporations, and zero otherwise. LEVERAGE is the
debt-to-total-assets ratio. BOARD equals one if lending banks are
represented on the board of directors, and zero otherwise. CEO equals
one if the chief executive officer is also chair of the board. FAMILY
equals one for firms controlled by a family group, and zero otherwise.
GROUP equals one if the firm belongs to an industrial group, and zero
otherwise. PUBLIC equals one for firms with issued public debt, and
zero otherwise. ICR is the interest coverage ratio. LBLLP is the loan
loss provision ratio of the largest lending bank. LBLEV and LBROE are
the debt-equity ratio and return on equity of the largest lending bank,
respectively. t-statistics are in parentheses.
Model 2 (N = 356) Model 3 (N = 356)
ROA LNUM ROA LFNUM
Intercept -4.97 -1.17 ** 7.44 -2.09 **
(-0.80) (-2.33) (1.26) (-2.13)
ROA -0.02 -0.14 ***
(-0.88) (-2.88)
LNUM -6.02 *
(-1.93)
LFNUM 4.57 **
(2.82)
LDNUM
BLR
LREP
AGE -0.14 0.02 0.40 -0.15
(-0.11) (-0.11) (0.31) (-0.72)
SIZE 3.33 *** 0.30 *** 0.57 0.42 ***
(3.80) (5.02) (0.85) (3.53)
DEP -0.10 *** -0.17 ***
(-2.96) (-4.59)
MNC 0.06 0.27
(0.68) (1.62)
LVG -17.47 *** 0.89 -26.09 *** -3.37 ***
(-3.17) (-1.47) (-8.79) (-2.70)
BOARD -1.35 -0.05 -1.16 -0.54
(-0.57) (-0.26) (-0.51) (-1.40)
CEO 0.77 1.71
(0.66) (1.47)
FAMILY 0.18 ** 0.20
(-2.41) (1.38)
GROUP -0.02 -0.05
(-0.17) (-0.30)
PUBLIC -0.04 0.46 **
(-0.64) (2.57)
ICR -0.00 0.01
(-1.32) (1.59)
LBLLP 0.02 0.18
(0.40) (1.59)
LBLEV -0.00 0.01 *
(1.53) (1.84)
LBROE -0.00 0.03 ***
(-0.00) (3.23)
Adjusted [R.sup.2] 0.21 0.22 0.23 0.07
t-statistic 14.20 *** 8.58 *** 15.82 *** 2.92 ***
Hausman -1.81 * -4.33 *** 2.40 ** -2.31 **
t-statistic
Model 4 (N = 356) Model 5 (N = 356)
ROA LDNUM ROA BLR
Intercept -5.80 -0.81 12.51 * 0.48 *
(-0.93) (-1.56) (-1.75) (3.27)
ROA 0.02 -0.01 ***
(0.70) (-1.82)
LNUM
LFNUM
LDNUM -9.06 ***
(-3.37)
BLR -19.22 ***
(-2.65)
LREP
AGE 0.78 0.08 -0.90 -0.04
(0.54) (0.78) (-0.70) (-1.06)
SIZE 3.31 *** 0.21 *** 1.04 *** 0.01
(4.95) (3.32) (1.73) (0.35)
DEP -0.12 *** -0.08 **
(-3.52) (-2.30)
MNC -0.07 0.00
(-0.82) (0.11)
LVG -10.85 * 1.98 *** -19.53 *** -0.02
(-1.91) (-3.19) (-4.92) (-0.08)
BOARD -1.18 0.18 -1.01 0.01
(-0.46) (0.88) (-0.45) (-0.21)
CEO 1.31 0.89
(1.04) (0.78)
FAMILY 0.13 * 0.03
(1.70) (1.32)
GROUP -0.03 -0.05 *
(-0.29) (-1.86)
PUBLIC -0.20 ** -0.08 ***
(-2.12) (-2.84)
ICR -0.01 ** -0.00
(-2.23) (-1.54)
LBLLP -0.05
(-0.77)
LBLEV -0.00
(-0.38)
LBROE -0.01 **
(-2.48)
Adjusted [R.sup.2] 0.20 0.22 0.22 0.16
t-statistic 13.35 *** 8.28 *** 15.16 *** 7.55 ***
Hausman -3.39 *** -4.04 *** -3.74 *** -4.23 ***
t-statistic
Model 6 (N = 356)
ROA LREP
