Table A1.
Birth Order and Number of Brothers as Instruments for Migration to the United States
| (1) | (2) | (3) | |
|---|---|---|---|
| Panel A. First stage | Dependent variable = In US in 1900 | ||
| Number of brothers | 0.016 (0.006) |
0.011 (0.006) |
|
| 2nd brother | −0.000 (0.012) |
— | |
| 3rd brother | 0.047 (0.019) |
0.037 (0.019) |
|
| 4th or higher brother | 0.076 (0.035) |
0.058 (0.036) |
|
| Panel B. OLS | Dependent variable = ln(earnings in 1900) | ||
| In US in 1900 | 0.642 (0.019) |
||
| Panel C. IV | Dependent variable = ln(earnings in 1900) | ||
| In US in 1900 | 0.669 (0.436) |
0.696 (0.381) |
0.668 (0.338) |
| Over-ID test (p-value) | 0.869 | ||
| N | 4031 | 4031 | 4031 |
Notes: Standard errors are reported in parentheses. The sample includes men in Match 1 who lived in a rural household that had some assets in 1865 and whose mother is 42 years old or younger in 1865. The regressions also include a quadratic in age and dummy variables for total number of siblings in the household (see equation (3) in the text). In column 3, we report the p-value from a Sargan (chi-squared) test of overidentification.