1、 1 外文翻译 原文 Capital structure influencing factor analysis research Material Source: Theory of Optimal Capital Structure Author : R. Barea Since the Modigliani and Miller (1958) since the academic structure of the capital a large number of theoretical and empirical research, trying to identify the pot
2、ential impact of capital structure choice factors. A lot of literature suggests that the choice of capital structure by the asset structure, firm size, non-debt tax shields, growth, volatility, product uniqueness, profitability and other firm characteristics factors. In addition, the choice of capit
3、al structure is also affected by industry characteristics, macroeconomic and institutional environment factors. Harris and Raviv (1991) from the experience of many U.S. companies to sum up: “leverage ratio of fixed assets, non-debt tax shields, growth and company size increases, with the volatility,
4、 advertising costs, bankruptcy the possibility of profitability and product uniqueness increases less. “ Chinese listed companies due to the particularity of the system, what factors determine the choice of capital structure? Characteristics of institutional factors influenced how the company capita
5、l structure choice? Experiences and things like that to be the model and empirical test. In recent years, researchers began to affect the capital structure of listed companies in an empirical study of factors, such as Lu Zhengfei, and Xin Yu (1998), Lishan Min and Su Yun (1999), Xiaozuo Ping and Wu
6、Shinong (2002), and achieved certain results, However, most studies are using a simple regression technique factors on capital structure for empirical analysis. Titman and Wessels (1988) pointed out the shortcomings of this approach: First, there is no wish to measure the sole representative of the
7、property; Second, it is difficult to find and other relevant property is not related to the measurement of a particular property; third, As can be observed variable is not perfectly representative of its properties should be measured, they are used in the regression analysis will lead to errors in v
8、ariable problem; fourth, the agent variables and measurement error 2 may be explained by variables related to measurement error will produce false (Spurious) related. In this paper, two-stage multiple procedures, application of factor analysis-based model to reduce measurement error, to expand the c
9、apital structure of Chinese listed companies Empirical Study. Capital Structure To build the empirical model, the author according to the capital structure theory and relevant empirical research on factors affecting capital structure analysis, and gives a proxy variable to capture these factors. I,
10、the asset structure Agency theory, balance theory and the theory of asymmetric information are considered assets for capital structure choice. According to agency theory, high-leverage the companys shareholders tend to sub-optimal investment (Jensen and Meckling, 1976; Myers, 1977). The assets of th
11、e company secured an opportunity to limit such behavior. Therefore, the value of assets and leverage are related to security. Another problem comes from a proxy service managers tend to consumption. Assets can be secured with fewer companies more vulnerable to such agency costs, because these compan
12、ies on the capital expenditure monitoring more difficult (Grossman and Hart, 1982). Companies can increase the level of debt as a monitoring tool to mitigate this problem. Therefore, security assets and leverage can be negative. Theory from the balance with debt secured creditors to reduce the poten
