Proceedings of 5th Asia-Pacific Business Research Conference
17 - 18 February, 2014, Hotel Istana, Kuala Lumpur, Malaysia, ISBN: 978-1-922069-44-3
The Catering Theory about Capital Investment in Chinese Area
Mei-Hung Huang*, So-De Shyu† and Huo-Lien Tsai‡
This study This study considers the increase of mispricing and financial
transparency for investors in the Taiwan, Hong Kong and Mainland three Chinese
areas, a firm’s managers would cater to the sentiment of investors by raising capital
investment. Under financial constraints, an increase in cash flow may raise the
amount of capital investment. Furthermore, the sensitivity of cash flow to capital
investment is higher for the less financially constrained firms in Taiwan. We find that
the increase of the capital investment raises the stock returns for less financially
constrained firms in Taiwan and Hong Kong.
JEL Codes: F15, G15 and H32
1. Introduction
Catering theory refers to any actions or strategies intended to boost stock prices above their
fundamental value. Baker and Wurgler (2004a), who first applied the catering theory, examine
the impact of investors’ demand on dividend payers, who in turn have to make decisions
regarding dividends. Therefore, firm manager may maximize shareholder wealth of short
holdings if managers’ short-sightedness occurs. This has limited relationship between investment
decisions and catering behavior of managers recently. One of the most important financial issues
is a firm’s capital expenditure decision. Capital investments play an essential role in firms’
business operations and often include a great deal of a firm’s assets. Another, Stein (1996), and
Baker, Stein and Wurgler (2003) proposed an equity-financing channel to examine whether stock
market mispricing affects investment decisions, and found that an equity-dependent firm will
issue equity and invest more if its stock price is above its fundamental value, but it will pass up
the investment if the stock price is below its fundamental value.
We extend this literature by an enhanced method that accounts for firm investment decisions
when faced with financial constraints. By focusing on the role played by investors mispricing
behavior, we provide a distinct perspective on the investment decision-making process. We
argue that managers’ short-sightedness may lead to inefficient investment.
We test this method using public listed firms over the period 2001 to 2011 and the empirical work
provides strong support for our model in particular. We examine the level of capital expenditure
under the impact of financial constraints. Titman, Wei and Xie (2004) used the relationship
between capital investment and equity returns to distinguish between the over-investment and
under-investment hypotheses, hence, we tests the status of mispricing and applies it to catering
theory in order to investigate the manager's investment decision-making. Furthermore, this study
takes the financial constraints into consideration to examine whether a firm’s managers would
cater to the sentiment of investors. Instead of using OLS method, we do quantile regression to
test the impact of investment in Taiwan, Hong Kong and China three Chinese areas. The plan of
this paper is as follows. Section 2 makes literature review. Section 3 takes an explanation for
*
Mei-Hung Huang, Department of Finance, Sun Yat-Sen University, Kaohsiung, Taiwan. Email: tiffanyhuang428@gmail.com
So-De Shyu, Department of Banking and Finance, Takming University of Science and Technology, Taipei, Taiwan. Email:
dshyu@takming.edu.tw
‡
Huo-Lien Tsai, Department of Finance, Sun Yat-Sen University, Kaohsiung, Taiwan. Email: huolien.tsai@gmail.com
†
Proceedings of 5th Asia-Pacific Business Research Conference
17 - 18 February, 2014, Hotel Istana, Kuala Lumpur, Malaysia, ISBN: 978-1-922069-44-3
data and methodology. Section 4 shows the empirical results. Section 5 gives concluding
remarks.
2. Literature Review
Tobin (1969) proposed the “Q” theory of investment, and suggested that markets are efficient
and firm has a high stock price reflects higher growth opportunities, that is, a high-priced firm will
invest more. Barberis and Thaler (2003) found that managers tend to make an investment that
has a negative net present value to increase the stock price in the short run when the investors
are in a highly mispriced market. Stein (1996) found that an equity-dependent firm will issue
equity and invest more if its stock price is above its fundamental value but it will pass up the
investment if the stock price is below its fundamental value. Baker, Stein and Wurgler (2003)
conducted a cross-sectional test on the hypothesis and found evidence that stock market
mispricing might influence firms’ investments through an equity issuance channel. Shleifer and
Vishny (2003) argued that overvaluation of a firm’s stock price lead to more investment in the
form of mergers because an overvalued firm may wish to acquire another firm by offering stock.
Gilchrist, Himmelberg and Huberman (2005) found that greater dispersion in analyst forecasts of
earnings is associated with higher aggregate equity issuance and capital expenditures. Baker,
Stein and Wurgler (2003) and Gilchrist, Himmelberg and Huberman (2005) provided model in
which rational managers may issue equity and increase investment in response to overvaluation
of their firm’s stock price. Dong, Hirshleifer and Teoh (2007) found that mispricing will affect the
investments through equity issuance and catering channels, especially for large firms. Chan et al.
(2007) supported that mispricing in the stock market has an impact on firm-level investment.
Nevertheless, Bolbol and Omran (2005) argued that stock price movements are inefficient,
because Arab managers do not take mispricing into consideration in investment decisions. Bakke
and Whited (2010) argued that firms with high levels of mispricing and large firms will consider
mispricing irrelevant for investment, while firms suffering from financial constraints make
investment decisions with market mispricing in mind.
Chen and Ho (1997) did not support free cash flow theory when assessing the value of corporate
investments on product strategies. Li (2004) found that future operating performance is lower for
firms engaging in investment expenditure and that this negative relation is increasing in
contemporaneous free cash flow. Abel and Eberly (2011) supposed that investment is more
sensitive to cash flow and smaller and faster growing firms are found to have larger cash flow
effects. Dechow, Richardson and Sloan (2008) found that cash flow retained within the firm are
usually associated with lower future operation performance and future stock return for internal
investors.
Polk and Sapienza (2009) found that the metric is positively related to investment, after
controlling for investment opportunities and financial slack. Titman, Wei, and Xie (2004) used the
relationship between capital investment and equity returns to distinguish between the overinvestment and under-investment hypotheses..
3. Data and Methodology
The data come from Taiwan Economic Journal (TEJ). Sample firms include: Taiwan listed
companies, Hong Kong listed companies and China listed companies. In addition, the dataset
includes firm-year observations from three Chinese areas (Taiwan, Hong Kong and China) over
the period of 2001–2011, because there is unavailable data for R&D expenses in China and in
Hong Kong before 2001. We select sample by first deleting any firm-year observations with
missing data, and also omit financial service firms since our investment model is inappropriate
Proceedings of 5th Asia-Pacific Business Research Conference
17 - 18 February, 2014, Hotel Istana, Kuala Lumpur, Malaysia, ISBN: 978-1-922069-44-3
for financial firms. Finally, we get 454 firms in Taiwan, 502 firms in Hong Kong, and 1033firms in
China.
After Koenker and Bassett (1978) proposed the method of quantile regression, some studies
used this method to analyze subjects. There is some quantile regression features which fit our
data better than traditional Ordinary Least Squares (OLS) regression. Because the quantile
regression estimator is derived by minimizing a weighted sum of absolute deviations, the
parameter estimates are less sensitive to outliers and long tails in the data distribution. This
makes the quantile regression estimator relatively robust to residuals heteroskedasticity. This
method is briefly illustrated as follows:
Following Polk and Sapienza (2009), the basic model of capital investment is used to regress
firm capital spending on discretionary accruals (the proxy for mispricing), Tobin’s Q and cash
flow, controlling for fixed effects of firms ( f i ) and years (  i ).
