4. Research Results
4.1 Descriptive Analyses
Table 2 provides the distribution of firms in my sample set. These sample firms distribute across fifteen industries and for the period 2000- 2009. Firms listed in TWSE and GTSM is increasing yearly. More than a half (about 57.77% or 5,008 firm-year observations) of the sample comes from electronics industry, consistent with the fact that electronic industry is the most important part of manufactures. 580 observations (about 6.69%) are from chemical industry, which is the second largest industry in my sample. The third and fourth largest industry is the electric machinery (about 6.52%) and textile (about 6.39%) industry. Four smallest percentages of sample firms comes from tourism, rubber, paper & pulp and automobile, each of which is less than 1%.
Table 3 presents the yearly distribution of conference calls. From the call firm column, there is an increasing trend from 2000 to 2005 and reaches the highest in year 2005. Although the number of firms that hold conference calls slightly decreases Taiwanese firms would hold conference calls overseas or attend calls that are invited by foreign security brokers, either of which are excluded from our sample. While year 2005 has the largest number of firms that hold conference calls, year 2006
18 http://www.fsc.gov.tw/ch/
‧
國立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
31
represents the highest frequency of conference calls, i.e. 321 calls within a year. A stable trend of conference calls lasts after 2006. For firms that hold conference calls, the average call per year is around 2 during the last four years in my sample period.
The relative stability of these numbers implies that the frequency of hold conference calls by listed firms in Taiwan has been relatively constant in recent years.
Table 4 illustrates the distribution of conference calls in different industries.
Around 89.56% of the calls are held by firms in the electronic industry. The second and third largest number of conference calls are in the electric machinery (2.15%) and the chemical (1.73%) industry. Less than 8% of conference calls come from other industries. This distribution matches the sample firm distribution in Table 2.
Conference calls held by firms in the electronic industry is up to around 90% in our samples, a situation consistent with the argument held in Chin et al. (2008). They indicate that firms in the electronics sector communicate corporate information through conference calls more often than other industries since their financial statements are unable to reflect their operation. In Tables 3 and 4, there are 1,288 firms that hold conference calls and 7,381 firms that do not. In percentage terms, 14.86% of the sample firms hold conference calls while 85.14% do not.
‧
國立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
32
Table 2. Sample distribution
Industry code Industry 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 Total
A Cement 11 10 10 10 9 9 9 10 10 9 97
B Plastic 18 19 21 21 23 21 21 25 25 26 220
C Textile 53 58 57 56 55 53 56 56 55 55 554
D Electric Machinery 38 43 48 54 58 62 63 64 66 69 565
E Chemical 33 41 44 50 61 65 67 69 72 78 580
F Paper & Pulp 7 5 7 6 6 6 6 7 7 6 63
G Iron & Steel 30 33 32 30 33 33 38 41 44 43 357
H Rubber 7 7 7 8 8 8 10 10 10 10 85
I Automobile 2 2 2 2 3 3 3 3 3 3 26
J Electronic 214 269 375 449 524 573 602 642 665 695 5,008
K Building Material and Construction 38 41 40 39 41 41 42 44 42 47 415
L Shipping and Transportation 16 19 18 18 15 15 15 15 15 16 162
M Tourism 6 7 7 9 9 9 8 10 10 10 85
N Trading and Consumers' Goods 7 9 11 10 10 12 13 13 13 12 110
O General 25 28 31 36 36 38 35 38 38 37 342
TOTAL 505 591 710 798 891 948 988 1,047 1,075 1,116 8,669
‧
Table 3. Yearly distribution of conference calls
Year Number of call Firm
a. The number of firms that hold conference calls. For example, the number of firms that hold conference calls in 2000 is 42+8+1+1=
52
b. The number of calls that are held in a year and is calculated based on call firms. For example, the number of calls that are held in 2000 is 42*1+8*2+1*3+1*4= 65
c. The average calls that is held by the conference call firm. In other words, for firms that hold conference calls, average calls is calculated by firm-year calls divided by call firms.
‧
Table 4. Industrial distribution of conference calls Industry
a. Firms that hold conference calls in each industry.
b. The number of calls that firms held in each industry during 2000-2009. For example, there are 11(=1*5+2*3) calls in industry A (Cement Industry) during 2000-2009.
c. The percentage of calls attributed to each industry. For example, 0.46% (=11/2367) of calls come from industry A (Cement Industry).
‧
國立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
35
Tables 5 and 6 report the yearly and industrial distribution of the number of other call firms linked to the focal firm. From table 5, there is an increasing trend in the number of firms connected to other call firms during the sample period and reaches the highest in year 2006 (271 firms, 27.43%). The number of firms connected to other call firms slightly decrease after year 2006 but the trend remains stable. While year 2006 has the most firms connected to other call firms (271 focal firms), year 2007 has the most call firms (720 call firms) connected to the focal firms. A slightly decrease but stable trend is also maintained after year 2007. The average number call firms connected to each focal firm is 2.67 in year 2007, the highest during the 10-year sample period. Overall, the average number of call firms linked to the focal firm during the sample period is 2.21. In other words, 2.21 call firms are connected to each focal firm during 2000- 2009. Table 6 shows that electronic industry has the most number of firms linked to other call firms and most number of call firms connected to focal firms, consistent with the industrial environment in Taiwan. On the contrary, paper & pulp industry has the fewest number of firms linked to other call firms and the fewest number of call firms connected to the focal firm. In percentage, 29.43% of firms in electronic industry are connected to other call firms through interlocked directors— the highest among 15 industries, while 3.17% of firms in paper & pulp industry are connected to other call firms— the lowest among 15 industries. Automobile industry has the highest average number of call firms connected to each focal firm (3.83 call firms), while the rubber industry has the lowest average number (1.25 call firms).
