Replication package for "Labor supply when productivity keeps growing"
by Timo Boppart, Per Krusell, and Jonna Olsson

Running the code
----------------

To replicate the main graphs and tables in the paper, proceed as follows:

1.  The code requires a Python environment with the following packages:
    Python: 3.10
    numpy: 1.24
    scipy: 1.10
    matplotlib: 3.7
    numba: 0.57
    pandas: 1.5
    lz4: 4.3

For users of Anaconda (https://www.anaconda.com/), the provided environment.yml
file located in this directory can be used to automatically create an
environment with the required packages and versions, e.g. by running (on Linux):

conda env create -f environment.yml
conda activate bko

On Window, the Anaconda Navigator contains an option to import an
environment by specifying the environment file environment.yml.

2.  The file model/src/main.py runs ALL functions required to compute
    and plot the main graphs. This file should be run from the directory
    model/src.

    It takes about 90 minutes to run the entire replication package.

3.  After the code finishes, the generated graphs are stored in the
    folder plots/.
    The generated file tables.txt contains the data for Table 1 and 2
    of the paper.


Description of files & directories
----------------------------------

model/src/main.py
    Main file to run ALL analyses. This is the ONLY file that needs to be run to
    replicate the results.

model/src/bko_im/
    Folder which ontains code to run incomplete markets models from section 4.2.

model/src/bko_im/main.py
    Solves steady-state model.

model/src/bko_im/main_trans.py
    Solves for transition path between two steady states.

model/src/bko_im/analyze_ss.py
    Generates graphs / tables for steady-state model results.

model/src/bko_im/analyze_trans.py
    Generates graphs / tables for transition path.

    All other files in this directory contain various helper functions.

model/src/common
    Contains environment setup and common plotting code called from other
    files.

model/src/ha_static
    Contains files to solve static model from section 3.2 and 3.3 in the paper.

model/src/ha_static/main.py
    Solves static model.

model/src/ha_static/analyze_res.py
    Generates graphs / tables for static model.

    All other files in this directory contain various helper functions.

model/input
    Contains the initial distribution over assets and productivity obtained
    from the PSID (see description in paper).


pydynopt/
    Contains 3rd-party library with various helper functions
    which is available from https://github.com/richardfoltyn/pydynopt
    and included here for convenience.