Huang yi
The Business Administration At The China-ASEAN
International College, Pundit University, Thailand
The College Of Economics And Management, Shandong
Yingcai University, China
E-mail: hy2924669@163.com
Yang Xiugang
the business administration at the China-ASEAN
International College, Dhurakij Pundit University, Thailand
E-mail: lyangxg@126.com
Submission: 23/06/2018
Revision: 28/06/2018
Accept: 04/07/2018
ABSTRACT
With
the
bounded rationality hypothesis, the psychological deviation of
managers often leads to non-efficient investment decision-making practices. The
study examines the impact of manager’s overconfidence, risk-preference and herd
behavior on non-efficient investment using the Chinese A-shares listed company
data as the research object, and finds that: (1) managers’ overconfidence and
herd behavior would lead to more non-efficient investment in Chinese listed
companies; and (2) managers’ risk preference restrains the increase of
non-efficient investment to some extent. Meanwhile, the influence of the
manager’s psychological deviation on the actual investment decision is a complicated
process and can have a comprehensive effect resulted from the interaction of
the above psychological biases, we also find that (3) managers' overconfidence
is an interactive term in the effects of herd behavior and risk preference on
non-efficient investment. That is, managers’ overconfidence can significantly
reduce the positive effect of herd behavior on non-efficient investment; and
can also significantly relieve the inhibition effect of risk
preference on non-efficient investment. These findings
reveal that it is important to understand managers’ irrational behaviors in
enterprise investment decision-makings.
Keywords: non-efficient
investment; irrational behaviors; overconfidence; risk-preference; herd
behavior
1. INTRODUCTION
Behavioral corporate theory believes that the managers’ irrational
behaviors often lead to non-efficient investment, namely, managers don’t choose
investment projects in accordance with the goal of maximizing shareholder
value, but the goal of maximizing their personal benefits. They may invest
projects with a negative NPV (Net Present Value) or withdraw from projects with
a positive NPV, called over-investment and under-investment respectively (JENSEN; MECKLING, 1976).
The irrational behaviors could result in low efficiency of capital
allocation and large waste of social resources.
With a special socio-economic environment and cultural background, managers’
irrational behaviors, such as overconfidence, risk-preference and blindly herd
behavior, are very common in Chinese listed companies, and have already been
affecting companies’ operation performance and investment efficiency (WANG, 2017; YONGZHUANG; LIJUAN, 2014).
The aim of this study is to
explore the comprehensive effects of various managers’ irrational behaviors on
the non-efficient investment of Chinese listed companies. The results provide further understanding and
empirical evidence relevant to the irrational behaviors of managers and
investment efficiency. Previous studies on this topic mainly relied on the data
from the US or other developed countries(KENNEDY et al., 2013; FACCIO et al., 2016; KREMER et
al., 2013),
and few studies reported empirical analysis with data from an emerging market.
Hence, this study fills the gap in the literature by investigating the impact
of managers’ irrational behaviors on non-efficient investment in China, which
has been regarded as the biggest developing country and the biggest emerging
market in the world.
Although the influence of
single irrational behavior on investment efficiency has been frequently
discussed in the literature (LIN; HUANG, 2012; KANG et al., 2018; HSIEH et al., 2014), few studies focused on the effects of various irrational behaviors on
non-efficient investment.
However, in practice,
managers may show several different irrational behaviors while decision making.
For instance, we think that overconfident managers usually have a
high-level risk appetite because of self-attribution, and managers
who have a herd mentality are often not overconfident.
Therefore, we inspect the comprehensive impacts of manager’s
overconfidence, risk-preference and herd behavior on non-efficient investment,
and further empirically analyze the interactive effects of managers’
overconfidence on their herd behavior and risk preference to non-efficient
investment. The results not only present the direction for managers’
psychological quality training, but are also critical to the management of
enterprise investment efficiency.
We contribute to the
literature in several ways. First, we show that, when managers' multiple
irrational psychology affects non-efficient investment, managers'
overconfidence and herd behavior will aggravate non-efficient investment, and
managers' risk preference can alleviate the non-efficient investment. Although
prior analytical work suggests this possibility (e.g., MALMENDIER; TATE, 2015;
HOLMES et al., 2013a; KAUFMANN et al., 2013), little empirical work exists on this topic. Second, we present
evidence that managers' herding behavior is most damaging to non-efficient
investment in China's listed companies.
Lastly, we empirically test
the moderating role of managers' overconfidence, that is, managers'
overconfidence can regulate the destructive effect of herd behavior on investment
efficiency, and also can regulate the inhibition of risk preference on non-efficiency.
This moderating effect has not been discussed in prior literature to our knowledge,
especially in emerging countries such as China.
This paper proceeds as
follows. The next section reviews prior literature and motivates our
hypotheses, section three describes research design, section four
presents the main results, and section five and six discuss the results and
concludes this paper.
2. LITERATURE REVIEW AND RESEARCH HYPOTHESIS
The emerging behavioral finance
theory states that with the bounded rationality hypothesis, irrational
mentality and behavior of managers’ cognitive bias are the main factors of
enterprises non-efficient investment (CALDAROLA,
2014; AHMED; DUELLMAN, 2013; RICHARDSON, 2006).
2.1.
