Do international earnings impact Chinese firms’ capital
structure decisions?
Student Number: S3882055
Student Name: Yang Yang
Thesis Supervisor: Dr. A. de Ridder
Do international earnings impact Chinese firms’ capital structure
decisions?
Abstract
This study aims to shed light on the determinants of capital structure in Chinese listed companies from an international perspective by identifying the foreign earnings proportion as a new
variable of interest. I document that, in general, foreign earnings proportions are negatively correlated with firms’ financing policy but also that the impact is minor. Perhaps of more interest is that only earnings proportions gained from the European market have a positive association with leverage ratios while the results of other general variables are similar to prior studies.
1.
Introduction
previous works of what determining the capital structure, pieces of literature such as Rajan and Zingales (1995), Wald (1999), Bevan and Danbolt (2000) only looked into the companies in the United States or other developed countries. Additionally, the majority of papers infer distinct results among various countries or regions which add more difficulty and interest to this research objective.
Lastly, it is acknowledged that, so far, in the world, China has been the second-largest economy and the second-largest emerging market as well. Chinese stock market is unique for the international financial market because of its policy and the negative relationship with the US stock market in terms of different institutional structures. To be specific, in China, companies and banks are primarily possessed by the government or state which is completely different from those in the US. Even nowadays, the economy of China is under transition and the China-United States trade war shows the growth of internationalization of China is so fast that it can pose threats over other developed regions, but also puts more challenge to the economic development of itself. Global companies and investors prefer to swarming into the Chinese rapidly expanding capital market. In other words, foreign income regarded as a part of internationalization should be paid attention to. As a result, it is worthy to conduct investigations on the Chinese financial market and get a deep insight into its characteristics of firms.
Europe. Such evidence may drive from the lower cost of transportation under the development of China’s Belt and Road Initiative between China and Europe. In this way, Chinese firms have a tendency to expand their market to Europe to gain higher profits at lower costs by using more debt.
This study is further set as follows: section 2 provides a brief literature review and theoretical predictions. Section 3 presents the characteristics of data and sample. Section 4 provides the research method and results analysis. Conclusions are described in Section 5.
2. Literature Review and Theoretical Predictions
borrowed from inside are insufficient. Next, Baker and Wurgler (2002) documented that market timing theory is needed to explain the capital structure. Generally, firms make full use of seasonal opportunities. Specifically, they tried to raise debt when stock prices are low and vice versa. And they mentioned that the time for the result of market timing on corporate financing is supposed to be quite long which is completely adverse to the outcome studied by Kayhan and Titman (2007) showing that market timing only temporarily affect the capital structure of firms. Frank and Goyal (2009) even pointed out that the two sides of equity market timing theory. For one side, it is an effective prediction for the effect of market-to-book ratio. For another side, equity market timing is not supposed to be restricted in the firms’ capital structure. Finally, these existing theories are not completely appropriate to understand the companies’ decisions of leverages, especially largely existing as short time debts. The underlying reason for this result is to put the same assumptions wrongly on different countries. Akman, Gokbulut, Nalin (2015) agreed that the effectiveness of those studies may be reduced when they are applied to developing countries instead of developed ones.
2.1 Leverage
Debt-to-Equity ratio is the most key measure for leverage ratios in empirical papers.
Total debt, short-term debt, and long-term debt are three main types of debt in financing. Also, there are two measures for the value of equity, book value and market value respectively. However, Toy et al. (1974) argued that book value is considered more than the counterpart for senior managers to make decisions. Chang, Chen, and Liao (2014) upheld this opinion as well by mentioning that due to the inefficient stock market in China, the financing policy of firms would be more difficult to decide with market value. Based on the discussion above, I finally use three measures of leverage which include book total debt ratio (BTD), book short-term debt ratio (BSTD) and book long-term debt ratio (BLTD).
2.2 Size
Theoretically, the answer to how size affects leverage is indefinite. Two paths diverge
here, one is the trade-off theory which illustrates that larger firms have higher leverage and vice versa since larger firms have wide variety of funding, contributing to lowering the likelihood of becoming bankrupt of larger companies (Rajan and Zingales, 1995). While in terms of the pecking order theory, the coefficient between firm size and leverage is negative. However, the majority of prior empirical studies identical with the trade-off theory have shown a positive relationship between US firm size and its leverage. Looking into Chinese firms, pieces of works still have the same conclusion of firm size as US twins. (see, e.g., Huang and Song, 2006; Bhabra, Liu, and Tirtiroglu, 2008; Chang, Chen, and Liao, 2014). The hypothesis in my paper thus remains the positive association.
A growing amount of works have already appeared without consistent results of the
relationship between profitability and leverage.In trade-off theory, firms that gained more profits tend to use more debt financing to be tax-deductible which is in line with the agency’s theoretical framework stating that higher leverage applied in lucrative firms with intense free-cash-flow issues is able to control senior management. Nonetheless, pecking order theory assumes the contradictory evidence that profitability and leverage of firms are negatively associated. In this case, they are more likely to use retained earnings to support investment and thus less likely to exert debt. In addition to that, most empirical studies (see, e.g., Pandey, 2001; Booth et al., 2001; Wiwattanakantang 1999) supported that the profitable companies in developing countries usually have lower leverage. Researches on the capital structure of China (see, e.g., Huang and Song, 2006; Bhabra, Liu and Tirtiroglu, 2008; Li, Yue, and Zhao 2009) also backing this prediction. Strebulaev (2007) creating the dynamic trade-off model by using cross-sectional tests proves again leverage and firms’ profitability are negatively correlated. Based on the above discussion, I assume there could be a negative coefficient of profitability and leverage.
