Does Market timing Affect Capital structure? Evidence ...
Paper #130012
Does Market timing Affect Capital structure? Evidence from Chinese listed companies Abstract Market timing theory of capital structure suggests that firms are more likely to issue equity when their stock price is high, and repurchase equity or issue debt when their stock price is low. So, the current capital structure is the cumulative outcome of past attempting to time the equity market. In this paper, we use the model in Baker and Wurgler(2002)’s study, and the variable of measuring market timing we use YT（Yearly Timing）constructed by Kayhan and Titman(2007). Furthermore, we improve the definition of the variables measuring capital structure, tangible assets and profitability respectively, by using the data from the adjusted management balance sheet as the basis of calculating the variables. The result indicates that Chinese listed companies do have marketing timing effect when making financing decision, however, the impact on the capital structure given by the market timing is not persistent.
Keywords: Market timing, Capital structure, Yearly timing, Long-term timing
1. Introduction Behavior finance has already proved that the capital market is not efficient, which means some stocks are likely to deviate from their intrinsic value for a long time. This gives the companies whose stocks are mispriced the opportunity of creating value by selling and purchasing their own stocks. So, Stein(1996) suggests a new theory of capital structure “market timing theory of capital structure”. This theory states that firms are likely to issue equity when their stock price is high, and repurchase equity or issue debt when their stock price is low. In reality, does market timing affect managers’ financing decision and change the firms’ capital structure？By reviewing the study of literature, we find that the research conclusions vary, the proxy for market timing is continually modified, so do the formulas for calculating the variables and financial statements. Barker and Wurgler(2002) examine the propensity of managers to “time the equity markets” by using the book-to-market ratios(M/B) and the historical market-to-book ratio(external finance weighted historical market-to-book ratio, EFWAMB) as the measure of market timing to study the American listed firms during the period of 1968-1999. They find that the market timing is significantly related to the capital structure and the fluctuations in market valuations have large effects on capital structure that persists for at least a decade. Mittoo and Zhang(2006), Bie and Haan(2007) examine market timing and its effects on capital structure based on a sample of
Paper #130012
Canada and Dutch listed firms by using EFWAMB as the measure of market timing respectively. Their results demonstrate that market timing does have an impact on the capital structure, but the impact in Canada and Dutch market does not persist for a long time. Kayhan and Titman(2007) improves the measure of market timing, through splitting Baker and Wurgler(2002)’s EFWAMB into two parts: Yearly timing(YT) and Long-term timing(LT). The former is used to measure the degree of stocks’ mispricing, the latter is the growth opportunity reflected in the stock price. Through analyzing the listed firms in American during the period of 1960-2003, they find that market timing does have an impact on the capital structure but the impact is not persistent. Mahajan and Tartaroglu(2008) use M/B and EFWAMB as the measure of market timing to test the market timing hypothesis of capital structure in major industrialized countries (G-7). They find that in all G-7 countries, except Japan, undo the effect of equity issuance and the impact of equity market timing attempts on leverage is short lived. In China, researches have been conducted based on the hypothesis of market timing, we summarize the current research from the following aspects: (1) Conclusions are not consistent. Liu Duan, Chen Shou and Chen Jian(2006) find that the market timing will affect the capital structure in a long time and this time length of the effect is five years. Li Guo-zhong(2006), Zhang Feng and Tang Hai-rong(2006) find that the market timing does have an impact on the capital structure but the impact is short lived. Hu Jin, Yan Yan-yang and Deng Ting(2008) suggest that market timing has little influence on the capital structure.(2) Former researchers are almost using M/B or EFWAMB as the measure of market timing. Kayhan and Titman(2007) point out that using M/B or EFWAMB as the measure of market timing is not accurate.(3) The sample data of the former studies are in the period before 2005. However, during the period of 2006-2007, stock markets in China are so active that market value of most stocks is high, that means market timing is likely to exist. Thus, it is necessary to study the market timing theory of capital structure in this period of time. Especially, in the former literature, it is not accurate to calculate the proxy based on the standard balance sheet. For example, using “total liabilities/total assets” as the measure of capital structure includes the influence on the capital structure given by operational liabilities. While using “financial liabilities/net operational assets” as the measure of capital structure, which is calculated based on the management balance sheet, is more effective to reveal the impact of market timing on capital structure in the financing decision-making. Therefore, in this paper, through using the empirical model frame of Baker and Wurgler(2002)’s and the testing method of market timing like Kayhan and Titman(2007), data of Chinese A share market from 1992 to 2007 is used as the research object. Based on the indicators
Paper #130012
which are calculated by using data from the balance sheet adjusted, we exam whether market timing has impact on the capital structure of the Chinese listed companies and the persistent of the impact The rest of the paper is organized as follows. In section 2, we choose some indicators to measure marketing timing and establish a regression model. In section 3, we introduce the samples and data source. In section 4 we present the empirical results. In section 5 ,we come up with a conclusion.