Managers’
Overconfidence and Non-efficient Investment
Managers’ overconfidence is the psychological
characteristic that managers overestimate their decision making ability and
underestimate the probability of failure (GERVAIS et al., 2003).
In other words, it is a perception
bias of managers that overestimates the company’s future performance and
underestimates future risks. Existing literature has demonstrated
that managers’ overconfidence will change the income and cost of enterprise
cash flow, which leads to a distortion of investment behavior (HEATON, 2002; GRINBLATT;
KELOHARJU, 2009).
Specifically, managers often overestimate the investment profits and undervalue
the risks and costs, resulting in over-investment (MALMENDIER; TATE, 2015).
Due to self-attribution cognitive
bias, managers may be over-confident in their own judgment and abilities on
investment. Thus, they will overvalue investment returns by setting a
relatively lower discount rate and underestimating investment risks, leading to
that projects with an NPV less than zero will be considered as proper target.
A more overconfident manager would
be more likely to expand the investment scale. In addition, as overconfident
managers tend to overestimate the likelihood of a good performance, they will
overestimate company market value and believe that the external market
participants would underestimate the intrinsic value of their company.
In addition, they will be reluctant
to have external financing because of the higher external financing cost.
Consequently, they may give up some investment projects with an NPV more than
zero, resulting in under-investment.
2.2.
Manager's
Herd Behavior and non-efficiency Investment
Managers’
herd behavior means that a manager makes his/her investment decisions based on
the information of similar managers from other companies, rather than on the
basis of their own information about the market in the face of uncertainty (HOLMES
et al., 2013b).
It
is a typical kind of blind following behavior in investment decision-makings.
Due to information uncertainty, managers tend to adopt the corresponding action
to other group members in order to avoid potential reputation loss caused by
the failure of individual decisions and to reduce the probability of missing
opportunities (QI-AN;
HONGFEI, 2015).
Although
herd behavior can reduce information costs and decision risks, and is
beneficial to maintain a professional reputation, irrational herd behavior will
lead to non-efficient investment, such as under-investment caused by
conservatism or over-investment caused by blind following (DEVENOW;
WELCH, 1996).
Under
an uncertain environment, managers tend to imitate and pursue other managers’
investment decisions, in order to maintain their reputation, salary and other
personal benefits. A manager with a more intensive herd mentality would be more
likely to imitate other managers in the same industry.
However, the investment direction
and scale learned from other companies are not necessarily suitable for the
actual situation of their own companies. Therefore, blindly following the
investment decisions of other companies will usually cause a lower investment
efficiency.
2.3.
Managers’
Risk-preference and Non-efficiency Investment
Managers’
risk-preference refers to the attitude of managers toward risks in the face of
many uncertain factors. Different managers often show different risk-preference
modes, including risk-loving, risk-averse and risk-neutral, in investment
decision-making processes. Different risk-preference modes have different
effects on investment behavior, and then produce different efficiencies.
Kremer
et al. (2013)
and Pattillo
and Söderbom (2000)
both find that companies with risk-loving managers make more investments and
grow faster than those with risk-averse managers. Tanaka
and Sawada (2015)
find that in Lao clothing industry, risk-averse managers have a tendency to
invest by using their internal assets rather than borrowing from banks or
informal sources. Moreover, the total investment amount of companies with
risk-averse managers is often lower than that of companies with risk-tolerant
managers. However, risk-averse managers have a tendency to invest more in
security equipment and facilities, such as fire exits and alarms.
Bromiley
et al. (2015)
state that managers who have a higher
risk-aversion level are usually more careful in investment and are more likely
to invest low-risk and low-income projects, and when managers’ risk-aversion
level rises, lower risk projects are more attractive to them than those with a
higher risk. Previous studies about Chinese stock market have demonstrated that
risk-loving managers tend to expand investment, and are prone to excessive
investment, but risk-averse managers usually have prudent and conservative
investment strategies and are easy to operate as underinvestment. As a result,
both risk-loving and risk-averse can lead to a lower investment efficiency (KAUFMANN
et al., 2013; KONGCHEN; CHENYAN, 2016; LIQING; FEIYUAN, 2015).
In
general, managers’ irrational behaviors, including overconfidence, herd
behavior and deviant risk preference, are positively correlated with the
enterprise’s non-efficient investment. These psychological deviations of
managers will lead to excessive or inadequate investment in enterprises.
The
psychological process of human beings is complex and changeable, and the
cognitive deviation is varied. When facing uncertainty in investment
decision-making process, managers will be affected by a variety of
psychological biases simultaneously. Thus, impacts of managers’ irrational
behaviors on enterprise’s non-efficient investment cannot be generally
concluded, but should be analyzed with specific conditions.
Because
over-confidence is regarded as a common psychological phenomenon in investment
decision-making processes, this study emphasizes the importance of the
interaction of managers’ overconfidence with the other two kinds of
psychological biases (herd behavior and deviant risk preference) on
non-efficient investment.
Overconfidence
has a negative impact on managers’ investment efficiency because they may
overvalue their true abilities and take over complex and difficult projects.
Overconfident managers want to prove their excellence by success, and are more
likely to invest in high-risk projects, leading to an increasing risk level of
the whole enterprise and causing some projects with a negative NPV can also be
implemented.