13、tial loss of the debtors insolvency and, therefore, limit the amount of shareholder wealth, occupation of the debtor. Meanwhile, in bankruptcy the value of tangible assets higher than the value of intangible assets. Therefore, the value of assets and leverage are related to security. Under asymmetri
14、c information theory, tangible assets, more businesses will face less information asymmetry, therefore, should issue equity rather than debt. And the existence of asymmetric information, to the sale of secured debt had a negative because it reduces information premium. For asset structure, we use st
15、ock / total assets (INV) and fixed assets / total assets (FIX) two proxy variables. II, firm size Many studies suggest that large companies tend to diversify, with more stable cash flow, so low probability of bankruptcy. Warner (1997), Angclua and Meconnel (1982) study found that direct costs of fin
16、ancial distress and negatively related to firm size. Fama and Jensen (1983) that large corporations to smaller companies tend to provide more information on lenders. Therefore, less monitoring costs of large 3 companies, large companies than small companies with high borrowing capacity. Therefore, f
17、irm size should be positively correlated with leverage. And Raja n and Zingales (1995) that the large companies than small companies tend to provide more information to the public, may be related to internal investment company size and level of external investment in human negative correlation of as
18、ymmetric information. Under asymmetric information theory, large companies should be inclined to equity financing and therefore have lower leverage. The size of the company, we use the natural logarithm of total assets (LN (TA) and the main business income of the natural logarithm (LN (S) of two pro
19、xy variables. III, the tax That the use of tax-based model of the main benefits of debt financing is tax credits. According to tax-based theory, companies with higher marginal tax rates should use more debt to get the tax shield benefits. Therefore, the effective marginal tax rates should be positiv
20、ely correlated with leverage. Unable to obtain relevant data to calculate the marginal tax rate, we use the average tax rate (TAX) to analyze the tax impact of capital structure choice. IV, non-debt tax shield DeAngelo and Masulis (1980) that non-debt tax shield can be used as an alternative to debt
21、 financing, tax benefits, the same as in other cases, the non-debt tax shields have more companies should use less debt. Barton et al (1989), P rowse (1990), Wald (1999), Kim and Sorensen (1986) research shows that non-debt tax shields and leverage negative. In this paper, depreciation / total asset
22、s as non-debt tax shield (DEP) of the proxy variables. V. Growth According to agency theory, equity-controlled companies tend to sub-optimal investment will be deprived of their wealth came from the hands of creditors. For high growth companies, because of its future investment opportunities in the
23、choice of greater flexibility, these companies may be more serious agency problems. Myers (1977) that high growth companies lower the future investment in growth companies have more options. If the high-growth companies need external equity financing options to implement in the future, then the comp
24、any has a large debt may give up this opportunity, because such investment will transfer wealth from shareholders to creditors of the body, which produces the problem of insufficient investment. Therefore, growth should be negatively correlated with leverage. For growth, this growth rate with total
25、assets (GRTA) and the equity value-added rate (GREQ) two 4 proxy variables. VI, volatility Regular payment of debt obligations involved, the highly leveraged company is more vulnerable to financial distress costs. Finance theory suggests that the risk of the company or bankrupt companies should not
