Table 4.
Propensity scores matching
| PSM N = 1 Without replacement | |
|---|---|
| (1) | |
| CVD |
-0.756*** (0.049) |
| Size |
1.563 (1.740) |
| Capital |
39.435*** (8.272) |
| Asset_Quality |
1.479 (1.572) |
| Loans |
9.570*** (2.494) |
| Deposits |
1.539 (1.388) |
| Growth |
-2.734** (1.079) |
| Earnings |
-25.666*** (7.281) |
| Healthcare |
-3.592 (3.106) |
| GDP |
0.309 (0.233) |
| Econ_Support |
-0.069*** (0.001) |
| Stringency |
0.001 (0.024) |
| Containment |
-0.071** (0.029) |
| Constant |
41.455 (30.601) |
| Observations | 1,953 |
| AdjR2 | 0.206 |
| #Banks | 547 |
| FE | Yes |
| F Statistic | 51.37*** |
In this table, we provide the results from our propensity scores matching test. Using a dummy variable CVD_Dummy, where CVD_Dummy is equal to 1 if a bank is from a country with above median CVD and 0 otherwise, we divide our sample into two groups, i.e., treatment and control groups. We match each bank from the treatment (CVD_Dummy = 1) group with one or more banks with similar properties from the control (CVD_Dummy = 0) group. We use one-to-one matching (N = 1) without replacement. We regress Z-Score on CVD, while controlling for all bank- and macro-specific variables, as well as country-year and bank-fixed effects. Standard errors are reported in parentheses. ***, **, and * indicate statistical significance at 1%, 5%, and 10% levels, respectively