Change in Log(Share of Female Immigrants Selecting a City). Source-Destination Pairs.
| OLS (1) | IV (2) | IV (3) | IV (4) | IV (5) | IV (6) | IV (7) | |
|---|---|---|---|---|---|---|---|
| Change in native female employment rate | −0.0208 (0.0190) | −0.155*** (0.0532) | −0.159*** (0.0507) | −0.120*** (0.0394) | −0.0989** (0.0441) | −0.0785** (0.0388) | −0.0820* (0.0445) |
| Change in decade average employment growth rate | 0.173*** (0.0415) | 0.120*** (0.0297) | 0.0902** (0.0372) | 0.0529 (0.0324) | 0.0456 (0.0361) | ||
| Ethnic group members as percent of MSA pop - decade start | −0.236*** (0.0270) | −0.222*** (0.0289) | −0.200*** (0.0264) | −0.176*** (0.0259) | |||
| Change in average increase in log(real minimum wage) | 0.738 (0.508) | 0.510 (0.440) | 0.923 (0.758) | ||||
| Change in log(share of immigrants - college degree) | 0.305*** (0.0479) | 0.275*** (0.0444) | |||||
| Maximum cash benefit, family of three | 0.0312 (0.0399) | ||||||
| Change in percent of low-skilled female immigrants receiving welfare | −0.0180 (0.0186) | ||||||
| MSA in state that restored Food Stamp benefits | −0.142 (0.107) | ||||||
| MSA in state that restored TANF benefits | 0.193 (0.143) | ||||||
| MSA in state that restored Medicaid benefits | −0.00741 (0.120) | ||||||
| MSA in state that restored SSI benefits | −0.293*** (0.111) | ||||||
| Constant | 0.0664 (0.115) | 0.430** (0.181) | 0.601*** (0.181) | 0.466*** (0.149) | 0.177 (0.277) | 0.147 (0.246) | −0.0414 (0.373) |
| Observations (Source-Destination Pairs) | 1,136 | 1,136 | 1,136 | 1,136 | 1,136 | 1,136 | 1,136 |
| First Stage Coefficient | 0.357 | 0.352 | 0.369 | 0.364 | 0.360 | 0.377 | |
| First Stage F-stat | 20.70 | 22.42 | 26.04 | 18.91 | 18.34 | 21.76 |
Standard Errors clustered by MSA in parentheses
↵*** p<0.01,
↵** p<0.05,
↵* p<0.1
Notes: Shares are calculated among immigrants meeting the selection criteria from Table 2. Destinations not selected by anyone from a source during one census year are omitted from the entire sample. Analysis limited to source-destination pairs with valid observations for all covariates. Pairs are weighted by the square root of the number of source observations nationwide.