5.1. Capital Structure Determinants of Large-Listed Firms
displays the descriptive statistics for capital structure variables, company variables, corporate governance factors and credit rating score. Recall that 182 companies are usually examined over 6 years, leading to N
= 1092 data points. A few points are missing, due to a lack of corresponding raw data. From Table 1
, one can see that the mean of total debt ratio, long-term debt ratio and short-term debt ratio distribution is 52%, 10% and 38%, respectively. The (integer) mean of the number of board members is 9; the percentage of outside directors is 37%, i.e., less than half of the board members are made of independent actors. The proportion of CEO duality occurs about 81%, i.e., in most companies the same person holds the CEO and board chair at the same time. About 9% of the companies hire Big-four auditors. The mean annual sales are about 789 millions (CNY). The tangibility, profitability, and non-debt tax shields mean is 22%, 6% and 2%, respectively. The mean number of operation time is about 32 months. The credit score mean is about equal to 3.
shows the Pearson correlation coefficient between these factors. It can be observed that both long-term debts and short-term debts are positively correlated with total debt, but short-term debt is negatively correlated to long-term debt.
As for corporate governance factors, both long-term debts and short-term debts are negatively correlated with board size. This indicates that the larger the board size, the less long-term debt and the less short-term debt would be, but there is more total debt. The result is consistent with above hypotheses and prior studies. Credit rating has a positive relationship with board size, CEO duality and auditor but a negative relationship with outsiders. Both Ashbaugh-Skaife et al.
) and Bradley et al.
) discovered that credit rating is positively related to board size, outsiders and auditor but negative related to CEO duality in the United States. Bhojraj and Sengupta
) found that outsiders have a positive relationship with credit ratings. Liu and Jiraporn
) observed a lowering of credit ratings when the CEOs have more decision-making power. Our results on the relationship between credit rating with outsiders and CEO duality are inconsistent with these researches. However, notice that Becker and Milbourn
) proved that credit rating coincides with good industry performance: a better performance leads to higher credit rating. Peng et al.
) supported the stewardship theory which represents that CEO duality is good for performance due to the unity of command it presents. Therefore, the CEO duality has a positive relationship with the performance which itself improves credit rating.
Institutional theory suggests that changes in organization, like, for example designating more outside directors, will result in a process that makes firms more similar rather than more efficient (DiMaggio and Powell 2000
). In this line of thought, Peng
) suggests that outsider directors have a different influence on firm performance in Chinese firms, if the performance is measured by sales growth; moreover, he finds that outsiders have little impact on financial performance measured as the return on equity (ROE). Thus, a greater number of outsiders will lower the performance, itself being positively related to credit rating, thus will be lowering credit rating!
There is also a significantly positive correlation between the percentage of outside directors and total debt and long-term debt, indicating that the more outside directors on the board has a company, the more so is total debt and long-term debt. Notice that Guest
) proved that there is a positive relationship between the outsiders proportion (and board size) and debt ratio, justifying the finding through the argument that the supervisory and the advisory persons are the same outside director. It can be understood as follows: China’s Company Law (2005) regulated that Chinese firms should have supervisory boards and management boards, both of them having a “social responsibility” in conducting financial operations. The monitoring functions impose that the management has the pursuit of the shareholder profits. Outsiders want to show their capability to the employer and have more independence than the insiders (Fama and Jensen 1983
; cited in Guest 2008
), whence a positive correlation factor.
The CEO duality is positively correlated to total debt and long-term debt, but weakly correlated to short-term debt, implying that the person who holds the CEO and board chairship, at the same time, usually prefers to choose long-term debt. Whether this is a merely psychological or financial corporate governance attitude is not further discussed here.
However, there is a positive correlation between the Big-four auditor choice and total debt, and interestingly, a negative correlation between the Big-four auditor and short-term debt. This means that the “high quality auditor” would rather influence an increase of long-term debt and a decrease of short-term debt. Notice that the correlation coefficient between CEO duality and auditor is positive but weakly significant (see below).
Consider the firm factors. As expected, the company size is positively correlated with total debt and short-term debt, whence it can be deduced that larger firms prefer short-term debt. The tangibility is positively but weakly correlated with long-term debt. It means that tangible assets may be regarded as the collateral to long-term debt. Profitability is negatively related to total debt and long-term debt, implying that if a company makes more profit, it would decrease its long-term debt. As expected, a negative correlation occurs between growth opportunities and debt. The non-debt tax shields are positively related to debt ratio, but negatively correlated to short-term debt and long-term debt ratios.
The firm age is also positively correlated with total debt and long-term debt—indicating that if a company operates has a longer operation time, the strategy is to prefer a long-term debt. This result contradicts above hypothesis. However, this seems to be explainable through an observation by the International Monetary Fund (Prasad and Ghosh 2005
) which comments that the reason might be found in the fact that “small and young firms in emerging markets are likely to find debt cheaper than equity, since they may have easy access to credit” (Huissan and Hermes 1997
; cited in International Monetary Fund, Prasad and Ghosh 2005
regression coefficients for the various debts, from 2010 to 2015, are found in Table 3
As for corporate governance factors, board size is negatively linked with long-term debts and short-term debts; the reason might be that a large board restricts the company to pursue the higher debts because higher debts could lower the company performance. This result is consistent with Anderson et al.