2. Methodology and variable construction In this paper, we use the empirical model frame in Baker and Wurgler(2002)’s study: (1)The main factors affecting the capital structure includes, marker timing, tangible assets, size, profitability.(2) Analyze timing measure and set up a regression model.(3) Select test samples and then make a regression. However, in this paper we use the modified measure constructed by Kayhan and Titman(2007) as the measure of market timing. Furthermore, in order to make our research more compatible with the essence of research and the reality of China，we modify others variables of the model.
2.1 Measure of Market timing Similar to Kayhan and Titman(2007), we suggest that the calculation of
EFWAMB is too
large (or too small) , this may be due to the stock is mispriced or the firms do have a lot of(or a little) growth opportunities, we describe below: t −1
EFWAMBt −1 = ∑ s =0
e +d ∑e + d s
r =0
=
× ( M / B) s =
s
t −1
r
YT + LT FD
cov( FD, M / B) + ( M / B) FD
r
（1）
t −1
t −1
s =0
s =0
LT(Long - term timing) t-1= ∑ (M / B) s / t ) * (∑ FDs / t ) = M / B * FD
（2）
t −1
YT(Yearly Timing) t-1= (∑ FDs * (M / B) s ) / t − M / B * FD = coˆv(FD, M / B) （3） s =0
Similar to Baker and Wurgler(2002), FD = (e + d ) presents the sum of external financing, denotes the average level over the period 0 to t − 1. Net equity issue (e) presents net equity issue and is defined as change between the beginning and the end of equity. Net debt issue (d) presents net debt issue and is defined as change between the beginning and the end of liabilities. Cov(FD, M/B) is the covariance between external financing and market valuation. M and B respectively
Paper #130012
represent market value and book value of equity which come from Wind financial database.
2.2 Measure of Leverage—D/A, net financial liabilities/net operational assets Similar to the former literatures, we use “total liabilities/total assets” (D/A) as the measure of capital structure. However, we find that former literatures have not specify whether the measures of “total assets” and “total liabilities” come from standard balance sheet directly released by the listed firms or management balance sheet which has been adjusted. If it comes from standard balance sheet, that means “total liabilities/total assets=(operational liabilities + financial liabilities)/ (operational liabilities + financial liabilities + book value of equity”. This shows that the data of “total liabilities/total assets” not only depends on the financial liabilities and equity arising from financing decision-making, but also affected by operational actives such as accounts payable, wages payable, taxes payable, notes payable. Thus, the measure of capital structure calculated above can not accurately reflect the original intention of this study—does market timing affect managers’ financing decision-making? In order to avoid the interference of operational liabilities, we use management balance sheet as the basis of calculation, and then “total liabilities in management balance sheet”=“net financial liabilities in standard balance sheet”=“financial liabilities-financial assets in standard balance sheet”, “total assets in management balance sheet ”=“net operational assets in standard balance sheet”=“operational assets-operational liabilities in standard balance sheet”. “D/A” bases on the management balance sheet can accurately reflect the managers’ financing decision-making.
2.3 Measure of Tangible assets—fixed assets/net operational assets If a large fraction of a firm’s assets are tangible, then assets should serve as collateral, diminishing the risk of the lender suffering the agency costs of debt. Therefore, the greater the proportion of tangible assets on the balance sheet, the more willing should lenders be to supply loans, and leverage should be higher (Rajan and Zingales, 1995; Baker and Wurgler,2002). Thus, the proportion of tangible assets may be one of the factors correlate with leverage. There are different measures for tangible assets. Someone use “fixed assets/total assets”(Rajan and Zingales,1995), and others use “(fixed assets + inventory)/total assets”(Hu Jun,etc,2008). They not only vary in the field of the scale of the numerator but also have not specify whether the measure of “total assets” comes from standard balance sheet directly released by the listed firms or management balance sheet which has been adjusted. In theory, accounts receivable、inventory are able to serve as collateral, however, in china, only fixed assets are served as collateral by banks, thus we adopt “fixed assets” as the numerator; The same with part2.2 in this paper, in the measure of tangible assets, the denominator “total assets” should not include financial assets and the current
Paper #130012
assets relate to the operational liabilities, so we use “fixed assets/net operational assets” as the measure of tangible assets.
