Poisson Male RD Effects
| Order 1 (1) | Order 2 (2) | Order 3 (3) | Birthday 1 (4) | Birthday 2 (5) | Birthday 3 (6) | |
|---|---|---|---|---|---|---|
| Panel A: Violent Crimes | ||||||
| All | 1.054*** (0.0165) | 1.069*** (0.0251) | 1.098*** (0.0343) | 1.067*** (0.0250) | 1.064*** (0.0241) | 1.068*** (0.0246) |
| Homicide | 0.998 (0.121) | 0.781 (0.148) | 0.705 (0.165) | 0.781 (0.148) | 0.769 (0.145) | 0.779 (0.147) |
| Sex offenses | 1.119 (0.234) | 1.744* (0.531) | 2.204* (0.948) | 1.753* (0.534) | 1.762* (0.539) | 1.742* (0.529) |
| Robbery | 1.048* (0.0278) | 1.078* (0.0424) | 1.139** (0.0585) | 1.078* (0.0424) | 1.080* (0.0426) | 1.078* (0.0425) |
| Assault | 1.058*** (0.0212) | 1.067** (0.0325) | 1.083* (0.0452) | 1.063** (0.0321) | 1.058* (0.0309) | 1.065** (0.0317) |
| Panel B: Property Crimes | ||||||
| All | 1.023 (0.0142) | 1.074*** (0.0222) | 1 077*** (0.0302) | 1.069*** (0.0221) | 1.066*** (0.0219) | 1.069*** (0.0217) |
| Burglary | 1.003 (0.0307) | 1.047 (0.0493) | 0.976 (0.0620) | 1.021 (0.0493) | 1.023 (0.0496) | 1.023 (0.0495) |
| Larceny | 1.032* (0.0181) | 1.079*** (0.0280) | 1.108*** (0.0385) | 1.066** (0.0297) | 1.063** (0.0294) | 1.066** (0.0293) |
| Motor vehicle theft | 1.007 (0.0402) | 1.091 (0.0682) | 1.089 (0.0914) | 1.121** (0.0542) | 1.113** (0.0534) | 1.121** (0.0533) |
Notes: This table contains IRR estimates for the RD effect of the minimum legal drinking age on male victimization rates for each crime type. The regressions in Columns 1–3 include first-through third-order polynomials in age fully interacted with an indicator for age over 21. The regressions in Columns 4–6 contain second-order polynomials in age fully interacted with an indicator for age over 21 and birthday effects 1–3, respectively. Birthday 1 includes indicator variables for exact birthdays. Birthday 2 includes indicator variables for exact birthdays and the following three days. Birthday 3 includes indicators for the week around each birthday. Each observation is the total number of victims in each age (days) relative to the 21st birthday. Robust standard errors in parentheses. All regressions use a two-year bandwidth.
↵* p < 0.10,
↵** p < 0.05,
↵*** p < 0.01.