) research that larger board adopts less short-term and long-term debt.
However, board size is positively related to total debt. According to Bragg
), debt consists of short-term debt and long-term debt, but it also could be regarded as the current liabilities when debts are supposed to be paid off within more than 1 year or through some operating cycle. Current liability includes accounting payable, customer deposits, accrued tax, accrued wage and vacation pay and other accruals connect with current operations. Therefore, debt not only includes short-term and long-term debt, but also consists of other liabilities. Moreover, according to Abor
), larger board sizes would have more non-executive directors who might bring better management decisions and assist the company to get more resource—since (or because) these external board people might have “greater knowledge” and “valuable information” on financing facilities. Thus, the larger the board size, the more possible are debts.
The CEO duality is positively related to total debt and long-term debt but the correlation between the CEO duality and leverage is insignificant. The result is consistent with Chen et al.
) who also discovered that the CEO duality is insignificantly related to firm performance.
The Big-four auditor is barely negatively correlated with short-term debt; this is not consistent with our expectation. However, Oxelheim
) also pointed out that the governance disclosure is negatively related to debt ratio. According to Qi et al.
), the firms employing high quality auditors could improve the reliability of their financial report in so doing, decrease the information asymmetry, whence leading toward cutting down debt cost.
In terms of the firm factors, the company size is negatively linked with long-term debt. According to Chen
), due to their reputation, large companies more easily obtain equity financing support from the secondary market. Moreover, the debt holder does not have the legal safeguard of their debt—because the legal safeguard is undefined in China and the bankruptcy cost is not high. Thus, the debt holders do not likely much appreciate long-term debts of the firm. It might be one of the main reasons why long-term debt is negatively related to the firm size.
There is a positive correlation between the tangibility and total debt and long-term debt, but a negative correlation between the tangibility and short-term debt. The result is the same as that found by Ortiz-Molina and Penas
), because the fixed assets are not appropriate for short-term loan guarantee (cited in Dasilas and Papasyriopoulos 2015
) also finds this negative relationship between short-term debt and tangible assets, in Ghana.
There is an inverse correlation between profitability and total debt. This is not consistent with our expectation. Notice that the result supports the pecking order theory. According to Chen
), the reason might stem in the limited banking financial resource, uncompleted bond markets, and a lack of legal protection. Moreover, the equity is more attractive than debt because capital gain is massive and debts are binding.
The non-debt tax shields are negatively correlated to total debt and long-term debt, but positively linked with short-term debt. This means that a company with a larger NDTS is likely to adopt more short-term debt and less long-term debt strategies.
Finally, credit rating is positively correlated with total debt and long-term debt, but is negatively related to short-term debt, indicating that if a company has a high credit rating, the company would hold more long-term debt but less short-term debt. According to Diamond
), the company with a low credit rating has no choice, but to adopt the short-term debt strategy. Notice that his theory recognizes two types of short-term borrowers: one is the firm with low credit rating—which has no choice, the other is the firm with high credit rating which adopts its debts to time of borrowing in order to make good use of information arrivals. Borrowers who have ratings “in between” depend more heavily on long-term debts.
5.2. Sensitivity Tests
In order to test the results’ robustness, some independent factors are to be replaced and thereafter one reruns a regression analysis. For example, profitability can be replaced by the return on equity (ROE), which is the ratio of retained profits to equities. Thus, we impose β7 = 0 in the model and take β12 as finite.
Three dummy factors are used for three credit groups: CR4 is the dummy representing the value of 1 for companies which rate at AAA and 0 otherwise. CR3 is the dummy representing the value of 1 for companies which rate at AA+, AA, AA− and 0 otherwise. CR2 is the dummy representing the value of 1 for companies which rate at A+, A, A− and 0 otherwise.
The results are shown in Table 4
, for the regression outputs with ROE and CR terms but without PROFIT term, in Table 5
, for the regression outputs with PROFIT term and CR4, CR3, CR2 but without ROE term, and in Table 6
, for the regression outputs with ROE term and CR4, CR3, CR2 but without PROFIT term, all for 2010 to 2015. For each regression coefficient, the statistical significance is given, as well as the appropriate F
The regression results between company variables and corporate governance factors show similar signs and relationships as those displayed in Table 3
. However, the three credit rating groups display different signs. CR4 is positively related to total debt, but both CR3 and CR2 are negatively correlated to total debt. All groups are negatively correlated with short-term debt. Only CR3 presents a positive relationship with long-term debt ratio. Therefore, an important conclusion arises: it can be concluded that there is a different significance between the high-rated credit rating and the low-rated credit rating for a firm capital structure. According to Becker and Milbourn
), the “let’s go shopping” for a better rating by some more accommodating auditor might be a cause of the weak linear relations between DR and credit rating dummy seen in Table 6