Table 9.
COVID-19 risk and the likelihood of loan default: testing for sampling bias
| Variables | DV = DEFAULT | ||
|---|---|---|---|
| (1) | (2) | (3) | |
| Panel A: Three big countries by the number of observations | |||
| PANDEMIC_ DUMMY |
0.405*** (0.008) |
||
| DAILY_CASES |
0.004*** (0.000) |
||
| DAILY_DEATHS |
0.045*** (0.001) |
||
| Loan originator individual effects | Yes | Yes | Yes |
| Controls | Yes | Yes | Yes |
| Pseudo-R-squared | 0.210 | 0.235 | 0.237 |
| N | 680,694 | 390,133 | 390,133 |
| Panel B: Bootstrap sampling | |||
| PANDEMIC_ DUMMY |
0.555*** (0.007) |
||
| DAILY_CASES |
0.004*** (0.000) |
||
| DAILY_DEATHS |
0.036*** (0.001) |
||
| Loan originator individual effects | Yes | Yes | Yes |
| Controls | Yes | Yes | Yes |
| Pseudo-R-squared | 0.129 | 0.173 | 0.170 |
| N | 814,872 | 503,167 | 503,167 |
| Panel C: Only unresolved loans | |||
| PANDEMIC_DUMMY |
0.387*** (0.010) |
||
| DAILY_CASES |
0.007*** (0.000) |
||
| DAILY_DEATHS |
0.067*** (0.001) |
||
| Loan originator individual effects | Yes | Yes | Yes |
| Controls | Yes | Yes | Yes |
| Pseudo-R-squared | 0.274 | 0.345 | 0.347 |
| N | 288,595 | 213,036 | 213,036 |
| Panel D: Heckman correction | |||
| PANDEMIC_DUMMY |
0.088*** (0.001) |
||
| DAILY_CASES |
0.001*** (0.000) |
||
| DAILY_DEATHS |
0.003*** (0.000) |
||
| Loan originator individual effects | Yes | Yes | Yes |
| Controls | Yes | Yes | Yes |
| N | 814,886 | 503,167 | 503,167 |
Table presents the results of regression analyses based on four panels. Panel A results are for logit regression analysis for the likelihood of loan default (DEFAULT) with the sample consisting of only three countries with the highest number of observations. Panel B reports the results after the application of bootstrap sampling with stratified sampling based on loan originators and each month of 2020. Panel C results are for logit regression analysis with the sample consisting of only unresolved loans. Panel D reports the results after the application of the Heckman selection model, where the selection in the sample is instrumentalised with loan amount and rating. All model specifications employ robust standard errors in parentheses (*p < 0.10, **p < 0.05, ***p < 0.01)