Journal of Social Sciences Studies, Volume 2, ISSN: 2664-0287

Research on the Relationship between Accounting Information Quality

and Enterprise Investment Efficiency
Aoxue Liu1,a,*
1
School of Economics and Management, Beijing Jiaotong University, China
a
17120658@bjtu.edu.cn
*Corresponding author

Keywords: Insufficient investment, Excessive investment, Accounting information quality.

Abstract. Based on the study of the relationship between the quality of accounting information and
investment efficiency, this paper empirically analyses the relationship between the quality of
accounting information and the efficiency of enterprise investment based on the statistics of A-share
listed company in 2013-2016. In this paper, the modified Jones model and Richardson investment
expectation model are used to measure the main variables, and descriptive statistical analysis,
correlation analysis and regression analysis are carried out to study the relationship between the two.
It is found that non-financial listed companies in China generally have the phenomenon of
inefficient investment and the phenomenon of underinvestment is more common, the quality of
high-quality accounting information can alleviate the underinvestment and overinvestment of listed
companies, the impact of high quality accounting information on inefficient investment of
state-controlled listed enterprises is stronger than that of the non-state-controlled listed companies.

1. Introduction
Investment decision is the core of enterprise financial decision, which determines the financing
behavior of enterprise, and it also has a vital influence on the investment income of the enterprise.
The investment decision of the enterprise can not only influence the market value of the enterprise
and the change of the shareholder's wealth, but also can influence the stability of the
macro-economy, the enterprise investment decision can be reflected from the enterprise's
investment efficiency.
However, Chinese capital market and corporate governance mechanism is not perfect enough,
making inefficient investment problem extremely serious. On the one hand, enterprises are short of
the grasp of the market or invest the enterprise resources to the unprofitable project for some special
purpose; On the other hand, some enterprises will face the financing dilemma and miss the
investment opportunity. Therefore, the suppression of inefficient investment has become an urgent
problem to be solved. The existing research found that free cash flow, the nature of the external
large shareholder and debt leverage have a certain impact on the investment efficiency of
enterprises [1,2,3]. In addition, high quality of accounting information can alleviate information
asymmetry and principal-agent problem and thus gradually become the research hotspot [4]. In
China, the accounting system gradually tends to unify in the development and improvement of the
market economy. Accounting standards, accounting reports and so connect with the international
standards on gradually, and gradually improve the degree of standardization. Therefore, whether the
accounting information quality is related to the investment efficiency of the enterprise becomes the
research content of this paper.

2. Literature Review and Research Hypothesis

In the previous study of investment efficiency, foreign scholars focused mainly on the influence of
external environment on investment efficiency. In recent years, many scholars began to focus on the
impact of internal environmental factors on investment efficiency, the focus shifted to the impact of

Copyright © (2019) The Authors and AEE 168 DOI: 10.35532/JSSS.V2.031

internal factors such as quality of accounting information. McNichols and Stubben documented that
the degree of overinvestment of enterprises with lower earnings management was correspondingly
lower [5].
Partha Sengupat documented that there was a negative relationship between quality of
accounting information and cost of debts [6]. And lower cost of debts could reduce the possibility of
under-investment for the lack of money. Also, Bushman and smith suggested that higher accounting
information quality could regulate investment decisions made by managers, mitigating agency
problems and optimize investment [7]. In addition, higher accounting information quality can
enhance the ability of investment decision makers to identify investment opportunities when he
making investment decisions, reducing the possibility of missing investment opportunities due to
insufficient recognition ability, and reduce or avoid blind over-investment, so improve the
efficiency of enterprise investment [4]. Therefore, the paper presents the following two hypotheses:
H1: There is a negative correlation between the quality of accounting information and
underinvestment.
H2: There is a negative correlation between the quality of accounting information and
overinvestment of enterprises.
In addition, non-state-controlled companies can change the mode of operation in accordance
with the provisions of law to meet the overall business objectives that pursue the maximization of
profit and shareholder value. But state-controlled listed companies are under heavier influence by
policy, which will reduce the impact of quality of accounting information in guiding the company's
resource allocation process. In view of this situation, this paper points out the third pending test
hypothesis:
H3: Compared with state-controlled listed companies, the quality of the accounting information
of the non-state-controlled listed companies has a deeper impact on the investment efficiency.