2.4 Growth opportunities
The pecking order theory states that the more growth opportunities firms have, the more
synchronicity environment of stock market in China, stock prices could offer irrelevant information to firm value. Therefore, MB ratios can be ignored when book leverage is used. Huang and Song (2006) argued that asset growth rate (labelled as TA) could be a sound proxy to capture historical growth, while Tobin’s Q (labelled as TQ) is employed to measure growth opportunities. Chang, Chen, and Liao (2014) agreed that asset growth could be an alternative to capture firms’ historical growth especially when firms are in ineffective market. All in all, I plan to include both 2 measures in my studies to check if it is a negative connection.
2.5 Tangibility
In general, theories illustrate that there is a positive coefficient of tangibility and leverage. In the trade-off theory, tangible assets function as collateral to protect lenders from moral issue caused by the severe conflict between shareholders and lenders during the financial distress. (Jensen and Mekling, 1976). On the other hand, no stable results of the relationship between tangibility and leverage of Chinese companies have been reflected. Specifically, in the finding of Chang, Chen, and Liao (2014), tangibility has a positive association with leverage among listed firms while negative relationship in non-listed companies. This evidence had been partly investigated by the research of Chen (2004) who illustrated that as debt can be collateralized by tangible assets, a positive coefficient is assumed. Most previous empirical studies defined asset tangibility as the ratio of fixed assets over total assets, I also measured this proxy in the same way.
2.6 Non-debt tax shields
Non-debt tax shields (NDTS) is the deductible tax from depreciation and investment tax
benefits of borrowings. Previous Chinese studies (Chen (2004), Huang and Song (2006), Qian, Tian, and Wirjanto (2009)) have reported the same results about NDTS and debt ratios consistent with trade-off theory. However, tax rates mentioned in the paper of Zou and Xiao (2006) are insignificant. In my paper, I employ depreciation scaled by total assets to measure non-debt tax shields which is the same as the one used in Huang and Song (2006) and my hypothesis is leverage is negatively connected to NDTS.
2.7 Liquidity
In consonance with off theory, the question of choosing capital structure is a
trade-off between interest tax shields and the costs of financial distress which is regarded as liquidity ratio. However, only one previous study (Liang, Li, and Song, 2014) included liquidity as a control variable in the model. I develop the same hypothesis that liquidity is expected to be positively correlated with the leverage ratio as they did in Chinese listed property companies.
2.8 Foreign earnings proportions
In addition to these common variables used in the majority of studies in the previous
years, I add another now factor called foreign earnings proportions, which is rare in other literature, into my regression model. As I stated in the former section, China is growing fast and expanding its market overseas, meaning that most Chinese companies have large foreign earnings through international sales or foreign subsidiaries which could be a considerable income of their annual report, but I do not have enough sources to verify if there is a positive or negative relationship during previous experiments, so the foreign earnings proportions on leverage is unclear in hypothesis.
3. Data and sample
My data sample covers Chinese listed companies from the year 2013 to the year 2017. The data used for my empirical analysis is derived from an authoritative Chinese database called China Stock Market and Accounting Research (CSMAR). I obtained all variables except book short-term debt ratio, book long-term debt ratio, non-debt tax shields, tangibility which were computed. Since I highlighted the foreign income from various parts, I dropped some vague definitions of overseas earnings, yielding a final specimen of 1996 firm-year observations among 221 different Chinese listed firms.
3.2. Descriptive Statistics of Leverages and Determinants
Table 2 reports the summaries of descriptive statistics for the variables used in my
much greater than those of companies from the United States, for example, both two ratios in the study of Frank and Goyal (2009) are only 0.07 and 1.76, respectively.
Table 3 provides summary statistics of foreign earnings proportions during the period 2013-2017. The mean (median) percentage of foreign earnings proportions in 2015 is 10.72 (6.59), which is slightly larger than any other years, while in 2013, the average (median) percentage is the lowest, about 9.12 (4.07). Additionally, foreign income percentage seems to play a limited part in the source of earnings among those Chinese-listed firms with approximately 10 percent in each year. However, the standard deviation between average quantities of foreign earnings and the median value is significant, about 12.5 % on average, inferring that the distribution of foreign revenue is unstable across 5 years.
In order to investigate further about foreign incomes, Table 4 presents more detailed information about where those foreign earnings proportions come from. Since a number of Chinese-listed firms do not open too much information about the source of foreign earnings, for example, only “overseas” or “other foreign regions” are available on the annual report of a few firms, I finally got access to 221 firms with specific names of various places. The foreign earnings proportions are mainly from eleven regions over the world in Table 4. According to the average (median) proportions, Chinese-listed firms tend to have much more incomes in East Asia1, Europe, North America, Hong Kong, Macau, and Tai Wan than other places, about 11.48% (7.99%), 15.30% (9.35%), 12.49% (8.55%), and 9.35% (8.61%) respectively. The average (median) percentages of foreign earnings proportions of the rest places, approximately 6% (4%), do not vary significantly.
1
The correlation matrix in Table 5 is provided for the sample of 1996 firm-year observations. A high and significant correlation of 0.77 between book total leverage and short-term debt ratios is shown while the coefficient of long-term debt ratios and total debt ratios is slightly lower, 0.43, uncovering the fact that short-term debt is the most general form of leverage reflected on the Chinese firms’ balance sheets. Besides, there is still a positive and significant correlation of 0.22 between the short-term debt ratio and the long-term debt ratio. Firm size is positively and significantly associated with its access to total debt ratios, short-term debt and long-term debt ratios respectively. Profitability has a negative and significant correlation with all three estimations of leverage, showing that whatever the debt firms use, it has a negative relation on firms’ risk-taking. The changes of total assets is positive and statistically significant with total debt to equity ratios and long term debt ratios whereas Tobin’s Q are all negatively connected with all three different asset-liability ratios, excluding the fact the growth of total assets has negative and insignificant correlation with short-term debt ratios. There is a significant and positive relationship between liquidity ratios and all measures of debt ratios. Tangibility has significantly positive connection with total debt and short-term debt ratios but insignificantly positive association with long-term debt ratios. The correlation of tax benefits is negative with total leverage, short-term debt ratio and long-term debt ratio. Additionally, those coefficients are all significant. Last but not the least, Foreign Earnings Proportion exhibits significant and negative interrelations to these three measures.
