| . | Unconditional risk () . | Conditional risk () . | ||||||
|---|---|---|---|---|---|---|---|---|
| pos. news . | neg. news . | pos. news . | neg. news . | |||||
| (1) . | (2) . | (3) . | (4) . | (5) . | (6) . | (7) . | (8) . | |
| News shock | 0.25* | 0.15 | 0.15 | 0.07 | 2.76*** | 2.72** | 1.51 | 1.65 |
| (0.14) | (0.16) | (0.20) | (0.20) | (0.88) | (1.12) | (1.16) | (1.14) | |
| Liquidity control | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ |
| Paris data only | ✓ | ✓ | ✓ | ✓ | ||||
| N (Observations) | 420 | 164 | 112 | 46 | 420 | 164 | 112 | 46 |
| N (Event-market) | 68 | 24 | 18 | 7 | 68 | 24 | 18 | 7 |
| Adjusted R2 | .94 | .98 | .88 | .78 | .91 | .95 | .87 | .89 |
| Within-R2 | .05 | .08 | .02 | .02 | .13 | .17 | .07 | .16 |
| . | Unconditional risk () . | Conditional risk () . | ||||||
|---|---|---|---|---|---|---|---|---|
| pos. news . | neg. news . | pos. news . | neg. news . | |||||
| (1) . | (2) . | (3) . | (4) . | (5) . | (6) . | (7) . | (8) . | |
| News shock | 0.25* | 0.15 | 0.15 | 0.07 | 2.76*** | 2.72** | 1.51 | 1.65 |
| (0.14) | (0.16) | (0.20) | (0.20) | (0.88) | (1.12) | (1.16) | (1.14) | |
| Liquidity control | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ |
| Paris data only | ✓ | ✓ | ✓ | ✓ | ||||
| N (Observations) | 420 | 164 | 112 | 46 | 420 | 164 | 112 | 46 |
| N (Event-market) | 68 | 24 | 18 | 7 | 68 | 24 | 18 | 7 |
| Adjusted R2 | .94 | .98 | .88 | .78 | .91 | .95 | .87 | .89 |
| Within-R2 | .05 | .08 | .02 | .02 | .13 | .17 | .07 | .16 |
This table presents the results of event study regressions for news pertaining to U.K. bondholders only. The liquidity control corresponds to the implicit bid-ask spread differential between the London market s and continental market j estimated using Roll’s (1984) method. See text for more details on the specification. Days for which the sources do not report prices are treated as missing. Internet Appendix B.4 reports qualitatively similar results for regressions where missing observations are replaced with the last previously available yield on any given day. Two-way clustered standard errors (Event-market & date dimension) are in parentheses.
p < .1;
p < .05;
p < .01.
| . | Unconditional risk () . | Conditional risk () . | ||||||
|---|---|---|---|---|---|---|---|---|
| pos. news . | neg. news . | pos. news . | neg. news . | |||||
| (1) . | (2) . | (3) . | (4) . | (5) . | (6) . | (7) . | (8) . | |
| News shock | 0.25* | 0.15 | 0.15 | 0.07 | 2.76*** | 2.72** | 1.51 | 1.65 |
| (0.14) | (0.16) | (0.20) | (0.20) | (0.88) | (1.12) | (1.16) | (1.14) | |
| Liquidity control | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ |
| Paris data only | ✓ | ✓ | ✓ | ✓ | ||||
| N (Observations) | 420 | 164 | 112 | 46 | 420 | 164 | 112 | 46 |
| N (Event-market) | 68 | 24 | 18 | 7 | 68 | 24 | 18 | 7 |
| Adjusted R2 | .94 | .98 | .88 | .78 | .91 | .95 | .87 | .89 |
| Within-R2 | .05 | .08 | .02 | .02 | .13 | .17 | .07 | .16 |
| . | Unconditional risk () . | Conditional risk () . | ||||||
|---|---|---|---|---|---|---|---|---|
| pos. news . | neg. news . | pos. news . | neg. news . | |||||
| (1) . | (2) . | (3) . | (4) . | (5) . | (6) . | (7) . | (8) . | |
| News shock | 0.25* | 0.15 | 0.15 | 0.07 | 2.76*** | 2.72** | 1.51 | 1.65 |
| (0.14) | (0.16) | (0.20) | (0.20) | (0.88) | (1.12) | (1.16) | (1.14) | |
| Liquidity control | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ |
| Paris data only | ✓ | ✓ | ✓ | ✓ | ||||
| N (Observations) | 420 | 164 | 112 | 46 | 420 | 164 | 112 | 46 |
| N (Event-market) | 68 | 24 | 18 | 7 | 68 | 24 | 18 | 7 |
| Adjusted R2 | .94 | .98 | .88 | .78 | .91 | .95 | .87 | .89 |
| Within-R2 | .05 | .08 | .02 | .02 | .13 | .17 | .07 | .16 |
This table presents the results of event study regressions for news pertaining to U.K. bondholders only. The liquidity control corresponds to the implicit bid-ask spread differential between the London market s and continental market j estimated using Roll’s (1984) method. See text for more details on the specification. Days for which the sources do not report prices are treated as missing. Internet Appendix B.4 reports qualitatively similar results for regressions where missing observations are replaced with the last previously available yield on any given day. Two-way clustered standard errors (Event-market & date dimension) are in parentheses.
p < .1;
p < .05;
p < .01.
This PDF is available to Subscribers Only
View Article Abstract & Purchase OptionsFor full access to this pdf, sign in to an existing account, or purchase an annual subscription.