Luckily, if the investment
gets a high return, overconfident managers will further confirm that their
abilities are the key factor in the success, and increase their confidence and
risk appetite in follow-up decisions. Therefore, managers’ overconfidence and
risk preference often interact with each other and finally influence the
enterprise investment efficiency.
However,
managers with a herd mentality will imitate other managers’ investment
behaviors. The causes of this are also various, including the lack of
confidence, the underestimating of the success likelihood of investment
projects, and the fear of failure or loss of reputation or pay.
If managers with a herd mentality can be more
confident about their abilities, make investment decisions independently,
revalue the influence of a successful project on their own reputation and pay
growth, and correctly and objectively estimate costs and risks of investment
projects, some over-investment caused by blindly following, and some
under-investment caused by conservative strategies would be avoided (PIKULINA
et al., 2017).
In other words, managers’ overconfidence can also play a positive role and can
reduce non-efficient investment caused by managers’ herd behavior.
To
sum up, we aim to test the effects of different irrational behaviors of
managers on investment efficiency. Following assumptions are proposed:
·
Hypothesis 1: Managers’
overconfidence, risk-preference and herd behavior are positively correlated
with non-efficient investment of enterprises.
·
Hypothesis 2: Managers’
overconfidence negatively moderates the relationship between herd behavior and
non-efficient investment.
·
Hypothesis 3: Managers’
overconfidence positively moderates the relationship between risk preference
and non-efficient investment.
According
to the hypothesis, we construct the following econometric models:
(1)
(2)
(3)
(4)
(5)
Where
ABSNE is enterprise non-efficient investment, MOC is managers’ overconfidence, MRP
is managers’ risk preference, MH*MOC is the interaction between MH and MOC,
MRO*MOC is the interaction between MRP and MOC, Controls are the control
variables that affect the enterprise's non-efficient investment, and is the annual dummy
variable.
3. RESEARCH DESIGN
3.1.
Data
Sources and Sample Selection
This study used
the data of China’s A-shares listed companies in Shanghai and Shenzhen Stock
Exchanges as the research sample. The data were collected from Wind Database.
The time frame is from 2009 to 2015. This time frame was selected, because
until 2008, when the “split share” regulation released, the liquidity and
fluidity of the Chinese stock market became normalized.
In addition, as
the VAT (value added tax) reform of the tax system was fully implemented in
2016, data after 2016 are no longer comparable to the previous. Thus, the
period from 2009 to 2015 is a proper choice. To ensure the representativeness
of the research sample, we excluded financial companies because their
investment behaviors are differ from those of non-financial companies, and
ignored firm-year observations with incomplete data, all samples of ST (Special
Treatment), PT (Particular Transfer) and the samples with negative net assets
and performance deterioration, and the samples that had been on the market for
less than eight years (SCHMELING,
2012),
Finally, we got a sample of 8809 observations across 1363 individual companies.
3.2.
Variable
Measurements
3.2.1. Non-efficient
Investment
Consistent
with prior research of Richardson (2006),
we measured non-efficient investment as deviations from expected investment
using a model that predicts investment as a function of growth opportunities,
leverage, the level of cash, firm age, firm size, return on assets and prior
firm investment level (JUNG et al., 2014).
Where
is
company’s new investment expenditure; α
is the constant; β is the regression
coefficient for each variable; ε is
the residual; i is the company index;
t is the time index; is the
growth of investment opportunities measured by the main business income growth
rate at year t-1; is the asset liability rate; is monetary capital stock; is years from listed; is natural log of total assets; is stock returns; and is new investment at year t-1. Year and industry are represented by dummy variables.
Following
Richardson (2006),
we employed the fixed effects regression models to estimate the above models.
The residuals from the regression model are the deviations from the expected
investment level, and can be used as the proxy variables of non-efficient
investment. Positive residuals measure over-investment and negative residuals
measure under-investment. Our proxy variables for non-efficient investment are
the absolute value of residuals(ABSNE),
and higher value means a higher degree of non-efficient investment.
Based
on the data above, the distribution features of over-investment and
under-investment are illustrated in the Table 1. It can be seen that both
over-investment and under-investment existed among Chinese listed companies.
Specifically, 3075 of the 8809 samples are over-investment, whereas 5733 are
under-investment. Compared with the results of Gongfu (2009) and Huangyi (2016) that used similar data from 2001 to
2008 and 2010 to 2014, respectively, the ratio of underinvestment is
increasing, showing that non-efficient investment of Chinese-listed companies
has not been improved since 2001.
Table
1: Degree of Non-efficient Investment
Index |
Sample |
Max |
Min |
Mean |
Std
Dev |
Rate(%) |
Over-investment |
3431 |
12.7511 |
0.0000 |
.5395 |
1.0646 |
38.95% |
Under-investment |
5378 |
0.0000 |
-16.4657 |
-.3999 |
1.1402 |
61.05% |
Total |
8809 |
- |
- |
- |
- |
- |
3.2.2. Managers’
Overconfidence
In this study, we employed the
financial earnings forecast (FERRIS
et al., 2013) to measure MOC. This method is currently widely used in financial studies.
Overconfidence occurs when the forecast net profit growth rate is greater than
the actual growth rate. In contrast, under-confidence occurs when the forecast
net profit growth rate is lower than the actual growth rate.