4. Empirical models and results
4.1 Empirical models
profitability, growth opportunities, tangibility, non-debt tax shields, liquidity and foreign earnings proportions. The regression model can be easily constructed as the following expression: LEV=ƒ (SZ, ROA, TAN, GR, CR, NDTS, EP). This study uses a panel data set of Chinese companies listed in China’s stock markets from 2013 to 2017. The empirical specification of the model is elaborated in Eq. (1):
𝐷/𝐸𝑖,𝑡 = 𝛼𝑖,𝑡+ 𝛽𝑖,𝑡𝑆𝑖𝑧𝑒𝑖,𝑡+ 𝛾𝑖,𝑡𝑅𝑂𝐴𝑖,𝑡+ 𝛿𝑖,𝑡𝐺𝑅𝑖,𝑡+ 𝜃𝑖,𝑡𝑇𝐴𝑁𝑖,𝑡+ 𝜇𝑖,𝑡𝑁𝐷𝑇𝑆𝑖,𝑡+
𝜋𝑖,𝑡𝐶𝑅𝑖,𝑡+ 𝜌𝑖,𝑡𝐸𝑃𝑖,𝑡+ 𝜀𝑖,𝑡 , (1)
Where 𝑖 is a subscript for each firm; is a subscript for each year. 𝑖,𝑡 denotes three different leverage measures (BTD, BSTD, BLTD); 𝑆𝑖𝑧𝑒𝑖,𝑡 , 𝑅𝑂𝐴𝑖,𝑡 , 𝐺𝑅𝑖,𝑡 , 𝑇𝐴𝑁𝑖,𝑡 , 𝑁𝐷𝑇𝑆𝑖,𝑡 , 𝐶𝑅𝑖,𝑡 , 𝐸𝑃𝑖,𝑡 are a set of factors explaining these leverages for the listed firms; 𝛼𝑖,𝑡 is the constant; 𝜀𝑖,𝑡 is the random disturbance term. In order to avoid the influence of outliers on the estimation of regression parameters, this paper is carried out winsorizing on the relevant variables with 1% and 99% quantile values. For BTD and BLTD as the explanatory variables, the F statistic, LM statistic, and Hausman statistic are all significant at least at the 5% significance level, indicating that a fixed effect model should be established. However, when BSTD was used as the explanatory variable, the Hausman statistic did not pass the significance test, and the F statistic and LM statistic were both significant at the 1% significance level, indicating that a random effect model is better for variable BSTD. Therefore, in the following regressions, a fixed effect model is used for BTD and BLTD while a random effect model is for BSTD.
4.2. Regression Results and Interpretation
(BTD), book short-term debt ratio (BSTD), book long-term debt ratio (BLTD) as explained variables and total assets growth (TA) as explanatory variable while in column (2), (4), (6), the results of the role of Tobin’s Q (TQ) on three different leverage ratios are reflected. It is obvious that when I use total assets growth (TA) as the main variable to measure growth opportunity (GR), foreign earnings proportions (EP) is significantly negative with leverage ratios. On the other hand, when I control Tobin’s Q (TQ) as the principal variable, foreign earnings proportions (EP) has negative and statistically significant correlations with total debt ratio (BTD) and book short-term debt ratio (BSTD). But in the column (6), on the contrary, there is no relationship between foreign earnings proportions (EP) and book long-term debt ratio (BLTD), which verifies the hypothesis foreign earnings proportions (EP) has unclear role on firms’ decision of their capital structure. Firm size (SZ) has positive effect on book total asset-liability ratios (BTD), book short-term debt ratios (BSTD) and book long-term debt ratios (BLTD). These coefficients are significant at 0.01 levels. Corporate risk-taking, which is profitability (ROA), is significantly negative with three diversified measures. Additionally, liquidity ratio (CR) has the same relationship. In terms of tangibility ratio (TAN), it plays a significantly negative role on book short-term debt ratio (BSTD) and insignificantly positive role on book total debt ratios (BTD), while it is significantly positive with book long-term debt ratio (BLTD) respectively. Non-debt tax shields (NDTS) has a significantly negative connection with book total D/E ratios, book short-term debt (BSTD), book long-term debt ratios (BLTD).
decision of capital structure of firms under the control of Tobin’s Q (TQ). We can see from this table, the increase on foreign earnings proportions from North America positively affects total asset-liability ratio (BTD) and book short-term debt ratio (BSTD), but these are all insignificant. Under the model (6), the relationship between foreign income proportions in North America and book long-term debt ratio (BLTD) is significant. However, the coefficient 0 shows that earnings proportions from North America have no influence on firms’ capital structure. The results indicate that there is no definite connection between earnings proportions from North America part and firms’ financing policy. Firm characteristics that explain the capital structure decision excluding foreign earnings proportions appear to play a similar role as what I have found in Table 6.
Model (1), (3), (5) in Table 8 shows the results of the role on determining companies’ capital structure with emphasize on the earnings proportions from European part and control the total asset growth (TA) for growth opportunities (GR). On the other hand, the regression results are presented in column (2), (4), (6) through the use of Tobin’s Q as the main variable for growth opportunities (GR). These statistics in the table prove the fact that foreign earnings proportions in Europe are statistically significant and positive with firms’ total leverage ratio (BTD) and book short-term debt ratio (BSTD) but the coefficient between earnings proportions of Europe and book long-term debt ratio (BLTD) is 0 and insignificant, which is consistent with the hypothesis that the role of foreign earnings proportions on firms’ decision of capital structure is unclear.
same time, the regression results are displayed in (2), (4), (6) through the use of Tobin’s Q as the main variable for growth opportunities (GR). From the first line of Table 9, firms which have foreign earnings outside mainland China, Hong Kong, Tai Wan, Macau are significantly and negatively related to their total leverage ratios (BTD) and book short-term debt ratios (BSTD), but book long-term debt ratios seem not to be affected by the variation of foreign earnings proportions from East Asia. These products prove the hypothesis again that foreign earnings proportions do not have definite relationship with companies’ debt decisions.