2.4 Measure of Profitability—EBIT/net operational assets In the former literatures, the measure of profitability is also inconsistence. Someone use “EBIT/total assets”(Baker and Wurgler,2002) and others use “revenue/total assets”(Hu Jun,etc,2008). Through analyzing the nature of profitability, then we give the variable to reflect the profitability. Trade-off theory(Modigliani and Miller,1963) states that a value-maximizing firm will pursue an optimal capital structure by considering the marginal costs and benefits of each additional unit of financing, and then choosing the form of financing that equates these marginal costs and benefits. Signal theory suggests that firms with better profitability are more willing to use financial leverage, therefore, profitability is positive related to the debt ratio. However, according to Pecking order theory(Myers and Majluf,1984), firms will prefer to finance with internal funds rather than debt. All the theories of capital structure have proved that profitability correlate with capital structure. Thus, profitability may be one of the factors correlate with leverage. However, it is not difficult to find that profitability stated in trade-off theory and signal theory refers to the company’s business profitability rather than the company’s profitability as a whole. Because only when the profit generated from operating activities (EBIT) is greater than zero there will be tax shield effect and also only when (EBIT/the capital invested in business) is greater than the interest rate the correlation between leverage and the company’s profitability is a positive. However, in pecking order theory, profitability refers to the profits the company is able to obtain including operating and financial activities. If a company does not have good profitability in one or the whole company operation, due to the huge capital investment, it may have huge cash flow after tax. In addition, the assumption of pecking order theory is that there is asymmetric information or other friction in the capital market. Above of all, in this paper, profitability refers to the company’s business profitability. According to financial statement analysis , “EBIT/net operational assets” is the most accurate measure proxy for the company’s business profitability. Therefore, we select this variable as the measure of profitability. It should be noted that if the value in Baker and Wurgler(2002)’s study origin from management balance sheet, then their “EBIT/total assets” is consistence with “EBIT/net operational assets” in this paper.
2.5 Regression model Similar with Kayhan and Titman(2007),we construct a regression model as follows: (D/A)t=α+β1（YT）t-1+β2（LT）t-1+β3（T）t-1+β4（EBIT/A）t-1+β5 ㏒(Sales）t-1+εt （4）
Paper #130012
Where, D/A measures for capital structure in the t year;（YT）t-1、（LT）t-1、（T）t-1、(EBIT/A） t-1、log
(Sales）t-1 respectively measures for market timing、growth opportunities、tangible assets、
profitability and size. In the empirical analysis, first of all we need to adjust the standard balance sheet to management balance sheet, and then calculate the measures above in the part 2.1-2.4. The introduction and adjustment of management balance sheet is specified in Penman(2005).
2.6 Data source and Sample selection Because there are only 12 listed firms in Chinese stock market during the period of 1990-1991, the sample is too small; Meanwhile, during the period of 2006-2007, China is in the fast developing era, most stocks’ market values are so high, that means market timing is likely to exist. So our sample consists of firms listed in Chinese Stock Exchange at any point between 1992 and 2007. The initial sample includes 1,514 listed firms. Then, we exclude 27 financial firms as well as 164 S, ST, SST, *ST and S*ST firms from the sample. In addition, we restrict the sample to include firms with missing data or firms whose capital structure is more than one or less than zero. Our filtering process yields a sample of 1,077 firms, through further sorted by IPO time; we finally yield a sample of 6,670 firm-year observations as a data sample in this study. Sample size of each group is shown in table 1. All of the data comes from Wind financial database and uses the data at the end of each year. Table 1 Research samples of IPO in every year
Research
IPO
IPO+
IPO+
IPO+
IPO+
IPO+
IPO+
IPO+
IPO+
IPO+1
+1
2
3
4
5
6
7
8
9
0
915
866
853
771
720
662
605
523
461
394
samples
3. Empirical results We make regression analysis for model(4) using statistical software SPSS13.0. statistics for the whole sample are represented in Table 2. Table 2 Determinants of Leverage Variables Prof YT T
IPO+ 1
IPO+ 2
IPO+ 3
IPO+ 4
IPO+ 5
IPO+ 6
IPO+ 7
IPO+ 8
IPO+ 9
IPO+1 0
-0.249
-0.313 ***
-0.281 ***
-0.283 ***
-0.210 ***
-0.158 ***
-0.178 ***
-0.199 ***
0.209 ***
-0.35
0
-0.037
-0.118 ***
0.005
-0.29
0.049
0.067
0.105 *
0.048
0.019
0.070
-0.018
0.075
0.096
0.36
0.051
0.089
0.102
0.064
0.098*
Paper #130012
**
***
**
**
-0.024
0.157 ***
0.126 ***
0.145 ***
0.151 ***
0.175 ***
0.226 ***
0.287 ***
0.218 ***
0.212* ***
0.181
0.122 ***
0.124 ***
0.045
0.68
0.037
-0.29
-0.16
0.007
-0.052
Adjusted R Square
0.099
0.131
0.111
0.106
0.068
0.047
0.067
0.089
0.066
0.038
N
915
866
853
771
720
662
605
523
461
394
LT Size
a. Dependent Variable: D/A. Values significantly different from zero at 10%, 5% and 1% are marked *, ** and *** respectively From the results in table 2, YT has a highly significant negative coefficient with D/A in the third year after IPO which means market timing is one of the determinants of leverage, firms are more likely to issue equity when their market value is high, and repurchase equity or issue debt when their market value is low. However, since the third year after IPO, YT no longer has significant negative coefficients with D/A that means there is no persistent impact of market timing. In addition, since the second year after IPO LT has highly significant positive coefficients with D/A which means firms with more growth opportunities will issue more debt. This is inconsistent with Baker and Wurgler(2002).