Ln(
I i ,t
Ki ,t 1
)  fi   i  1DACCRi ,t   2Qi ,t 1  3 (
CFi ,t 1
Ki ,t 2
)   i ,t
(1)
The dependent variable is the firm’s investment-capital ratios ( Ln(
I i ,t
Ki ,t 1
) ), where investment ( Ii ,t )
is capital expenditure for firm i at year t and capital ( K i ,t 1 ) is the value of net property, plant, and
equipment of the beginning of the year for firm i at year t-1. Q stands for the market-to-book ratio
for firm i at year t-1. Cash flow item (
CFi ,t 1
K i ,t  2
) equals the sum of earnings before extraordinary
items and depreciation for firm i at year t-1, and deflated by the beginning-of-year capital for firm i
at year t-2.  i ,t is an error term for firm i at year t.
The extended specifications consider other controlling variables such as the equity issuance
(
EQISSi ,t 1
K i ,t  2
) , R&D intensity (
R & Di ,t 1
Ai ,t 1
) , share turnover ( TURNi ,t1 ) and high level of DACCR
( HighDACCR i ,t1 ) as follows:
Ln(
I i ,t
Ki ,t 1
)  fi   i  1DACCRi ,t   2Qi ,t 1  3 (
CFi ,t 1
K i ,t  2
 7TURNi ,t 1   i ,t
)  4 (
EQISSi ,t 1
K i ,t  2
)  5 HighDACCRi ,t  6 (
R & Di ,t 1
Ai ,t 1
)
(2)
Where equity issuance (
EQISSi ,t 1
K i ,t  2
) is the new issuance of equity for firm i at time t-1 over
beginning-of-year capital, which is defined as net property, plant, and equipment for firm i at year
t-2. The R&D intensity is measured by the R&D expense for firm i at year t-1 over the book value
of assets for firm i at year t-1. Share turnover is the average of the daily ratio of shares traded to
shares outstanding at the end of the day in Decembert-1 for firm i and year t-1. The high level of
DACCR is a dummy value equal to one if the firm has discretionary accruals in the top 20 th
percentile, and zero otherwise for firm i at year t-1.  i ,t is an error term for firm i at year t.
The expected coefficient of investment on return performance is assumed to be negative
because firm business investment is linked to the market’s misvaluation of the firm’s equity. The
cross-sectional regression of yearly stock returns ( R i ,t 1 ) on investment, Tobin’s Q and a control
for cash-flow sensitivity.
Ri ,t  fi   i  b1Ln(
I i ,t 1
K i ,t  2
)  b2 LnQi ,t 1  b3 (
CFi ,t 1
K i ,t  2
)   i ,t
(3)
This study includes firm’s characteristics that are associated with cross-sectional differences in
average returns that may or may not be associated with risk: size (market capitalization) and
Proceedings of 5th Asia-Pacific Business Research Conference
17 - 18 February, 2014, Hotel Istana, Kuala Lumpur, Malaysia, ISBN: 978-1-922069-44-3
book-to-market equity. These characteristics are known anomalies that this analysis wants to
control for. We also includes the control variable for equity issuance, (
Ri ,t  fi   i  b1 ln(
I i ,t 1
K i ,t  2
)  b2 ln Qi ,t 1  b3 ln(
CFi ,t 1
K i ,t  2
)  b4 (
EQISSi ,t 1
K i ,t  2
EQISSi ,t 1
K i ,t  2
).
)  b5 DACCRi ,t  b6 ln(MEi ,t 1 )
 b7 (
BEi ,t 1
MEi ,t 1
)   i ,t
(4)
where MEi ,t 1 is firm market equity for firm i at year t-1, and
BEi ,t 1
MEi ,t 1
is firm book-to-market equity
for firm i at year t-1. In addition, we define BEi ,t 1 as stockholders’ equity and measure
stockholders’ equity as the book value of common equity, plus the book value of preferred stock
for firm i at year t-1. Moreover, other variables are described as above.
According to the Modified Jones model (Dechow et al. 2005), Authors consider a modified
version of the Jones Model in the empirical analysis. The modification is designed to eliminate
the conjectured tendency of the Jones Model proposed by Jones (1991), who relaxes the
assumption of constant nondiscretionary accruals and attempts to control for the effect of
changes in a firm's economic circumstances on nondiscretionary accruals. The Modified Jones
model is used to measure discretionary accruals with errors when discretion is exercised over
revenues.
To determine the discretionary accruals (DACCR), we begin by estimating an annual OLS
regression of total accruals (TA) for each firm as follows:
TAi ,t
Ai ,t 1
 1
REVi ,t
PPEi ,t
1
 2
 3
  i ,t
Ai ,t 1
Ai ,t 1
Ai ,t 1
(5)
Where TAi ,t is total accrual for firm i at year t (defined as net income minus cash flow from
operations),  REVi ,t is the change in sales revenues for firm i from year t-1 to year t and PPEi ,t is
gross property, plant, and equipment for firm i at year t. All variables are scaled by the lagged
total assets (Ai ,t 1 ) for firm i at year t-1.
Using the estimated regression coefficients in equation (5), then we calculate the nondiscretionary accruals (NDACCR) as follows:
NDACCRi ,t  ˆ1
REVi ,t ARi ,t
PPEi ,t
1
 ˆ 2 (

)  ˆ3
Ai ,t 1
Ai ,t 1
Ai ,t 1
Ai ,t 1
(6)
where ARi ,t is the change in accounts receivable for firm i from year t-1 to year t, and ̂1 , ̂ 2 and
̂ 3 are the estimated coefficients. DACCRi ,t for firm i at year t is estimated as follows:
DACCRi ,t 
TAi ,t
Ai ,t 1
 NDACCRi ,t
(7)
Larger values of DACCRi ,t for firm i at year t indicate a higher probability of earning-increasing
manipulation, while firms with smaller DACCRi ,t are more likely managing earnings downward.
4. Empirical Results
A. The Results of Discretionary Accruals and Capital Investment
The empirical investigation is conducted by estimating Equation (1) to Equation (2) at 5 quantiles,
namely the 10th, 25th, 50th, 75th, and 90th quantiles, using the same list of explanatory
variables for each of these quantiles. On the one hand, we do not account for financial
constraints. On the Basic model (1), the coefficient of discretionary accruals, we find that the
discretionary accruals have a significant positive relationship with capital investment under the
Proceedings of 5th Asia-Pacific Business Research Conference
17 - 18 February, 2014, Hotel Istana, Kuala Lumpur, Malaysia, ISBN: 978-1-922069-44-3
OLS estimation in Taiwan, and under the 25th, 50th and 75th quantiles in China, The coefficient
of cash flow, we find that cash flow has a significant positive relationship with capital investment
under the OLS estimation and the different quantiles in Taiwan, but the results in China are
under the 25th and 50th quantiles, and also display results in panel A of table 1, table 2, and
table 3 for three Chinese areas, respectively. The outcome of basic model represents that the
firms that are mispricing and have more cash flow may increase the level of capital expenditures.
Furthermore, On the extend model (2), the coefficient of discretionary accruals, the result shows
the outcome that the discretionary accruals have a significant positive relationship with capital
investment under the OLS estimation in Taiwan, and under the 50th, 75th and 90th quantiles in
China. We display estimation results in panel B of table 1, table 2, and table 3 for three Chinese
areas, respectively. However, the discretionary accruals have a significant negative relationship
with capital investment under the OLS estimation, and under 25th and 75th quantiles in Hong
Kong. The coefficient of cash flow, there is a statistically significant positive correlation between
the cash flow and investment holdings in OLS estimation, and the 10th to 90th quantiles in
Taiwan. The result is the same under the 25th, 50th and 90th quantiles in China. Another, there
is a statistically significant negative correlation between the cash flow and investment holdings in
OLS estimation, the 75th, and 90th quantiles in Hong Kong. The coefficient of R&D, there is a
statistically significant positive correlation between R&D and investment holdings in lower
quantiles in Taiwan and Hong Kong. There is a significantly positive correlation between R&D
and investment holdings in the 10th to 90th quantiles in China. The coefficient of share turnover,
there is a statistically significant positive correlation between share turnover and investment
holdings in higher quantiles in Hong Kong. There is a statistically significant negative correlation
between share turnover and investment holdings in the 10th to 90th quantiles in Taiwan, and in
the 25th, 50th, and 75th quantiles in China. On the other hand, we take financial constraints into
consideration, and divide financially constrained firms into less constrained firms and high
constrained firms. First, using KZ index stands for financial constraints, and also display results
in table 7, table 8, and table 9 for three Chinese areas, respectively. We find that there are
positive relationships between the discretionary accruals and investment for high financially
constrained firms under OLS in Taiwan. In addition, there are positive relationships between the
discretionary accruals and investment for less financially constrained firms in lower quantiles, or
for high financially constrained firms in higher quantiles in China and there are significantly
negative relationships on higher quantiles for less financially constrained firms in Hong Kong.
