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 ,t1 ) and high level of DACCR ( HighDACCR i ,t1 ) 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.