26、have a high probability of higher leverage. Therefore, the main business income volatility or commercial risk as the possibility of occurrence of financial distress proxy variables, which should be negatively correlated with leverage. Bradley et al (1984), Titman and Wesssels (1988), Wald (1999) and
27、 Booth et al (2001) and other studies have shown that volatility negatively correlated with leverage. In this paper, the main business of the standards slip (VOL) as a proxy for volatility. VII ability to generate internal resources Trade-off theory is that ability to generate internal resources to
28、leverage a positive correlation, because a strong ability to generate internal resources, companies choose higher leverage to get more debt tax shield. Jensen (1986) pointed out that instead of borrowing to pay dividends to ensure that the management discipline empire method. The benefits of debt “c
29、an improve the efficiency of managers and their organizations“, which act as a “control effect“ role. Therefore, the company has a large free cash flow should have higher debt to limit managements discretion. According to the Theory of Optimal Financing (Pecking order theory), because the existence
30、of asymmetric information, the company follows the financial pecking order model: companies prefer internal resources, internal resources have been exhausted if the company was to issue debt, and finally the issue of equity. Therefore, the ability to generate internal resources, negatively correlate
31、d with leverage. The ability to generate internal resources, this paper, the cash rate of sale (NOCFS) and total assets of cash recovery rate (NOCFA) two proxy variables, but to test the Jensen (1986) free cash flow hypothesis proposed in this paper with a cash rate of sales / total Asset growth rat
32、e (FCFS) and total assets of cash recovery rate / total assets growth rate (FCFA), as free cash flow (Note: free cash flow is difficult to quantify, can not be obtained directly from the financial data, must be used in other empirical research cash flow concept, and in line with the growth of the co
33、mpany (such as TobinQ, growth rate of total assets), investment opportunities, free cash flow and other indicators in order to explain the problem.) proxy variables. VIII, product uniqueness 5 From the stakeholder theory of capital structure and product / factor market theory perspective, the compan
34、y has a unique product should have less leverage. Titman and Wessels (1988) that, in liquidation, the production of unique or specialized products company, its customers, suppliers, workers will suffer from higher costs. Their workers and suppliers may have the skills and capital, job characteristic
35、s, and the customer service more difficult to find a replacement phase. From the agency cost perspective, the expected cost of employees looking for work products and services depends on whether there is unique. Employees working on the implementation of mass-specific work with respect to employees
36、engaged in the former expected to find lower cost. Therefore, when other conditions being equal, and human-related costs for the agency to provide specialized products and services relative to the companies higher. Due to higher leverage will have higher agency costs and bankruptcy costs, so the uni
37、queness of products and services will affect the degree of capital structure choice. These companies promote their unique products will suffer more sales costs and advertising costs. In this paper, operating expenses / Income from principal operations (SEXP) as a proxy for product uniqueness. IX liq