The coefficient of equity issuance, there are significantly positive relationships for less and high
financially constrained firms in Taiwan, Hong Kong, and China, respectively. The coefficient of
R&D expense, the coefficient is more sensitive and the relationship significantly positive between
the R&D expense and investment for high constrained firms than for less constrained firms in
lower quantiles in Taiwan. The coefficient is also significantly positive between above the two
variables for high constrained firms in lower quantiles in Hong Kong. The coefficient is less
sensitive and the relationship less significantly positive between the R&D expense and
investment for high constrained firms than for less constrained firms in lower quantiles in China.
Second, using WW index stands for financial constraints, there are some inconsistent in three
Chinese areas. We also display results in table 10, table 11, and table 12 for three Chinese
areas, respectively.
B. The Results of Capital Investment and Stock Return
In the basic model (3), the coefficient on investment is positively significantly to stock returns in
Taiwan and in Hong Kong. The result is inconsistent with Polk and Sapienza (2009), that is, firm
investment and valuation of investors is consistent. In the extend model (4), there are the same
result. In addition, book-to-market equity predicts average returns with a positive coefficient,
while size has a negative coefficient in three Chinese areas. Another, equity issuance,
Proceedings of 5th Asia-Pacific Business Research Conference
17 - 18 February, 2014, Hotel Istana, Kuala Lumpur, Malaysia, ISBN: 978-1-922069-44-3
(
EQISSi ,t 1
K i ,t  2
) , consistent with previous research, we find that firms issuing equity subsequently
underperform in Taiwan and China.
5. Conclusion
This study tests the status of mispricing and applies it to catering theory in order to investigate
managers’ investment decision-making. We further take financial transparency and cash flow
into consideration to broadly examine the relationship with capital investment. Moreover, we take
the financial constraints into consideration to examine whether a firm’s managers would cater to
the sentiment of investors, and do quantile regression to test the impact of capital investment. In
addition, we test the relationship between stock returns and capital investment under the
financial constraints and splits the sample by different share turnover level in the three Chinese
areas.
According to Stein (1996), stock prices deviations from fundamental value may have influence on
capital investments. We use the OLS and quantile regression to estimate the relationship
between discretionary accruals and the capital investment in order to completely observe the
relationship. When the conditional distribution is heterogeneous, the ordinary least squares
method can only provide the estimation of the mean of depending variable. Therefore, the
usefulness of the estimated results is limited and may even be biased.
The empirical results of study show that the data in all three Chinese areas have inconsistent
patterns in the various coefficients obtained from different quantile functions. The coefficients are
significantly positive between discretionary accruals and capital investment under the OLS
estimation in Taiwan and in the 50th to 90th quantiles estimation in China. These findings are
consistent with Polk and Sapienza (2009). We predicted that managers would take market
participants’ sentiment into consideration when they seem to lack more accurate information
about the market, and that managers tend to make an investment that has a negative net
present value to increase the stock price in the short run when investors are in the highly
mispriced market. However, there is positive relationship between share-turnover and capital
investment in Hong Kong under the higher quantiles estimation. The firms with shorter
shareholder horizons, and those with assets which are more difficult to value, cater more in Hong
Kong.
The efficiency of investment decisions examined provides empirical evidence of the pecking
order hypothesis and the free cash-flow hypothesis by taking the impact of cash flow into
consideration. The coefficients between cash flow and capital investment are significantly
positive in Taiwan and in China.
Combining the catering behavior in investment decisions and financial constraints, we are unable
to obtain consistent results in all three Chinese areas. However, this study shows that the
coefficients between the cash flow and capital investment are more sensitive for less financial
constrained firms in Taiwan and in Hong Kong. This finding is consistent with Kaplan and
Zingales (1997) who analyze both quantitative and qualitative information on firms and find that
less constrained firms would exhibit significantly higher cash flow sensitivity of investment.
This study also shows the relationship between capital investment and stock returns in order to
examine whether the firms would cater more. When we control for investment opportunities and
other characteristics, the investment of financially unconstrained firms predicts significantly
positive stock returns in Taiwan and Hong Kong under the KZ index. Hence, we can conclude
that firms would raise capital investment and cause the higher stock returns.
When they make decisions regarding capital investment, firm’s managers can’t cater to the
investor to take overinvestment. However, taking the condition of financial constraints into
consideration, the increase of capital investment may lead to higher stock returns.
Proceedings of 5th Asia-Pacific Business Research Conference
17 - 18 February, 2014, Hotel Istana, Kuala Lumpur, Malaysia, ISBN: 978-1-922069-44-3
References
Baker, M. and Wurgler J. 2000, The equity share in new issues and aggregate stock
Returns, Journal of Finance, Vol.55, pp.2219-2259.
Baker, M. and Wurgler, J. 2004a, A catering theory of dividends, Journal of Finance,
Vol.59, pp.271–288.
Baker, M. and Wurgler, J. 2004b, Appearing and disappearing dividends: The link to
catering incentives, Journal of Financial Economics, Vol.73, pp.271–288.
Dechow, P. M., Sloan, R. G. and Sweeney, A. P. 1995, Detecting earnings management,
The Accounting Review, Vol.70, pp.193-225.
Fazzari, S. M., Hubbard, R. G. and Petersen, B. 1988, Financing constraints and
corporate investment, Brookings Papers on Economic Activity, Vol.1, pp.141-195.
Jensen, M. and Meckling, W. 1976, Theory of the firm: managerial behavior, agency costs,
and ownership structure, Journal of Financial Economics, Vol.3, pp.305-360.
Kaplan, S. N. and Zingales, L. 1997, Do investment-cash flow sensitivities provide useful
measures of financing constraints? Quarterly Journal of Economics, Vol.20, pp.169215.
Modigliani, F. and Miller, M. 1958, The cost of capital, corporation finance, and the theory
of investment, American Economic Review, Vol.48, pp.261-297.
Polk, C. and Sapienza, P. 2009, The stock market and corporate investment: a test of
catering theory, Review of Financial Studies, Vol.22, pp.187-217.
Stein, J. C. 1996, Rational capital budgeting in an irrational world. Journal of Business,
Vol.69, pp.429–455.
Titman, S. K., Wei, C. J. and Xie, F. 2004, Capital Investment and Stock Returns, Journal
of Financial and Quantitative Analysis, Vol.39, pp.677-700.
Whited, T. 1992, Debt, liquidity constraints, and corporate Investment: evidence from
Panel Data, Journal of Finance, Vol.47, pp.1425-1460.
Whited, T. and Wu, G. 2006, Financial Constraints Risk, The Review of Financial Studies,
Vol.19, pp.531- 559.
Proceedings of 5th Asia-Pacific Business Research Conference
17 - 18 February, 2014, Hotel Istana, Kuala Lumpur, Malaysia, ISBN: 978-1-922069-44-3
Appendix
Table 1: Summary statistics for Taiwan
Country/Variable
Full sample
Less
financially
constrained
firms (KZ
index)
Ln(
Obs.
)
Highly
Financially
Constrained
firms (KZ
index)
Less
Financially
Constrained
firms (WW
index)
Highly
Financially
Constrained
firms (WW
index)
Mean
Std.
dev.