4. Conclusion In this paper, we use the empirical model frame in Baker and Wurgler(2002)’s study, in fact we use the YT which is adopted by Kayhan 和 Titman(2007) to measure market timing. We also modify the definition of the model variable based on the research content and Chinese reality, to exam the influence of market timing given to capital structure of the Chinese listed companies. Baker and Wurgler(2002) suggests that the fluctuations in market valuations have large effects on capital structure that persist for at least a decade. However, the result in this paper shows that YT just has s highly significant negative coefficient with D/A in the third year after IPO which means market timing in Chinese market doesn’t have a persistent impact on capital structure. Why market timing in Chinese market doesn’t have a persistent impact on capital structure? The possible reasons are as follows: it needs the government’s carefully check and give official approval when the firms issue new stocks, also, it is difficult for firms to refinancing in a short time after the latest financing, so it is hard for managers to create value for their firms by using market timing even they can observe the market timing exiting in stock market. The reason may be that there is market friction in time and approach aspects while China firms make use of market timing.
Paper #130012
Reference Baker M. and Wurgler J., 2002, “Market timing and capital structure” Journal of Finance, No. 57, pp. 1–32. Bie T. and Haan L., June 2007, “Market Timing and Capital Structure: Evidence for Dutch Firms” De Economist, Volume 155, No. 2 , pp. 183-206. Bie , T , and Haan , L. 2004, “Does market timing drive capital structures A panel data study for Dutch firms.” Netherlands : De Nederlandsche Bank , Working Paper . Franco Modigliani and Merton H.Miller , June 1963, “Corporate income taxes and the cost of capital: a correction” The American Economic Review, Vol.53,No.3 pp.433-443. Hu Jun,Yan Yan-yang,Deng Ting, 2008, “Market timing and capital structure:Evdence from China” Journal of Finance Theory and Practice, No.3 7-10. Kayhan A. and Titman S, 2007, “Firms’ histories and their capital structure”, Journal of Financial Economics, No. 83, pp. 1–32. Liu Duan, Chen Shou and Chen Jian, Jan. 2006 ,“Study on the sustained effects of markets timing on capital structure,” Chinese Journal of Management, Vol. 3, No. 1, pp. 85-90. LI Guo-zhong, 2006, “Capital structure and market timing: evidence from cross-section data of Chinese listed companies,” Journal of Central University of Finance & Economics, No. 8, pp. 22-28. Li Xiao-ping, Wan Di-Fang, Yue Liang, 2007, “Market timing theory of capital structure review”, Foreign Economics & Management, 59-65. Mittoo ,U R ,and Zhang , Zhou. 2006, “Market timing , capital structure and cross listing : Canadian evidence” . University of Manitoba ,Working Paper . Michael C. Jensen and Wiiliam H.Meckling, 1976, “Theory of the firm: managerial behavior, agency costs”, Journal of Financial Economics ,Vol.3,No.4. Myers,Stewart C.,and Nicholas S.Majluf, 1984, “Corporate financing and investment decisions when firms have information that investors do not have”, Journal of Financial Economics 13,187-221. Rajan , R. , and L. Zingales , 1995, “What Do We Know about Capital Structure Some Evidence from International Data” Journal of Finance , 50 .
Paper #130012
Ross S., 1977 ,“The determination of financial structure: the incentive signalling approach” Journal of Economics, Spring, 23-40. Wang Zheng-wei,Zhu Wu-xiang,Zhao Dong-Qing, 2007, “Market Timing in Seasoned Equity Offerings with Security Issue Regulation and Its Impact on Capital Structure”, Journal of Nankai Management Review, Vol.10,No.6,40-46. ZHANG Feng and TANG Hai-rong, 2006 ,“An empirical analysis of timing and its persistent effect on financing decision of listed firms,” East China Economic Management, Vol.20, No. pp. 131-135, Feb. Stephen H. Penman, 2005, “Financial Statement Analysis and Securities Valuation”, China Financial & Economic Publishing House.
