Addressing endogeneity using 2SLS regression
| (1) | (2) | (3) | (4) | (5) | (6) | |
|---|---|---|---|---|---|---|
| Panel A: First stage | Panel B: Two stage regression results | |||||
| Dependent variable | DIR | CEOIR | EMPS | DD | DD | DD |
| Industry average | 6.885*** | 0.305*** | 0.855*** | |||
| (9.38) | (9.93) | (4.88) | ||||
| DIR | −0.438*** | |||||
| (−7.00) | ||||||
| CEOIR | −0.213*** | |||||
| (−11.47) | ||||||
| EMPS | −0.210*** | |||||
| (−13.53) | ||||||
| LNS | −1.500*** | −1.168*** | −0.206*** | 0.823*** | 0.407*** | 0.128*** |
| (−3.01) | (−3.73) | (−7.22) | (8.58) | (16.71) | (16.23) | |
| K/S | −1.973*** | −1.592*** | −0.451*** | 1.884*** | 1.372*** | 0.781*** |
| (−10.08) | (−8.98) | (−7.26) | (10.96) | (18.44) | (16.87) | |
| Y/S | 0.689 | 1.336 | −2.975*** | 4.849*** | 5.280*** | 4.561*** |
| (0.72) | (1.55) | (−9.80) | (7.83) | (15.92) | (21.04) | |
| R&D/K | −0.087*** | −0.050*** | −0.042*** | 0.064*** | 0.045*** | 0.022*** |
| (−3.97) | (−2.55) | (−5.90) | (4.10) | (4.51) | (3.60) | |
| RDUM | −5.156* | −4.350 | 1.015 | 3.192*** | 1.819*** | 1.075*** |
| (−1.66) | (−1.56) | (0.97) | (7.23) | (6.63) | (4.25) | |
| I/K | 0.359** | 0.562*** | −0.857*** | −0.246 | −0.216** | −0.288*** |
| (2.40) | (4.15) | (7.41) | (−1.57) | (−2.02) | (−2.71) | |
| ROA | 0.212*** | 0.166*** | −0.012*** | −0.047*** | 0.008 | 0.043*** |
| (19.19) | (16.51) | (−3.71) | (−3.01) | (1.52) | (18.03) | |
| LEVERAGE | 8.523*** | 7.762*** | −0.501*** | −9.426*** | −7.405*** | −5.864*** |
| (24.47) | (24.64) | (−4.58) | (−16.33) | (−37.49) | (−70.56) | |
| Age | −5.311*** | −3.802*** | −0.074*** | 2.408*** | 0.941*** | 0.007 |
| (−3.49) | (−4.78) | (−3.50) | (6.95) | (11.32) | (0.50) | |
| Constant | 41.867*** | 29.958*** | 2.527** | −17.030*** | −5.283*** | 2.882*** |
| (13.26) | (10.49) | (2.39) | (−6.27) | (−7.67) | (10.06) | |
| Observations | 31,621 | 31,621 | 36,156 | 31,621 | 31,621 | 36,156 |
| R-squared | 0.421 | 0.422 | 0.422 | |||
| Firm fixed effects | Yes | Yes | Yes | Yes | Yes | Yes |
| Year effects | Yes | Yes | Yes | Yes | Yes | Yes |
| F-statistics | 58.23** | 45.21** | 46.21** | |||
| Cragg–Donald F Statistics | 42.34 | 38.19 | 38.90 | |||
| Stock–Yogo Weak ID test value | 16.38 | 16.38 | 16.38 | |||
| (1) | (2) | (3) | (4) | (5) | (6) | |
|---|---|---|---|---|---|---|
| Panel A: First stage | Panel B: Two stage regression results | |||||
| Dependent variable | DIR | CEOIR | EMPS | DD | DD | DD |
| Industry average | 6.885*** | 0.305*** | 0.855*** | |||
| (9.38) | (9.93) | (4.88) | ||||
| DIR | −0.438*** | |||||
| (−7.00) | ||||||
| CEOIR | −0.213*** | |||||
| (−11.47) | ||||||
| EMPS | −0.210*** | |||||
| (−13.53) | ||||||
| LNS | −1.500*** | −1.168*** | −0.206*** | 0.823*** | 0.407*** | 0.128*** |
| (−3.01) | (−3.73) | (−7.22) | (8.58) | (16.71) | (16.23) | |
| K/S | −1.973*** | −1.592*** | −0.451*** | 1.884*** | 1.372*** | 0.781*** |
| (−10.08) | (−8.98) | (−7.26) | (10.96) | (18.44) | (16.87) | |
| Y/S | 0.689 | 1.336 | −2.975*** | 4.849*** | 5.280*** | 4.561*** |
| (0.72) | (1.55) | (−9.80) | (7.83) | (15.92) | (21.04) | |
| R&D/K | −0.087*** | −0.050*** | −0.042*** | 0.064*** | 0.045*** | 0.022*** |
| (−3.97) | (−2.55) | (−5.90) | (4.10) | (4.51) | (3.60) | |
| RDUM | −5.156* | −4.350 | 1.015 | 3.192*** | 1.819*** | 1.075*** |
| (−1.66) | (−1.56) | (0.97) | (7.23) | (6.63) | (4.25) | |
| I/K | 0.359** | 0.562*** | −0.857*** | −0.246 | −0.216** | −0.288*** |
| (2.40) | (4.15) | (7.41) | (−1.57) | (−2.02) | (−2.71) | |
| ROA | 0.212*** | 0.166*** | −0.012*** | −0.047*** | 0.008 | 0.043*** |
| (19.19) | (16.51) | (−3.71) | (−3.01) | (1.52) | (18.03) | |
| LEVERAGE | 8.523*** | 7.762*** | −0.501*** | −9.426*** | −7.405*** | −5.864*** |
| (24.47) | (24.64) | (−4.58) | (−16.33) | (−37.49) | (−70.56) | |
| Age | −5.311*** | −3.802*** | −0.074*** | 2.408*** | 0.941*** | 0.007 |
| (−3.49) | (−4.78) | (−3.50) | (6.95) | (11.32) | (0.50) | |
| Constant | 41.867*** | 29.958*** | 2.527** | −17.030*** | −5.283*** | 2.882*** |
| (13.26) | (10.49) | (2.39) | (−6.27) | (−7.67) | (10.06) | |
| Observations | 31,621 | 31,621 | 36,156 | 31,621 | 31,621 | 36,156 |
| 0.421 | 0.422 | 0.422 | ||||
| Firm fixed effects | Yes | Yes | Yes | Yes | Yes | Yes |
| Year effects | Yes | Yes | Yes | Yes | Yes | Yes |
| 58.23** | 45.21** | 46.21** | ||||
| Cragg–Donald | 42.34 | 38.19 | 38.90 | |||
| Stock–Yogo Weak ID test value | 16.38 | 16.38 | 16.38 | |||
Note(s): Table 4 represents the regression results for the relationship between insider ownership and firm default risk using 2SLS. Panel A presents the result of the first stage regression and panel B presents the second stage regression. T-statistics are in parentheses. Definitions of all variables are provided in Appendix. Superscripts ***, ** and * denote statistical significance at 1, 5 and 10% level, respectively
Source(s): Table created by the authors