3. Research design
3.1 Data and sample selection
All the data in this paper are from CSMAR database, from which we obtain the financial data in the
2011-2016 period of all shares listed companies of Shanghai and Shenzhen Stock Exchange. The
following companies are excluded in the process of selecting samples: (1) Listed abroad and in
domestic at the same time; (2) PT, ST, *ST and companies where main business income or
shareholder rights and interests of the company is not positive; (3) Classified into financial industry,
insurance industry or real estate industry; (4) With incomplete or abnormal financial data during the
study period. Finally, a total of 7,840 annual observations were obtained in the empirical testing
phase. In addition, all variables are indented at 1% and 99% to avoid the effects of extreme values.
3.2 Variable definition
3.2.1 Explained variables: inefficient investments (OverInv/UnderInv)
The Richardson model is used as an alternative variable to investment efficiency. The model refers
to Eq. 1:
Invi.t =β0 +β1 Growthi,t-1 +β2 Levi,t-1 +β3 Cashi,t-1 +β4 Agei,t-1 +β5 Sizei,t-1 +β6 Returnsi,t-1 +β7 Invi,t +β8 Year
+β9 Industry+εi,t . (1)
ε i,t ,the residual of the model, represents the investment efficiency. When ε i,t >0, it means
over-investment, which is represented by OverInv. When ε i,t <0, it means that the investment is
insufficient, which is represented by UnderInv. For ease of understanding and explanation,
Underlnv is taken as an absolute value. Therefore, the greater the value of Overlnv and Underlnv,
the lower the investment efficiency of the listed company. In addition, the listed company’s own
main business growth, debt ratio, cash inventory, company establishment period, company size,

169
return rate for investment, industry's own variables and annual variables are represented separately
by Growth, Lev, Cash, Age, Size, Return, Year, and Industry.
3.2.2 Explanatory variable: Quality of accounting information (AQ)
In this paper, the accounting information quality is measured by the controllable total accrual profit
of the modified Jones model. The difference between the actual total accrual profit and the
uncontrollable accrual profit predicted by the regression model can be regarded as controllable
accrual profit manipulated by the management. When the absolute value of the difference amount is
taken as the profit management proxy variable, the greater the absolute value is, the greater the
company's earning management degree is, and the poorer the company's earnings quality. Its
estimated process is as follows:
First, calculate the total accrual profit:
TAi,t =NIi,t -CFOi,t . (2)
Then make a regression based on Eq. 3 by industry and year:
TAi,t α1 α2 ∆REVi,t α3 PPEi,t
= + + +εi,t . (3)
Asseti,t-1 Asseti,t-1 Asseti,t-1 Asseti,t-1

The coefficient estimated by Eq. 3 is put into the f Eq. 4 to calculate Discretionary Accruals:
TAi,t α
�1 α
��∆REV
2 i,t -∆RECi,t � α
�PPE
3 i,t
DAi,t = - - - . (4)
Asseti,t-1 Asseti,t-1 Asseti,t-1 Asseti,t-1

Among them, NI, CFO, △REV, △REC, PPE, NDA, and DA respectively represent the
operating profit, net cash flow generated from operating activities, total assets at the end of the year,
increments in sales revenue, increments in accounts receivable, original value of fixed assets (net
fixed assets in this paper), non-operating accruals profit and discretionary accruals.
Since the discretionary accruals should be counted as positive or negative, for the sake of
analysis, we multiply the absolute value by -1 as the earnings quality to reflect the quality of the
listed company's accounting information. The larger the value, the higher the quality of the
accounting information (AQ).
3.2.3 Control variables
Based on the research methods of related literatures, this paper chooses the control variables of
enterprise scale (Size), free cash flow (FCF), Total asset turnover (SAR), equity concentration (Herf)
and the nature of property (Soe) to test the model. In addition, this paper also includes indicator
variables for industry membership and temporal effects. The main variable as shown in Table 1.
Table 1. Variable definitions.
Types Variable Variable definitions
OverInv Absolute value of regression residual which is greater than 0 in Richardson model
Explained variables:
UnderInv Absolute value of regression residual which is less than 0 in Richardson model
Explanatory variable AQ The negative value of the absolute value of discretionary accruals
Size Natural logarithm of total assets
FCF Free cash flow / total assets at the end of the year
Control variables
SAR Main business income/average total assets
Herf Top 10 shareholding ratio
Soe Nature of property, divided into state-owned (0) and non-state-owned (1)
Based on the research methods of related literatures, this paper chooses the control variables of
enterprise scale (Size), free cash flow (FCF), Total asset turnover (SAR), equity concentration (Herf)
and the nature of property (Soe) to test the model. In addition, this paper also includes indicator
variables for industry membership and temporal effects. The main variable as shown in Table 1.

170
3.3 Model design
Based on the previous discussion, the following regression models are proposed to test hypothesis 1
and hypothesis 2:
OverInvi,t �UnderInvi,t �=β0 +γ1 AQi,t-1 +γ2 Sizei,t +γ3 FCFi,t +γ4 SARi,t +γ5 Herfi,t +γ6 Soei,t +εi,t . (5)
If hypothesis 1 and hypothesis 2 are confirmed to be valid, the corresponding coefficients of AQ
should be negative.