Tables 7 and 8 further indicate the yearly and industrial distribution of board interlocking ties through two different board positions. In table 7, the percentage that firms linked to other call firms through independent directors is decreasing from year 2000 and reaches the lowest in year 2005 (45.88%). After year 2005, the trend goes up in the last three years in
‧
國立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
36
the sample period. Table 8 shows that, all the firms connected to other call firms are through independent directors in cement and trading and consumers’ goods industry. None of the firm is connected to other call firms through independent directors in automobile industry. Overall, for firms connected to other call firms, 55.09% are connected through independent directors.
‧
Table 5. Yearly distribution of the number of other call firms linked to the focal firm
Year Number of call firms Firm
a. The number of firms that are linked to other call firms through board interlocks during the year. For example, 19 (= 11+6+2) firms are connected to at least 1 other firm that holds conference calls in year 2000.
b. The total number of other call firms linked to the focal firm through board interlock. For example, in year 2000, 11 firms connected to 1 other call firm, 6 firms connected to 2 others and 2 firms connected to 3 others. 29= 11*1+6*2+2*3
c. The average number of other firms that hold conference calls and are tied to the focal firms. For example, for each firm that is linked to other call firms, 1.53 (=29/19) other call firms on average are connected to the focal firm in year 2000.
‧
Table 6. Industrial distribution of the number of other call firms linked to the focal firm Industry
code
Number of call firms Firm
Industry
a. The number of firms that are linked to other call firms through board interlocks in each industry during 2000-2009. For example, 14 (= 8+3+1+2) firms are linked to at least 1 other firm that holds conference calls in industry A (Cement Industry) during 2000-2009.
b. The total number of other call firms linked to the focal firm through board interlocks. For example, in industry A (Cement Industry), 8 firms link to 1 other call firm, 3 firms link to 2 others, 1 firm link to 3 others and 2 firms link to 4 others. 29= 11*1+6*2+2*3 c. The average number of other call firms tied to the focal firms. For example, for each firm that is linked to other call firms in
‧
國立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
39
industry A (Cement Industry), 1.79 (=25/14) other call firms on average are connected to the focal firm.
‧
國立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
40
Table 7. Yearly distribution of independent director links
Year Firms linked to no other call firms
Firms linked to other
call firms Firm year
Firms linked to other call firms
Through independent director Through non-independent directors No. of firms Percentage No. of firms Percentage
2000 486 19 505 16 84.21% 3 15.79%
2001 560 31 591 26 83.87% 5 16.13%
2002 609 101 710 68 67.33% 33 32.67%
2003 622 176 798 100 56.82% 76 43.18%
2004 683 208 891 98 47.12% 110 52.88%
2005 693 255 948 117 45.88% 138 54.12%
2006 717 271 988 131 48.34% 140 51.66%
2007 777 270 1,047 146 54.07% 124 45.93%
2008 809 266 1,075 152 57.14% 114 42.86%
2009 847 269 1,116 174 64.68% 95 35.32%
Total 6,803 1,866 8,669 1,028 55.09% 838 44.91%
‧
Table 8. Industrial distribution of independent director links Industry
code
Firms linked to no other call firms
Firms linked to
other call firms Firm industry
Firms linked to other call firms
Through independent director Through non-independent directors No. of firms Percentage No. of firms Percentage
A 83 14 97 14 100.00% 0 0.00%
‧
國立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
42
Table 9 provides the descriptive statistic of the main variables. To control for extreme values, all control variables are winsorized at the top and bottom one percent of observations. The median and third quartile of CALL and NCALL are both zero, consistent with the distribution in Tables 3 and 4 that only 1,288 (14.86%) firms hold conference calls while 7,381 (85.14%) firms do not. The mean of NCALL is 0.251, indicating that each firm in the sample holds 0.251 calls. Turning to board interlocks, the median and third quartile of DLINK and LINK are zero, consistent with the distribution in Table 5 and 6 that only 1,866 (21.52%) firms are linked to other call firms. The mean of LINK is 0.475, demonstrating that each firm is connected to 0.475 call firms averagely. The mean of INDLINK is 0.119, pointing out that 11.9%
of the sample firms are connected to other call firms through independent directors.
Regarding other control variables, the mean (median) of MB is 1.557 (1.220).
The mean (median) value of SIZE is 14.941 (14.791). The mean (median) value of LEV is 0.428 (0.435). The mean (median) value of ROA is 0.035 (0.04). The mean (median) value of directors’ shareholding (DSHARE), shareholdings of the largest ten shareholders (LSHARE) and institutional shareholdings (ISHARE) are 24.438 (21.660), 17.155 (15.720) and 32.837 (29.390), respectively.
‧
Table 9. Descriptive statistic
Mean Stdev. Min Q1 Median Q3 Max
CALL- Dummy variable that equals to 1 if a firm holds conference calls in a year, and 0 otherwise
NCALL- The number of conference call that a firm holds in the year
DLINK- Dummy variable that equals to 1 if the focal firm connected to other call firms through interlocked directors, and 0 otherwise
LINK- The number of other call firms that are connected to the focal firm through board interlock
INDLINK- Dummy variable that equals to 1 if the focal firm is connected to other call firms through interlocked independent directors, and 0 otherwise
ROA- Net income divided by total assets SIZE- Nature log of a firms sales
LEV- Total liabilities divided by total assets MB- Market value divided by book value DSHARE- Directors’ shareholdings
LSHARE- Shareholdings of the largest ten shareholders ISHARE- Institutional shareholdings
‧
國立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
44