3.2.3.
Managers’ Herd Behavior
Following Bo
et al. (2016),
we employed the deviation of the firm investment level from the industry
average investment level to measure managers’ herd behavior. If the deviation
is higher, the difference between firm’s investment level and the industry
average investment level is greater and the herd investment behavior of
managers is less. The MH is computed
as follows:
Where is the managers’ herd behavior; i is firm index; t is the time index; is the new investment amount in firm’s fixed
assets, construction projects and intangible assets; is the average value of new investment in the
industry including the firm; and is the
value of total assets of the enterprise. Because this index is a reverse
index, in order to make the research results easier to understand, this study
used the reciprocal of this index.
3.2.4.
Managers’ Risk Preference
We employed the most commonly used
index, namely, “the proportion of risk assets to total
assets” to measure managers’ risk preference (CHEN
et al., 2011). The principle of the index is
that risk preference is linked to the composition of personal income. The
income includes salary which is relatively safe and the contingent reward which
is relatively risky (i.e. the company stock price volatility returns). When the
proportion of contingent reward in total income is higher, managers prefer to
accept more risk. The MRP is
therefore computed as follows:
Where is the risk assets ratio; i is the firm index; t is
the time index; is the contingent reward; is the firm’s shares value held by managers in
year t, is the firm’s shares value held by managers in
year t-1; and is the salary income in year t. When the index rises, the degree of
managers’ risk preference increases, and vice versa.
3.2.5.
Control Variables
Because the
enterprise investment behavior, enterprise future profitability, performance
level, enterprise risk and financially troubled possibility will be affected by
other factors, according to related theories and literature (Liuyan, 2016), we set up some
control variables including: company size (Size), financial leverage (Lev), growth opportunity (Growth),
free cash flow (Cf), total assets
profit rate (Roa), Tobin Q value (Q), and industry category. The specific measurement method of control variables is shown
in Table 2.
Table 2: the
Control Variables Definition
Variable symbol |
Variable name |
Variable definitions |
Size |
Enterprise size |
Log (the final total assets) |
Lev |
ratio of liabilities to assets |
Liability/asset |
Growth |
Increase rate of main business
revenue |
(Current turnover-previous
turnover)/turnover *100% |
Cf |
free cash flow |
Net cash flow in operating
activities /final total assets |
Roa |
returns on total assets |
Net income/ final average total
assets |
Q |
Tobin Q |
(year-end liabilities +
Circulating stock market value + non-tradable Stock quantity * Net assets per
share) / (initial total assets + total assets) ÷2. |
4. RESULTS
4.1.
Descriptive
Statistics
From
the descriptive statistic results (Table 3), the maximum of ABSNE is 10.3987, and the minimum is
-2.7791, showing that over-investment in Chinese listed companies is far more
common than underinvestment. The average absolute value of ABSNE is 0.1533. Combined with the data in Table 1, it can be seen
that although there are more samples of under-investment, the degree is not
very large.
Table 3: Descriptive
Statistics of Major Variables
Variable |
Mean |
Std Dev |
Min |
Max |
NE |
6.58e-11 |
0.4131 |
-2.7791 |
10.3987 |
ABSNE |
0.1533 |
0.3836 |
0.0000 |
10.3987 |
MOC |
70.5014 |
2770.41 |
0.0001 |
101375.7 |
MH |
-0.3666 |
3.3012 |
-162.7235 |
0.7160 |
MRP |
0.3039 |
2.0575 |
-19.7865 |
102.9913 |
Size |
22.1086 |
1.3873 |
14.9416 |
28.5087 |
Lev |
50.7675 |
20.2512 |
0.7080 |
99.8124 |
Cf |
0.0451 |
0.0947 |
-1.0324 |
0.9319 |
Growth |
16.8097 |
113.2210 |
-100.00 |
5835.6730 |
Roa |
3.6376 |
6.5799 |
-99.8602 |
92.8513 |
Q |
2.5842 |
6.8708 |
0.3374 |
495.7741 |
The average MOC
is 70.5014, indicating that managers’ profit forecast growth rate exceeds the
company’s real profit growth rate of 70.5%. In general, Chinese listed company
managers tend to be overconfident. The absolute values of both maximum and
minimum are relatively large, showing that the characteristics of
overconfidence and under-confidence are obvious.
The
average of MH is -0.3666, indicating that the average deviation of listed
companies’ investment from their industry average is small and the herd
behavior is serious. The absolute value of the minimum is much higher than that
of the maximum, indicating that the investment scales of listed companies are
generally lower than that of the same industry.
The
average of MRP is 0.3039, indicating that the average ratio of contingent
income of Chinese listed company managers is 30.39%, and the degree of their
risk preference is not high.
The
average value of the natural logarithm of the total assets (Size) is 22.1086,
the difference between the size of the company is hard to see from the natural
logarithm, but because the original value is based on the index of e, so the
scale difference between companies is quite large.
The
average of leverage (Lev) maintains at 50.7675% level which is a high ratio.
From a higher leverage ratio, it can be seen that most of enterprises are
confident about their future development. On the other hand, companies should
also be careful about financial troubles.