Mean
Std.
dev.
Mean
Std.
dev.
Mean
Std.
dev.
Mean
Std.
dev.
37.759
1128.8
0.740
7.653
74.778
1595.6
28.253
1019.1
47.266
1228.8
-0.144
0.920
-0.053
0.455
-0.235
1.213
-0.211
1.163
-0.077
0.577
3.542
16.126
1.541
2.005
5.542
22.543
3.887
17.294
3.196
14.862
29.518
1269.1
15.952
1.361
1.193
15.934
8.052
1.383
57.844
15.970
1794.5
1.339
16.910
16.620
769.595
1.396
42.127
15.284
1621.5
0.930
0.839
51.219
0.144
2.802
1.534
72.381
0.143
3.386
1.535
72.356
0.013
0.023
0.016
0.026
0.010
0.019
0.013
0.022
0.013
0.024
2.154
1.999
2.145
2.009
2.164
1.990
2.345
2.083
1.964
1.893
3.891
16.788
4.834
21.529
2.948
9.926
5.566
22.109
2.216
8.330
15.368
1.502
15.281
1.369
15.454
1.620
15.813
1.579
14.922
1.274
4994
4994
2497
2497
2497
2497
2497
2497
2497
2497
Table 1 shows the summary statistics for Taiwan and splits the sample according to firms’ degrees of financial
constraints using the KZ index and the WW index. The firms with a below-median KZ index are less financially
constrained firms, and those with an above-median KZ index are highly financially constrained firms. The firms with a
below-median WW index are less financially constrained firms and those with an above-median WW index are highly
financially constrained firms.
Proceedings of 5th Asia-Pacific Business Research Conference
17 - 18 February, 2014, Hotel Istana, Kuala Lumpur, Malaysia, ISBN: 978-1-922069-44-3
Table 2: Summary statistics for Hong Kong
Country/Variable
Full sample
Less
financially
constrained
firms (KZ
index)
Ln(
)
Highly
Financially
Constrained
firms (KZ
index)
Less
Financially
Constrained
firms (WW
index)
Highly
Financially
Constrained
firms (WW
index)
Mean
Std.
dev.
Mean
Std.
dev.
Mean
Std.
dev.
Mean
Std.
dev.
Mean
Std.
dev.
0.612
8.826
0.323
1.581
0.902
12.375
0.417
2.676
0.808
12.189
-0.030
0.354
-0.016
0.183
-0.044
0.466
-0.022
0.313
-0.039
0.391
1.409
2.719
1.073
1.226
1.745
3.613
1.235
2.404
1.584
2.991
-15.8
14.651
1440.3
1.787
3.422
14.742
989.195
1.686
-35.1
14.560
1780.6
1.879
18.691
16.032
769.759
1.320
-50.4
13.270
1885.4
0.911
41.147
1904.2
74.040
2688.6
8.254
149.871
50.797
2445.7
31.498
1127.6
0.033
0.006
0.045
0.007
0.023
0.004
0.021
0.004
0.050
0.007
1.132
7.236
1.143
9.201
1.122
4.479
0.970
3.308
1.295
9.682
2.280
3.386
2.238
2.645
2.322
3.991
2.736
3.403
1.824
3.307
13.799
1.829
14.023
1.817
13.574
1.815
14.964
1.694
12.633 1.054
2761
2761
2761
2761
2761
2761
2761
2761
5522
5522
Obs.
Table 2 shows the summary statistics for Hong Kong and splits the sample according to firms’ degrees of financial
constraints using the KZ index and the WW index. The firms with a below-median KZ index are less financially
constrained firms and those with an above-median KZ index are highly financially constrained firms. The firms with a
below-median WW index are low financially constrained firms and those with an above-median WW index are highly
financially constrained firms.
Proceedings of 5th Asia-Pacific Business Research Conference
17 - 18 February, 2014, Hotel Istana, Kuala Lumpur, Malaysia, ISBN: 978-1-922069-44-3
Table 3: Summary statistics for China
Country/Variable
Full sample
Less
financially
constrained
firms (KZ
index)
Ln(
)
Highly
Financially
Constrained
firms (KZ
index)
Less
Financially
Constrained
firms (WW
index)
Highly
Financially
Constrained
firms (WW
index)
Mean
Std.
dev.
Mean
Std.
dev.
Mean
Std.
dev.
Mean
Std.
dev.
Mean
Std.
dev.
1.802
157.079
0.289
3.646
3.314
222.103
3.346
222.147
0.257
1.438
0.052
0.803
0.018
0.415
0.087
1.056
0.106
0.971
-0.001
0.584
2.432
4.944
1.970
1.523
2.894
6.793
1.769
1.527
3.095
6.759
0.649
14.563
16.100
1.171
0.550
14.488
5.493
0.981
0.749
14.638
22.096
1.329
0.958
15.452
7.942
0.839
0.341
13.674
21.335
0.675
0.266
4.027
0.258
4.337
0.275
3.691
0.385
5.387
0.147
1.842
0.048
0.002
0.003
0.000
0.082
0.002
0.048
0.002
0.000
0.000
4.277
3.426
3.800
3.150
4.753
3.619
4.254
3.388
4.299
3.464
1.352
5.212
1.743
5.212
0.962
5.183
2.033
6.771
0.672
2.750
14.531
1.251
14.450 1.264 14.613
1.232
15.008
1.261
14.054 1.041
5682
5682
5681
5681
5682
5682
5681
5681
11363
11363
Obs.
Table 3 shows the summary statistics for China and splits the sample according to firms’ degrees of financial
constraints using the KZ index and the WW index. The firms with a below-median KZ index are less financially
constrained firms and those with an above-median KZ index are highly financially constrained firms. The firms with a
below-median WW index are less financially constrained firms and with those an above-median WW index are highly
financially constrained firms.
Proceedings of 5th Asia-Pacific Business Research Conference
17 - 18 February, 2014, Hotel Istana, Kuala Lumpur, Malaysia, ISBN: 978-1-922069-44-3
Table 4: Regression results with discretionary accruals and capital investment in Taiwan
Panel A: discretionary accruals
OLS
Quantile regressions
0.1
0.25
0.5
0.75
0.9
DACCRi,t
182.054*** -0.437*** -0.398*** -0.423*** -0.004
1.807
(20.6386) (0.0627) (0.068)
(0.136)
(0.221)
(1.975)
Qi,t-1
39.099*** 0.054*** 0.111*** 0.281*** 0.310**
2.115***
(1.274)
(0.010)
(0.004)
(0.006)
(0.149)
(0.231)
CFi,t-1/Ki,t-2
0.137***
0.027*** 0.027*** 0.026*** 1.235*** 1.225***
(0.012)
(0.000)
(0.000)
(0.000)
(0.001)
(0.001)
EOISSi,t-1/Ki,t-2
HighDACCRi,t-1
R&Di,t-1/Ai,t-1
TURNi,t-1
2
R /Pseudo R
2
0.307
Obs.
4994
Panel B: capital investment
OLS
DACCRi,t
Qi,t-1
CFi,t-1/Ki,t-2
EOISSi,t-1/Ki,t-2
HighDACCRi,t-1
R&Di,t-1/Ai,t-1
TURNi,t-1
R2/Pseudo R2
Obs.