Does Market timing Affect Capital structure? Evidence from Chinese listed companies Abstract Market timing theory of capital structure suggests that firms are more likely to issue equity when their stock price is high, and repurchase equity or issue debt when their stock price is low. So, the current capital structure is the cumulative outcome of past attempting to time the equity market. In this paper, we use the model in Baker and Wurgler(2002)’s study, and the variable of measuring market timing we use YT（Yearly Timing）constructed by Kayhan and Titman(2007). Furthermore, we improve the definition of the variables measuring capital structure, tangible assets and profitability respectively, by using the data from the adjusted management balance sheet as the basis of calculating the variables. The result indicates that Chinese listed companies do have marketing timing effect when making financing decision, however, the impact on the capital structure given by the market timing is not persistent.
Keywords: Market timing, Capital structure, Yearly timing, Long-term timing
1. Introduction Behavior finance has already proved that the capital market is not efficient, which means some stocks are likely to deviate from their intrinsic value for a long time. This gives the companies whose stocks are mispriced the opportunity of creating value by selling and purchasing their own stocks. So, Stein(1996) suggests a new theory of capital structure “market timing theory of capital structure”. This theory states that firms are likely to issue equity when their stock price is high, and repurchase equity or issue debt when their stock price is low. In reality, does market timing affect managers’ financing decision and change the firms’ capital structure？By reviewing the study of literature, we find that the research conclusions vary, the proxy for market timing is continually modified, so do the formulas for calculating the variables and financial statements. Barker and Wurgler(2002) examine the propensity of managers to “time the equity markets” by using the book-to-market ratios(M/B) and the historical market-to-book ratio(external finance weighted historical market-to-book ratio, EFWAMB) as the measure of market timing to study the American listed firms during the period of 1968-1999. They find that the market timing is significantly related to the capital structure and the fluctuations in market valuations have large effects on capital structure that persists for at least a decade. Mittoo and Zhang(2006), Bie and Haan(2007) examine market timing and its effects on capital structure based on a sample of
Paper #130012
Canada and Dutch listed firms by using EFWAMB as the measure of market timing respectively. Their results demonstrate that market timing does have an impact on the capital structure, but the impact in Canada and Dutch market does not persist for a long time. Kayhan and Titman(2007) improves the measure of market timing, through splitting Baker and Wurgler(2002)’s EFWAMB into two parts: Yearly timing(YT) and Long-term timing(LT). The former is used to measure the degree of stocks’ mispricing, the latter is the growth opportunity reflected in the stock price. Through analyzing the listed firms in American during the period of 1960-2003, they find that market timing does have an impact on the capital structure but the impact is not persistent. Mahajan and Tartaroglu(2008) use M/B and EFWAMB as the measure of market timing to test the market timing hypothesis of capital structure in major industrialized countries (G-7). They find that in all G-7 countries, except Japan, undo the effect of equity issuance and the impact of equity market timing attempts on leverage is short lived. In China, researches have been conducted based on the hypothesis of market timing, we summarize the current research from the following aspects: (1) Conclusions are not consistent. Liu Duan, Chen Shou and Chen Jian(2006) find that the market timing will affect the capital structure in a long time and this time length of the effect is five years. Li Guo-zhong(2006), Zhang Feng and Tang Hai-rong(2006) find that the market timing does have an impact on the capital structure but the impact is short lived. Hu Jin, Yan Yan-yang and Deng Ting(2008) suggest that market timing has little influence on the capital structure.(2) Former researchers are almost using M/B or EFWAMB as the measure of market timing. Kayhan and Titman(2007) point out that using M/B or EFWAMB as the measure of market timing is not accurate.(3) The sample data of the former studies are in the period before 2005. However, during the period of 2006-2007, stock markets in China are so active that market value of most stocks is high, that means market timing is likely to exist. Thus, it is necessary to study the market timing theory of capital structure in this period of time. Especially, in the former literature, it is not accurate to calculate the proxy based on the standard balance sheet. For example, using “total liabilities/total assets” as the measure of capital structure includes the influence on the capital structure given by operational liabilities. While using “financial liabilities/net operational assets” as the measure of capital structure, which is calculated based on the management balance sheet, is more effective to reveal the impact of market timing on capital structure in the financing decision-making. Therefore, in this paper, through using the empirical model frame of Baker and Wurgler(2002)’s and the testing method of market timing like Kayhan and Titman(2007), data of Chinese A share market from 1992 to 2007 is used as the research object. Based on the indicators
Paper #130012
which are calculated by using data from the balance sheet adjusted, we exam whether market timing has impact on the capital structure of the Chinese listed companies and the persistent of the impact The rest of the paper is organized as follows. In section 2, we choose some indicators to measure marketing timing and establish a regression model. In section 3, we introduce the samples and data source. In section 4 we present the empirical results. In section 5 ,we come up with a conclusion.
