Robustness test using entropy balancing - The effects of politically-connected boards (PCBs) on over-(under-)investment in labor (OVER_LAB or UNDER_LAB)
| OVER_LAB | UNDER_LAB | |||
|---|---|---|---|---|
| Coefficient | t-values | Coefficient | t-values | |
| #1 | #2 | #3 | #4 | |
| PCBsit | −0.046*** | (−5.14) | 0.002 | (1.47) |
| TOP5it | −0.015 | (−0.61) | 0.009** | (2.36) |
| LEVit | −0.106*** | (−4.07) | 0.007* | (1.86) |
| ROAit | −0.404*** | (−5.21) | −0.010 | (−0.97) |
| DPRit | −0.028** | (−2.47) | −0.003 | (−1.59) |
| TANGit | 0.084*** | (5.45) | −0.006*** | (−3.89) |
| OPER_CCit | 0.027*** | (4.83) | −0.004*** | (−3.53) |
| Qit | 0.012** | (2.53) | 0.001 | (1.22) |
| Constant | 0.255*** | (4.83) | 0.096*** | (12.69) |
| Observations | 1,439 | 1,017 | ||
| F | 14.5*** | 28.6*** | ||
| R2 | 0.280 | 0.431 | ||
| OVER_LAB | UNDER_LAB | |||
|---|---|---|---|---|
| Coefficient | Coefficient | |||
| #1 | #2 | #3 | #4 | |
| −0.046*** | (−5.14) | 0.002 | (1.47) | |
| −0.015 | (−0.61) | 0.009** | (2.36) | |
| −0.106*** | (−4.07) | 0.007* | (1.86) | |
| −0.404*** | (−5.21) | −0.010 | (−0.97) | |
| −0.028** | (−2.47) | −0.003 | (−1.59) | |
| 0.084*** | (5.45) | −0.006*** | (−3.89) | |
| 0.027*** | (4.83) | −0.004*** | (−3.53) | |
| 0.012** | (2.53) | 0.001 | (1.22) | |
| 0.255*** | (4.83) | 0.096*** | (12.69) | |
| 1,439 | 1,017 | |||
| 14.5*** | 28.6*** | |||
| 0.280 | 0.431 | |||
Note(s): Entropy balancing regression model. Over-(under)investment in labor (OVER_LABit or UNDER_LABit) in firm i occurs when the ratio of the number of employees at the end of financial year t to the annual sales of firm i during year t is greater (lower) than the ratio of the total number of employees of all firms in the same industry sector in year t to the total annual sales for all firms in the same industry sector in year t. In the sub-sample of over-investment in labor, a higher (lower) positive number indicates a higher (lower) over-investment in labor. For ease of interpretation, the negative signs of the sub-sample for under-investment in labor are multiplied by −1 to obtain a positive number. Hence, a higher (lower) positive number indicates a higher (lower) under-investment in labor. The presence of PCBsit is a dummy variable set to 1 if firm i in year t has politically-connected board member(s) and zero (0) otherwise. Columns 1 and 2 report the OVER_LAB regression coefficients and t-values in parentheses, columns 3–4 report the UNDER_LAB regression coefficients and t-values in parentheses. *, **, and *** indicate statistical significance at the 10, 5 and 1% levels, respectively (two-tailed). The definitions of variables are presented in Appendix
Source(s): Table 7 by authors