Intercept -31.33 *** 6.49 ***
(-1.91) (4.43)
ROA -0.02 **
(-2.12)
LNUM
LFNUM
LDNUM
BLR
LREP 4.87 *
(-1.91)
AGE -1.14 0.01
(-0.75) (0.36)
SIZE 0.97 0.04 *
(1.25) (1.90)
DEP -0.18 ***
(-3.48)
MNC 0.06 **
(-2.09)
LVG -15.20 * -0.63 ***
(-2.23) (-3.29)
BOARD -0.25 -0.05
(-0.09) (-0.78)
CEO 2.84 ***
(1.67)
FAMILY -0.01
(-0.48)
GROUP -0.01
(-0.35)
PUBLIC 0.05 **
(-1.70)
ICR 0.00
(1.55)
LBLLP
LBLEV
LBROE
Adjusted [R.sup.2] 0.16 0.03
t-statistic 10.83 *** 2.14 **
Hausman 0.44 1.41
t-statistic
*** Significant at the 0.01 level.
** Significant at the 0.05 level.
* Significant at the 0.10 level.
Table VI. Percentage of Total Foreign Bank Loans by Country
This table shows the ratio of a particular foreign country banks' loans
to the total foreign bank loans (in terms of dollar value) for
Taiwanese firms between 1994 and 1998. Numbers of loans provided by
banks in different countries are in parentheses. The last two columns
show the average for the pre-crisis period (1994-1996) and the crisis
period (1997-1998). The ratio for Europe is equal to the sum of the
ratios for Belgium, France, Germany, the Netherlands, Switzerland,
and the UK Others includes banks from Australia, New Zealand, Hong
Kong, and South Africa. The last row (Foreign BL/Total BL) indicates
the ratio of foreign bank loans to total bank loans for our sample of
Taiwan firms. Totals may not equal 100% for reasons of rounding.
Country 1994 1995 1996 1997 1998 Pre-Crisis Crisis
Belgium 0.47% 1.15% 0.94% 0.96% 1.55% 0.85% 1.26%
(11) (19) (26) (30) (43) (56) (73)
Canada 8.25% 9.57% 10.58% 10.44% 7.45% 9.47% 8.95%
(141) (179) (162) (127) (101) (482) (228)
France 26.00% 19.60% 15.34% 11.04% 10.30% 20.13% 10.67%
(274) (261) (201) (138) (169) (736) (307)
Germany 2.30% 3.21% 5.61% 5.19% 3.30% 3.71% 4.25%
(17) (20) (18) (15) (15) (55) (30)
Japan 7.10% 7.74% 8.95% 10.14% 6.06% 7.93% 8.10%
(72) (92) (108) (131) (139) (272) (270)
Netherlands 12.86% 14.84% 17.09% 14.85% 9.56% 14.93 12.21%
(145) (182) (161) (143) (162) (488) (305)
Singapore 1.42% 1.09% 1.25% 1.39% 1.08% 1.25% 1.24%
(38) (32) (34) (37) (40) (104) (77)
Switzerland 2.43% 2.52% 3.19% 3.98% 7.20% 2.72% 5.59%
(11) (18) (19) (21) (17) (48) (38)
UK 8.32% 11.69% 9.39% 9.60% 10.63% 9.80% 10.12%
(100) (108) (99) (119) (137) (307) (256)
US 25.23% 21.95% 21.54% 27.74% 37.40% 22.91% 32.57%
(264) (299) (241) (223) (254) (804) (477)
Others 5.60% 6.82% 5.89% 4.71% 5.45% 6.10% 5.08%
(103) (104) (95) (83) (101) (302) (184)
Europe 52.38% 53.01% 51.59% 45.62% 42.54% 52.33% 44.08%
(558) (608) (524) (466) (543) (1090) (1009)
Foreign BL/ 14.78% 17.45% 15.97% 16.57% 14.86% 16.07% 15.72%
Total BL