The
net cash flow (Cf) from operating activities is accounts for 4.51% of the total
assets, despite the fact that the value is small, but it reflects the net cash
flow generated by the company's operating activities. However, the mean of the
net cash flow is positive, indicating that the inflow company's business
activities are greater than outflow. The cash situation of China's listed
companies in general is relatively stable, and has the
"self-hematopoiesis" function, which is the investment capital to
expand the invest scale.
The
overall average of the main business growth rate (Growth, the company growth
Opportunity) is 16.81%, indicating that most of the company's products are in
the growth period, will continue to maintain a good growth momentum, the growth
of enterprises more opportunities. However, because of its high standard
deviation, it shows high volatility.
The
mean value of the Tobin Q value is 2.5842, indicating that the market price of
the company is more than twice times its basic book value of the company. At the
same time, the Q-value gap between enterprises is also very large, the smallest
only 0.3374, and the maximum value of 495.7741.
In
addition, to avoid the influence of outliers, we standardized each continuous
variable.
4.2.
Model
Regression Results
Before regression modeling,
we made a multicollinearity diagnosis for each variable, and the results show
that all the variance inflation factors (vif) are less than 2. Thus, there is
no collinearity between the variables. Then, we conducted the Hausman test, and
the results demonstrated that the panel data should be analyzed with fixed
effect models.
Table 4: the Results of Fix Effect Regression
models |
(1) |
(2) |
(3) |
(4) |
(5) |
variables |
ABSNE |
ABSNE |
ABSNE |
ABSNE |
ABSNE |
MOC |
0.0586*** |
|
|
0.0623*** |
0.0429* |
|
(0.0204) |
|
|
(0.0202) |
(0.0249) |
MH |
|
0.8240*** |
|
0.8265*** |
0.8771*** |
|
|
(0.0712) |
|
(0.0712) |
(0.0738) |
MRP |
|
|
-0.0086* |
-0.0083* |
-0.0122*** |
|
|
|
(0.0044) |
(0.0044) |
(0.0047) |
MH*MOC |
|
|
|
|
-0.5540** |
|
|
|
|
|
(0.2378) |
MRP*MOC |
|
|
|
|
0.0475** |
|
|
|
|
|
(0.0224) |
Size |
0.0529*** |
0.1057*** |
0.0568*** |
0.1068*** |
0.1066*** |
|
(0.0081) |
(0.0092) |
(0.0082) |
(0.0093) |
(0.0093) |
Lev |
-0.0557*** |
-0.0551*** |
-0.0555*** |
-0.0563*** |
-0.0560*** |
|
(0.0045) |
(0.0044) |
(0.0045) |
(0.0044) |
(0.0044) |
Growth |
0.0156* |
0.0139 |
0.0167* |
0.0141 |
0.0147 |
|
(0.0091) |
(0.0091) |
(0.0091) |
(0.0091) |
(0.0091) |
Cf |
-0.0145*** |
-0.0133*** |
-0.0139*** |
-0.0137*** |
-0.0136*** |
|
(0.0030) |
(0.0030) |
(0.0030) |
(0.0030) |
(0.0030) |
Roa |
0.0028 |
0.0033 |
0.0028 |
0.0040 |
0.0038 |
|
(0.0043) |
(0.0042) |
(0.0043) |
(0.0042) |
(0.0042) |
Q |
0.0990*** |
0.0676*** |
0.1018*** |
0.0708*** |
0.0701*** |
|
(0.0193) |
(0.0193) |
(0.0194) |
(0.0194) |
(0.0194) |
Year(dummy) |
included |
included |
included |
included |
included |
Constant |
0.2764*** |
0.3273*** |
0.2582*** |
0.3016*** |
0.3045*** |
|
(0.0052) |
(0.0064) |
(0.0051) |
(0.0066) |
(0.0067) |
Observations |
8,809 |
8,809 |
8,809 |
8,809 |
8,809 |
Number of zqdm |
1,363 |
1,363 |
1,363 |
1,363 |
1,363 |
Adj R-squ |
0.3935 |
0.4036 |
0.3932 |
0.4045 |
0.4052 |
Note:Standard errors in
parentheses; *** p<0.01, ** p<0.05, * p<0.1
The
results of Table 4 show that the adjustment R-square of each model is about
40%, showing that these models fit the data well. All the control variables
have similar effects as reported in the literature except that the Roa is not significant. MOC and MH are positively correlated with ABSNE in Model 1 and Model 2, both statistically significant. The
coefficient of MH is 0.8240.
Although
the value is not very high, considering that the mean of ABSNE is 0.1533, the coefficient of MH indicates that one unit increase of MH leads to an average increase of about 5.38 times of ABSNE (0.8240 ÷ 0.1533 ≈ 5.38). In Model
3, the MRP is negatively correlated
with the ABSNE and is statistically
significant, which is opposite to the hypothesis.
The
reason for this result may be that the risk preference level of Chinese
managers is not high during the sample period. This shows that appropriate risk
preference can release enterprise’s non-efficient investment to a certain
extent. However, from the result of Model 3, the degree of this relief is not
high. One unit increase of MRP will
only reduce 5.61% of the ABSNE (0.0086÷0.1533≈5.61%).
Model
4 integrates the effects of the three kinds of irrational behaviors on ABSNE. All the explanatory variables in
Model 4 are statistically significant (P<0.05) and the directions are also
consistent with the previous three models. This result shows that in China’s
capital market, MOC, MH and MRP can simultaneously influence ABSNE.