73.244***
(17.926)
22.845***
(1.139)
0.164***
(0.010)
10.619***
(0.234)
15.369
(28.717)
-1083.5**
(490.693)
4.975
(5.650)
0.510
4994
0.023
4994
0.026
4994
0.030
4994
0.238
4994
0.1
-0.621**
(0.262)
0.046***
(0.017)
0.027***
(0.000)
0.905***
(0.009)
0.057
(0.063)
0.032
(0.108)
-0.007***
(0.003)
0.033
4994
Quantile regressions
0.25
0.5
0.75
-0.702*** -0.831*** -0.087**
(0.151)
(0.123)
(0.036)
0.086***
0.262***
0.337***
(0.011)
(0.006)
(0.006)
0.027***
0.026***
1.234***
(0.000)
(0.000)
(0.000)
13.072*** 13.056*** 9.831***
(0.001)
(0.000)
(0.001)
0.097***
0.118***
-0.032*
(0.034)
(0.025)
(0.018)
0.315***
0.769***
-2.457***
(0.087)
(0.211)
(0.541)
-0.009*** -0.008*** -0.019***
(0.002)
(0.002)
(0.004)
0.149
0.237
0.414
4994
4994
4994
0.367
4994
0.9
0.567
(1.972)
1.747***
(0.203)
1.227***
(0.000)
9.707***
(0.015)
-0.199
(0.409)
0.841
(1.033)
-0.026***
(0.010)
0.5499
4994
Table 4 shows regression results with discretionary accruals and capital investment in Taiwan. Dependent variable:
, The standard errors reported in parentheses are corrected clustering of the residual at the firm level.
Coefficients starred with *, **, and *** are statistically significant at 10%, 5%, and 1% levels, respectively.
Proceedings of 5th Asia-Pacific Business Research Conference
17 - 18 February, 2014, Hotel Istana, Kuala Lumpur, Malaysia, ISBN: 978-1-922069-44-3
Table 5: Regression results with discretionary accruals and capital investment in Hong Kong
Panel A: discretionary accruals
OLS
Quantile regressions
0.1
0.25
0.5
0.75
0.9
DACCRi,t
-0.311
-0.006
-0.015**
-0.028
0.001
0.014
(0.198)
(0.005)
(0.007)
(0.019)
(0.089) (0.023)
Qi,t-1
0.069***
0.002***
0.008**
0.015***
0.035
0.099*
(0.026)
(0.001)
(0.003)
(0.003)
(0.027) (0.052)
CFi,t-1/Ki,t-2
-0.005***
0.000***
0.000*** -0.001***
-0.003
-0.006***
(0.000)
(0.000)
(0.000)
(0.000)
(0.062) (0.000)
EOISSi,t-1/Ki,t-2
HighDACCRi,t-1
R&Di,t-1/Ai,t-1
TURNi,t-1
R2/Pseudo R2 0.654
Obs.
5522
Panel B: capital investment
OLS
DACCRi,t
Qi,t-1
CFi,t-1/Ki,t-2
EOISSi,t-1/Ki,t-2
HighDACCRi,t-1
R&Di,t-1/Ai,t-1
TURNi,t-1
R2/Pseudo R2
Obs.
-0.380*
(0.198)
0.069***
(0.024)
-0.005***
(0.000)
-0.001***
(0.000)
0.264
(0.176)
-7.097
(11.253)
0.002
(0.009)
0.693
5522
0.001
5522
0.002
5522
Quantile regressions
0.1
0.25
-0.008
-0.020**
(0.007)
(0.010)
0.002***
0.008***
(0.001)
(0.002)
0.000
0.000
(0.000)
(0.000)
0.000
0.000
(0.000)
(0.000)
0.003
0.011**
(0.003)
(0.005)
1.518***
1.506***
(0.208)
(0.205)
0.000
-0.000
(0.000)
(0.000)
0.002
0.003
5522
5522
0.009
5522
0.5
-0.052
(0.035)
0.015***
(0.003)
-0.000
(0.000)
0.000*
(0.000)
0.040***
(0.014)
0.917***
(0.284)
0.002***
(0.000)
0.0111
5522
0.024
5522
0.75
-0.089**
(0.037)
0.034***
(0.008)
-0.007***
(0.000)
-0.001***
(0.000)
0.151***
(0.024)
0.252
(0.514)
0.003***
(0.000)
0.033
5522
0.069
5522
0.9
-0.113
(0.098)
0.107***
(0.038)
-0.006***
(0.000)
0.001***
(0.000)
0.370***
(0.078)
-1.617***
(0.453)
0.012***
(0.001)
0.076
5522
Table 5 shows regression results with discretionary accruals and capital investment in Hong Kong.
Dependent variable:
, The standard errors reported in parentheses are corrected clustering of the
residual at the firm level. Coefficients starred with *, **, and *** are statistically significant at 10%, 5%, and 1% levels,
respectively.
Proceedings of 5th Asia-Pacific Business Research Conference
17 - 18 February, 2014, Hotel Istana, Kuala Lumpur, Malaysia, ISBN: 978-1-922069-44-3
Table 6: Regression results with discretionary accruals and capital investment in China
Panel A: discretionary accruals
OLS
Quantile regressions
0.1
0.25
0.5
0.75
0.9
DACCRi,t
0.327
0.004
0.009***
0.032
0.052***
0.060
(1.835) (0.005)
(0.002)
(0.023)
(0.019)
(0.069)
Qi,t-1
-0.083
-0.002**
-0.002**
-0.000
0.013**
0.056***
(0.298) (0.001)
(0.001)
(0.000)
(0.006)
(0.021)
CFi,t-1/Ki,t-2
-0.002
0.002
0.002***
0.003***
0.005
0.017
(0.092) (0.002)
(0.000)
(0.000)
(0.014)
(0.038)
EOISSi,t-1/Ki,t-2
HighDACCRi,t-1
R&Di,t-1/Ai,t-1
TURNi,t-1
R2/Pseudo R2
0.000
Obs.
11363
Panel B: capital investment
OLS
DACCRi,t
Qi,t-1
CFi,t-1/Ki,t-2
EOISSi,t-1/Ki,t-2
HighDACCRi,t-1
R&Di,t-1/Ai,t-1
TURNi,t-1
R2/Pseudo R2
Obs.
-0.626
(1.892)
-0.079
(0.299)
-0.005
(0.092)
0.050
(0.367)
7.822**
(3.800)
1.703
(953.449)
0.011
(0.431)
0.000
11363
0.001
11363
0.1
0.001
(0.003)
-0.002**
(0.001)
0.001
(0.005)
0.002*
(0.001)
0.004
(0.003)
5.069***
(0.320)
0.000
(0.000)
0.001
11363
0.001
11363
0.001
11363
0.001
11363
Quantile regressions
0.25
0.5
0.75
0.005
0.006**
0.037***
(0.018)
(0.003)
(0.002)
-0.002**
-0.000*** 0.011
(0.001)
(0.000)
(0.008)
0.002***
0.003***
0.004
(0.000)
(0.000)
(0.032)
0.003***
0.013*
0.090*
(0.000)
(0.007)
(0.054)
0.012*
0.034***
0.063***
(0.007)
(0.006)
(0.017)
5.225***
4.382***
2.900***
(0.220)
(0.208)
(0.310)
-0.001*** -0.002*** -0.002*
(0.000)
(0.001)
(0.001)
0.001
0.001
0.003
11363
11363
11363
0.002
11363
0.9
0.032***
(0.008)
0.046**
(0.019)
0.009*
(0.005)
0.106***
(0.001)
0.120***
(0.031)
0.671**
(0.288)
-0.002
(0.003)
0.004
11363
Table 6 shows regression results with discretionary accruals and capital investment in China. Dependent variable:
, The standard errors reported in parentheses are corrected clustering of the residual at the firm level.
Coefficients starred with *, **, and *** are statistically significant at 10%, 5%, and 1% levels, respectively.
