Immigration and Labor-Market Outcomes

What is the impact of large inflows of immigrants on inequality across regions in the United States?

Data Sources

• DATA I: 1980 Census and "2007" data from the American Community Survey (ACS) (i.e. 2006-2008 3 year ACS) https://usa.ipums.org/usa/
• DATA II: Comuting zones (CZ) from https://ddorn.net/data.htm

Data Transformation

Step I

By CZ $c$ and year $y$, native average wages, native unemployment and labor force participation rates are constructed.

Step II

The immigration inflow is defined and calculated as:

$$x_c = \dfrac{1}{N_{c,1980}}\left(I_{c,2007} - I_{c,1980}\right)$$

where

• $N_{c,1980}$ is the total population of $c$ in 1980.
• $I_{c,y}$ is the population of immigrant in $c$ in year $y$.

Step III-a

The first instrument is constructed as:

$$z_c^{(1)} = \dfrac{1}{N_{c, 1980}} \sum_{s} f_{c, 1980}^{s} \left(I_{2007}^s - I_{1980}^s\right)$$

where

• $I_{y}^s$ is the number of immigrants from source region $s$ in US in year $y$.
• $f_{c,1980}^{s} = \dfrac{I_{c, 1980}^s}{I_{1980}^s}$ is the share of immigrants from source $ s$ who are in $c$ in year 1980.

Step III-b

The second instrument is constructed as:

$$z_{c}^{(2)} = \dfrac{1}{I_{c, 1980}} \sum_{s} f_{c, 1980}^{s} \left(I_{2007}^s - I_{1980}^s\right)$$

Step IV

Using 2SLS, project changes in CZ outcomes (percentage point for unemployment and labor force participation rates; percent or log for wages) on immigrant inflow.

• Instrument for $x_c$, with either $z_c^{(1)}$ or $z_c^{(2)}$, depending on which has a stronger first stage when controls are included.
• Controls are included (like in Author Dorn, and Hanson) measured in 1980 data: key control is the share of population that is immigrant in 1980.

Create the conda Environment from the environment.yml File

To run the code in this repository, cloning the conda environment is recommended. To do so, run the following command in the terminal:

conda env create -f environment.yaml

This will create a new conda environment called 'imm-mk-env'. To activate the environment, run the following command in the terminal:

conda activate imm-mk-env

For updating the environment from the environment.yml file, run the following command in the terminal:

conda env update -f environment.yml --prune

This action should be done if the environment.yml file is updated.

Warning

Before running the code, make sure to activate (or create) the conda environment by running the command conda activate imm-mk-env.