Model
5 verifies the interactive effects of the three irrational behaviors on ABSNE. The interaction coefficient of MH*MOC
is -0.5540 and statistically significant at 5% level, indicating that MOC can negatively moderate the effect
of MH on ABSNE. The result validates our hypothesis 2 that MOC is a significant negative moderating
term between MH and ABSNE.
The
result of Model 5 also shows that the interaction coefficient of MRP*MOC
is 0.0475 and statistically significant at 5% level, indicating that MOC can positively moderate the effect
of MRP on ABSNE. The result also validates our hypothesis 3 that MOC is a significant positive moderating
term between MRP and ABSNE.
5. DISCUSSION
A
large and growing body of evidence suggests that
a substantial share of managers exhibit symptoms of overconfidence in their
decisions (ANTONIOU et al., 2013, DUELLMAN et al., 2015).
In this study, the main measure of managers’ overconfidence is whether the
forecast net profit growth rate is greater than the actual growth rate (FERRIS
et al., 2013),
the risk preference of managers is measured by the proportion of risky assets
to total assets (BO
et al., 2016),
and the managers’ herd behavior is measured by the deviation of the level of
investment and the average investment level of the industry (CHEN
et al., 2011).
This
study has examined whether managers’ overconfidence, risk preference and herd
behavior have positive effects on enterprise's non-efficient investment, and
results have shown that managers' overconfidence and herd behavior have
positive effect on non-efficient investment.
This
is consistent with the study of Malmendier and Tate (2015).
Managers' overconfidence and herd behavior can lead them to make some
irrational decisions of over-investment or under-investment, especially the
managers' herd mentality has the greatest loss to enterprise investment
efficiency.
However,
the results have also shown that the relationship between managers' risk
preference and non-efficient investment is just the opposite to previous views,
which indicate that risk-loving managers are more inclined to increase
over-investment, while risk-averse managers take more prudent and conservative
investment strategies and are prone to under-investment (KREMER
et al., 2013).
However, the results of this
study have shown that risk preference and non-efficient investment are negatively
correlated. That is, manager's risk preference can reduce the non-efficient
investment to some extent, because managers who are risk-loving or risk-averse will
reduce under-investment or over-investment. This finding adds to the
generalisability of previous research on the relationship between managerial
herd behaviors and non-efficient investment.
In
investment decision-making processes, various irrational mentalities will
interact with and influence each other. Rational use of these irrational mentalities
can also effectively inhibit and mitigate the enterprise’s non-efficient
investment. And managers' overconfidence phenomenon is universal.
Therefore,
this study have set managers' overconfidence as the moderator variable, to
examine whether managers’ overconfidence has a moderating effect on the
influences of managers' risk preference and herd behavior on non-efficient
investment, which has not been investigated in previous literature. The results
have shown that managers’ overconfidence negatively moderates the positive
correlation between herd behavior and non-efficient investment, and positively
moderates the negative correlation between risk preference and non-efficient
investment.
That
is to say, managers’ overconfidence can reduce the deterioration of manager's
herd behavior on non-efficient investment, and also promote the negative effect
of risk preference on non-efficient investment. As a result, managers’
overconfidence as a single irrational behavior will deteriorate the efficiency
of enterprise investment, but it can play a positive role in alleviating
non-efficient investment when it is combined with the herd mentality.
Therefore,
we believe that, the managers with a herd mentality, should properly cultivate
their self-confidence. They should believe that their independent investment
could be more profitable than imitation of other people’s investment, which
could help to filter out those projects which do not fit the enterprise’s
strategy and to improve the efficiency of investment.
In
addition, the managers with a higher risk preference level, must control their
overconfidence, correctly assess their abilities, reasonably estimate the risks
and costs of the project, control their over-investment desires and impulses,
and finally make right investment decisions to reduce potential
over-investment. Meanwhile, the managers with a higher risk aversion level,
should appropriately cultivate their confidence in their abilities to avoid
underinvestment.
In
summary, to improve the investment efficiency, Chinese listed companies must
provide necessary psychological training to their managers. Training managers
learn to analyze specific issues, to take responsibility bravely, to avoid
“follow suit” and “imitation” behavior, and to prevent herd behavior regardless
of the target and the “doing nothing” mentality of going back and forth.
Managers should adapt to the “new normal” based on the actual situation of
their own enterprises, and innovate and invest scientifically and rationally to
reduce the loss of investment efficiency and to maximize the value of
enterprises.
6. CONCLUSIONS
On
the basis of verifying the influence of managers' overconfidence, risk
preference and herd behavior on the enterprise's non-efficient investment, this
study has analyzed the moderating effect of managers’ overconfidence between
risk preference, herd behavior and non-efficient investment.
The
results have shown that managers' overconfidence and herd behavior are
positively correlated with non-efficient investment, and risk preference is
negatively correlated with non-efficient investment, while overconfidence can
negatively moderates the positive correlation between herd behavior and
non-efficient investment, and can positively moderates the negative correlation
between risk preference and non-efficient investment.
This
study also has some limitations. There are many kinds of irrational
psychological manifestations of managers, such as managers’ excessive optimism certain errors, control illusion,
representative, easy to take, anchor qualitative, affective, and so on, which are not mentioned in this study.