Proceedings of 5th Asia-Pacific Business Research Conference
17 - 18 February, 2014, Hotel Istana, Kuala Lumpur, Malaysia, ISBN: 978-1-922069-44-3
Table 7: Regression results with discretionary accruals and capital investment in Taiwan by KZ
index
Panel A: Less financially constrained firms (KZ index)
OLS
Quantile regressions
0.1
0.25
0.5
0.75
0.9
DACCRi,t
-0.347
-0.139
-0.379**
-0.624**
-0.705
-1.148
(0.358)
(0.221)
(0.1619)
(0.258)
(0.664) (0.719)
Qi,t-1
0.193**
0.026**
0.078***
0.135***
0.218
0.407**
(0.078)
(0.011)
(0.007)
(0.022)
(0.140) (0.174)
CFi,t-1/Ki,t-2
0.339*** 0.003*** 0.004
0.082
0.305* 0.745*
(0.017)
(0.000)
(0.015)
(0.076)
(0.174) (0.408)
EOISSi,t-1/Ki,t-2 0.919*** 0.123
0.373***
0.832***
1.335
1.130**
(0.048)
(0.097)
(0.012)
(0.098)
(1.029) (0.500)
HighDACCRi,t-1 -0.320
0.019
0.056
0.097*
0.111
0.290*
(0.330)
(0.043)
(0.032)
(0.051)
(0.140) (0.173)
R&Di,t-1/Ai,t-1
-0.667
0.076*
0.241**
0.547
0.982
1.966
(4.767)
(0.044)
(0.117)
(0.693)
(0.775) (1.817)
TURNi,t-1
0.001
-0.001
-0.003**
-0.004**
-0.005
-0.008
(0.062)
(0.001)
(0.001)
(0.002)
(0.004) (0.013)
R2/Pseudo R2 0.349
0.029
0.085
0.085
0.313
0.465
Obs.
2497
2497
2497
2497
2497
2497
Panel B: Highly financially constrained firms (KZ index)
Taiwan
OLS
Quantile regressions
0.1
0.25
0.5
0.75
0.9
DACCRi,t
74.719***
-1.088*** -1.162*** -0.786
-0.209
2.670
(28.312)
(0.250)
(0.402)
(0.681)
(0.458)
(6.196)
Qi,t-1
23.683***
0.039**
0.084**
0.298*** 0.384*** 2.489***
(1.695)
(0.017)
(0.034)
(0.096)
(0.028)
(0.106)
CFi,t-1/Ki,t-2
0.160***
0.027*** 0.027*** 0.026*** 1.234*** 1.223**
(0.014)
(0.000)
(0.000)
(0.000)
(0.000)
(0.002)
EOISSi,t-1/Ki,t-2
10.571***
13.075*** 13.071*** 13.053*** 9.826*** 9.645***
(0.331)
(0.001)
(0.002)
(0.007)
(0.002)
(0.001)
HighDACCRi,t-1 55.040
0.061
0.166*
0.084
0.004*** -0.658
(57.052)
(0.191)
(0.088)
(0.141)
(0.109)
(1.149)
R&Di,t-1/Ai,t-1
3522.442*** 1.080
0.471*
0.398
1.498
1.186
(1210.751)
(0.664)
(0.273)
(0.372)
(1.259)
(2.705)
TURNi,t-1
12.689
-0.117*** -0.011*** -0.011*** -0.015*** -0.018
(11.362)
(0.038)
(0.003)
(0.004)
(0.005)
(0.023)
2
2
R /Pseudo R
0.512
0.110
0.221
0.261
0.424
0.554
Obs.
2497
2497
2497
2497
2497
2497
Table 7 shows the regression results with discretionary accruals and capital investment in Taiwan and splits the
sample according to firms’ degrees of financial constraints using the KZ index. The firms with a below-median KZ
index are less financially constrained firms and those with an above-median KZ index are highly financially
constrained firms. Dependent variable:
, The standard errors reported in parentheses are corrected
clustering of the residual at the firm level. Coefficients starred with *, **, and *** are statistically significant at 10%, 5%,
and 1% levels, respectively.
Proceedings of 5th Asia-Pacific Business Research Conference
17 - 18 February, 2014, Hotel Istana, Kuala Lumpur, Malaysia, ISBN: 978-1-922069-44-3
Table 8: Regression results with discretionary accruals and capital investment in Hong Kong by
KZ index
Panel A: Less financially constrained firms (KZ index)
OLS
Quantile regressions
0.1
0.25
0.5
0.75
0.9
DACCRi,t
0.054
-0.036
-0.104**
-0.179***
-0.130***
-0.228***
(0.183)
(0.022)
(0.044)
(0.057)
(0.015)
(0.075)
Qi,t-1
0.000
0.013***
0.028***
0.059***
0.065***
0.072
(0.026)
(0.002)
(0.005)
(0.008)
(0.013)
(0.057)
CFi,t-1/Ki,t-2
0.000*
0.000
0.000
0.000*
0.000***
0.000
(0.000)
(0.000)
(0.000)
(0.000)
(0.000)
(0.001)
EOISSi,t-1/Ki,t-2 0.000***
0.000***
0.000*
0.000***
0.000***
0.000*
(0.000)
(0.000)
(0.000)
(0.000)
(0.000)
(0.000)
HighDACCRi,t-1
0.188**
0.006
0.022**
0.046**
0.106***
0.415**
(0.083)
(0.005)
(0.009)
(0.016)
(0.023)
(0.176)
R&Di,t-1/Ai,t-1
-0.240
0.547
0.376
0.050
-0.055
-1.051**
(4.276)
(0.487)
(0.676)
(1.412)
(0.516)
(0.454)
TURNi,t-1
0.001
0.000
0.001
0.002***
0.002***
0.011***
(0.003)
(0.000)
(0.006)
(0.000)
(0.000)
(0.001)
2
2
R /Pseudo R
0.058
0.014
0.020
0.038
0.048
0.054
Obs.
2761
2761
2761
2761
2761
2761
Panel A: Highly financially constrained firms (KZ index)
OLS
Quantile regressions
0.1
0.25
0.5
0.75
0.9
DACCRi,t
-0.426**
-0.006
0.003
-0.034
-0.073
-0.041
(0.203)
(0.018)
(0.021)
(0.029)
(0.193)
(0.083)
Qi,t-1
0.066***
-0.000
0.002**
0.011***
0.017*
0.183**
(0.025)
(0.004)
(0.001)
(0.002)
(0.009)
(0.087)
CFi,t-1/Ki,t-2
-0.006***
-0.002
-0.006***
-0.006***
-0.006***
-0.003*
(0.000)
(0.009)
(0.000)
(0.000)
(0.000)
(0.002)
EOISSi,t-1/Ki,t-2 0.004***
0.001
0.001***
0.001***
0.005
0.071*
(0.001)
(0.000)
(0.000)
(0.000)
(0.009)
(0.037)
HighDACCRi,t-1
0.636***
0.006
0.006
0.036**
0.171*
0.223
(0.237)
(0.010)
(0.009)
(0.014)
(0.091)
(0.355)
R&Di,t-1/Ai,t-1
-9.411
1.479***
1.469***
3.805***
3.482
-5.015
(22.089)
(0.402)
(0.452)
(0.782)
(2.800)
(3.060)
TURNi,t-1
0.006
-0.000
-0.000
-0.001**
0.002
0.010***
(0.020)
(0.000)
(0.000)
(0.000)
(0.004)
(0.002)
2
2
R /Pseudo R
0.858
0.020
0.156
0.220
0.259
0.306
Obs.
2761
2761
2761
2761
2761
2761
Table 8 shows the regression results with discretionary accruals and capital investment in Hong Kong and splits the
sample according to firms’ degrees of financial constraints using the KZ index. The firms with a below-median KZ
index are less financially constrained firms and those with an above-median KZ index are highly financially
constrained firms. Dependent variable:
, The standard errors reported in parentheses are corrected
clustering of the residual at the firm level. Coefficients starred with *, **, and *** are statistically significant at 10%, 5%,
and 1% levels, respectively.
.