2. Methodology and variable construction In this paper, we use the empirical model frame in Baker and Wurgler(2002)’s study: (1)The main factors affecting the capital structure includes, marker timing, tangible assets, size, profitability.(2) Analyze timing measure and set up a regression model.(3) Select test samples and then make a regression. However, in this paper we use the modified measure constructed by Kayhan and Titman(2007) as the measure of market timing. Furthermore, in order to make our research more compatible with the essence of research and the reality of China，we modify others variables of the model.
2.1 Measure of Market timing Similar to Kayhan and Titman(2007), we suggest that the calculation of
EFWAMB is too
large (or too small) , this may be due to the stock is mispriced or the firms do have a lot of(or a little) growth opportunities, we describe below: t −1
EFWAMBt −1 = ∑ s =0
e +d ∑e + d s
r =0
=
× ( M / B) s =
s
t −1
r
YT + LT FD
cov( FD, M / B) + ( M / B) FD
r
（1）
t −1
t −1
s =0
s =0
LT(Long - term timing) t-1= ∑ (M / B) s / t ) * (∑ FDs / t ) = M / B * FD
（2）
t −1
YT(Yearly Timing) t-1= (∑ FDs * (M / B) s ) / t − M / B * FD = coˆv(FD, M / B) （3） s =0
Similar to Baker and Wurgler(2002), FD = (e + d ) presents the sum of external financing, denotes the average level over the period 0 to t − 1. Net equity issue (e) presents net equity issue and is defined as change between the beginning and the end of equity. Net debt issue (d) presents net debt issue and is defined as change between the beginning and the end of liabilities. Cov(FD, M/B) is the covariance between external financing and market valuation. M and B respectively
Paper #130012
represent market value and book value of equity which come from Wind financial database.
2.2 Measure of Leverage—D/A, net financial liabilities/net operational assets Similar to the former literatures, we use “total liabilities/total assets” (D/A) as the measure of capital structure. However, we find that former literatures have not specify whether the measures of “total assets” and “total liabilities” come from standard balance sheet directly released by the listed firms or management balance sheet which has been adjusted. If it comes from standard balance sheet, that means “total liabilities/total assets=(operational liabilities + financial liabilities)/ (operational liabilities + financial liabilities + book value of equity”. This shows that the data of “total liabilities/total assets” not only depends on the financial liabilities and equity arising from financing decision-making, but also affected by operational actives such as accounts payable, wages payable, taxes payable, notes payable. Thus, the measure of capital structure calculated above can not accurately reflect the original intention of this study—does market timing affect managers’ financing decision-making? In order to avoid the interference of operational liabilities, we use management balance sheet as the basis of calculation, and then “total liabilities in management balance sheet”=“net financial liabilities in standard balance sheet”=“financial liabilities-financial assets in standard balance sheet”, “total assets in management balance sheet ”=“net operational assets in standard balance sheet”=“operational assets-operational liabilities in standard balance sheet”. “D/A” bases on the management balance sheet can accurately reflect the managers’ financing decision-making.
2.3 Measure of Tangible assets—fixed assets/net operational assets If a large fraction of a firm’s assets are tangible, then assets should serve as collateral, diminishing the risk of the lender suffering the agency costs of debt. Therefore, the greater the proportion of tangible assets on the balance sheet, the more willing should lenders be to supply loans, and leverage should be higher (Rajan and Zingales, 1995; Baker and Wurgler,2002). Thus, the proportion of tangible assets may be one of the factors correlate with leverage. There are different measures for tangible assets. Someone use “fixed assets/total assets”(Rajan and Zingales,1995), and others use “(fixed assets + inventory)/total assets”(Hu Jun,etc,2008). They not only vary in the field of the scale of the numerator but also have not specify whether the measure of “total assets” comes from standard balance sheet directly released by the listed firms or management balance sheet which has been adjusted. In theory, accounts receivable、inventory are able to serve as collateral, however, in china, only fixed assets are served as collateral by banks, thus we adopt “fixed assets” as the numerator; The same with part2.2 in this paper, in the measure of tangible assets, the denominator “total assets” should not include financial assets and the current
Paper #130012
assets relate to the operational liabilities, so we use “fixed assets/net operational assets” as the measure of tangible assets.