In Table 10, the results of determinants of capital structure are regressed through above mentioned variables and emphasized the foreign earnings proportions which come from Hong Kong, Macau, and Tai Wan. Column (1), (3), (5) are the regression results while regarding total assets growth (TA) as my main variable measuring as the impact of Tobin’s Q on firms’ gearing activities is presented in Column (2), (4) and (6). We can see that from the first result of foreign earnings proportions coming from Hong Kong, Macau and Tai Wan have the same negative coefficients, 0.002, on these three gearing measures. These interaction terms are significant excluding the book long-term debt ratios.
Continuous with above regressions, I further focus on the remaining places in Table 11 to figure out if it has a similar associations with firms’ financing as mentioned regions above. I find that the coefficients (0.001) of book total D/E ratios and book short-term D/E ratios on foreign earnings proportions are positive, but these coefficients are not significant. While the interaction terms of book long-term D/E ratios on foreign earnings proportions are 0 and insignificant. In a word, earnings proportions coming from the rest areas are independent of Chinese listed companies’ gearing activities.
4.3.1 Foreign Earnings Proportions
My main interest is in the new variable, foreign earnings proportions. In table 6, I firstly
regressed the overall foreign earnings proportions from eleven various regions outside mainland China. I find that foreign incomes percentages are negatively correlated with all types of debt ratios, and all these coefficients are significant except for book long-term D/E ratios. But the more interesting result is that coefficients are similar to each other, approximately (-0.001). In other words, the more sales Chinese listed firms gain from overseas market, the less likely they are to use debt financing, even though the impact of foreign earnings on firms’ capital structure is minor. The underlying reason for that would be when Chinese firms find out more investment opportunities abroad, they have a preference sequence to apply retained earnings for investment funds first and then bonds and new equity only if it is necessary.
Next, foreign earnings proportions coming from four parts, North America, Europe, East Asia, Hong Kong, Macau, and Tai Wan, are regressed respectively in following tables. The most significant reason for me to choose these regions is that Chinese listed companies have much larger foreign sales in these mentioned places than in other counterparts. In Table 7, I analyzed the effect of earnings proportions from North America on firms’ gearing activities. And I observed that the coefficients are similar to those in Table 6 but results are not statistically significant, suggesting that foreign incomes from North America do not help to decide the determinants of firms’ capital structure.
foreign earnings percentage from European market, the higher likelihood for those Chinese firms to use more debt, which is completely contrary to overall results and other places. The reason for that may be the growing foreign trade services of Chinese firms in Europe with the development of China’s Belt and Road Initiative2. The transportation cost between China and Europe is lower, contributing to trade strength for Chinese firms to Europe. Thus, Chinese firms tend to exhibit comparatively higher foreign earnings in Europe. Debt can be regarded as a managerial tool to avoid the malfeasance of managers, such as build up their authority. Therefore, it is possible for high debt to restrain management discretions which are also pointed out in some studies (see, e.g., Jensen, 1986; Williamson, 1988).
After that, in Table 9 and Table 10, results of earnings proportions from East Asian and Hong Kong, Macau, Tai Wan areas appear to play a similar role in leverage ratios. Incomes from both two parts are negatively associated with various debt ratios. While significance of former are at 1% level and of latter are at 5%, 10% level. Then I can infer from that the foreign incomes in overall East Asia tend to negatively impact Chinese firms’ decision of capital structure.
In the end, Table 11 shows the results from other remained areas. Earnings proportions from those areas do not have relationship with firms’ leverage decisions, although the coefficients (0.001) on book total D/E ratios and book short-term debt ratios are positive. The reason for this may be the slightly smaller foreign sales of Chinese listed firms in those areas make them not be able to impact firms’ capital structure.
All in all, foreign earnings proportions have limited role in explaining determinants of capital structure of Chinese listed firms and it may diversify across different regions which is in line with my hypothesis that the relationship between foreign earnings proportions and firms’
2
financing policy is unclear. Besides, in all these six tables, it is completely independent of firms’ long-term debt ratios on capital structure which shed light on the uncommon example of using long-term debt in Chinese firms gearing activities.
4.3.2 Total assets
The first table presents regression results using the full sample. It says that the impact of
firm size on the gearing activities is similar from some existing studies and my previous theoretical predictions. Although interaction terms for all measures are below 0.1, they are at 0.01level of significance. It also confirms the trade-off theory indicating that the direct cost of larger companies to finance their operations is smaller because they have various sources of fund and thus less possible to pay for bankruptcy. Another important reason why the coefficient of book long-term debt ratios, 0.007, is larger than that of book short-term debt ratios, 0.001, is that larger firms more often choose long-term debt whereas small firms choose short-term debt consistent with evidence in Marsh (1982).
4.3.3 Profitability
The output in table 1 illustrates that corporate risk-taking measure as ROA is
correspondingly, then those firms would have more diversified choices to raise capital not just rely on debt.
4.3.4 Tangibility
In line with the maturity matching principle, my results indicate tangibility ratio is
negatively correlated with book short-term debt ratios (BSTD) and this relationship is significant. The fact that the coefficient of tangibility ratio and book long-term debt ratios (BLTD) is positive, which verifies the trade-off theory since fixed assets always act as collateral in debt issues. Overall, a positive relation (0.071) of tangibility ratios with book total leverage ratios (BTD) suggests that if a Chinese firm’s tangible assets are high enough to be used as collaterals, then they are likely to uses these assets to diminish lender’s risk of suffering such agency costs of debt. However, this relationship is statistically insignificant.