38、uidity Current ratio of capital structure choice is mixed. On the one hand, high flow rate paid by the company short-term debt due ability. Therefore, liquidity should be positively correlated with leverage. On the other hand, companies with more liquid assets may use these assets to finance its inv
39、estments. Therefore, the flow of state assets would negatively affect leverage. And, as Prowse (1990) points out, can be used to indicate the liquidity of the assets to creditors, the interests of shareholders to manipulate the expense of the extent of these assets. In this paper, the current ratio
40、(CR) and the quick ratio (QR) as a proxy for liquidity. Ten, industry characteristics The asset risk, asset type, and the demand for external funds vary by industry, the average leverage will vary with the industry. Industry characteristics and capital structure characteristics of the fact that the
41、leverage within the same industry in different sectors of the lever more than the similar, leverage levels to remain relatively the same industry (Bowen et al, 1982; Bradley et al 1984). Bradley et al (1984) studies have shown that regulated industries (telecommunications, electronics, utilities and
42、 aviation industry) with higher leverage. This article uses the industry dummy variables to control the impact of industry factors on the lever. 6 译文 资本结构影响因素的分析研究 资料来源 : 最优资本结构原理 作者：巴里亚 自 Modigliani 和 Miller（ 1958）以来，学术界对资本结构进行了大量的理论和实证研究，试图辨别影响资本结构选择的潜在因素。大量的文献表明，资本结构的选择受资产结构、公司规模、非债务税盾、成长性、波动性、产品
43、独特性、获利能力等公司特征因素影响。此外，资本结构的选择还受行业特性、宏观经济以及制度环境等因素影响。 Harris 和 Raviv(1991)对来自美国公司的许多经验研究进行总结：“杠杆随着固定资产比率、非债务税盾、成长性和公司规模的增加而增加，随着变动性、广告费用、破产可能性、获利能力和产品独特性的增加而减少。” 中国上市公司由于制度的特 殊性，究竟什么因素决定资本结构的选择？制度因素如何通过公司特征因素影响资本结构的选择？等等这些问题有待于经验模型的建立和实证检验。近年来，中国研究者开始对上市公司资本结构影响因素进行实证研究，如陆正飞和辛宇 （ 1998） 、李善民和苏贇 （ 1999）
44、 、肖作平和吴世农 （ 2002） 等，取得了一定的成果，但大多数研究都是使用简单的回归技术对资本结构影响因素进行实证分析。 Titman 和 Wessels(1988)指出了这种方法的缺点：第一，不存在希望测量属性的唯一代表；第二，很难找到与其他相关属性不相关的某一特定属性的计量 ；第三，由于可观察变量不能完美地代表其所应测量的属性，它们在回归分析中的使用将引起变量问题中的误差；第四，代理变量的计量误差也许与被解释变量的计量误差相关将产生谬误 (Spurious)相关。本文采用二阶段多元程序，应用基于因子分析模型来减缓计量误差，以拓展中国上市公司资本结构影响因素的实证研究。 资本结构影响因素
45、：为了构建经验模型，作者根据有关的资本结构理论和相关的经验研究对资本结构影响因素进行分析，并给出捕捉这些因素的代理变量。 一、资产结构 代理理论、权衡理论和信息不对称理论都认为资产构成影响资本结构的 选择。根据代理理论，高杠杆公司的股东倾向于次优投资 （ Jensen 和Meckling,1976;Myers,1977）。而有担保资产的公司能限制这种机会行为。因此，资产担保值与杠杆正相关。另外一个代理问题来自经理们倾向在职消费。具有较少可担保资产的公司更易遭受这种代理成本，因为这些公司对资本支出的监7 督较困难 （ Grossman 和 Hart,1982） 。公司可以提高债务水平作为监督工具
46、以减缓这个问题。因此，可担保资产与杠杆负相关。从权衡理论看，带抵押的债务减少了债权人在债务人破产时的潜在损失，因而也就限制了股东侵占 债务人财富的数额。同时，在公司破产时有形资产的价值比无形资产的价值高。因此，资产担保值与杠杆正相关。根据信息不对称理论，有形资产较多的企业将面临较少的信息不对称问题，因此，应该发行权益而不是债务。且信息不对称的存在，给出售担保债务带来了不利，因为其减少了信息溢价。对于资产结构，本文使用存货 /总资产 (INV)和固定资产 /总资产 (FIX)两个代理变量。 二、公司规模 许多研究认为大公司倾向于多角化经营，具有较稳定的现金流，因此破产的概率较小。 Warner(
47、1997)、 Angclua 和 Meconnel(1982)的研究发现财务困境的直接成本与公司规模负相关。 Fama 和 Jensen(1983)认为大公司较小公司倾向于向贷款人提供更多的信息。因此大公司的监督成本较少，大公司比小公司具有较高的借贷能力。因此，公司规模应与杠杆正相关。而 Rajan 和 Zingales (1995)认为大公司比小公司倾向于向公众提供更多的信息，公司规模可能与内部投资人和外部投资人间信息不对称水平负相关。根据信息不对称理论，大公司应倾向于权益融资，因此具有更低的杠杆。对于公司规模，本文使用总资产的自然对数 (LN(TA)和主营业务收入的自然对数 (LN(S)两
48、个代理变量。 三、税收 基于税收模型认为使用债务融资的主要益处在于税收抵减。根据基于税收理论，具有较高边际税率的公司应使用更多的债务以获得税盾收益。因此，有效边际税率应与杠杆正相关。由于无法获得相关数据来计算边际税率，本文采用平均税率 (TAX)来分析税收对资本结构选择的影响。 四、非债务税盾 DeAngelo 和 Masulis(1980)认为非债务税盾可作为债务融资税收利益的替代，在其他情况相同下，拥有较多非债务税盾的公司应更少地使用债务。 Barton et al(1989)、 Prowse(1990)、 Wald(1999)、 Kim和 Sorensen(1986)的研究表明非债务税盾
49、和杠杆负相关。本文用折旧 /总资产作为非债务税盾 (DEP)的代理变量。 五、成长性 根据代理理论，权益控制型公司倾向于次优投资而将财富从债权人手中剥夺过来。对于成长性较高的公司，由于其在选择将来投资机会时具有更大的灵活性，这些公司的代理问题可能更严重。 Myers(1977)认为高成长性公司较低成长性公司对将来投资具有更多的选择权。如果高成长性公司需要外部权益融资来执行将来选择权，那么具有大量债务的公司可能会放弃这个机会，因为 这样的投资将使财富从股东转移到债权人身上，即产生投资不足问题。因此，成长8 性应与杠杆负相关。对于成长性，本文使用总资产增长率 (GRTA)和权益增值率(GREQ)两个代理变量。 六、波动性 由于债务涉及定期支付义务，高杠杆公司更易受财务困境成本的影响。财务理论认为风险性公司或破产概率较高公司不应该具有较高的杠杆。因此，主营业务收入波动性或商业风险作为财务困境发生可能性的代理变量，其应与杠杆负相关。 Bradley et al(1984)、 Titman 和 Wesssels(1988)、 Wald(1999)和 Booth et al(2001)等的研究表明波动性与杠杆负相关。本文用主营业务的标准差率(VOL)作为波动性的代理变量。 七