2.4 Measure of Profitability—EBIT/net operational assets In the former literatures, the measure of profitability is also inconsistence. Someone use “EBIT/total assets”(Baker and Wurgler,2002) and others use “revenue/total assets”(Hu Jun,etc,2008). Through analyzing the nature of profitability, then we give the variable to reflect the profitability. Trade-off theory(Modigliani and Miller,1963) states that a value-maximizing firm will pursue an optimal capital structure by considering the marginal costs and benefits of each additional unit of financing, and then choosing the form of financing that equates these marginal costs and benefits. Signal theory suggests that firms with better profitability are more willing to use financial leverage, therefore, profitability is positive related to the debt ratio. However, according to Pecking order theory(Myers and Majluf,1984), firms will prefer to finance with internal funds rather than debt. All the theories of capital structure have proved that profitability correlate with capital structure. Thus, profitability may be one of the factors correlate with leverage. However, it is not difficult to find that profitability stated in trade-off theory and signal theory refers to the company’s business profitability rather than the company’s profitability as a whole. Because only when the profit generated from operating activities (EBIT) is greater than zero there will be tax shield effect and also only when (EBIT/the capital invested in business) is greater than the interest rate the correlation between leverage and the company’s profitability is a positive. However, in pecking order theory, profitability refers to the profits the company is able to obtain including operating and financial activities. If a company does not have good profitability in one or the whole company operation, due to the huge capital investment, it may have huge cash flow after tax. In addition, the assumption of pecking order theory is that there is asymmetric information or other friction in the capital market. Above of all, in this paper, profitability refers to the company’s business profitability. According to financial statement analysis , “EBIT/net operational assets” is the most accurate measure proxy for the company’s business profitability. Therefore, we select this variable as the measure of profitability. It should be noted that if the value in Baker and Wurgler(2002)’s study origin from management balance sheet, then their “EBIT/total assets” is consistence with “EBIT/net operational assets” in this paper.
2.5 Regression model Similar with Kayhan and Titman(2007),we construct a regression model as follows: (D/A)t=α+β1（YT）t-1+β2（LT）t-1+β3（T）t-1+β4（EBIT/A）t-1+β5 ㏒(Sales）t-1+εt （4）
Paper #130012
Where, D/A measures for capital structure in the t year;（YT）t-1、（LT）t-1、（T）t-1、(EBIT/A） t-1、log
(Sales）t-1 respectively measures for market timing、growth opportunities、tangible assets、
profitability and size. In the empirical analysis, first of all we need to adjust the standard balance sheet to management balance sheet, and then calculate the measures above in the part 2.1-2.4. The introduction and adjustment of management balance sheet is specified in Penman(2005).
2.6 Data source and Sample selection Because there are only 12 listed firms in Chinese stock market during the period of 1990-1991, the sample is too small; Meanwhile, during the period of 2006-2007, China is in the fast developing era, most stocks’ market values are so high, that means market timing is likely to exist. So our sample consists of firms listed in Chinese Stock Exchange at any point between 1992 and 2007. The initial sample includes 1,514 listed firms. Then, we exclude 27 financial firms as well as 164 S, ST, SST, *ST and S*ST firms from the sample. In addition, we restrict the sample to include firms with missing data or firms whose capital structure is more than one or less than zero. Our filtering process yields a sample of 1,077 firms, through further sorted by IPO time; we finally yield a sample of 6,670 firm-year observations as a data sample in this study. Sample size of each group is shown in table 1. All of the data comes from Wind financial database and uses the data at the end of each year. Table 1 Research samples of IPO in every year
Research
IPO
IPO+
IPO+
IPO+
IPO+
IPO+
IPO+
IPO+
IPO+
IPO+1
+1
2
3
4
5
6
7
8
9
0
915
866
853
771
720
662
605
523
461
394
samples
3. Empirical results We make regression analysis for model(4) using statistical software SPSS13.0. statistics for the whole sample are represented in Table 2. Table 2 Determinants of Leverage Variables Prof YT T
IPO+ 1
IPO+ 2
IPO+ 3
IPO+ 4
IPO+ 5
IPO+ 6
IPO+ 7
IPO+ 8
IPO+ 9
IPO+1 0
-0.249
-0.313 ***
-0.281 ***
-0.283 ***
-0.210 ***
-0.158 ***
-0.178 ***
-0.199 ***
0.209 ***
-0.35
0
-0.037
-0.118 ***
0.005
-0.29
0.049
0.067
0.105 *
0.048
0.019
0.070
-0.018
0.075
0.096
0.36
0.051
0.089
0.102
0.064
0.098*
Paper #130012
**
***
**
**
-0.024
0.157 ***
0.126 ***
0.145 ***
0.151 ***
0.175 ***
0.226 ***
0.287 ***
0.218 ***
0.212* ***
0.181
0.122 ***
0.124 ***
0.045
0.68
0.037
-0.29
-0.16
0.007
-0.052
Adjusted R Square
0.099
0.131
0.111
0.106
0.068
0.047
0.067
0.089
0.066
0.038
N
915
866
853
771
720
662
605
523
461
394
LT Size
a. Dependent Variable: D/A. Values significantly different from zero at 10%, 5% and 1% are marked *, ** and *** respectively From the results in table 2, YT has a highly significant negative coefficient with D/A in the third year after IPO which means market timing is one of the determinants of leverage, firms are more likely to issue equity when their market value is high, and repurchase equity or issue debt when their market value is low. However, since the third year after IPO, YT no longer has significant negative coefficients with D/A that means there is no persistent impact of market timing. In addition, since the second year after IPO LT has highly significant positive coefficients with D/A which means firms with more growth opportunities will issue more debt. This is inconsistent with Baker and Wurgler(2002).