4.3.5 Liquidity
Currency ratios are all positively associated with all types of debt ratios, and those
coefficients are significant at 0.01 levels. It is apparent to find out that this result is totally consistent with my previous hypothesis and the trade-off theory. In addition to that, T-statistics are also high compared to other variables except for ROA.
4.3.6 Growth opportunities
Since I apply both total assets growth rate and Tobin’s Q to measure Chinese listed
stating the opposing evidence that equity financing is highly constrained in Chinese market because of intense governmental scrutiny. On the other hand, Chinese firms who are going to have brighter growth opportunities in the future are more likely to invest in profitable programs. In order to gain profit and benefit from these investments, they need to decrease debt in their financing policy presently.
4.3.7 Non-debt tax shields
The coefficients of non-debt tax shields are negatively interacted with all kinds of debt
ratios. Some previous works (Chen (2004), Huang and Song (2006), Zou and Xiao (2006), Qian, Tian and Wirjanto (2009)) have reported the same results. The significance of those interaction terms is at 0.01 level. However, the relationship between non-debt tax shields and book short-term debt is insignificant. My findings on this variable prove the truth that tax benefits from raising debt can be replaced by non-debt tax shields. In this way interest that accrues on debt funding cannot be tax deductible.
5. Conclusions
References
Akman, E., Gokbulut, R., Nalın, H., Gokbulut, E., 2015. Capital structure in an emerging stock market : The case of Turkey. Journal of The Faculty of Economics and Administrative Sciences 5, 639-660.
Allen, F., Qian, J., Qian, M., 2005. Law, finance and economic growth in China. Journal of Financial Economics 77, 57–116.
Baker, M., Wurgler, J., 2002. Market timing and capital structure. Journal of Finance 57, 1-32. Baskin, J., 1989. An empirical investigation of the pecking order hypothesis. Financial Management 18, 26-42.
Bevan, A., Danbolt, J., 2000. Dynamics in the determinants of capital structure in the UK. Unpublished Working Paper. University of Glasgow.
Bhabra, H.S., Liu, T., Tirtiroglu, D., 2008. Capital structure choice in a nascent market: Evidence from listed firms in China. Financial Management 37, 341–364.
Booth, L., Aivazian, V., Demirguc-Kunt, A., Maksimovic, V., 2001. Capital structures in developing countries. Journal of Finance 56, 87–130.
Chang, C., Chen, X., Liao, G., 2014. What are the reliably important determinants of capital structure in China? Pacific-Basin Finance Journal 30, 87-113.
Chen, J., 2004. Determinants of capital structure of Chinese listed companies. Journal of Business Research 57, 1341–1351.
DeAngelo, H., Masulis, R., 1980. Optimal capital structure under corporate and personal taxation. Journal of Financial Economics 8, 3-29.
Demirguc-Kunt, A., Maksimovic, V., 1999. Institutions, financial markets and firm debt maturity. Journal of Financial Economics 54, 295–336.
Diamond, D., 1989. Reputation acquisition in debt markets. Journal of Political Economy 97, 828-62.
Fama, E., 1970. Efficient capital markets: A review of theory and empirical work. Journal of Finance 25, 383-417.
Fama, E., French, K., 2002. Testing trade-off and pecking order theory about dividend and debt. Review of Financial Studies 15, 1-33.
Fan, J., Huang, J., Zhu, N., 2009. Distress without bankruptcy: an emerging market perspective. Working Paper, Chinese University of Hong Kong.
Fan, G., Wang, X., Zhu, H., 2010. NERI index of marketization of China's provinces 2009 Report. Beijing Financial Economics 26, 3-27.
Fischer, E., Heinkel, R., Zechner, J., 1989. Dynamic capital structure choice: Theory and tests. Journal of Finance 44, 19-40.
Frank, M., Goyal, V., 2009. Capital structure decisions: Which factors are reliably important. Financial Management 38, 1-37.
Gul, F., 1999. Growth opportunities, capital structure and dividend policies in Japan. Journal of Corporate Finance 5, 141-168.
Hackbarth, D., Hennessy, C., Leland, H., 2007. Can the trade-off theory explain debt structure? Review of Financial Studies 20, 1389-1428.
Huang, G., Song, F.M., 2006. The determinants of capital structure: Evidence from China. China Economic Review 17, 14–36.
Jensen, M., Meckling, M., 1976. Theory of the firm: managerial behavior, agency costs and ownership structure. Journal of Financial Economics 3, 305-60.
Jensen, M., 1986. Agency costs of free cash flow, corporate finance, and takeovers. American Economic Review 76, 323-339.
Kayhan, A., Titman, S., 2007. Firms’ histories and their capital structures. Journal of Financial Economics 83, 1-32.
Kester, W., 1986. Capital and ownership structure: A comparison of United States and Japanese manufacturing corporation. Financial Management in Japan 15(1) 5-17.
Kim, W., Sorensen, E., 1986. Evidence on the impact of the agency costs of debt in corporate debt policy. Journal of Financial and Quantitative Analysis 21, 131-144.
Kraus, A., Litzenberger, R., 1973. A state preference model of optimal financial leverage. Journal of Finance 28, 911-922.
Liang, J., Li, L., Song, H., 2014. An explanation of capital structure of China’s listed property firms. Property management 32, 4-15.
Li, K., Yue, H., Zhao, L., 2009. Ownership, institutions, and capital structure: Evidence from China. Journal of Comparative Economics 37, 471–490.
Marsh, P., 1982. The choice between equity and debt: An empirical study. Journal of Finance 37, 121-144.