According
to Tombaugh
(2005),
‘Optimistic leaders are more likely to see problems as challenges, exert
greater effort for longer periods to reach their goals, and seek out and
appreciate the positive aspects of difficult situations.’ Managers' overoptimism
and overconfidence are similar, but they are different from overconfidence (MOORE;
HEALY, 2008).
Hilary
et al. (2016)
empirically examine a setting in which over-optimism is a relevant but different
bias from overconfidence, emerges dynamically in a rational economic framework,
and generates higher managerial effort. In addition to
overconfidence and over optimism, these other psychological investment
behaviors are few and immature in the current literature.
In
addition, there are many measures for managers' irrational psychological
indicators, but none of them can accurately quantify the "fitness"
and "extreme" of these psychological characteristics.
Although
this study has researched the interaction of managers of irrational psychology
on the efficiency of investment, but not thoroughly analyzed the impact of
these irrational psychological factors. Therefore, future studies should expand
the scope to investigate the influence of more comprehensive irrational
psychological factors, and the interactions of these irrational factors, to
explore the most fundamental and direct source of the impact of an enterprise's
non-efficient investment.
REFERENCE
AHMED, A. S.; DUELLMAN, S. (2013)
Managerial overconfidence and accounting conservatism. Journal of Accounting Research,
v. 51, p. 1-30.
ANTONIOU,
C.; DOUKAS, J. A.; SUBRAHMANYAM, A. (2013) Cognitive dissonance, sentiment, and
momentum. Journal of Financial and
Quantitative Analysis, v. 48, p.
245-275.
BO,
H.; LI, T.; SUN, Y. (2016) Board attributes and herding in corporate
investment: evidence from Chinese-listed firms. European Journal of Finance,
v. 22, p. 432-462.
BROMILEY,
P.; MCSHANE, M.; NAIR, A.; RUSTAMBEKOV, E. (2015) Enterprise risk management:
Review, critique, and research directions. Long
range planning, v. 48, p.
265-276.
CALDAROLA,
R. A. L. (2014) The intersection of emotional intelligence and corporate
financial decision making. The Journal
of Corporate Accounting & Finance,
v. 25, n. 2, p. 67-72.
CHEN,
S.; SUN, Z.; TANG, S.; WU, D. (2011) Government intervention and investment
efficiency: Evidence from China. Journal
of Corporate Finance, v. 17, p.
259-271.
DEVENOW,
A.; WELCH, I. (1996) Rational herding in financial economics. European Economic Review, v. 40, p. 603-615.
DUELLMAN,
S.; HURWITZ, H.; SUN, Y. (2015) Managerial overconfidence and audit fees. Journal of Contemporary Accounting &
Economics, v. 11, p. 148-165.
FACCIO,
M.; MARCHICA, M.-T.; MURA, R. (2016) CEO gender, corporate risk-taking, and the
efficiency of capital allocation. Journal
of Corporate Finance, v. 39, p.
193-209.
FERRIS,
S. P.; JAYARAMAN, N.; SABHERWAL, S. (2013) CEO overconfidence and international
merger and acquisition activity. Journal
of Financial and Quantitative Analysis,
v. 48, p. 137-164.
GERVAIS,
S.; HEATON, J.; ODEAN, T. (2002) Overconfidence, investment policy, and
executive stock options. The Wall Street
Journal, v. 4, p. 1-52.
GONGFU,
Z. (2009) Excessive or insufficient investment of Listed Companies in China --
Based on empirical measurement of inefficient investment in Shanghai and
Shenzhen Industrial Listed Companies. Accounting
Research, v. 5, p. 69-77+97 (in
Chinese).
GRINBLATT,
M.; KELOHARJU, M. (2009) Sensation seeking, overconfidence, and trading
activity. Journal of Finance, v. 64, p. 549-578.
HEATON
(2002) Managerial Optimism and Corporate Finance[J]. Journal of Financial Management,
v. 31, n. 6, p. 33-45.
HILARY,
G.; HSU, C.; SEGAL, B.; WANG, R. (2016) The bright side of managerial
over-optimism. Journal of Accounting and
Economics, v. 62, p. 46-64.
HOLMES,
P.; KALLINTERAKIS, V.; FERREIRA, M. (2013a) Herding in a concentrated market: a
question of intent. European Financial
Management, v. 19, p. 497-520.
HOLMES,
P.; KALLINTERAKIS, V.; FERREIRA, M. P. L. (2013b) Herding in a concentrated
market: A question of intent. European
Financial Management, v. 19, n. 3, p. 497-520.
HSIEH,
T. S.; BEDARD, J. C.; JOHNSTONE, K. M. (2014) CEO overconfidence and earnings
management during shifting regulatory regimes. Journal of Business Finance & Accounting, v. 41, p. 1243-1268.
HUANGYI;
YUMINH, ZH. (2016). Research on investment efficiency from three angles of
input, output and input output -- a case study of Listed Companies in Shandong
Province. Finance and Accounting Monthly, v. 15, p. 79-84 (in Chinese).
JENSEN,
M. C.; MECKLING, W. H. (1976) Theory of the firm: Managerial behavior, agency
costs and ownership structure. Journal
of Financial Economics, v. 3, p.