4. Conclusion In this paper, we use the empirical model frame in Baker and Wurgler(2002)’s study, in fact we use the YT which is adopted by Kayhan 和 Titman(2007) to measure market timing. We also modify the definition of the model variable based on the research content and Chinese reality, to exam the influence of market timing given to capital structure of the Chinese listed companies. Baker and Wurgler(2002) suggests that the fluctuations in market valuations have large effects on capital structure that persist for at least a decade. However, the result in this paper shows that YT just has s highly significant negative coefficient with D/A in the third year after IPO which means market timing in Chinese market doesn’t have a persistent impact on capital structure. Why market timing in Chinese market doesn’t have a persistent impact on capital structure? The possible reasons are as follows: it needs the government’s carefully check and give official approval when the firms issue new stocks, also, it is difficult for firms to refinancing in a short time after the latest financing, so it is hard for managers to create value for their firms by using market timing even they can observe the market timing exiting in stock market. The reason may be that there is market friction in time and approach aspects while China firms make use of market timing.
Paper #130012
Reference Baker M. and Wurgler J., 2002, “Market timing and capital structure” Journal of Finance, No. 57, pp. 1–32. Bie T. and Haan L., June 2007, “Market Timing and Capital Structure: Evidence for Dutch Firms” De Economist, Volume 155, No. 2 , pp. 183-206. Bie , T , and Haan , L. 2004, “Does market timing drive capital structures A panel data study for Dutch firms.” Netherlands : De Nederlandsche Bank , Working Paper . Franco Modigliani and Merton H.Miller , June 1963, “Corporate income taxes and the cost of capital: a correction” The American Economic Review, Vol.53,No.3 pp.433-443. Hu Jun,Yan Yan-yang,Deng Ting, 2008, “Market timing and capital structure:Evdence from China” Journal of Finance Theory and Practice, No.3 7-10. Kayhan A. and Titman S, 2007, “Firms’ histories and their capital structure”, Journal of Financial Economics, No. 83, pp. 1–32. Liu Duan, Chen Shou and Chen Jian, Jan. 2006 ,“Study on the sustained effects of markets timing on capital structure,” Chinese Journal of Management, Vol. 3, No. 1, pp. 85-90. LI Guo-zhong, 2006, “Capital structure and market timing: evidence from cross-section data of Chinese listed companies,” Journal of Central University of Finance & Economics, No. 8, pp. 22-28. Li Xiao-ping, Wan Di-Fang, Yue Liang, 2007, “Market timing theory of capital structure review”, Foreign Economics & Management, 59-65. Mittoo ,U R ,and Zhang , Zhou. 2006, “Market timing , capital structure and cross listing : Canadian evidence” . University of Manitoba ,Working Paper . Michael C. Jensen and Wiiliam H.Meckling, 1976, “Theory of the firm: managerial behavior, agency costs”, Journal of Financial Economics ,Vol.3,No.4. Myers,Stewart C.,and Nicholas S.Majluf, 1984, “Corporate financing and investment decisions when firms have information that investors do not have”, Journal of Financial Economics 13,187-221. Rajan , R. , and L. Zingales , 1995, “What Do We Know about Capital Structure Some Evidence from International Data” Journal of Finance , 50 .
Paper #130012
Ross S., 1977 ,“The determination of financial structure: the incentive signalling approach” Journal of Economics, Spring, 23-40. Wang Zheng-wei,Zhu Wu-xiang,Zhao Dong-Qing, 2007, “Market Timing in Seasoned Equity Offerings with Security Issue Regulation and Its Impact on Capital Structure”, Journal of Nankai Management Review, Vol.10,No.6,40-46. ZHANG Feng and TANG Hai-rong, 2006 ,“An empirical analysis of timing and its persistent effect on financing decision of listed firms,” East China Economic Management, Vol.20, No. pp. 131-135, Feb. Stephen H. Penman, 2005, “Financial Statement Analysis and Securities Valuation”, China Financial & Economic Publishing House.