Proceedings of 5th Asia-Pacific Business Research Conference
17 - 18 February, 2014, Hotel Istana, Kuala Lumpur, Malaysia, ISBN: 978-1-922069-44-3
Table 9: Regression results with discretionary accruals and capital investment in China by KZ
index
Panel A: Less financially constrained firms (KZ index)
OLS
Quantile regressions
0.1
0.25
0.5
0.75
0.9
DACCRi,t
-0.071
0.008*
0.012*** 0.033
0.071
0.083*
(0.138)
(0.005)
(0.004)
(0.054)
(0.058)
(0.042)
Qi,t-1
0.118***
-0.002**
0.002
0.012**
0.036*** 0.086***
(0.036)
(0.001)
(0.001)
(0.005)
(0.006)
(0.019)
CFi,t-1/Ki,t-2
0.053***
0.001
0.004*** 0.016
0.070*** 0.069***
(0.009)
(0.001)
(0.000)
(0.016)
(0.000)
(0.000)
EOISSi,t-1/Ki,t-2 0.082***
0.001*** 0.014
0.065*** 0.095*** 0.095
(0.011)
(0.000)
(0.013)
(0.016)
(0.000)
(0.242)
HighDACCRi,t-1 0.243*
-0.002
0.001
0.005
-0.005
0.006
(0.127)
(0.003)
(0.004)
(0.015)
(0.017)
(0.037 )
R&Di,t-1/Ai,t-1
9.6125
32.057*** 27.483*** 17.336*** -0.067
-25.994***
(557.678) (2.391)
(3.138)
(4.160)
(3.536)
(4.541)
TURNi,t-1
-0.020
-0.000
-0.001*** -0.004*** -0.007*** -0.014***
(0.016)
(0.000)
(0.000)
(0.001)
(0.001)
(0.003)
2
2
R /Pseudo R
0.0197
0.002
0.006
0.028
0.070
0.107
Obs.
5681
5681
5681
5681
5681
5681
Panel B: Highly financially constrained firms (KZ index)
OLS
Quantile regressions
0.1
0.25
0.5
0.75
0.9
DACCRi,t
0.643
-0.000
0.007
0.002
0.036*** 0.034***
(2.890)
(0.002)
(0.005)
(0.007)
(0.002)
(0.011)
Qi,t-1
-0.118
-0.002
-0.003** -0.001**
0.005
0.028
(0.435)
(0.001)
(0.001)
(0.001)
(0.006)
(0.026)
CFi,t-1/Ki,t-2
-0.006
0.001
0.002*** 0.002***
0.002*** 0.001***
(0.134)
(0.001 )
(0.000)
(0.000)
(0.000)
(0.000)
EOISSi,t-1/Ki,t-2
0.038
0.002*** 0.001*** 0.004***
0.006**
0.079***
(0.801 )
(0.000)
(0.000)
(0.001)
(0.003)
(0.016)
HighDACCRi,t-1
-48.468
5.087*** 5.174*** 4.181***
2.770*** 0.156
(1350.3) (0.469)
(0.254)
(0.300 )
(0.317)
(0.337)
R&Di,t-1/Ai,t-1
-4.080
0.007*
0.013**
0.051***
0.107*** 0.256***
(7.630)
(0.004)
(0.006)
(0.010)
(0.016)
(0.040)
TURNi,t-1
-0.112
0.000
-0.001*
-0.003*** -0.001
0.004
(0.816)
(0.000)
(0.001)
(0.001)
(0.001)
(0.005 )
2
2
R /Pseudo R
0.000
0.001
0.001
0.001
0.001
0.001
Obs.
5682
5682
5682
5682
5682
5682
Table 9 shows the regression results with discretionary accruals and capital investment in China and splits the
sample according to firms’ degrees of financial constraints using the KZ index. The firms with a below-median KZ
index are less financially constrained firms and those with an above-median KZ index are highly financially
constrained firms. Dependent variable:
, The standard errors reported in parentheses are corrected
clustering of the residual at the firm level. Coefficients starred with *, **, and *** are statistically significant at 10%, 5%,
and 1% levels, respectively.
.
Proceedings of 5th Asia-Pacific Business Research Conference
17 - 18 February, 2014, Hotel Istana, Kuala Lumpur, Malaysia, ISBN: 978-1-922069-44-3
Table 10: Regression results with discretionary accruals and capital investment in Taiwan by WW
index
Panel A: Less financially constrained firms (WW index)
OLS
0.1
DACCRi,t
42.891***
2.530***
(8.288)
(0.145)
Qi,t-1
8.092***
-0.208***
(0.557)
(0.024)
CFi,t-1/Ki,t-2
1.211***
1.239***
(0.008)
(0.000)
EOISSi,t-1/Ki,t-2
52.541***
-0.542***
(1.770)
(0.033)
HighDACCRi,t-1
-19.007
-1.338***
(14.193)
(0.242)
R&Di,t-1/Ai,t-1
-303.560
-20.003***
(261.069)
(3.127)
TURNi,t-1
0.871
-0.012
(2.725)
(0.019)
2
2
R /Pseudo R
0.925
0.579
Obs.
2497
2497
Panel B: Highly financially constrained firms (WW
OLS
0.1
DACCRi,t
680.862***
-0.462***
(24.567)
(0.038)
Qi,t-1
67.082***
0.040***
(1.230)
(0.016)
CFi,t-1/Ki,t-2
-0.226***
0.027***
(0.009)
(0.000)
EOISSi,t-1/Ki,t-2
8.725***
13.077***
(0.174)
(0.001)
HighDACCRi,t-1
-182.750*** -0.034
(27.711)
(0.070)
R&Di,t-1/Ai,t-1
-936.123**
0.338
(436.233)
(0.211)
TURNi,t-1
-4.200
-0.104**
(5.573)
(0.051)
R2/Pseudo R2
0.819
0.100
Obs.
2497
2497
Quantile regressions
0.25
0.5
0.75
2.266***
1.202***
0.082
(0.038)
(0.308)
(1.500)
0.019**
0.054***
0.363
(0.008)
(0.002)
(0.231)
1.237***
1.237***
1.234***
(0.000)
(0.000)
(0.001)
0.008
0.531***
2.922***
(0.009)
(0.002)
(0.158)
-0.694***
-0.303***
-0.040
(0.054)
(0.065)
(0.317)
-9.682***
-3.817***
-1.349
(1.869)
(0.639)
(1.348)
-0.011
0.001
-0.005
(0.007)
(0.004)
(0.0058)
0.651
0.687
0.709
2497
2497
2497
index)
Quantile regressions
0.25
0.5
0.75
-0.524***
-0.616**
-0.464**
(0.181 )
(0.279)
(0.203)
0.077***
0.202***
0.642***
(0.023)
(0.041)
(0.010)
0.027***
0.026***
0.024***
(0.000)
(0.000)
(0.000)
13.073*** 13.062*** 13.022***
(0.002)
(0.004)
(0.000)
0.081**
0.092
0.040
(0.036)
(0.060)
(0.050)
0.284**
0.812***
2.425***
(0.135)
(0.310)
(0.788)
-0.010***
-0.009**
-0.013**
(0.004)
(0.004)
(0.004)
0.311
0.383
0.411
2497
2497
2497
0.9
-0.633
(2.481)
0.951
(0.710)
1.231***
(0.003)
3.090
(3.321)
0.133
(0.473)
2.326
(2.694)
-0.012
(0.010)
0.727
2497
0.9
36.218***
(3.314)
35.056***
(0.043)
-0.136***
(0.000)
9.998***
(0.008)
-9.304***
(1.110)
54.376**
(23.897)
-0.494**
(0.211)
0.481
2497
Table 10 shows the regression results with discretionary accruals and capital investment in Taiwan and splits the
sample according to firms’ degrees of financial constraints using the WW index. The firms with a below-median WW
index are less financially constrained firms and those with an above-median WW index are highly financially
constrained firms. Dependent variable:
, The standard errors reported in parentheses are corrected
clustering of the residual at the firm level. Coefficients starred with *, **, and *** are statistically significant at 10%, 5%,
and 1% levels, respectively.