Miller, M., 1977. Debt and taxes. Journal of Finance 32, 261-75.
Modigliani, F., Miller, M., 1958. The cost of capital, corporation finance and the theory of investment. American Economic Review 48, 261-297.
Modigliani, F., Miller, M., 1963. Corporate income taxes and the cost of capital: A correction. American Economic Review 53, 433–443.
Morellec, E., Nikolov, B., Schurhoff, N., 2012. Corporate governance and capital structure dynamics. Journal of Finance 67, 803-848.
Myers, S., Majluf, N., 1984. Corporate financing and investment decisions when firms have information that investors do not have. Journal of Finance 39, 575-592.
Pandey, I., 2001. Capital structure and the firm characteristics: evidence from an emerging market. IIMA working paper WP2001-10-04. Indian Institute of Management Ahmedabad, Research and Publication Department.
Qian, Y., Tian, Y., Wirjanto, T., 2009. Do Chinese publicly listed companies adjust their capital structure toward a target level? China Economic Review 20, 662–676.
Smith, C., Watts, R., 1992. The investment opportunity set and corporate financing, dividend and compensation policies. Journal of Financial Economics 32, 263-292.
Strebulaev, I., 2007. Do tests of capital structure theory mean what they say? Journal of Finance 62, 1747–1787.
Stulz, R., 1990. Managerial discretion and optimal financing policies. Journal of Financial Economics 26, 3-27.
Swanson, Z., Srinidhi, B.N., and Seetharaman, A., 2003. The Capital structure paradigm: Evolution of debt/equity choices. Greenwood Publishing Group, Westport, USA.
Titman, S., Wessels, R., 1988. The determinants of capital structure choice. Journal of Finance 43, 1-19.
Toy, N., Stonehill, A., Remmers, L., Wright, R., Beekhuisen, T., 1974. A comparative international study of growth profitability and risk as determinants of corporate debt ratios in the manufacturing sector. Journal of Financial and Quantitative Analysis 9, 875-886.
Wald, J., 1999. How firm characteristics affect capital structure: An international comparison. Journal of Financial Research 22(2), 161-187.
Warner, J., 1977. Bankruptcy costs: some evidence. Journal of Finance 32, 16-18.
Williamson, O., 1988. Corporate finance and corporate governance. Journal of Finance 43, 567-591.
Wiwattanakantang, Y., 1999. An empirical study on the determinants of the capital structure of Thai firms. Pacific-Basin Finance Journal 7, 371-403.
Zou, H., Xiao, J., 2006. The financing behaviour of listed Chinese firms. The British Accounting Review 38, 239–258.
Table 1 Summaries of Determinants of Capital Structure, and their
hypotheses
Proxy(Abbreviation) Definitions Hypothesis
Size(SZ) Natural logarithm of Sales ﹢
Profitability(ROA) Earnings before Interest and tax divided by total assets ﹣
Growth Opportunities(GR) Assets growth (TA) and Tobin’s Q (TQ) ﹣
Tangibility(TAN) Fixed Assets divided by Total Assets ﹢
Non-debt tax shields(NDTS) Depreciation divided by Total Assets ﹣
Liquidity(CR) Current Assets divided by Current Liabilities ﹢
Foreign Earnings Proportions(EP) Foreign Earnings divided by Total earnings ？
Note: This table gives detailed information of variables in my paper. “﹢” sign means that leverage increases with the factor. “﹣” sign means that leverage decreases with the factor. “?” sign means there is no clear hypothesis between leverage and the factor.
Table 2 Descriptive Statistics of Explained and Explanatory Variables of
Chinese Listed Firms
Note: This table reports the summary statistics of variables for the sample 221 firms between 2013 and 2017. Variables definitions are elaborated in Table 1. All continuous variables measured are winsorized at 1% and 99% levels. BTD, book total debt ratio, is defined as total debt (short-term plus long-term) scaled by total assets (total debt plus book value of equity). BSTD, book short-term debt ratio, is construed as short-term debt divided by short-term debt plus book value of equity. BLTD, book long-term debt ratio, is interpreted as long-term debt divided by long-term debt plus book value of equity.