4. Empirical tests
4.1 Descriptive statistics
As the Table 2 shows, 2,974 of the 7,840 samples belong to over-investment, which accounted for
37.93% of the total number of samples, and 4,866 of the total sample belong to under-investment
and are accounted for 62.07%.
Table 2. Descriptive statistics of main variables.
Variable Obs Mean Std.Dev. Min Max
OverInv 2974 0.0416647 0.0452804 0.0000597 0.2660503
UnderInv 4866 0.0254646 0.0207515 1.24e-06 0.2034439
AQ 7840 -0.0620272 0.1352972 -6.497737 -1.05e-06
Size 7840 22.13237 1.198223 17.6413 27.96173
FCF 7840 -0.0012595 0.1372052 -3.719524 1.764304
SAR 7840 0.6715986 0.5900773 0.000989 12.37286
Herf 7840 0.5553067 0.1482738 0.0132 0.96504
Soe 7840 0.3989796 0.4897198 0 1
The mean value of over-investment is 0.0417, and the average value of investment is 0.0255.
We can infer that there are more companies with under-investment than those which invest
excessively, and the degree of over-investment is greater than the lack of investment. In short, the
phenomenon of inefficient investment in China's listed companies are common.
4.2 Correlation analysis
According to the results of the correlation analysis, AQ is negatively correlated with OverInv, and
the correlation coefficient is 0.0788, which indicates that the improvement of accounting
information quality can alleviate the excessive investment of enterprises. A similar result can be
found when it comes to UnderInv, with a correlation coefficient of 0.0967, which suggests that the
higher the quality of accounting information, the lower the degree of enterprise under-investment,
indicating that the quality of accounting information can alleviate the under-investment of
enterprises.
4.3 Multiple regression analysis
4.3.1 Regression analysis on overinvestment
The first two of the eight regressions in Table 3 presents the results of regression in case of
over-investment based on Eq. 5. The regression results show that there is a negative correlation
between the AQ and OverInv, which is highly significant at 1% level. This indicates that the
improvement of accounting information quality can inhibit the excessive investment of enterprises.
Both two regressions pass F test, and the overall regression equation is very significant. Therefore,
this result validates the hypothesis 1.
4.3.2 Regression analysis on underinvestment
Regression 3 and Regression 4 in Table 3 presents the results of regression in case of
over-investment based on Eq. 5. The regression results show that there is a negative correlation
between the AQ and UnderInv, which is highly significant at 1% level. This indicates that the
improvement of accounting information quality can inhibit the under-investment of enterprises.

171
Both two regressions pass F test. Therefore, this result validates the hypothesis 2.
Analyze Regression 1-4 comprehensively we can found: (1) The results above fully support the
hypothesis 1 and hypothesis 2, that is, high quality of accounting information can inhibit the
enterprise's inefficient investment, (2) the size of the enterprise, the amount of cash flow and the
turnover speed of the enterprise is negatively correlated with the degree of the inefficient
investment, but the concentration of equity is positively correlated with inefficient investment, and
the results are all very significant.
4.3.3 Actual controller, accounting information quality and inefficient investment
It is assumed that the impact of accounting information quality on the enterprise's inefficient
investment is different when it comes to different actual controller in hypothesis 3. Studying 3,128
state-owned listed companies selected from the 7,840 samples, we can find that the phenomenon of
under-investment is more common than over-investment whether in state-owned enterprises or
non-state-owned enterprises. However, the phenomenon of under-investment in state-owned
enterprises is more serious than that in non-state-owned enterprises.
In order to study the influence of the quality of accounting information on the inefficient
investment of state-owned enterprises, the paper perform regression on Eq. 5 and the results of the
regressions is shown in Table 3.
In the aspect of over-investment (Reg5 and Reg6), the regression result of state-owned
enterprise (Reg5) is superior to the non-state-owned enterprise (Reg6) both in terms of the
magnitude of the coefficient and the significance level. In the aspect of under-investment (Reg7 and
Reg8), the coefficient of accounting information quality in Reg 7 is smaller than that in Reg 8. The
four regressions show that compared with non-state-owned enterprises, the quality of accounting
information of listed companies in state-owned enterprises is more capable of relieving the
inefficient investment, which is the opposite of hypothesis 3.
Table 3. Results of regressions on Eq. 5.
variable Reg 1 Reg 2 Reg 3 Reg 4 Reg 5 Reg 6 Reg 7 Reg 8
0.039*** 0.081*** 0.025*** 0.101*** 0.081*** 0.043* 0.010*** 0.103***
Constant
(0.000) (0.000) (0.000) (0.000) (0.001) (0.093) (0.000) (0.000)
-0.044*** -0.035*** -0.016*** -0.013*** -0.077*** -0.020 -0.023*** -0.013***
AQ_l
(0.000) (0.001) (0.000) (0.000) (0.000) (0.109) (0.000) (0.000)
-0.002*** -0.003*** -0.003*** -0.001 -0.003*** -0.004***
Size
(0.004) (0.000) (0.009) (0.329) (0.000) (0.000)
-0.015** -0.008*** -0.049*** 0.0005 0.0003 -0.013***
FCF
(0.023) (0.004) (0.000) (0.955) (0.944) (0.001)
-0.010*** -0.003*** -0.007*** -0.013*** -0.003*** -0.003***
SAR
(0.000) (0.000) (0.004) (0.000) (0.000) (0.000)
0.040*** 0.009*** 0.030*** 0.048*** 0.010*** 0.007***
Herf
(0.000) (0.000) (0.001) (0.000) (0.001) (0.007)
-0.009*** -0.002***
Soe - - - -
(0.000) (0.000)
Year indicators No Yes No Yes YES YES YES YES
Industry
No Yes No Yes YES YES YES YES
indicators
18.55*** 16.42*** 45.95*** 36.53*** 7.82*** 9.79*** 16.05*** 17.14***
F-test
(0.000) (0.000) (0.000) (0.000) (0.0000) (0.0000) (0.0000) (0.0000)
Adjusted R2 0.006 0.0493 0.0092 0.0681 0.0533 0.0403 0.0624 0.0489
Note: *, **, ***indicates significance at the 10%, 5%,1% level respectively.
4.4 Robustness test
In order to test the robustness of the results of empirical analysis, the paper uses the method of
Xuan Shi (2013) to measure the efficiency of investment once again, that is the value which reduces
the ratio between the investment level and the average of the investment level in the past 3 years by
1, the specific formula is:

172
 CEi,t
CIi,t = CE +CE +CEi,t -1. (6)
i,t-2 i,t-1
3

CE i,t is calculated as dividing the amount of investment of the company I in year of t by its
current year's main business income. The amount of investment is expressed by the amount of cash
paid for the purchase of fixed assets, intangible assets and other long-term assets in the cash flow
statement. CI i,t on behalf of inefficient investment of Company I in the year of t. If CI>0, it means
that the enterprise invests excessively; CI<0 suggests that the company is not investing enough. The
greater the value of CI, the higher the level of the enterprise's inefficient investment. Replacing the
measurement of efficiency, the similar results are found after performing regression 1-4 again.

5 Conclusion
This paper firstly analyzed the mechanism of the quality of accounting information on investment
efficiency in theory, and then verifies the influence of accounting information quality on investment
efficiency with the data of China’s listed companies. The main conclusions are as follows:
There is a widespread phenomenon of inefficient investment in China, where under-investment
is more common than over-investment; State-controlled listed companies are in deeper degree of
under-investment than non-state-controlled listed companies.
High quality of accounting information can significantly alleviate the inefficient investment,
whether it is under-investment or over-investment, supporting hypothesis one and hypothesis two.
The mitigating effect of the accounting information quality on under-investment in
state-controlled listed companies is significantly stronger than that of the state-controlled listed
companies. The conclusion is not in accord with the hypothesis 3, the reason may be that the
state-owned enterprises occupy the advantage in the financing platform and scarce resources, which
form the government's support to the state-owned enterprises and enhance the degree of the relief of
the high quality of accounting information to the inefficient investment.

References
[1] C. G. Liu, Corporate governance mechanism, free cash flow and over-investment behavior of
listed companies, Economic Science, vol.4, pp. 50-58, 2006.
[2] L. P. Xu, Z. Xin, and G. M. Chen, Equity concentration and equity checks and balances and
their impact on the company's operating performance, Economic Research, vol.1, pp. 90-100,
2006.
[3] Q. Q. Xin and B. Lin, Debt leverage and corporate investment: A dual budget soft constraint
perspective, Financial Research, vol.7, pp. 73-83, 2006.
[4] B. Y. Luo, The Mechanism of the effect of accounting information quality on enterprise
investment efficiency, Business research, vol.6, pp. 64-75, 2014.
[5] M. F. McMichols and S. R. Stubben, Earnings management affect firm's investment decisions,
The Accounting Review, vol.83, pp. 1571-1603, 2008.
[6] P. Sengupta, Corporate disclosure quality and the cost of debt, The Accounting Review,
vol.73(4), pp. 459-474,1998.
[7] P. M. Bushman and A. J. Smith, Financial accounting information and corporate governance,
Journal of accounting and Economics, vol.32(1), pp.237-333, 2001.

173