3. RESEARCH DESIGN
3.2 Variable Choice
3.2.3 Control Variables Measurement
國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
15
In addition, I also follow the definition of TEJ database and define a test variable,
ACCEXP, equal to one if the borrower’s AC member with prior experience as
accountant, auditor, chief financial officer, controller, financial controller, or chief accounting officer. Even though the qualification of financial expertise adopted by SEC includes accounting and non-accounting financial expertise, I specifically focus on accounting financial expertise and also follow the definition from TEJ database. I classify accounting experts into two groups and construct two indicator variables:1. D: if firm’s AC with at least one accounting expert, then ACCEXP-D is coded 1, and 0 otherwise.
2. ACCEXP-O: if firm’s AC with only accounting experts, then ACCEXP-O is coded 1, and 0 otherwise.
Accordingly, under these classifications, this study can provide direct evidence to highlight the different situation in the presence of financial and accounting expert and their separate impacts on enhancing financial reporting quality, reducing the cost of debt, and then rewarding better loan terms.
3.2.3 Control Variables Measurement
Built upon the findings of literature (Felo et al., 2003; Li, Xie, and Zhou, 2010;
Dhaliwal et al., 2010), I incorporate three groups of control variables into my analysis:
governance, firm, and bond characteristics. The first group is related to governance characteristics, including number of meetings, number of directors, and average attendance. The second group is firm characteristics, including firm’s size, return on
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
16
asset, earning before tax margin, liability ratio, and the size of external auditor firm.
The third group is bond characteristics.
1. Governance characteristic variables
(1) Audit committee size (ACSIZE)
In Taiwan, based on Corporate Governance Best Practice Principles for TWSE/GTSM Listed Companies, the number of AC shall not be fewer than three. The Blue Ribbon Committee also proposes that an AC should have a minimum of three members. This suggests that a minimum of three persons is necessary for a functional AC. Felo et al. (2003) find that there is a positive relationship between the size of AC and financial reporting quality.
Accordingly, I expect AC size to be positively correlated with loan terms.
(2) Audit committee meeting frequency (ACMEET)
Strong AC governance attributes may even lead to more effective use of any expertise possessed by individual AC members (Cohen et al., 2004). As a result, it is possible that AC that meets more frequently is more likely to enhance AC governance and more effective at overseeing the financial reporting process. Anderson et al. (2004) also indicate that AC meeting frequency exhibits a positive relation with loan terms. Accordingly, I control for this variable.
(3) Average attendance of AC members (ATTENDANCE)
Average attendance is average percentage that audit each committee member attends the meetings. Some studies argue that the financial quality is affected by not only meeting frequency but also by attendance at meetings
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
17
(Stewart and Munro, 2007). Therefore, I control for this variable and expect a positive correlation between ATTENDANCE and loan terms.
(4) Board size (BDSIZE)
Klein (1998) indicates that the number of directors on the board affects committee assignments and board monitoring. Furthermore, Anderson et al.
(2004) find that debt yields are negatively related to board size since larger boards may increase the level of managerial monitoring and enhance the financial accounting process. These studies suggest that larger board leads to higher financial quality. However, Eisenberg, Sundgren, and Wells (1998) indicate that larger board of directors reduces firm value. Since there is divergence of board size effects in prior studies, I make no prediction about the sign of the board size coefficient.
(5) Board meeting frequency (BDMEET)
Cohen et al. (2004) indicate that strong AC governance attributes may even lead to more effective use of any expertise possessed by individual AC members. Furthermore, Magena and Pike (2004) find that board meeting frequency may improve the financial accounting process. Accordingly, I control for this variable.
(6) Establish the AC under regulatory requirement (ACLAW)
Prior research suggests that an effective AC can contribute to high financial reporting quality (Wild, 1996; McMullen, 1996; Felo et al., 2003;
Krishnan, 2005; Anderson et al., 2004; Zaman et al., 2011; Madawaki et al.,
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
18
2013). In Taiwan, the policymaker requires certain kind of public companies to establish the AC, including finance company and TWSE/GTSM listed company whose paid-up capital is more than 10 billion dollars. Therefore, I control for this variable and expect a positive relation between ACLAW and loan terms.
2. Firm characteristics variables
(1) Firm size (SIZE)
Since firm size is positively related to financial reporting quality ( Felo et al., 2003), and the creditors perceive larger firms as less risky and smaller economies of scale in debt production costs (Carey, Prowse, Rea, and. Udell, 1993), I control for this variable. In this study, SIZE is measured as the natural log of total assets at year-end.
𝑆𝐼𝑍𝐸 = 𝐿𝑛(𝑇𝑜𝑡𝑎𝑙 𝐴𝑠𝑠𝑒𝑡𝑠)
(2) Return on asset (ROA)
Return on asset is an important indicator of firm’s performance. It is used to evaluate how effectively and efficiently firm uses its assets to generate earnings. In this study, I measure this control variable as
𝑅𝑂𝐴 =Continuing Operations Income+Interest Expense∗(1−17%)
Average Total Assets * 100%
(3) Earnings before interest, tax, depreciation, and amortization(EBITDA) EBITDA is a measure of firm’s performance and an important indicator to creditors and investors. Therefore, I expect a positive relationship between EBITDA and loan terms. In this study, EBITDA is calculated as
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
19
𝐸𝐵𝐼𝑇𝐷𝐴 =𝐼𝑛𝑐𝑜𝑚𝑒 𝐵𝑒𝑓𝑜𝑟𝑒 𝑇𝑎𝑥
Operating Revenue * 100%
(4) Liability ratio (LEVERAGE)
Leverage is an essential tool to maximize firm’s returns and an important ratio for potential creditors and investors. Some studies conclude that a negative relationship between leverage and profitability (Babalola, 2013;
Kebewar, 2012; Dogan, 2013). Accordingly, I control for this variable and expect a negative relationship between LEVERAGE and loan terms.
𝐿𝐸𝑉𝐸𝑅𝐴𝐺𝐸 =Debt+Equity𝐷𝑒𝑏𝑡 * 100%
(5) Big four audit firms (BIG4)
Prior studies suggest that larger firms tend to choose Big Six auditors (Francis and Krishnan, 1999) and pay lower interest rates (Petersen and Rajan, 1994). Following prior studies (e.g., Ernst & Young, PricewaterhouseCoopers, KPMG, Deloitte), I control for this dummy variable BIG4, equal to 1 if auditor is big 4 audit firm, and expect a negative association between BIG4 and loan terms.
3. Bond Characteristics Variables
(1) The risk-free rate of interest (RF)
In this study, I use Central bank announced average rate of one-year time deposit account in top 5 banks as control variable. The risk-free rates of interest from 2009 to 2013 are 0.9, 1.135, 1.355, 1.355, and 1.355, respectively.
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
20
(2) The frequency of interest payments per year (INTERESTPAY)
INTERESTPAY is a dummy variable in this study. I find some firms pay their interests only at maturity and period is more than one year, so I control for the number that a firm pays interest in a year.
(3) Ways to pay off (METHODPAY)
There are many ways to pay off, including lump sum, installment, and other ways. I expect that different way will lead to different loan terms.
Therefore, I control for this variable.
Detailed definitions of the below variables are summarized in Appendix A.