Research on the Influence of Capital Structure on R&D Investment

based on Big Data of Listed Companies

Xiaotang Huang

and Chunying Ma

Business School of Shenyang University, Liaoning, Shenyang, China

Keywords: Capital Structure, R&D Investment, Big Data of Listed Companies.

Abstract: Based on the big data of China's computer, communication and other electronic equipment manufacturing

listed companies from 2017 to 2020. This paper empirically analyzes the impact of capital structure on R&D

investment.

1 INTRODUCTION

The R&D level of an enterprise reflects its

competitiveness. A slight change in the allocation of

property right ratio, ownership structure and debt

structure that reflects the capital structure will affect

the amount of R&D investment of enterprises.

Therefore, this paper selects computer,

communication and other electronic equipment

manufacturing industries to conduct in-depth

research, and empirically analyzes the impact of

capital structure on R&D investment intensity and

scale, hoping that the conclusion will have theoretical

reference for this industry and even other industries

(Duan 2020).

2 BUILDUP OF MODEL

This paper constructs two multiple regression models

to verify the impact of capital structure on R&D

investment intensity and scale (Liu 2018).

2.1 R&D Intensity

RDI=β0+β1DER+β2OCD+β3CDR+β4SIZE+β5RI+

β6GRO+β7AGE+ε

2.2 R&D Investment Scale

RD=β0+β1DER+β2OCD+β3CDR+β4SIZE+β5RI+β

6GRO+β7AGE+ε

Where β0 is a constant and ε represents random

error

Table 1: Variable Description and Measurement Description.

Variable type variable symbol meaning variable value method and description

The explained variable

RDI R&D intensity

R&D expenditure /end-of-period total

assets

RD R&D investment scale R&D costs take natural logarithm

Explanatory variable

DER equity ratio

Total liabilities/ total shareholders '

equity

OCD

Ownership

Concentration

Square sum of top ten shareholders

CDR current liabilities ratio current liabilities / total liabilities

Control variable

SIZE enterprise scale Ln (yea

-end total assets)

rofitabilit

net profit / total assets

GRO Enterprise Growth Main Business Income Growth Rate

AGE Listing Time Natural logarithm of listing time

Huang, X. and Ma, C.

Research on the Inﬂuence of Capital Structure on RD Investment based on Big Data of Listed Companies.

DOI: 10.5220/0011350900003440

In Proceedings of the International Conference on Big Data Economy and Digital Management (BDEDM 2022), pages 849-853

ISBN: 978-989-758-593-7

849

3 EMPIRICAL RESEARCH

This paper selects 39 types of computer,

communication and other electronic equipment

manufacturing enterprises in the industry category

code under A-share manufacturing (C) in Shenzhen

and Shanghai Stock Exchanges from 2017 to 2020 as

the research sample selection range.

3.1 Descriptive Statistics

3.1.1 Descriptive Statistics of the Explained

Variables

First of all, the average value of the relative index

(RDI) of R&D investment intensity of enterprises in

each year in Table 2 is between 1% and 2%. It can be

seen that the R&D investment level of China ' s

computer, communication and other electronic

equipment manufacturing industry is low. Overall,

RDI (mean) has an increasing trend year by year, but

the increase is small; the scale of corporate R&D

investment (the natural logarithm of R&D

expenditure) gradually increased from 16.34 in 2017

to 16.62 in 2020, which also showed a slight upward

trend. Therefore, based on the above, the conclusion

is that the scale of R&D shows an upward trend in

terms of both R&D investment intensity and R&D

investment scale, which benefits from China’ s

emphasis on R&D and innovation in recent years. To

develop into an innovative country, we should start

with increasing R&D investment (Lv 2010).

Table 2: Descriptive statistics of the R&D input intensity (RDI) and scale (RD).

year

sample capacity ndex inimum maximum mean standard deviation

2017

25 DI 0.000040 0.046580 0.011266 0.012174

RD 11.8204 21.1061 16.3445 2.2133

2018 42 RDI 0.000072 0.082393 0.014952 0.018850

RD 12.4070 19.6921 16.4025 1.7525

2019 59 RDI 0.000062 0.128730 0.016478 0.022528

RD 11.8578 21.6181 16.5763 2.0924

2020 60 RDI 0.000065 0.167144 0.016174 0.024453

RD 11.8578 21.7990 16.6209 1.9149

ample overall 186 RDI 0.000040 0.167144 0.015335 0.021230

RD 11.8204 21.7990 16.5203 1.9672

3.1.2 Descriptive Statistics on Explanatory

and Control Variables

The above table is the result of descriptive statistics

on explanatory and control variables from 2017 to

2020. There is still a big gap between the minimum

value of 0.0248 and the maximum value of 6.1816 in

the property right ratio (DER), that is, there are great

differences in the use of debt financing and equity

financing among enterprises. It is generally believed

that the financial structure with the average value of

the property right ratio close to 1 is stable. From the

average value of 0.7649, the ratio is less than 1, that

is, in general, equity financing is more selected than

debt financing. Ownership concentration (OCD)