305-360.
JUNG,
B.; LEE, W. J.; WEBER, D. P. (2014) Financial reporting quality and labor
investment efficiency. Contemporary
Accounting Research, v. 31, p.
1047-1076.
KANG,
J.; KANG, J.-K.; KANG, M. (2018) Curbing Managerial Myopia: The Role of
Managerial Overconfidence in Owner-Managed Firms and Professionally Managed
Firms. Managed Firms and Professionally Managed Firms, v.19,
p. 1-61.
KAUFMANN,
C.; WEBER, M.; HAISLEY, E. (2013) The role of experience sampling and graphical
displays on one's investment risk appetite. Management Science, v.
59, p. 323-340.
KENNEDY,
J. A.; ANDERSON, C.; MOORE, D. A. (2013) When overconfidence is revealed to
others: Testing the status-enhancement theory of overconfidence. Organizational Behavior and Human Decision
Processes, v. 122, p. 266-279.
KONGCHEN;
CHENYAN (2016) The Empirical Study on the Overconfidence, Risk Appetite and
Tendency for Fraud of Management of State-owned Enterprises. Journal of Shanxi Finance and Economics
University, v. 38, p. 77-87 (in
Chinese).
KREMER,
M.; LEE, J.; ROBINSON, J.; ROSTAPSHOVA, O. (2013) Behavioral biases and firm
behavior: Evidence from Kenyan retail shops. The American Economic Review,
v. 103, p. 362-368.
LIN,
S.-W.; HUANG, S.-W. (2012) Effects of overconfidence and dependence on
aggregated probability judgments. Journal
of Modelling in Management, v. 7,
p. 6-22.
LIQING,
Q.; FEIYUAN, L. (2015) An Analysis of the Influence of Manager 's Background
Characteristics on Investment Efficiency. Modern
commerce industry, v. 10, p.
94-95 (in Chinese).
LIUYAN
(2016) Effect of Equity Incentive on Inefficient Investment in Chinese
State-owned Enterprises. International
Journal of Economics and Finance,
v. 8, n. 8, p. 222-228.
MALMENDIER;
TATE, G. (2015) Behavioral CEOs: The role of managerial overconfidence. The
Journal of Economic Perspectives,
v. 29, n. 4, p. 37-60.
MOORE,
D. A.; HEALY, P. J. (2008) The trouble with overconfidence. Psychological review, v. 115, n. 2, p. 502.
PATTILLO,
C.; SöDERBOM, M. (2000) Managerial risk attitudes and firm performance in
Ghanaian manufacturing: An empirical analysis based on experimental data. University of Oxford, Institute of
Economics and Statistics, Centre for the Study of African Economies.
PIKULINA,
E.; RENNEBOOG, L.; TOBLER, P. N. (2017) Overconfidence and investment: An
experimental approach. Journal of
Corporate Finance, v. 18, p.
1-33.
QI-AN,
L.; HONGFEI, Y. (2015) Corporate Governance, Institutional Environment,
Behavioral Corporate Finance and Inefficient Investment. Journal of Service Science and Management, v. 08, p. 452-459.
RICHARDSON,
S. (2006) Over-investment of free cash flow. Review of accounting studies,
v. 11, p. 159-189.
SCHMELING,
M. (2012) Investor sentiment and stock returns: Some international evidence. Journal of empirical finance, v. 16, p. 394-408.
TANAKA,
M.; SAWADA, Y. (2015) Risk preference of managers and firm investments in Lao
PDR. In: SAWADA, Y.; S. OUM (eds.), Disaster
Risks, Social Preferences, and Policy Effects: Field Experiments in Selected
ASEAN and East Asian Countries, ERIA Research Project Report, Jakarta:
ERIA, p. 265-283.
TOMBAUGH,
J. R. (2005) Positive leadership yields performance and profitability: Effective
organizations develop their strengths. Development
and Learning in Organizations: An International Journal, v. 19, p. 15-17.
WANG,
Y.; ZHOU, Y. 2017. The role of managers' overconfidence on the irrational
investment. In: SERVICE SYSTEMS AND
SERVICE MANAGEMENT, 7, Dalian, China, International
Conference on Service
Systems and Service Management, Dalian, IEEE,
2017.
YONGZHUANG,
L.; LIJUAN, L. Top managers' confidence, ultimate controller and firm value: An
empirical study based on panel data of a shares on Shanghai stock exchange. 11,
Beijing, China, International Conference on Service
Systems and Service Management, Beijing, IEEE, 2014.
APPENDIX
Table 5: Variable abbreviation table
Variable symbol |
Variable name |
NE |
Non-efficient
investment |
ABSNE |
The
abslout of non-efficient investment |
MOC |
The
managers’ overconfidence |
MH |
The
managers’ herd behavior |
MH*MOC |
The
interact tem of managers’ herd behavior and managers’ overconfidence |
MRP*MOC |
The
interact tem of managers’ risk preference and managers’ overconfidence |
MRP |
The
managers’ risk preference |
Size |
company size: Log (the final total assets) |
Lev |
financial
leverage: ratio of liabilities to assets |
Cf |
The free cash flow |
Growth |
growth
opportunity :Increase rate of main business revenue |
Roa |
returns on total assets: Main business growth rate |
Q |
Tobin Q |