Proceedings of 5th Asia-Pacific Business Research Conference
17 - 18 February, 2014, Hotel Istana, Kuala Lumpur, Malaysia, ISBN: 978-1-922069-44-3
Table 11: Regression results with discretionary accruals and capital investment in Hong Kong by
WW index
Panel A: Less financially constrained firms (WW index)
OLS
Quantile regressions
0.1
0.25
0.5
0.75
DACCRi,t
-0.538***
-0.019
-0.078
-0.106
-0.281*
(0.168)
(0.013)
(0.073)
(0.073)
(0.165)
Qi,t-1
0.156***
0.011***
0.019***
0.027**
0.057***
(0.021)
(0.001)
(0.003)
(0.013)
(0.019)
CFi,t-1/Ki,t-2
0.000
0.000
0.000
0.000**
0.000
(0.000)
(0.000)
(0.000)
(0.000)
(0.001)
EOISSi,t-1/Ki,t-2
0.000***
0.000***
0.000***
0.000***
0.000***
(0.000)
(0.000)
(0.000)
(0.000)
(0.000)
HighDACCRi,t-1
0.268**
0.005
0.027**
0.046**
0.157***
(0.131)
(0.005)
(0.0145)
(0.019)
(0.047)
R&Di,t-1/Ai,t-1
-9.898
0.715*
1.407***
1.249
2.200
(11.814)
(0.420)
(0.513)
(1.394)
(1.746)
TURNi,t-1
0.000
0.000
0.000
0.003
0.002
(0.015)
(0.000)
(0.001)
(0.007)
(0.013)
2
2
R /Pseudo R
0.047
0.028
0.035
0.036
0.043
Obs.
2761
2761
2761
2761
2761
Panel B: Highly financially constrained firms (WW index)
OLS
Quantile regressions
0.1
0.25
0.5
0.75
DACCRi,t
-0.391*
-0.007
-0.008
-0.035**
-0.006
(0.223)
(0.007)
(0.011)
(0.014)
(0.066)
Qi,t-1
0.015
-0.000
0.002*
0.008**
0.030
(0.027)
(0.001)
(0.001)
(0.004)
(0.241)
CFi,t-1/Ki,t-2
-0.007***
-0.001***
-0.007***
-0.007***
-0.007***
(0.000)
(0.000)
(0.000)
(0.000)
(0.001)
EOISSi,t-1/Ki,t-2
-0.004***
-0.001***
-0.004***
-0.004***
-0.001
(0.000)
(0.000)
(0.000)
(0.000)
(0.030)
HighDACCRi,t-1
0.604***
0.002
-0.000
0.043***
0.130
(0.219)
(0.006)
(0.007)
(0.015)
(0.134)
R&Di,t-1/Ai,t-1
-4.564
1.574***
1.465***
0.948***
0.366
(11.306)
(0.185)
(0.234)
(0.340)
(4.700)
TURNi,t-1
0.002
0.000
-0.000
0.002***
0.004
(0.008)
(0.000)
(0.000)
(0.000)
(0.042)
2
2
R /Pseudo R
0.880
0.017
0.118
0.208
0.257
Obs.
2761
2761
2761
2761
2761
0.9
-0.430*
(0.227)
0.176**
(0.087)
0.002
(0.015)
0.000
(0.002)
0.418***
(0.154)
-6.997***
(1.935)
-0.004
(0.011)
0.064
2761
0.9
-0.051
(0.010)
0.072
(0.105)
-0.006***
(0.000)
0.002***
(0.000)
0.379**
(0.166)
-1.596***
(0.585)
0.012***
(0.002)
0.322
2761
Table 11 shows the regression results with discretionary accruals and capital investment in Hong Kong and splits the
sample according to firms’ degrees of financial constraints using the WW index. The firms with a below-median WW
index are less financially constrained firms and those with an above-median WW index are highly financially
constrained firms. Dependent variable:
, The standard errors reported in parentheses are corrected
clustering of the residual at the firm level. Coefficients starred with *, **, and *** are statistically significant at 10%, 5%,
and 1% levels, respectively.
Proceedings of 5th Asia-Pacific Business Research Conference
17 - 18 February, 2014, Hotel Istana, Kuala Lumpur, Malaysia, ISBN: 978-1-922069-44-3
Table 12: Regression results with discretionary accruals and capital investment in China by WW
index
Panel A: Less financially constrained firms (WW index)
OLS
Quantile regressions
0.1
0.25
0.5
0.75
DACCRi,t
1.315
-0.003*
-0.009***
0.013
0.027***
(3.323)
(0.002)
(0.003)
(0.063)
(0.003)
Qi,t-1
-1.224
0.002*
0.009***
0.026***
0.059***
(2.111)
(0.001)
(0.001)
(0.005)
(0.011)
CFi,t-1/Ki,t-2
0.000
0.001***
0.003***
0.006
0.039
(0.383)
(0.000)
(0.000)
(0.011)
(0.028)
EOISSi,t-1/Ki,t-2
0.059
0.002***
0.002***
0.004
0.056
(0.559)
(0.000)
(0.000)
(0.017)
(0.106)
HighDACCRi,t-1
-3.912
0.005
0.016***
0.027
0.047***
(7.597)
(0.004)
(0.006)
(0.022)
(0.015)
R&Di,t-1/Ai,t-1
-42.590
4.830***
4.817***
3.842***
2.126***
(1349.391)
(0.425)
(0.270)
(0.301)
(0.352)
TURNi,t-1
0.101
-0.000
-0.001
-0.003***
-0.003*
(0.888)
(0.000)
(0.001)
(0.001)
(0.002)
R2/Pseudo R2
0.000
0.000
0.001
0.001
0.003
Obs.
5681
5681
5681
5681
5681
Panel B: Highly financially constrained firms (WW index)
OLS
Quantile regressions
0.1
0.25
0.5
0.75
DACCRi,t
-0.051
0.002
0.001
0.002
-0.010**
(0.034)
(0.002)
(0.002)
(0.002)
(0.005)
Qi,t-1
0.004
-0.001
-0.001*
-0.000***
0.008**
(0.003)
(0.000)
(0.000)
(0.000)
(0.004)
CFi,t-1/Ki,t-2
-0.006***
0.001***
0.002***
0.002***
0.002***
(0.001)
(0.006)
(0.000)
(0.000)
(0.000)
EOISSi,t-1/Ki,t-2
0.056***
0.001***
0.010
0.055**
0.106
(0.010)
(0.000)
(0.009)
(0.026)
(0.241)
HighDACCRi,t-1
0.041
0.001
0.003
0.020***
0.062***
(0.050)
(0.002)
(0.003)
(0.006)
(0.014)
R&Di,t-1/Ai,t-1
-0.250
0.012
-0.006
-0.056**
-0.174***
(1.430)
(0.013)
(0.020)
(0.026)
(0.025)
TURNi,t-1
0.004
0.000
-0.001***
-0.003***
-0.006***
(0.006)
(0.000)
(0.000)
(0.001)
(0.001)
2
2
R /Pseudo R
0.014
0.002
0.004
0.013
0.023
Obs.
5682
5682
5682
5682
5682
0.9
0.006**
(0.003)
0.140***
(0.016)
0.069***
(0.013)
0.094***
(0.003)
0.106***
(0.036)
-0.416**
(0.209)
-0.006**
(0.003)
0.006
5681
0.9
-0.048***
(0.012)
0.040**
(0.018)
0.001***
(0.000)
0.251*
(0.089)
0.142***
(0.028)
-0.422***
(0.045)
-0.009***
(0.003)
0.039
5682
Table 12 shows the regression results with discretionary accruals and capital investment in China and splits the
sample according to firms’ degrees of financial constraints using the WW index. The firms with a below-median WW
index are less financially constrained firms and those with an above-median WW index are highly financially
constrained firms. Dependent variable:
, The standard errors reported in parentheses are corrected
clustering of the residual at the firm level. Coefficients starred with *, **, and *** are statistically significant at 10%, 5%,
and 1% levels, respectively.