minimum is 0.0137, the maximum is 0.4699, the

average is 0.1440, the difference is not big. The

maximum value of current liabilities ratio (CDR) is

1, indicating that all liabilities of the company are

current liabilities. From the perspective of the

average value of 0.8500, the sample companies focus

more on short-term current liabilities in debt

financing. From the perspective of financing sources,

short-term financing is easier, the company ' s

repayment period is short, the risk is small from the

perspective of creditors, banks or other financial

institutions are more inclined to short-term lending to

enterprises. The minimum value, minimum value and

mean value of enterprise scale (SIZE) have little

difference, indicating that the selected company scale

is basically balanced. The minimum value of

profitability (RI) is negative, but the absolute value of

the minimum and maximum is almost the same. The

minimum value of enterprise growth (GRO) is

negative, but it is almost the same as the absolute

value of the maximum value, and the situation is

good. Listing time (AGE) minimum, maximum,

mean change little, the situation is good.

BDEDM 2022 - The International Conference on Big Data Economy and Digital Management

850

Table 3: Descriptive statistics of explanatory variables and control variables of sample companies.

index minimum

maximum mean

standard deviation

DER 0.0248 6.1816 0.7649 0.8250

OCD 0.0137 0.4699 0.1440 0.0877

CDR 0.1432 1.0000 0.8500 0.1675

SIZE 19.5411 25.4003 21.5779 1.1969

RI -0.4042 0.1917 0.0306 0.0604

GRO -0.5690 2.1575 0.1955 0.3546

AGE 1.3863 3.0910 2.2535 0.4965

N 186 186 186 186

3.2 Correlation Analysis

Table 4: Correlation between explanatory and explained variables (Model 1).

RDI DER OCD CDR

RDI Pearson correlation 1

Significance

(bilateral)

DER Pearson correlation 0.090 1

Significance

(bilateral)

0.220

OCD Pearson correlation 0.249** -0.128 1

Significance

(bilateral)

0.001 0.083

CDR Pearson correlation -0.240** -0.049 -0.024 1

Significance

(bilateral)

0.001 0.507 0.749

**. Significant correlation was significant at 0.01 level (bilateral).

Table 5: Correlation of explanatory and explained variables (Model 2).

RD DER OCD CDR

RD Pearson correlation 1

Significance (bilateral)

DER Pearson correlation 0.455** 1

Significance (bilateral) 0.000

OCD Pearson correlation 0.057 -0.128 1

Significance (bilateral) 0.443 0.083

CDR Pearson correlation -0.211** -0.049 -0.024 1

Significance (bilateral) 0.004 0.507 0.749

**. Significant correlation was significant at 0.01 level (bilateral).

Table 4 shows the ownership ratio (DER) and R&D

intensity (RDI) were positively correlated, but the

significant (bilateral) value was 0.220, and the

correlation was not significant. Ownership

concentration (OCD) is positively correlated with

R&D investment intensity (RDI), and current debt

ratio (CDR) is negatively correlated with R&D

investment intensity (RDI), and both are at 1% level.

Table 5 shows the correlation between the three

independent variables representing the capital

structure and the dependent variable R&D investment

scale (RD) in Model 2. For the correlation between

independent variables and dependent variables, first

of all, the Pearson correlation coefficient between the

ratio of property rights (DER) and the scale of R&D

investment (RD) is 0.455, that is, the two are

Research on the Inﬂuence of Capital Structure on RD Investment based on Big Data of Listed Companies

851

positively correlated and are significantly indigenous

at the 1% level. Ownership concentration (OCD) is

positively correlated with R&D investment scale

(RD), but the coefficient of significant (bilateral) is

0.443, so the correlation is not significant. The

Pearson correlation coefficient between current debt

ratio (CDR) and R&D investment scale (RD) is-

0.211, that is, the two are negatively correlated at the

1% level.

3.3 Regression Analysis

Table 6: Results of model I and II regression analysis.

variable Model 1（RDI） Model 2（RD）

Standard

coefficient

（B）

T value sig. VIF Standard

coefficient

（B）

T value sig. VIF

Constant

2.790 0.006***

-0.453 0.651

DER 0.273 2.909 0.004*** 1.925 0.184 2.337 0.021** 1.925

OCD 0.246 3.465 0.001*** 1.100 0.102 1.716 0.088* 1.100

CDR -0.308 -4.262 0.000*** 1.144 -0.129 -2.117 0.036** 1.144

SIZE -0.146 -1.501 0.135 2.056 0.571 7.009 0.000*** 2.056

RI -0.029 -0.383 0.702 1.266 -0.117 -1.832 0.069* 1.266

GRO -0.168 -2.222 0.028** 1.245 -0.076 -1.199 0.232 1.245

AGE -0.118 -1.414 0.159 1.533 -0.179 -2.545 0.012** 1.533

Adj-R2 0.155 0.403

F-Value

(Sig.)