Variable Obs Mean Median Std. Dev. Min Max
Table 3 Summaries of Foreign Earnings Proportions in Each Year
Foreign Earnings Proportions (%)
Year Obs Mean Median Std.Dev Min Max
2013 366 9.12 4.07 12.54 0 74.45 2014 361 9.88 4.93 13.16 0 81.03 2015 367 10.72 6.59 13.06 0 83.20 2016 419 9.76 5.39 12.35 0 98.41 2017 483 9.65 4.68 13.24 0 97.23 All 1996 9.81 5.21 12.88 0 98.41
Table 4 Descriptive Statistics of Foreign Earnings Proportions (%) for
Chinese Listed Firms around the World
Variable Obs Mean Std.Dev Min Max Median
Africa 141 6.05 11.99 0 57.33 1.56 East Asia 310 11.48 13.46 0.01 80.48 7.99 Europe 332 15.30 15.95 0 98.02 9.35 North America 329 12.49 14.99 0 79.02 8.55 West Asia 40 5.43 13.22 0 83.20 1.70 Oceania 117 4.01 10.46 0 81.03 0.92 HKMOTW 327 9.35 9.73 0 64.61 8.61 North Asia 50 7.05 6.85 0.02 28.31 6.29 South America 223 6.56 6.29 0 29.54 4.28 Southeast Asia 93 7.48 7.30 0.01 32.99 5.54 South Asia 34 4.03 4.51 0.02 17.03 2.29 ALL 1996 9.81 12.88 0 98.41 5.21
Table 5 Correlation Matrix of Leverage and Independent Variables for
Chinese Listed Firms
BTD BSTD BLTD SZ ROA TA TQ CR TAN NDTS EP BTD 1 BSTD 0.77* 1 BLTD 0.43* 0.22* 1 SZ 0.01* 0.06* 0.58* 1 ROA -0.38* -0.27* -0.10* -0.02 1 TA 0.02* -0.03 0.07* 0.01 0.04 1 TQ -0.37* -0.23* -0.18* -0.02 0.15* 0.15* 1 CR 0.32* 0.28* 0.12* -0.03 0.04 -0.01 0.15* 1 TAN 0.07* 0.16* 0.05 0.07* 0.01 -0.12* -0.10* -0.10* 1 NDTS -0.14* -0.14* -0.07* 0.09* 0.00 -0.21* -0.03 -0.06* 0.79* 1 EP -0.03* -0.05* -0.00* 0.10* -0.07* 0.10* 0.07* 0.05* 0.02 0.07* 1
Table 6 Overall Results of the impact of Foreign Earnings Proportions on
Capital Structure over Diversified Measures of Leverage and Growth
Opportunities
(1) (2) (3) (4) (5) (6) BTD BTD BSTD BSTD BLTD BLTD EP -0.001*** -0.001*** -0.001*** -0.001** -0.001 -0.000 (-4.56) (-3.74) (-2.80) (-2.50) (-2.82) (-1.82) SZ 0.010*** 0.010*** 0.001*** 0.001*** 0.007*** 0.007*** (3.60) (3.50) (1.08) (1.33) (2.77) (2.26) ROA -1.272*** -1.122*** -0.759*** -0.687*** -0.279*** -0.215*** (-19.30) (-17.76) (-12.98) (-11.80) (-6.31) (-4.86) TA 0.001** 0.009** 0.013*** (0.05) (2.04) (3.80) CR 0.008*** 0.007*** 0.007*** 0.006*** 0.002*** 0.002*** (15.96) (14.76) (14.34) (13.30) (6.39) (5.49) TAN 0.071 0.002 -0.199*** -0.241*** 0.259*** 0.242*** (1.68) (0.05) (-5.33) (-6.54) (9.19) (8.64) NDTS -2.810*** -2.340*** -0.364*** -0.048*** -2.343*** -2.397*** (-6.05) (-5.40) (-0.88) (-0.12) (-7.51) (-7.91) TQ -0.026*** -0.013*** -0.010*** (-15.05) (-8.56) (-8.00) _cons 0.608*** 0.657*** 0.441*** 0.464*** 0.332*** 0.355*** (50.82) (56.36) (41.60) (43.20) (41.40) (43.55) N 1996 1996 1996 1996 1996 1996 Adj.R2 0.368 0.343 0.294 0.221 0.393 0.408Table 7 Results of Foreign Earnings Proportions from North America impact
on Capital Structure
(1) (2) (3) (4) (5) (6) BTD BTD BSTD BSTD BLTD BLTD EP 0.001 0.000 0.001 0.000 0.000 0.000* (1.28) (0.82) (1.15) (0.42) (1.26) (1.77) SZ 0.045*** 0.042*** 0.017*** 0.013*** 0.025*** 0.026*** (10.98) (9.60) (4.14) (2.92) (12.42) (12.24) ROA -1.129*** -1.118*** -0.983*** -0.969*** -0.156** -0.161*** (-9.09) (-8.97) (-7.91) (-7.74) (-2.57) (-2.64) TA 0.012* 0.018** 0.005 (1.75) (2.57) (1.48) CR 0.007*** 0.007*** 0.007*** 0.007*** 0.000 0.000 (6.21) (6.13) (6.29) (6.17) (0.26) (0.20) TAN 0.036 0.011 -0.190** -0.225*** 0.246*** 0.256*** (0.43) (0.13) (-2.28) (-2.68) (6.04) (6.27) NDTS -1.557* -1.247* -0.525* -0.074* -1.306*** -1.436*** (-1.81) (-1.47) (-0.61) (-0.09) (-3.10) (-3.45) TQ -0.005* -0.007* 0.002 (-1.42) (-1.92) (1.26) _cons -0.917*** -0.810*** -0.081 0.067 -0.756*** -0.802*** (-6.67) (-5.41) (-0.59) (0.45) (-11.25) (-10.96) N 329 329 329 329 329 329 Adj.R2 0.367 0.345 0.312 0.306 0.422 0.421T
able 8 Results of the Foreign Earnings Proportions from Europe impact on
Capital Structure
(1) (2) (3) (4) (5) (6) BTD BTD BSTD BSTD BLTD BLTD EP 0.003*** 0.003*** 0.002*** 0.002*** 0.000 0.000 (4.92) (4.89) (4.13) (4.02) (1.48) (1.61) SZ 0.049*** 0.045*** 0.019*** 0.015*** 0.026*** 0.027*** (12.62) (10.85) (5.04) (3.67) (13.48) (12.78) ROA -1.068*** -1.007*** -0.803*** -0.750*** -0.293*** -0.293*** (-6.58) (-6.17) (-5.01) (-4.64) (-3.62) (-3.56) TA 0.006** 0.016** 0.009 (0.54) (1.40) (1.55) CR 0.007*** 0.007*** 0.007*** 0.007*** -0.000 -0.000 (6.81) (6.67) (6.83) (6.64) (-0.08) (-0.16) TAN 0.013 0.002 -0.205** -0.226*** 0.240*** 0.249*** (0.16) (0.02) (-2.45) (-2.74) (5.67) (5.91) NDTS -1.264*** -1.209*** -0.321*** -0.069*** -1.289*** -1.464*** (-1.38) (-1.37) (-0.35) (-0.08) (-2.82) (-3.30) TQ -0.009** -0.009** 0.001 (-2.27) (-2.41) (0.67) _cons -1.056*** -0.918*** -0.181 -0.033 -0.773*** -0.797*** (-8.22) (-6.52) (-1.43) (-0.24) (-12.07) (-11.23) N 332 332 332 332 332 332 Adj.R2 0.506 0.513 0.389 0.397 0.470 0.467T
able 9 Results of the Foreign Earnings Proportions from East Asia impact
on Capital Structure
(1) (2) (3) (4) (5) (6) BTD BTD BSTD BSTD BLTD BLTD EP -0.002*** -0.002*** -0.002*** -0.002*** -0.000 -0.000 (-4.37) (-4.38) (-3.71) (-3.80) (-1.38) (-1.25) SZ 0.041*** 0.038*** 0.014*** 0.009** 0.024*** 0.025*** (10.91) (9.46) (3.57) (2.28) (12.28) (12.07) ROA -1.239*** -1.215*** -1.088*** -1.064*** -0.179*** -0.181*** (-10.16) (-10.05) (-8.82) (-8.73) (-2.81) (-2.83) TA 0.001*** 0.010*** 0.010 (0.02) (-0.88) (1.62) CR 0.006*** 0.006*** 0.007*** 0.006*** 0.000 0.000 (6.07) (6.01) (6.21) (6.11) (0.43) (0.31) TAN 0.027 -0.002 -0.185** -0.230*** 0.232*** 0.249*** (0.34) (-0.02) (-2.34) (-2.94) (5.67) (6.07) NDTS -1.647* -1.564* -0.954** -0.629** -1.087** -1.320*** (-1.81) (-1.80) (-1.04) (-0.72) (-2.29) (-2.88) TQ -0.009** -0.010*** 0.002 (-2.56) (-3.04) (1.28) _cons -0.763*** -0.625*** 0.063 0.232* -0.730*** -0.770*** (-6.05) (-4.60) (0.50) (1.70) (-11.09) (-10.72) N 310 310 310 310 310 310 Adj.R2 0.369 0.341 0.368 0.385 0.427 0.425T
able 10 Results of the Foreign Earnings Proportions from Hong Kong,
Macau, and Tai Wan impact on Capital Structure
(1) (2) (3) (4) (5) (6) BTD BTD BSTD BSTD BLTD BLTD EP -0.002** -0.002* -0.002* -0.002* -0.002 -0.002 (-2.17) (-1.68) (-1.67) (-1.47) (-1.67) (-1.47) SZ 0.947*** 0.815*** -0.030 -0.061 -0.030 -0.061 (10.46) (9.39) (-0.30) (-0.60) (-0.30) (-0.60) ROA -0.927*** -0.858*** -0.078 -0.050 -0.078 -0.050 (-7.19) (-7.14) (-0.55) (-0.35) (-0.55) (-0.35) TA 0.004*** 0.014*** 0.014*** (0.25) (0.87) (0.87) CR 0.005*** 0.005*** 0.004*** 0.003*** 0.004*** 0.003*** (6.39) (5.60) (3.90) (3.65) (3.90) (3.65) TAN -0.187 -0.209 -0.145*** -0.149*** 0.145*** 0.149*** (-1.64) (-1.96) (-1.15) (-1.19) (1.15) (1.19) NDTS -1.092*** -1.086*** -1.353*** -1.255*** -1.353*** -1.255*** (-0.81) (-0.87) (-0.92) (-0.85) (-0.92) (-0.85) TQ -0.032*** -0.007*** -0.007 (-6.81) (-1.37) (-1.37) _cons 0.545*** 0.614*** 0.371*** 0.391*** 0.371*** 0.391*** (28.50) (30.70) (17.62) (16.64) (17.62) (16.64) N 327 327 327 327 327 327 Adj.R2 0.349 0.319 0.236 0.239 0.336 0.375
T
able 11 Results of the Foreign Earnings Proportions from Other Areas
3impact on Capital Structure
(1) (2) (3) (4) (5) (6) BTD BTD BSTD BSTD BLTD BLTD EP 0.001 0.001 0.001 0.001 0.000 0.000 (1.26) (1.08) (1.05) (1.02) (1.25) (1.37) SZ 0.040*** 0.032*** 0.027*** 0.013*** 0.025*** 0.026*** (8.08) (7.60) (4.13) (2.72) (6.26) (6.34) ROA -1.228*** -1.117*** -0.887*** -0.862*** -0.154** -0.156*** (-9.10) (-9.01) (-7.71) (-7.22) (-2.25) (-2.31) TA 0.021* 0.028** 0.005 (1.53) (2.55) (1.17) CR 0.006*** 0.006*** 0.005*** 0.005*** 0.000 0.000 (6.11) (6.02) (5.59) (5.37) (0.14) (0.10) TAN 0.033 0.013 -0.192** -0.215*** 0.255*** 0.269*** (0.39) (0.14) (-2.32) (-2.88) (5.11) (5.26) NDTS -1.545*** -1.326*** -0.084*** -0.072*** -1.321*** -1.426*** (-1.80) (-1.58) (-0.66) (-0.49) (-1.11) (-1.46) TQ -0.004* -0.005* -0.002* (-1.53) (-1.95) (1.34) _cons -0.811*** -0.801*** -0.172*** -0.166*** -0.658*** -0.632*** (-5.67) (-5.43) (-0.55) (-0.47) (-10.15) (-10.04) N 698 698 698 698 698 698 Adj.R2 0.276 0.285 0.301 0.206 0.404 0.412
Note: Variable definitions are in table 1. BTD and BLTD choose fixed effect model, BSTD chooses random effect model. Model (1), (3), (5) are the results when control variable GR is estimated by TA; Model (2), (4), (6) are the results when control variable GR is estimated by TQ. EP in this table is the foreign earnings proportions from the remaining areas. T-statistics are reported in parentheses and are computed with robust standard errors adjusted for clustering at both the firm level and the year level. ***, **, and * indicate significance at the 1%, 5%, and 10% levels, respectively.