5.847 18.832

(0.000) (0.000)

Note : *. , * *. , * * *. respectively, indicating that the regression coefficients are significant at 10%, 5%, 1% levels.

Table 6 shows that the F values of Model 1 and Model

2 are 5.847 and 18.832, respectively, and the sig

values are 0.000, indicating that both the regression

models have significant statistical significance.

First of all, the data results of the first analysis

model show that the three independent variables of

property rights ratio (DER), ownership concentration

(OCD), current liabilities ratio (CDR) have passed

the test of the regression coefficient, which are at the

level of 1%, indicating that under the control of other

variables, the above three independent variables have

a significant impact on the dependent variable. The

strongest explanatory power is the current debt ratio

(CDR), and the coefficient is -0.308. Observing the

regression coefficient, the first two are positive, the

latter is negative, which is consistent with the

correlation analysis results of the previous model one,

that is, the ownership ratio (DER), ownership

concentration (OCD) and R&D investment intensity

are positively correlated; current debt ratio (CDR) is

negatively correlated with R&D investment intensity.

Then look at the variance expansion factor (VIF) in

the table, VIF values are between 0 and 10, so the

independent variables in model 1 do not have serious

collinearity problem.

The data results of the second analysis model

show that the three independent variables have passed

the significance test, and the property right ratio

(DER), ownership concentration ratio (OCD), current

liabilities ratio (CDR) and R&D investment scale

(RD) are significantly indigenous at the levels of 5%,

10% and 5%, respectively. It also shows that the

above three independent variables have significant

indigenous effects on the dependent variables when

other variables are controlled. Looking at the

standardization coefficient, the independent variable

with the strongest explanatory power of R&D

investment scale (RD) is the property right ratio

(DER), the coefficient is 0.184. By observing the

regression coefficient, the first two are positive, and

the latter is negative, which is consistent with the

correlation analysis results of Model 2, namely, the

property right ratio (DER) and ownership

concentration (OCD) are positively correlated with

the scale of R&D investment. The current debt ratio

(CDR) is negatively correlated with the scale of R&D

investment. The variance expansion factor VIF in the

table is between 0 and 10, so there is no serious

collinearity problem in model 2 (Lv 2018).

BDEDM 2022 - The International Conference on Big Data Economy and Digital Management

852

4 CONCLUSIONS AND POLICY

RECOMMENDATIONS

Through the above research, the following

conclusions can be drawn: the proportion of property

rights and ownership concentration are positively

correlated with the intensity and scale of R&D

investment, and the current debt ratio is negatively

correlated with the intensity and scale of R&D

investment (Zhang 2019).

First, enterprises should choose more debt

financing in the choice of financing methods, so that

the management control rights of existing

shareholders will not be diluted. The existing large

shareholders will stand in the company’ s long-term

development and pay more attention to R&D

investment. Second, in terms of ownership structure,

it is necessary to appropriately enhance the ownership

concentration of enterprises. The higher the

ownership concentration is, the larger shareholders

controlled by enterprises will have more discourse

power, and they are also the real owners of enterprises

(small shareholders generally make short-term

investments). For their own interests and the good

development of enterprises, they will operate and

manage more seriously. Third, in terms of debt

structure, if the enterprise needs debt for R&D

investment, it is difficult for the author to recommend

long-term borrowing. Because of the lag of R&D

investment, the potential economic return funds can

not be recovered in the short term. The current debt

will make the enterprise have a lot of repayment

pressure, which is likely to cause the rupture of

capital flow and bring risks to the normal operation.

REFERENCES

Duan, Xin, Ren, QunLuo, 2020. The impact of physical

capital, human capital and R&D investment on

economic growth — from the perspective of industrial

structure optimization [J]. Journal of Hubei University

of Arts and Sciences, 2020, 41 (05): 40 – 48.

Liu, Wei, 2018. Literature Review on the Impact of Capital

Structure on R&D Investment [J]. Shandong Textile

Economy, 2018 (03): 16-19.

Lv, Minle, 2010, Wang, Xiaohu. Corporate capital structure

and R&D investment - Based on the empirical research

of automobile manufacturing listed companies [J].

China Science and Technology Forum, 2010, (1) : 62-

66.

Lv, Wenping, 2018. A review of the correlation between

capital structure, R&D investment and corporate

performance [J]. Accounting for township enterprises

in China, 2018 (05): 10-11.

Zhang, Rongming, 2019. The relationship between family

control, venture capital and corporate R& D

investment [J]. Shanxi Agricultural Economics,

2019(05):30-31.

Research on the Inﬂuence of Capital Structure on RD Investment based on Big Data of Listed Companies

853